Im unologická diagnostika a monitorování léčeb né odpovědi u nádorových onemocnění krvetvorby v dětském věku MUDr. Ester Mejstříková Školitel: Doc. MUDr. Ondřej Hrušák, Ph.D
Univerzita Karlova v Praze 2. lékařská fakulta 2008
UNIVERZITA KARLOVA V PRAZE ,
v
,
2. LEKARSKA FAKULTA V PRAZE
DIZERTAČNÍ PRÁCE
2008
MUDr. Ester Mej stříková
E. Mejstříková, strana 1
Imunologická diagnostika a monitorování léčebn é odpovědi u nádorových krvetvorby v dětském věku 1.
o~
onemocnění
4
-------------------------------------------------------------------- 4
l.l. Akutní lymfoblastická leukémie 1.1 . 1. Prognostické faktory a klasifikace ALL 1.1.1.1. Imunologická klasifikace ALL 1.1.1.1. I. ALL z prekurzorů B Iymfocytů (BCP ALL) 1.1.1.1.2. ALL z prekurzorů T Iymfocytů (T ALL) 1.1.1.2. Genotypová klasifikace ALL 1.1.1 .2.1. Genetické podtypy BCP ALL 1.1.1.2.2. Genetické podtypy T ALL 1.1.1.3. Klinické a biologické projevy ALL při diagnóze 1.1.1.4. Biologické projevy ALL 1.1.1.5 . Fannakogenetika AL L 1.1 .1.6. Princip terapie AL L 1.1 . 1.6. 1. Odpověď na léčbu ALL 1.1. 1.6.1.1. Kompletní remise 1.1.1 .6.1.2. Korelace počtu blastů v den 8 v periferní krvi po kortikoidové před fázi prednisonová odpověď 1.1.1.6.1 .3. Hodnocení poklesu bla s tů v kostní dřeni před dosažením remise 1.1 . 1.6.1.4. Minimální reziduální nemoc (MRN) 1.1.2. P řístup k pacient m s relapsem ALL a k primárně vysoce rizikové ALL 1.1.2.1. Indikace transplantace kostní dřeně (SCT) u pacientů v první kompletní remisi ALL _ Přístup k pacientům s fúzním genem BCR! ABL 1.1.2.2. 1.1.2.3. Přístup k pacientům s vysoce rizikovou T ALL I. J .2.4. P řístup k pacient ů m s relapsem ALL
17 18 18 18 J8 18 19 20
1.2.
22
Akutní myeloidní leukémie (AML)
5 5 6 8 8 II 12 14 14 14 15 15 17 17
1.2. J. Prognostické faktory a klasifikace AML 23 1.2.1.1. Imunologická klasifikace AML 24 1.2.1.2. Genotypová klasifikace, epigenetické změny u AML a jej ich význam pro terapii _____ 24 1.2.1.2.1. AML s fúzním genem AMLI /ETO, CBF~/MYH II a PMLlRARa 24 1.2.1.2.2. AM LlMDS u pacientů s konstitutivní trizómií 21 (m. Down) 25 1.2.1.2.3. AM L spřestavbamigenuMLL(llq23) 26 1.2.1.2.4. AM L s FLT3 interní tandemovou duplikací (ITD) nebo aktivační mutací FLT3(D835)26 26 1.2.1.3. Klinické a biologické projevy AML, primárně extramedulární leukémie 1.2.1.4. Farmakogenetika AML 27 27 1.2.1.5. Princip léčby AML, perspektivn í nová léčiva 28 1.2.1.5.1. Léčba AML M3 s fúzním genem PML/RA Ra J .2. J .5.2. Imunoterapie AML 28 1.2.1.5.3. Inh ib itory kináz 29 1.2.1.5.4. Inh ibitory histon deacetyláz, demetylující látky, terapeutické ovlivnění epigenetických změn u leukemických buněk 29 1.2.1.5.5. Nová cytostatika 30 1.2.1.5.6. Odpověď na léčbu AML, terapeutické ovlivnění leukemické kmenové buňky _____ 30 1.2.1.5.6.1. Odpověď na léčbu definovaná morfologicky 30 30 1.2.1.5.6.2. Molekulárně genetické a cytogenetické cíle detekce MRN 31 1.2.1.5.6.3. Princip detekce MRN cytometricky u AML 1.2.2. Přístup k pacientům s relapsem AML a k primárně vysoce rizikové AML 33
2 3.
-------------------------------------------------------------------- N Výsledky a Diskuse 35
~
3.1. 3. J .1. 3. I .2. 3.1.3.
Imunologická diagnostika ALL ALL z prekurzorů B Iymfocytů (BCP ALL) Aberantní exprese antigenů Akutní hybridní leukémie
35 35 37 38
E. Mejstříková, strana 2 3.IA. Leukémie s nálezem blastů z různých linií a liniový přesmyk během časné fáze léčby před dosažením kompletní remise. 3.1.5 . Imunologická diagnostika terapeutických cílů na leuke mické buňce . Anti-C D33 léčba, perspektiva využití dalších monoklonálních protilátek v léč bě dě tské ALL
~
44
účinnosti léčby
46
3 .2. 1. Hodnocení minimální reziduální nemoci 3.2.1.1. Imunofenotypová detekce MRN u ALL 3.2.1.1.1. Cytometrická reziduální nemoc v mez inárod ní studii Mini-Mini 3.2.1.2. Molekulárně genetická detekce MRN u ALL 3.2.1.2 . 1. Sledování MRN u ALL pomocí přestaveb imunoreceptorových gen ů (l gITC R) _ _ 3.2 . 1.2.2. Sledování MRN u ALL pomocí fúzních genů Nepřímé sledování MRN pomocí chimérismu po aloge nní SeT 3.2.1.3. 3.2.2. Léčebný protokol ALL IC-BFM 2002 pro primární léčbu de novo ALL 3.2.3. Detekce MRN po relapsu ALL a u pacientů indikovaných k alogenní SCT 3.2A. Detekce MRN u pacientů s AML
46 47 49 50 51 52 52 53 55 56
3.2.
Léčba
42
Závěr
leukémie, monitorování
_ __ _ _ _ _ _ _ _ _ _ _ _ __ _______________ 56
4.1.
Myeloidní antigeny u ALL _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 56
4.2.
Význa m
4.3.
Minimální reziduální nemoc
diferenciačního
antigenu CD10
57 57
5.
Seznam zkratek_ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ 57
~
Lliff~ura
M
6.1.
Použitá literatura _____ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _____ 60
6.2.
Přiložené
publikace simpact faktorem
77
Příloha
I 77 Correlation of CD33 w ith poorer prognosis in childhood ALL implicates a potential of anti-C D33 frontline therapy 77 Př íloha 2 82 Transfer of genomics information to tlow cytometry: expression of C D27 and C D44 discriminates subtypes of acute lymphoblastic leukemia 82 Příloha 3 86 Mye loid antigens in childhood Iymphoblastic leukemia:clinical data point to regulation ofCD66c distinct from other myeloid antigens 86 Příloha 4 98 Res idual disease monitoring in childhood acute myeloid leukemia by multiparameter tlow cytometry: the MRD-AML-BFM study group 98 Příloha 5 106 Detectable minimal residua l disease before allogeneic hematopoietic stem cell transplantation predicts extremely poor prognosis in children with acute Iymphoblastic leukemia 106 Příloha 6 115 B-cell recoostitution after allogeneic stem cell transplantation impairs minimal residual disease (MRD) monitoring in children with ALL 115 Příloha 7 116 Minimal residual disease (MRD) analysis in the non-M R D-based ALL lC-B FM 2002 protocol for childhood ALL: is it possible to avoid MRD testing? 116 Příloha 8 126 CD44 and CD27 del ineate B-precursor stages with di fferent recombination status and with an uneven distribution in nonmalignant and malignant hematopoiesis 126 Příl o ha 9 137 Childhood secondary ALL after ALL treatment 13 7 Příloha 10 142 Allogeneic stem cell transplantation in children with leukemia using human leukocyte antigenmismatched unrelated donors 142
6.3.
Přiložené Příloha
II
publikace
před
odesláním _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 151 151
E. Mejstříková, strana 3 Detection of residual B precursor lymphoblastic leukemia by uniform gating flow cytometry _ _ 151
6.4.
Přiložené
publikace bez impakt faktoru _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 163 12 163 lmunofenotypizace d ětskýc h leukémií 163
Příloha
E. Mejstříková, strana 4
Im unolog ická diagnosti ka a monitorování od povědi u nádorových on e m o c n ě ní krvetvorby v
léčebné
dětském
vě ku
1. Úvod onemocnění
Mezi základní podtypy maligních dětí
leukémie. U
tvoří nejčastější
leukémie
skupinu chorob, u kterých genetické výhodě
a/nebo poruše apoptózy.
prekurzorů
(ALL), která
onemocnění
krvetvorby u
Nejčastějším
tvoří
onemocnění.
Leukémii lze definovat jako buňce
v nezralé krvetvorné podtypem u
dětí
akutní
vedou k
růstové
je leukémie z lymfocytámích
asi 77% ze všech leukémií. Další podtypy maligních tvoří
asi 11 % ,
chronická myeloidní leukémie (2 - 3%) a myelodysplastický syndrom (1 - 2%).
Léčebný
úspěch
u
dětské
ALL
dětí
maligní
změny
dětí patří především
krve tvorby u
představuj e
v průběhu 40 let, kdy na současnosti
jsou: akutní myeloidní leukémie (AML), která
více než 80%
jednu z
počátku
nejúspěšnějších
kapitol medicíny minulého století,
pacientů dlouhodobě vyléčit.
Již od
u
dětí
neuspokojivé. onemocnění,
vzácná a výsledky jsou
s odlišnou biologií nemoci od dřeně
dospělých
av
masy (reziduální nemoci) metodikami
protokolech. genů
dál
Nejčastěji
dětí
onemocnění
autoimunitního poškození kmenových buněk kostní dřeně
čím
u
daří
v
ALL je snaha
oproti ALL stále
věku
se
velmi vzácné
uplatňuje především
(SCT). V rámci diferenciální diagnózy MOS a akutních
leukémií je nutné zmínit i aplastickou anémii,
mikroskopie, je
dětském
léčbě
se
AML je na rozdíl od
přes podstatně agresivnější léčbu
Myelodysplastický syndrom (MOS) je v
alogenní transplantace kostní
počátku léčby
l éčby.
pacienty co n jvíce stratifikovat podle míry rizika selhání dospělých
vyléčení ,
byla prakticky nulová šance na
důlež itějším
přesnějšími
a
J.
vznikající na
podkladě
Sledování redukce nádorové
citlivějšími,
než je klasická optická v moderních
léčebných
používanými metodikami jsou u ALL sledování klonálních
přestaveb
parametrem stratifikace
pacientů
pro imunoglobuliny a T -buněčné receptory (lg/TCR) pomocí kvantitativní PCR (RQ
PCR) a průtoková cytometrie. U AML lze u části pacientů použít pro detekci MRN fúzní geny (AMLlIETO, PMLlRARa, CBF~/ MYHl1) nebo průtokovou cytometrii. Přes svoji relativní ekonomickou
nenáročnost
a dostupnost je ale cytometrie stále metodou používanou
pro MRN poměrně málo, všechny dosud publikované práce jsou bud' omezené na jednu instituci, či jednu centrální laboratoř. Jedním z důvodů je i nedostatečná standardizace této komplexní metodiky. V rámci svého postgraduálního studia jsem se zabývala zapojením
E. Mejstříková, strana 5 prutokové cytometrie a její standardizací v kontextu dalších metod u hematologických malignit v dětském
věku.
1.1. Akutní lymfoblastická leukémie
v
České republice onemocní ročně přibližně 65-70 dětí. Prognóza dětí v ČR je srovnatelná se
světovými
výsledky. U naprosté
ke kumulaci
onemocnění
většiny
dětí
u
pacientll neznáme
předškolním věku
v
pře sně
lze
očekávat,
v leukemogenezi hraje postupné setkávání se s infekcemi a závislé na lymfocytech
2.
etiologii nemoci, vzhledem že významnou úlohu
získávání adaptivní imunity
U menšího počtu pacientů jsme schopni identifikovat genetickou
poruchu, která je spojená s vyšší incidencí ALL. Typicky se jedná o geneticky poruchy kontroly
buněčného
breakage syndrome) onkologické
léčbě
3. 4 .
cyklu
(např. Bl oom ů v
5. 6.
současnosti
na struktury leukemických léčiva
(včetně
ALL) (sekundární ALL po
léčbě
ALL
Za postupným zlepšováním prognózy dětí s ALL stály především
randomizované studie. Hlavní podíl na rokem 1980. V
syndrom, ataxia teleangiectasia, Nijmegen
ALL rovněž může vzniknout jako sekundární onemocnění po
pro jinou malignitu
shrnujeme v příloze 9)
podmíněné
se
postupně
buněk
nynější úspěšnosti
do
léčby
mají
dostávají nová
přitom
léky objevené
před
léčiva, často cíleně namířená
(monoklonální protilátky, inhibitory tyrosinkináz) . Tato
v současnosti se ale používají u specifických podtypů leukémie (např. BCRlABL
pozitivní ALL) ne bo u nemoci refrakterní na klasickou
léčbu.
1.1 .1. Prognostic ké fa ktory a klasifi kace ALL Morfo logická klas ifikace - rozlišení zralé a prekurzorové BALL Nej starší
dělení
je na Ll, L2 a L3 subtyp (L 1 - homogenní malé blasty, L2 - blasty
veli kosti, L3 - typická je vakuo lizace s intenzivní bazofilní cytoplazmou Wright - Giemsy). Diagnosticky a
léčebně
je
důležitý
při
různé
barvení podle
zejména podtyp L3, který odpovídá
imunologicky zralé B leukémii (leukemizovaný non hodgkinský Iymfom (NHL) Burkittova typu s více než 25% infiltrací kostní BALL se používá jiná bloků
léčebná
7.
nebo periferní krve nádorovými
buňkami).
U zralé
strategie oproti typické ALL, s použitím krátkých intenzivních
chemoterapie, nesprávné
onemocnění
dřeně
zařazení může
vést k selhání
léčby
a následnému relapsu
Cytogeneticky asi u 85% pacientů se zralou B leukémií nacházíme typickou
translokaci t(8; 14)(q24;q 32), u zbývajících t(8;22)(q24;q l l)
8. 9.
pacientů
nacházíme translokace t(2;8)(pll;p24) a
Translokace mají společné to, že se protoonkogen MYC přesouvá do
oblasti gen ů, včetně jejich promotorů, pro těžký nebo lehké řetězce imunoglobulinů. Diagnosticky svízelný m ůže být nález tzv. "transitional" praeB
ALL s pozitivitou
E. Mejstříková, strana 6 povrchového IgM (ll řetězec)
/0
a někdy i s expresí jednoho z lehkých řetězců (kapa nebo
lambda), zpravidla však ne na všech blastech. vzhled
připomínající
Někd y souč asně nicméně
L3. Publikovaných dat je málo,
morfologicky nacházíme i cytogeneticky nenacházíme
výše zmíněné přestavby MYC protoonkogenu typické pro zralou BALL sporných
případů
onemocnění
je
klíčová
informace z cytogenetického
je indikována podle ní.
zpravidla artefaktem, rozpaky
při podezření
v kostní
dřeni
u
způsobeným
na leukémii
pacientů
pozitivita lehkého
nízkou teplotou
při
připomínat
lymfocytů,
zvýšený
(např. při
případná
a
/3.
počet
virovém infektu nebo po
těchto
představují
situacích dokáže
správně
Nejzralejší stádia těchto B prekurzorů (zpravidla
buňky odpovídají populaci tzv. "transitional" B lymfocytů
/4,
částečně tyto
/5.
Imunologická klasifikace ALL
Leukemické částečně
hematogonů
tzv.
slabě CDlOpoZ a jasně CD20 P0 7.) lze nalézt v malém počtu i v periferní krvi
1.1.1.1.
léčba
kapa a lambda je
blasty ALL. Tyto hematogony v podstatě
imunofenotypizace je v
odlišit od fenotypu leukemických buněk
řetězce
U takto
transportu vzorku. Diagnostické
může rovněž přinést
s regenerující hematopoézou
chemoterapii), které mohou zmnožené prekurzory B
Současná
vyšetření
12.
ll.
buňky
jsou
částečně imunofenotypově
nacházíme v expresi jednotlivých
od nemaligních
buněk.
antigenů četné
Klasifikaci, kterou v
vychází z upravené a aktualizované
podobné svým nemaligním
současné
správné imunologické odlišení od AML.
asynchronie a aberace odlišující je
chvíli používáme u
klasifikace EGIL
lmmunological Characterization of Acute Leukemias/ 6.
protějškům,
/7.
dětských
(European Group for
ALL the
Diagnosticky důležité je rovněž
'1:::s \.::).. ~ ...... Os:.:.CIl~ >::l... CIl- ~ "l
(rj
~ ~ F?..~, ~ ..... C\) ~o t', ::a::~ ::::, Jo... __ 5' t', t',
'ť5
. II
o.s
[ t 1
C\)
pre-pre-B-cell HLA-DR (CD34) CD19 CD22 C010 (C020)
~'~ N
~ o O· 5:. (:), ~ ~ ~ ::s-~
~
\.::) ~ ::::-:
......
N
" ~ o "'"
'"~~~ .
~ ~~. ~ \.::)
_ON ~ t',
Iymphoid
~ progenitor cell
~
-.
C\)
\.::)
\.::):::-
~< ~. ~ ~ ~ "2
C\)
-8
f.t ....
~Oe Vl
::s- :3. ~
,<:,
-<:
~
C\)
-<:
\.::)
O
~
<-2. ~§ :::=-. t', ~'~ "'= - . ..... ~ -<: ~ O ' '-- ....,
-C\)
t',
intermediate B-Iymphocyte HLA-DR CD19/CD22 CD20 CD37 Smlg-C079 (COS/CD6) (C023)
C\).
mature (fol.) post-follicular B-Iymphocyte B-Iymphocyte HLA-DR HLA-DR CD19/CD22 CD19 CD20 (COllc) CD22 CD31 ~g,2~) C020 Smlg-CD79( C031 CD10 Smlg-CD79 (CD38) ~
§,
~ ~ ~
Ili!l,
"-ir-I!
f(
'j
1...,11
immunocyte HLA-DR CD19 CD22 (CD20) (C031) (C0138) Smlg-CD79
HLA-DR CD34 (C0117)
(~D3)
/TCF-l
prothymocyte HLA-DR CD34 CD7 C02
--8CF-1eLJ~
mature thymocyte CD7 CD2 CD5 CD4 TCR-CD3
~
helperl activated helperl i1ducer inducer T-Iymphocyte T-Iymphocyte (CD7) (CD7) CD2 CD2 CDS C05 CD4 CD4 TCR-CD3 TCR-CD3 HLA-DR CD2S I I L ... L
I
,....1
I ~ ~ r"-
__ I .
-, I
11
C\)
_ 0 O ."'"' ...... '<:'
~ C\)
~.
"",
C\) t', ::s::s__
t'"
'" ..... ~~ e;:;~ (=;: \.::) .::"..,
~
~
C\)
\.::)..
\.::),~
~'
lC ~~ .
r"' ,
plasma cell (HLA-DR) CD38 CD138
CD7 C02 CDS
COS TCR-CD3 (CD16/CD56ICD57) HLA-DR CD25
N
,
•
1
~ I I ~
~
~.
....
C/>
:J ~
:;.;-O
<:
J>'
.... ....
C/> ~
::l ~
'-.J
C'l
'" >::l...;:'-. ;::,- ~ t',
"'"' ~ ::::-:'1:::s ...."
,"' 1
'ífMIIee." _ _
~
.-<:
transitional pre-B-cell HLA-DR CD19 CD22 C010 (C020)
t--l< ~
:::- ~.~ t','~, ~ 6 "'; ;:::
t'1 E.i
pre-B-<:el! HLA-DR (CD34) CD19 CD22 CD10 (CD20)
pra-B compl&x
;:::... ~ ~ V) ('\:) ~
j
§--~--e-------o~o~ 8 ------@-----GY
"'~. = Cll ;:s~ __ "l< o_ . "N"'= ;:". -oe \.::)..
II
-I II
II
'>
i.'" I
I
fl (r
N f<_CL'( II~ (>'l l-
@ NKcell CD7 (CD2) (C08) CD16 CD56 (CD 57) (HLA-DR)
I
j
E.
Mejstříková,
strana 8
ALL
1.1.1.1.1.
Z prekurzorů
B
lymfocytů
(BCP A LL)
B prekurzorová leukémie je dominujícím podtypem ALL jak u Základní imunofenotypové
rozdělení
je v tabulce 1,
dětí ,
lastní výsledky a
tak i u
dospělých.
podrobnější
rozbor
tohoto podtypu uvádím v kap itole 3.1. 1.
Kategorie
Podtř í da
Kritéria
• B prekurzorová
• • • • • • •
zralá B
,
skupiny BFM (Berlin - Frankfurt - M unster) pro
ALL
CD I Opoz intralgM"eg intralgMPOZ
2 nebo 3 z následuj ÍCÍch: CD I 9 Poz, (intra)CD79a Poz a CD22 poZ CD3 neg Bez další subklasifikace intraCD3 1leg PoZ nebo APo > K v
1.1.1.1.2.
CD I OJ"!; CD20 lleg
2 nebo 3 z následuj ících : CD I 9 PO" proB ALL (intra)CD79apoz a CD22 PoZ CD3 1leg cALL intra CD3" eg llog a AJIt!; K praeB ALL
Tabulka 1. Klasifikace ALL Brady, adaptovana podle EGIL protokolů
Kritéria
Z prekurzorů
lb
" , kteraJe pOUZlvana v ramCI
léčbu dětské
ALL.
T lymfocytů (T A LL)
T ALL jsou méně častým podtypem, z dětských leukémií tvoří asi 10 až 15% . Často při diagnóze nacházíme nádorovou infiltraci thymu, vedoucí k tumoru mediastina, který způsobovat /9.
syndrom horní duté žíly, respektive dušnost na
podkladě
může
obstrukce dýchacích cest
Tato nádorová infiltrace v sobě odráží normální vývoj T lymfocytů, který právě z větší
části
probíhá v thymu. N ení neobvyklé, že u
nacházíme je ště normální krevní obraz a do periferní krve a kostní arbitrárně
dřeně.
procentem 25%
těchto pacientů při
během
prvním kontaktu s
lékařem
krátkého intervalu dochází k tzv. leukemizaci
Hranice mezi lymfoblastickým lymfomem a ALL je tak dána
blastů
v kostní
k oběma jednotkám velmi podobný
20.
dřeni .
Terapeutický
přístup
je ale v
současné době
Retrospektivní analýzou přežití dětí diagnostikovaných
s de novo ALL mezi zářim 1996 a srpnem 2006 nevidíme v ČR signifikantní rozdíl mezi T a Bep A LL (obrázek 2). Postupné zlepšování provedené kolegy z
Německa
je
například
dokumentováno v analýze
u vysoce rizikových T ALL, kde se mezi protokoly ALL BFM
90 a ALL BFM 95 došlo zlepšilo přežití bez události v 5 letech od diagnózy o 20% imunofenotypu lze T ALL děl it podle EGIL klasifikace
/6.22
2/.
Podle
na proT, praeT, intermediární T
a zralou T (tabulka 2)23, toto členění má částečně i prognostický význam.
E. Mejstříková, strana 9 o 1,00 0,95 0,90 0,85 0,80 0,75 0,70 _ 0,65 ~ 0,60 ~ 0,55 ~ 0,50 ~ 0,45 ~ 0,40 UJ 0,35
Complele
+ Censored
11Ir---~--~---------~---.
'"
0,30
0,25 0,20
0,15
T -- --- B
0,10
0,05
0,00
L-..o._ _ _
~
_ _ __ _ __ _ _
~
__
~
8
_
_
10
-----'
12
roky
Obrázek. 2. Analýza přežití dětí diagnostikovaných mezi lety 1996 až 2006. Rozdíl mezi bez události s T ALL a BCP ALL není signifikantní (Cox-Mantelův test). Celkem do analýzy zahrnuto 582 pacientů (83 T ALL, 499 BCP ALL). Událos/je definována jako relaps, sekundární malignita nebo smrt. Medián sledování kohorty pacientů s T ALL je 3,9 let a BCP ALL 4,4 let. Pravděpodobnost přežití (PEFS) pacientLI sT ALL v 5 letech je 72%±5%, s BCP ALL 75%±2, I%. přežitím pacientů
Kategorie
Kritéria
(intra)CD3 POZ a CD7 PoZ
I
TALL
Podtřlda
Kritéria
proT ALL
CD2 neg CDS neg CD8 neg
preT ALL
CD2 poZ a/nebo CDS PoZ a/nebo CD8 poZ
intermediámí T ALL
CD 1aPo >.
zral á T ALL
CD3 PoZCD 1a"e s
TCR a~Poz
TCRa~Poz
TCRyů v
POZ
T ALL
T ALL ,
TCRyoPoZ
Tabulka 2. Klasifikace ALL T rady, adaptovana podle EG!L protokoHI skupiny BFM pro léčbu dětské ALL.
. pOUZlvana v ramCl ,ktera' Je v,
T ALL lze
rovněž členit
ALL)
Lepší prognóza u dětí i u dospělých je spojena s podtypem intermediámí, resp.
24-26 _
tzv. "thym ic" T AL L 27, imunofenotypu u
28.
podle typu genetických
změn
lb
(viz kapitola Genetické podtypy T
Srovnání přežití s pacienty s Bep ALL a rozdělení podle zralosti T
českých dětí
je na obrázku 2 a 3. Studie z konce 80. let v rámci studie POG
7865 a 8035 neprokázala asociaci zralosti podle imunofenotypu s prognózou , ukázala však , že pacienti s nezralou formou T ALL častěji nedosáhnou kompletní remise
29.
Naše data
dokumentují prognostický význam zralosti T ALL (obrázek 3), i když jsou omezena velikostí
· vlk ova, strana 10 E . M eJstn I
souboru.
V rámci
protokolů
GM ALL
pro
dospělo u
ALL
(Německo)
imunofenotyp intermediární T ALL jedním z kritérií standardního rizika
30.
jsou
Je
Alternativní
imunofenotypovou klasifikaci, která lépe odráží vývoj T řady z pohledu přestaveb T buněčného
receptoru podle našich
profesorky Macintyrové (tabulka 3)
současných
znalosti, navrhla skupina pod vedením
3/-34.
Podtyp T ALL
Exprese antigenů
nezralá T ALL
Cyt_~neg,
TCRy8
povrchová CD3 neg, TCRa~neg a neg
povrchová CD3 neg, TCRa~neg a neg
Cyt_~poz,
pre-ap T A LL
TCRy8 TCR-ap T ALL
povrchová CD3 Poz , TCR-a~Poz
TCR-yo T ALL
povrchová CD3 Poz , TCR- y8 poz
Tabulka 3. Klasifikace T ALL podle Macintyrové
Skupina profe sorky Macintyrové
rovněž
prokázala, že s
věkem
až do pozdní
dospělosti
přibývá nezralých T ALL a ubývá zralých prekurzorových forem T ALL, zejména TCR-a~ Tento jev pravděpodobně souvisí s postupnou involucí thymu v
o
C o mplell!
průběhu
33.
života.
+ Cem;Dr.d
1.00 O , 9~
0,90 0,85 0,80 0.75
..
0.70 o,~s
2:: 0,60
~
.!_
!
0,55 0,50
0,45 Q40
0.35 031)
Ol-S 020 0 ,1 5
- mBI~i!'T
o iO
-
lntermedJme
-es.rtyT
0.05
0.00 10
12
roky
Obrázek 3. Prognóza dětí s de novo T ALL (n=80); podtypy zralá [ma ture] T n=33, intermediární [intermediary] T n=31 a nezralá [early] T n=17 ('Ihrnující: praeT n=16 a pro T n=l), Je zřetelná sign~fikantně horší prognóza skupiny s nezralou T ALL (pEFS v 5 letech 44%± /4%) oproti skupině s intermediární T ALL (Cox-Mantel, p =O.Ol , pEFS skupiny s intermediární T v 5 letech 84%±6, 6%). Rozdíly mezi skupinou nezralou T a zralou T, resp. mezi zralou a intermediární signifikantní nejsou (pEFS skupiny se zralou T v 5 letech 76%±7,5%).
E. Mejstříková, strana 12
Genetické podtypy BCP ALL
1.1.1.2.1.
Nej častějšími změnami
genotypu u
genu TEL/AMU. Tyto
dvě
dětských
aberace jsou
korelují s dobrou prognózou
36-38 .
leukém ií je hyperdiploidie a
přítomny
celkem asi u 50%
(např.
chromozómů
fluorescenčních
kromě
(poměr
40
diskrepancí
můž
vyjadřujeme
vyšetřením
tzv. DNA
buněk
není stoprocentní.
být jak horší senzitivita cytometrického
proliferovat in vitro, která
často
poměrně
vyšetření
u
špatná schopnost
vede k selhání cytogenetického
Hodnota DNA indexu nad 1, 16 a pod 1,6 koreluje s dobrou prognózou u BCP ALL
(obrázek 5 a 6). Naopak hypodiploidie pod 0,8 je spojena s horší prognózou, (hypodiploidie
«45 chromozómů) až haploidie (23 - 29 chromozómů) «44
chrom ozómů
genetického
4/.
nebo DNA index <0,81) kvalifikuje
léčebných protokolech do vysokého rizika
materiálu
části
39.
Nález signifikantní hypodiploidie příslušného
pacienta v
pacientů
až 39
4/.
pacientů
s
s oučasně
hypodiploidního klonu a nacházíme pak
počtem chromozómů
24 až 29 a
některých
U části pacientů dochází ke zdvojení
hypodiploidní tak hyperdiploidní klon. Ve studii Nachmana et a!. se toto zdvojení u
rovněž
barviv, schopných stechiometricky vázat nukleové kyseliny
výrazného zmnožení genetického materiálu, tak i
vyšetření.
být
modu GOl fáze aneuploidnÍ populace ku modu G01 fáze euploidní kontrol y).
Důvodem ojedi ně lých
leukemických
může
klasických cytogenetických metod
Korelace cytometrického DNA indexu s cytogenetickým
méně
4, 10 a 17, což
propidium jodid, DAPl nebo HOECI-IST). Zmnožení DNA
indexem
BCP ALL a
39.
Hyperdiploidii , respektive hypodiploidii lze stanovit pomocí
dětských
fúzního
Podle některých studií jsou pro dobrou prognózu
hyperdiploidní ALL specificky významné trizómie zohledněno i léčebně
přítomnost
méně často
u
pacientů
s počtem
jak
častěji děje
chromozómů
33
Prognóza pacientů se zdvojeným hypodiploidním klonem je stejně špatná jako u bez tohoto zdvojení. Správná identifikace hypodiploidního klonu je
správnou stratifikaci pacientů
42 .
důležitá
pro
Nález velmi vysoké hyperdiploidie až téměř tetraploidie je
velmi vzácná (incidence ze všech ALL asi 1% až 2%), u BCP ALL koreluje s fúznÍm genem TEL! AML 1
43-45 .
Špatná prognóza je rovněž spojena s přestavbami genu MLL (11 q23)
zejména u d ětí mladších 1 roku
46,
u starších dětí s MLL přestavbou je zřejmá ni žš í
kumulativní incidence relapsů oproti kojencům v rámci BFM protokolů
47.
Negativní vliv na
prognózu má translokace (9;22) vedoucí ke vzniku fúzního genu BCRJABL narozdíl od prognózy i u
dospělých, dětí ,
j e tento fúzní gen
poměrně
vzácný.
neproj evuje se pro celkovou vzácnost
Přestože
významně
48. 49 .
U dětí,
je znakem špatné
na celkové prognóze u
E. Mejstříková, strana 13 ALL. U v této
dospělých tvoří
až 30% (obrázek 4) a podílí se na
významně
horší prognóze ALL
věkové skupině . Cumulatl ve Proponion Surviving (Kaplan-Meier) o Com plete + Censored
0 ,8
0,6 (j)
"ll! 0,4
tém ěr t e traploid ni (01)= 1.6) euploidní (01=1,0) DI nezměre n - hyperdlploidni (01)= 1.16 a <1.6) _.- nízce hyperdlploidni (01) 1,0 a <1.16) hypodiploidnl (01 <1.0 a >0.8) nízce hypodiploidni (01<=0.8)
-
0,2
0,0
10
12
ro ky
Obrázek 5. Přežití českých pacientů s BCP ALL podle obsahu DNA. Kategorie jsme definovali pomoci DNA indexu (DI). (euploidní n=307, hyperdiploidní n=112, nízce hyperdiploidní n=35, hypodiploidní n=l , nízce hypodiploidní n=6 a téměř tetraploidní n=6. Analýza v§ech podskupin najednou potvrzuje signifikantní rozdíly (p =0, Ol 7). Pro úplnost je ukázáno přežití bez události i u pacienttt, kde pro nedostatek materiálu nebo technickou závadu na pNstroji nebyl DNA index změřen (n=34). Rozdíl mezi podskupinou euploidní a hyperdiploidní je sign~fzkantní (p = 0,023). Prognóza ostatních podskupin se od euploidní BCP ALL neliší. C urn.J ail.e Prqx:rtion&.l'\.il.ing
, 1,0 0,9 0.8 0.7 0,6 ~
Ol
0,5 0.4
,
Corrp" e
+
(~Ia(l.Meer)
Ce-lsored
t,
~,
I
"':>~ "'r. , I
1
I-
-
--
I I
I
-
,'Jl} '"
',I . ,:
"'
0,3 0,2 0,1
bez~twerdpodeanootetl.m1 - - t?,pErd poimi a noo te /an 1 .... tet -.m1 j- ' ;I h)poápode a m l teU.m 1
0,0 10
12
rol<\l
Obrázek 6. Prognostický význam DNA indexu v kontextu TELIAJv/Ll genotypu: vysoká hyperdiploidie (DNA index -:::'1 ,16 a <1,6) n=112, pacienti s jakoukoli hypodiploidií n=6, TELlAMLlPo= pacientll n=141 a ostatní n=218, Rozdíly jsou celkově signifikantní (p =0,00007). nechny svorky znázorňují signifikantní rozdíly mezi dvěma podtypy (p
E. Mejstříková , strana 14 1.1.1.2.2.
Genetické podtypy T ALL
Pro relativní vzácnost T ALL je vztah genotypu a prognózy stále téměř
než u BCP ALL. Zhruba u 5% T ALL nacházíme
poněkud méně objasněný
tetraplodii, která nemá prognostický
význam. Zhruba u 50% dětí s T ALL nacházíme rekurentní chromozomální trans lokace vě tšiny
těc hto
translokací
je
T-buněč ný
zavzat
s předpokladem, že se jedná o chybu vzniklo u často
Další geny
zapojené do
přestaveb
při
r ceptor
fyzi logickém
TCR-a/o
Do
TCR-~
nebo
přestavování genů
50.
pro TCR.
jsou MYC, TAL1, TAL2, L YL1, bHLHB I, LMO I,
LM02, homeoboxové geny (HOXll/TLXl, HOXII L2/TLX3 nebo HOXA cluster). Vzácnou translokací u T ALL, která demonstruje složitý vztah genotypu a fenotypu, je fúzní gen CALM/AFlO, vzniklý na
podkladě
pojí s přestavbam i TCR yo histiocytární Iymfom
52.
Zhruba u poloviny
pacientů
5/.
translokace (10 ;11). UT ALL se fúzní gen CALM/AFIO Tato translokace však způsobuje i AML nebo např.
sT ALL nacházíme mutace v genu NOTCH 1, které vedou
k upregulaci signalizační dráhy, do níž je NOTCHl zapojen že existují inhibitory této dráhy (inhibitory y-sekretázy)
53 .
Důležité pro terapii T ALL je,
54.
Nej eví se zatím jednoznačná asociace genotypových změn a zralosti dle imunofenotypu
1.1.1.3.
Klinické a biologické projevy ALL
Leukémie se klinicky krvetvorby,
tzv .
trombocytopen ie
a
trias:
anémií,
imunodeficitem buněk.
kompresi
fyziologických důsledky
Rovněž při
leukocytóze
může
viskozity
krve).
Typicky
na
Jak již bylo
s typickými klinickými
(hepatosplenomegalii),
1.1. 1.4. Věk
především příznaky
krvácivostí,
podkladě
zmíněno
často
případně
normálních
infiltruje thymus a
přítomných
v mediastinu
(syndrom horní duté žíly, obstrukce dýchacích cest apod.).
nastat leukostáza (porucha mikrocirkulace na u
podkladě
na
počtu
redukce
struktur
ze selhání nemaligní
zejména
výše, T ALL
anatomických
ALL
dále
nacházíme
infiltraci
lymfatických uzlin. Dále u
bolestivost kostí a subfebrilie až febrilie, pravděpodobn ě
diagnóze
diagnóze nemoci projevuje
leukemickou
imunokompetentních způsobuje
při
při
26.
vyp lavují jak leukemické
část těchto příznaků
buňky,
je
tak i aktivované
části
podkladě
jater
pacientů
způsobená
buňk y
zvýšené
a
sleziny
Je
zřetelná
i cytokiny, které
imunitního systému.
Biologické projevy ALL
a leukocytóza při diagnóze jsou již dlouhou dobu známým důležitým prognostickým
faktorem. Poměrně často jsou zmiňována tzv. NCI (National Cancer Institute) kritéria Podle
těchto
NCI kritérií standardní riziko
splňuje
pacient 1 až 10 let
věku
39. 55.
a s iniciální
E. Mejstříková, strana 15 leukocytózou nižší než 50000/IlL, ostatní pacienty (děti
Kojenci
nepříznivou
NCI klasifikace do vysokého rizika.
mladší 1 roku) jsou naproti tomu velmi rizikovou skupinou ALL s odpovědí
cytogenetikou a špatnou
částečně hůře
řadí
odpovídají na
léčbu
na
léčbu.
56.
rovněž nepříznivým
u
léčby;
oba tyto faktory
Postižení CNS při diagnóze ALL je tě chto pacientů intenzivnější
cílenou
3.
Farmakogenetika ALL
Snížená akumulace aktivních
metabolitů
zvýšenou c1earence, inaktivací publikována
řada
či
jinými
cytostatik v leukemické
důvody,
podání
některých
významně
zvyšuje c1earence
antiepileptik některých
(např.
buňce, ať
už je
genů důležitých
pro metabolizaci
odpovědi
v léčebných výsledcích
3.
léků.
fenytoin, fenobarbital nebo karbamazepin)
cytostatik produkcí cytochromu P-450. Polymorfismy
které kódují metaboli zující enzymy, transportéry, receptory jsou
významné rozdíly v
způsobena
je spojena s horší prognózou. Nedávno byla
studií zabývající se odlišnou expresí
Současné
genů,
překonat
faktorem, který ale lze
léčbou, jak chemoterapií, tak i radioterapií
1.1.1.5.
O adolescentech je známo, že
a zejména dívky trpí vyšší toxicitou
se podílejí na horší prognóze dětí mezi 15 až 18 lety
často
zodpovědné
za
na jednotlivé léky. Tyto polymorfismy se ne vždy odrážejí
Jedinou významnou výjimkou je vztah polymorfismů thiopurin
methyltransferázy. Pacienti s vrozenou homozygotní
či
heterozygotní deficiencí enzymu
thiopurin methyltransferázy, který katalyzuje S-metylaci (inaktivaci) merkaptopurinu, mají signifikantně
vyšší riziko toxických nežádoucích
na léčbu v protokolech obsahujících tento lék 1.1.1.6.
účinků,
odpověď
5 7.
Princip terapie ALL
Většina léků
používaných v
léčbě
leukémií interferuje s
syntézu DNA nazýváme antimetaboJity a poškozují např.
na druhou stranu mají lepší
buněčným
cyklem. Léky postihující
především dělící
metotrexát blokující enzym dihydrofolát reduktázu, která je
kyseliny listové na kysel inu tetrahydrolistovou, která hraje
klíčovou
se bunky.
zodpovědná
v syntéze
Patří
sem
za redukci
purinů.
Dalším
dů ležitým
antimetabolitem je cytosin arabinosid (ara-C), 6-merkaptopurin a 6-thioguanin.
Alkylační
cytostatika poškozují více
se
buňce (např . odloučení
sem
např.
se
buňky,
narušují ale i strukturu DNA v
purinových bází, zlomy v jednom
či
především
vinblastin a vinorelbin se léčbě
nedělící
obou vláknech DNA).
cyklofosfamid a ifosfamid. Antracyklinová antibiotika
daunorubicin) blokují
lékem v
dělící
(např.
Patří
doxorubicin a
enzym topoizomerázu II. Rostlinné alkaloidy vinkristin,
především
váží na mikrotubuly
dělícího
ALL je asparagináza, která necílí na syntézu
či
se
vřeténka.
Významným
strukturu DNA, ale
způsobuje
E. Mejstříková, strana 16 depleci extrace1ulární aminokyseliny asparaginu, což postihuje především proteosyntézu buněk ,
leukemických
které nedokáží
kompenzovat
nedostatek této
aminokyseliny.
Kortikoidy obecně po pasivní difúzi buněčnou membránou se váží na specifický kortikoidový receptor a s ním jsou transportovány do jádra. Glukok0l1ikoidy mají mimo jiné výrazný ú či nek na buňky imunitního sysť mu zejména na lymfoidní buňky, u kterých ve farmakologických dávkách indukují apoptózu. Léčba
a dodržení léčebného schématu bez dlo uhých prodl ev v terapeutickém schématu je
velmi důležitým prognostickým faktorem. Velkým zdrojem poučení jsou studie srovnávající prognózu
adolescentů
a mladých
dospělých
léčených
pediatrickým protokolem nebo
protokolem pro dospě lé. Podle publikovaných studií je protokolem pro dospělé horší dospělé
prognóza pacientů léčených
Za tímto rozdílem může stát nižší intenzita protokolů pro
58-60 .
i vyšší protokolární compliance a
úspěšnější podpůrná léčba
na pediatrických
pracovištích. Současná
léčba
ALL v jednotlivých protokolech J
postavena na
několika
léčebných
principech: 1)
indukční léčba normálně
ALL má za cíl dosáhnout kompletní remlse «5%
blastů
v KD a
např .
tumoru
fungujíc í hematopoéza, redukce extramedulárního postižení
mediastina, sanace mozkomíšním moku od uplatňuje
V rámci indukce se zpravidla
blastů při
kombinace
iniciální CNS infiltraci apod.). kortikoidů
(prednisonu nebo
dexamethasonu) s vinkristinem a dále s asparaginázou nebo antracykliny (resp. s
oběma).
V rámci BFM
protoko lů
před fáze
(stejná kortikoidová
pokračování
metotrexátu,
je strategie v indukci u všech
s prednisonem a jednou dávkou
pacientů
stejná
intrathekálního
v kortikoterapii spolu s vinkristinem, asparaginázou a
daunorubicinem. 2) velmi (např.
časně
podaná
pomocí
léč ba
CNS , jak prokazatelného, tak
intrathekálně
3) konsolidace remise a
podaného metotrexátu)
rein dukční léčba
normální krvetvorbu s cílem 4) udržovací
léč ba, např.
metotrexátem. zkrácení
např.
Přesný
v
léčit
mechanismus
intensivní
intenzivnější
49.
kdy pacient již má
protokolů účinku
dobře
fungující
buněk
s perorálním 6-merkaptopurinem a
udržovací
léčby
není znám, ale její nárůstu
Je možné, že dlouhodobé podávání cytostatik po
části léčby ničí
prvek v
době,
studie ALL BFM 1990 vedlo k signifikantnímu
relapsů ve skupině SR rizika skončení
v
zbytky resistentních leukemických
v rámci BFM
průběhu
subklinického postižení
podobě pulsů
leukemické
buňky
s pomalejším
vinkristinu a dexamethasonu v
dělením.
průběhu
Naopak
udržovací
E. Mejstříková, strana 17 léčby
se ukázal jako
95/ 1
62.
neúčinný
v rámci
součas ného
typ u
léčby
(v protokolu ALL BFM
Historický význam těchto pulsů mohl vyplývat z obecně nižší intenzity
předchozí léčby.
5)
ozařování
CNS je v
současné době
(např.
pro CNS relaps
používáno j en u podskupiny s
největším
rizikem
pacienti sT ALL, vysoké riziko ALL, infiltrace CNS
při
diagnóze).
1.1.1.6.1.
Odpověď
1.1.1.6.1.1.
Kompletní remise
na
léčbu
Stav, kdy nacházíme v kostní
ALL
dřeni
optickou mikroskopií
k rekonstituci fyziologické hemat poézy a v mediastinum) je
zřetelná
metodách o více než Standardně
daří
se
sanace od
třetinu
blastů,
případě
než 5%
resp. redukce tumoru mediastina na zobrazovacích
indukčním
špatnou prognózou, tito pacienti jsou vždy
dětských pacientů
řazeni
do nejvyššího rizika a v
j sou indikováni k alogenní transplantaci kostní
dárcé9, 63.
64.
prednisonová
odpověď
počtu blastů
dřeně
příbuzného
od
řadě léčebných
i
nepříbuzného
v den 8 v periferní krvi po kortikoidové
známé je hodnocení tzv . prednisonové
s ALL.
bloku je vzácné a je spojeno s velmi
protokol ů
Korelace
a kdy došlo
extramedulárního postižení (CNS,
kompletní remise dosáhnout u více než 98%
1.1.1.6.1.2.
blastů
od iniciální velikosti, nazýváme termínem kompletní remise.
Nedosažení kompletní remise po prvním
Dobře
méně
odpovědi,
skupiny BFM . Tento parametr zavedl profesor Riehm a
předfázi
které se používá v protokolech
spočívá
v hodnocení úbytku
blastů
v periferní krvi po týdnu léčby prednisonem a jedné dávky íntrathekálního metotrexátu Pacienti s více než 1000
blastůJ~L
-
v periferní krvi v den 8 mají
signifikantně
65.
horší prognózu a
v protokolech BFM skupiny jsou řazeni do vysokého rizika. Špatná prednisonová odpověď vzhledem k vysokému riziku relapsu
těchto pacientů
tak neukazuje jen farmakorezistenci na
prednison, ale i vlastně na další cytostatika použitá následně v léčbě ALL. Špatná prednisonová
odpověď
predikuje horší prognózu i u
například kojenecká AU.
66
prokázala horší prognózu
pacientů
nepříznivých
podskupin ALL, jakou je
nebo ALL s fúznÍm genem BCR! ABL
kombinací cytostatik a kortikoidů
68.
67.
Studie Gajjara et al.
s prokazatelnými blasty v periferní krvi po týdnu
léčby
E. Mejstříková, strana 18
1.1.1.6.1.3.
Hodn ocení poklesu
Další parametry, které se
buď
blastů
v kostní
dřeni před
používaly, nebo používají v
dosažením remise léčebných
protokoly Chjldren Cancer Group), jsou odvozené od hodnocení vlastním dosažením kompletní remise
1.1.1.6.1.4.
odpovědi
(např.
v kostní
dřeni před
Léčba
leukémie,
39, 69, 70
Minimální reziduální nemoc (MRN)
P roblematice MRN je monitorování
1.1 .2.
protokolech
věnována
kapitola ve Výsledcích a diskusi (3.2 -
účinnosti léčby)
Přístup
k pacientů m s relapsem All a k
p rimárně
vysoce rizikové
pacientů
v první kompletní
ALl 1.1.2.1.
dřeně
Indikace transplantace kostní
(SCT) u
remisi ALL V rámci ALL je alogenní transplantace krvetvorných kmenových podskupiny
pac ientů
s velmi vysokým rizikem selhání
léčby
s
buněk
indikována u
předpokládaným přežitím
události nižším než 50%. Studie Balduzziové et aJ. prokázala lepšÍ prognózu transplantovaných od
příbuzných
léčených pouze chemoterapií
nedosažení remise po MLL! AF4, špatná nad 100.1 09/L nepříbuzenská
dárců
s velmi vysokým rizikem na rozdíl od
dětí dětí
Velm i vysoké riziko bylo v této studii definováno jako
7/.
indukčním
odp ověď
7/ .
shodných
bez
bloku nebo
přítomnost
fúzního genu BCRI ABL nebo
na prednison spojená sT imunofenotypem a/nebo s leukocytózou
Jedinou nadějí pacienta s vysoce rizikovou fonnou AL je někdy
transplantace kostní
dřeně
( příloha
s neúplnou shodou
10).
Autologní transplantace není u ALL indikována. Přístup
1.1.2.2. Prognóza
k
pacien t ů m
těchto pacientů
vysokého rizika
67.
s fúzním genem BCRIABL
je velmi
nepříznivá
a ve
většině léčebných protokolů
jsou
řazeni
do
V současné době lze těmto pacientů přidat ke standardní chemoterapii
specifickou terapii inhibitorem BCRIABL kinázy imatinib mesylátem (imatinib). V
současné
době
jsou dostupné na trhu i dalšÍ inhibitory BCRIABL kinázy, které se testují pro použití i u
dětí
s rezistencí
či
intolerancí imatinibu. V
současné době
kombinovat chemoterapii s léčbou imatinibem zvyšuje
signifi k antně
72.
několik přístupů,
procento dosažených kompletních remisí po iniciální
onemocnění,
tyto remise jsou jen
jak
Časné podání imatinibu do indukční léčby
pacientů, nicméně dopad na celkové přežití je omezený
navodit remisi
existuje
léčbě
u
dospělých
73.
Přestože i samotný imatinib může
dočasné
a velmi rychle dochází k rozvoji
E. Mejstříková, strana 19 resistence na tuto léčbu
74.
Mechanismus vzniku resistence není zcela objasněn, ale klíčovou doméně,
roli hrají bodové mutace v tyrosinkinázové často
tyrosinkinázy. Tyto mutace jsou
přítomny
resp. ve vazebné
doméně
v subklonu leukémie a
tyrosinkinázovým inhibitorem dochází kj ej ich selektivní expanzi
pro inhibitor během
léčby
Od září 2004 jsou děti
75.
s nově diagnostikovanou ALL s prokázaným fúzním genem BCR! ABL od dne 33
přeřazeny
do protokolu EsPhALL (European Intergroup Study on Post Induction Treatment of Philadelphia Positive Acute Lymphoblastic Leukaem ia with Imatinib). Pacienti s vysokým rizikem dostávají v protokolu EsPhALL ke standardní chemoterapii
ještě
imatinib a SCT.
Pacienti s lepší prognózou jsou randomizováni do ramene s a bez imatinibu, a podle hladin MRN a dostupnosti dárce SCT je prognóza řada,
jednou z
rovněž
části pacientů
nejdůležitějších
podávaný inhibitor na
přes léčbu
kombinací imatinibem, chemoterapií a
s tímto fúznÍm genem velmi špatná.
Příčin
selhání
léčby
je
je již výše popsaný dynamický vznik resistence BCR! ABL na
podkladě
forem CML popisována
SCT. I
mutace v ABL kinázové
kompenzační
doméně.
Dále je u
pokročilých
hyperexprese BCRlABL na úrovni mRNA,
podíl tohoto jevu na klinické resistenci na imatinib není znám
76,
přesný
pravděpodobně hraje i roli
v resistenci na imatinib u PhPozALL. Další typy resistence označujeme jako BCR! ABL nezávislé, jedná se zejména o konstitutivní aktivaci kináz z rodiny SRC dostávají na trh další inhibitory tyrosin kináz
(např.
schopností inhibovat BCRlABL tyrosinkinázu a resistenci na imatinib
77.
překonávat část
mutací,
způsobujících
U dasatinibu na rozdíl od imatinibu je navíc popisována i schopnost
inhi bovat kinázy z rodiny SRC může
Postupně se
dasatinib, nilotinib) s větší terapeutickou
pronikat hematoencefalickou bariérou do CNS při systémovém podání
SRC kináz
76.
76.
78
a schopnost
Zatím není zcela objasněn vliv působení, zdá se že inhibice
vést k poruše funkce zejména T
imunodeficitu. výzvou pro výzkum nových
lymfocytů
inhibitorů
a vést tak k sekundárnímu
tyrosinkináz je mutace T3151, která
vede k rezistenci na všechny dosud používané inhibitory (imatinib, dasatinib, nilotinib) jak u CML tak i ALL. Přístup
1.1.2.3. Prognózu
k
pacientům
části pacientů
zlepší SCT
21.
s vysoce rizikovou T ALL
se špatnou
odpovědí
na prednisonovou
předfázi
a T imunofenotypem
vývoj léčby T ALL dobře demonstruje nutnost stavět nové protokoly na datech
MRN. Pro relativní vzácnost T ALL srovnávala Schrauderova studie efekt SCT u s vysoce rizikovou T ALL
léčených
podle
protokolů
pacientů
ALL BFM 90 a 95. Nutno zmínit, že
protokol ALL BFM 90 znamenal zhoršení prognózy u všech vysoce rizikových ALL oproti předchozímu
protokolu ALL BFM 86,
pravděpodobně
z
důvodu
snížení celkové dávky
E. Mejstříková, strana 20 alkylačních cytostatik a z důvodu léčby krátkými rotačním i bloky chemoterapie po indukci působením
bez delšího protokolu II s prolongovaným výsledk ů
bylo reflektováno v protokolu ALL BFM 95, který
prognózy vysoce rizikové T ALL oproti zřejmé, kteří
cytostatik. Toto zhoršení
že pacienti sT ALL
jsou 3
nerelabují
měsíce
obecně
odpovídají na
léčby
od zahájení
předchozímu
přinesl
léčebných
signifikantní zlepšení
protokolu. Ze studie Willemse et a1. je
léčbu
pomaleji,
nicméně
pacienti sT ALL,
v rámci BFM protokolu MRN negativní, prakticky
Předběžné výsledky z ALL-BFM 2000 protokolu, které zatím nebyly
79.
publikovány pro krátkou dobu sledování celé kohorty, ukazují, že negativita 3 diagnózy znamená výbornou prognózu i u předfázi.
49
základě těchto výsledků
Na
v
pacientů
se špatnou
současné době
odpovědí
od
na prednisonovou
těchto pacientů
u
měsíce
není indikována
transplantace kostní dřeně v první remisi, shodně se postupuje i u dětí v ČR. Naopak pacienti s vysokou hladinou MRN 3
měsíce
od diagnózy jsou ohroženi vysokým rizikem
relapsu a je u nich indikována SCT i od dostanou vysoce intenzivní během
následných
bloků
léčbu
nepříbuzného
dárce.
s monitorováním MRN.
chemoterapie,
kdy
SCT tito pacienti
Při nedostatečném
už je
chemorezistentní, jsou takoví pacienti kandidáty pro
Ještě před
zřejmé,
že
časného
poklesu MRN
nemoc je
změnu léčby včetně
velmi
experimentální
léčby. V současné době se na trh dostávaj í nové léky namířené nejen na léčbu T ALL Příklady těchto
80.
léku jsou nelarabin (ARA-G) a forodesin, inhibitor purin nucleosid
fosforylázy (PNP). Forodesin navodí situaci podobnou vrozenému poškození PNP. Mutace genu pro PNP vedoucí k jejímu vrozenému poškození je kombinovaného imunodeficitu. Absence na
podkladě
zvýšené kumulace
funkční
příčinou
PNP vede ke zvýšené apoptóze T
metabolitů purinů.
účinek.
spuštěné aktivační
těžkého
Iymfocytů
Zajímavá je myšlenka kombinovat u
refraktemí T ALLlLBL nelarabin s forodesinem, a prohloubit tak antileukemický
varianty
vzájemně
jejich
Inhibitory gama-sekretáz nejen blokují zvýšenou aktivaci dráhy
mutací v proteinu NOTCH 1
(přitornno
asi u 50% T ALL), ale i zatím
podle in vitro dat dokáží zvýšit senzitivitu na dexamethason u T ALL buněčných linií 81, 82. 1.1.2.4.
Přístup
k
pacientům
s relapsem ALL
Prognóza relapsu ALL j e závislá na 30 - 40% pacientů
83.
S vyšší efektivitou
léčby
pacienti. zařazení
Mění
několika
faktorech a
dlouhodobě
se
daří vyléčit přibližně
Prognóza relapsů je mimo jiné závislá na efektivitě front line protokolů. klesá celkový
se i skladba
pacientů
počet relapsů
a kumulují se
rizikovější
a
resistentnější
v relapsových protokolech (zpravidla je kritériem pro
do protokolu 1. relaps ALL po chemoterapii bez předchozí alogenní SeT) související
s rozšiřující se indikací alogenní transplantace v 1. kompletní remisi (Ph POZ ALL, již zmíněná
E. Mejstříková, strana 21 vysoká hladina MRN (~ 10-3) 3 měsíce od diagnózy apod.) . Nejdůležitější prognostické faktory pro úspěch léčby relapsu jsou doba od diagnózy primárního onemocnění (velmi časný méně
18
než 18
měsíců
měsíců
od zahájení
léčby, časný méně
od diagnózy a pozdní více než 6 onemocnění
místo relapsu
kombinovaný). Podle
(v kostní
dřeni
těchto tří faktorů
měsíců
měsíců
než 6
od konc
od konce
léčby), dřeň
nebo mimo kostní
léčby
a více než
imunofenotyp (BIT) a (extramedulární) nebo
v rámci proto kolu ALL REZ BFM 2002, podle
kterého se léčí pacienti i v České Republice, se pacienti rozdělují do 4 skupin Sl až S4 (rozdělení
tabulka 4). BCP ALL
T ALL
izolovaný
kombinovaný
izolovaný
izolovaný
jakýkoliv
mimodřeňový
dřeňový
dřeňový
mimodřeňový
dřeňový
S2 S2
S2 S2
S1
S1
Tabulka 4. Plehled prognostických podskupin relapsu ALL. Sl skupina Pacienti v S 1 skupině
splňují
kritéria pozdního izolovaného extramedulárního relapsu B nebo přežití
T imunofenotypu, mají nejlepší prognózu a dosahuj í použití transplantace kostní
bez události vyšší 75 % bez
dřeně.
S2 skupina U této skupiny
(zařazení
v tabulce 4) je
45%. Indikace k transplantaci kostní
pravděpodobnost přežití
dřeně
se v této
skupině řídí
v 5 letech od relapsu asi
hladinou reziduální nemoci
po dvou blocích chemoterapie. Publikace Eckertové et a!. prokázala u podle
protokolů
ALL REZ BFM 90, 95 a 96
signifikantně
lepší prognózu
pacientů léčených
pacientů
s hladinou
MRN nižší než 10- po dvou blocích chemoterapie (pravděpodobnost přežití v 6 letech od 3
relapsu skupiny s nízkou až negativní zrelabovali)
84.
MRN bylo 86%, pacienti s vysokou MRN všichni
Na základě studie Eckertové et a!. byl designován protokol ALL REZ BFM
2002, kde v podskup ině S2 transplantace kostní
dřeně
je indikována u
pacientů
s hladinou
MRN ~ 10- 3 po dvou blocích chemoterapie.
S3/S4 skupina Do této skupiny JSou
řazeni
v tabulce 4). Remise se pacientů měsíce
daří
pacienti s nejvíce rizikovým typem relapsu ALL dosáhnou u 80%
pacientů
ze skupiny S3 a asi jen u poloviny
z S4 podskupi ny, medián trvání remise je ale jen asi 8
(S4). Do S4 skupiny jsou
leukémie a jakýmkoli
dřeňovým
řazeni
pacienti s velmi
relapsem T leukémie. U
(zařazení
měsíců
časným
(S3), případně pouze 3
relapsem B prekurzorové
těchto dětí
se remise
daří
dosáhnout
E. Mejstříková, strana 22 jen asi v 50 až 60%
případů .
U
těchto pacientů,
je jednoznačně indikována transplantace kostní
pokud se
podaří
dosáhnout kompletní remise,
dřeně.
1.2. Akutní myeloidní leukémie (A ML) AML je
především onemocněním
starších
dospělých.
V
dět ské
populaci je incidence
přibližně
8 x 10-6. Analýza incidence AML podle věku ukazuje kumu laci pacientů do 2 let (12 x 10-6), dále mírný pokles incidence do 9 let věku (3 ,8 x 10-6) a dále kontinuální vzestup incidence s každým rokem věku s určitou kumulací incidence kolem 16. roku (9 x 10-6) incidence AML M3 je popisována např . v Itálii a v Latinské Americe zjištěním
o incidenci M3 podtypu u
českých dětí
je její
dlouhodobě
85.
Vyšší
Zajímavým
86. 87.
vysoké (13%) zastoupení
mezi AML, sice nedosahující hodnot výskytu u dětí v Itálii, Španělsku či Latinské Americe, ale v
významně
či
vyšší než v SRN, Skandinávii
(~6%)
Velké Británii
(Starý et aJ. , publikace
přípravě) .
léčbě
Významnou podskupinou jsou pacienti se sekundární AML po
onemocnění
nádorového
chemoterapií nebo radioterapií, problematika se překrývá se sekundárním MOS
88-95.
Z cytostatik se na vzniku sekundárních AML podílejí nejvíce inhibitory topoizomeráz a alkylační pacientů
řadíme rovněž
cytostatika. K sekundárním AML s definovaným
kongenitálním
selháním
AML vzniklé jako komplikace u dřeně
kostní
agranulocytóza, Fanconiho anémie, Diamond Blackfanova anémie,
(např.
Kostmannova
Shwachman-Diamondův
syndrom, amegakaryocytární trombocytopénie s radioulnární synostózou) republice je každý rok asi u 4
dětí ročně
nově
diagnostikován relaps
více než polovina dětí
97.
přibližně
diagnostikováno
onemocnění.
léčby
jsou
biologické chování nemoci a toxicita vlastní 98, 99.
vyšetření,
mladších 18 let a dále je
současné době přežívá dlouhodobě
zodpovědné
léčby,
jak relapsy, tak i
agresivnější
významně intenzivnější
která je
než u
Duležité pro prognózu pacienta je rovněž správné zhodnocení všech vstupních z j ména
při hraničním počtu blastů
CBFWMYH l l , zda se nejedná o u kterého je u
dětí jednoznačně
pokročilou
a
negativitě
jednotek u sporných
případů
blastů
genů
indikována alogenní transplantace kostní 100, 101.
je uveden na obrázku 7.
rekurentních genetických abnormit AMLl ! TO a libovolném procentu
fúzních
v KD.
AMLl!ETO a
formu myelodysplastického syndromu (RAEB),
intenzivní léčba není zcela jednoznačně indikována
při
V
dětí
V České
Zlepšení prognózy u dětí bylo stejně jako u ALL dosaženo pomocí
randomizovaných studií. Za selhání
ALL
10 - 18
96.
dřeně
a u kterého
Algoritmus pro odlišení těchto Při
CBF~!MYHll
pozitivitě
je pacient
již
zmíněných
léčen
jako AML
E. Mejstříková, strana 23
AM L ..................... . .
.. .
. . . . . ·0 . . . . . . . . . . . . . . . . . . . . . -
........ - _ .... - - ~ .................. - _ ...... - _ .. _
. . . . . . . . . . . . . . . . . . . . . . - . ,-
- - . . . . . . . . . . . . . . . ,- ... - - . . . . . . . . . . . .
4 -- ......... ..
t(8;21) 1(15;17)
Blasta
Inv(16)
>30%
BM
R8~t BM
WBC.>1S-20
aftor
Organomogaly
2wook.
1(0;11) _ ............... - .... - - - - - - ______ ...... - _ . - __ .. .. _ .. _ ........ .. "'_ .... ____
.. __ •• _ •• •• •• , . .... _ .......... __ .... _ .. __ • ___ .. •• _
~
_ ....... .. tI ... .. _~. ..... _
MOS
Obrázek 7. (převzato z review Hasleho et al. /(2). Schéma diagnostického algoritmu k rozlišení AML a MDS u pacientů s hraničním počtem blastLI v kostní dřeni. Schéma nepřímo demonstruje blízkost obou onemocnění, zejména pN absencí typických cytogenetických změn.
1.2.1. Prognostické faktory a klasifikace AML Morfologická klasifikace Na rozdíl od ALL, kde morfologie
blastů
kromě
L3 podtypu prakticky nekoreluje
s imunologickým nálezem, u A ML má morfologické rozdělení stále své významné místo Nejznámějším
a dosud používaným
rozdělením
103.
je tzv. F AB klasifikace (Proposals for the
c\assification of the acute leukaemias. French-American-British (F AB) co-operative group) 104,
která hodnotí morfol ogickou a cytochemickou zralost leukémie (MO - s minimální
diferenciací, Ml - bez vyzrávání, M2 - s vyzráváním, M3 - promyelocytární, M4 myelomonocytární, MS - monocytární, M6 - erytroleukémie, M7 - megakaryoblastická). Podtyp M2 koreluje s genem
přítomností
C B F~/MY H 11.
Down. U
dětí
fúzního genu AMLlIETO. Podtyp M4eo koreluje s fúznÍm
Podtyp M7 specificky v
dětském věku
s AML M7 a bez Downova syndromu
nacházíme u malých
nejčastějšÍ
MDS
106.
dětí
bez m. Down se
částečně překrývá
s m.
strukturální abnormitou je
t(l;22)(p13 ;q13), která je asociována s velmi nízkým mediánem věku (4,2 měsíce) M7 u starších
dětí
s problematikou
105.
pokročilých
M6 a forem
U podtypu M4 a MS často nacházíme přestavby MLL genu. Později FAB
klasifikace zahrnula i hodnocení imunofenotypu
107,
zejména pro definici podtypu AML M7 a
MO. Současná klasifikace W HO IOB hodnotí nález podle přítomnosti cytogenetických abnormit (t(8 ;21)(q22;q22) (AMLlIETO), t(lS;17)(q22 ;qI2) (PMLIRARa), inv(l6) AML s abnormitami
11q23
s multilineámí dysplázií,
lokusu, komplexní karyotyp),
rovněž předchozí
podle
(CBF~/MYHll),
morfologie (AML
F AB kategorie jsou ponechány pro cytogeneticky
E. Mejstříková, strana 24 neklasifikované AML) a podle geneze vzniku (sekundární AML po onemocnění
nebo po
předchozím
léčbě
pro jiné nádorové
myelodysplastickém syndromu).
Imun ologická klasifikace AML
1.2.1.1.
Dosud nejrozsáhlejší studie zabývající se hodnocením korelace morfologie s imunofenotypem a prognostickým významem exprese Creutzigové z roku 1995 antigenů.
některých antigenů
u
dětské
Tato práce nenašla prognostický význam žádného z vyšetřených
109.
Imunofenotyp koreluje
dobře
se specifickými genotypovými podskupinami AML,
jakými jsou například AML 1/ETO, CBF~/MYH I I a PMLlRARo. 1.2.1.2.
AML je práce profesorky
změny II
Genotypová kJasifikace, epigenetické
110.
AML a jejich význam pro
terapii Genotypová klasifikace v
sobě
odráží heterogenitu AML. V
hypotéza definuje tzv. leukemickou kmenovou V této leukemické kmenové
buňce
jako
buňku
se schopností sebeobnovy.
jsou nakumulované genetické
diferenciaci a proliferaci . Tyto genetické skupin
buňku
současné době často zmiňovaná
změny
změny ovlivňující
současných
podle
jejich
dělíme
znalostí
do 3
IIi:
a) mutace typu I indukují proliferaci a mutace
(např .
vytvářejí růstovou
bez
FLT3/ITD resp. FLT3 mutace, c-KIT, RAS, PTPNll, JAK2) .
b) mutace typu II blokují myeloidní diferenciaci a dávají (např.
buňkám
výhodu proti
buňkám
schopnost sebeobnovy
AML l, CEBPo., WTI , PML-RARo.)
c) mutace v genech zapojených do kontroly
buněčného
cyklu a apoptózy
(např.
NPMl
(nukleofosmin), TP53) často
Mutace typu I a II
nacházíme u
pacientů
s AML
hypotézu vzniku AML pomocí dvou zásahů do genomu V
současné době
změny
jsou
Acetylace exprese
změny,
genů,
exprese
jsou pro
které nejsou
typicky
histonů
léčbu pacientů důležité důsledkem změny
posttranslační
modifikace
zpravidla vede k expresi
genů,
současně,
což
částečně
podporuje
112 .
tzv. epigenetické
změny.
sekvence DNA, ale histonů,
Epigenetické
ovlivňují
regulaci
metylace cytosinových bazí.
deacetylace vede zpravidla k utlumení
gen ů .
1.2.1.2.1. Př ítomnost
A ML s fúzním genem AMLl/ETO, CBFplMYHll a PML/RARa fúznÍch
genů
AMLlIETO a
CBF~/MYH Il
je spojena s lepší prognózou
onemocnění a celkem tyto dvě translokace nacházíme asi u 7 až 8% dětí s AML
113.
E. Mejstříková, strana 25 součástí
Gen AMLl kóduje protein CBFa2, který je zodpovědného
za regulaci velkého
CBF (core binding factor) heterodimeru
počtu genů . S oučástí
CBF komplexu je
rovněž CBF~.
Fúzní protein způsobí špatnou funkci CBF a nepřímo tak represi velkého počtu genů leukémie jsou někdy označovány jako tzv. core binding fac tor leukémie jsou v řadě studií dřeně
řazeny
obecně
do lepšího rizika a
115.
114.
Tyto
Tyto leukémie
se u nich neindikuje transplantace kostní
v první kompletní remisi. U AML s fúzním genem AMLl/ETO nacházíme
imunologicky
často
heterogenitou
blastů,
diferenciaci leukemických
a typicky aberantní expresi CD 19,
TdT a intra-CD79a. Tyto pozitivity lymfoidnich rozpakům blastů
11 6. 117,
která koreluje s morfologickou
někdy
antigenů
nacházíme i pozitivitu intra-
mohou
někdy
vést k diagnostickým
ale podle našich zkušeností jsou často přítomny jen v menší subpopulaci
a zpravidla ani
Imunohistochemicky je PAX5
buněk,
nevedou k překročení AHL skóre podle EGIL klasifikace.
často
u tohoto podtypu prokazatelný B lymfoidní
transkripční
faktor
AML s fúzním genem CBF~/MYH ll typicky spadají do morfologické
1/8-120.
podskupiny AML M4eo, imunologicky je prokazatelná diferenciace pozitivitou aberantní CD2 a CD7,
rovněž
je
zřetelná
blastů
s častou
i podle imunofenotypu atypická
eosinofilie. Přítomnost
fúzního genu PMLlRARa velmi léčena
leukémie je dnes
(především
koreluje s AML M3 podtypem a tato
spolu s klasickou chemoterapií derivátem all- trans retinové kyseliny
(ATRA), která indukuje diferenciaci blastů promyelocytům
těsně
Imunofenotypově blasty odpovídají částečně
J21 .
(pozitivita CDl 17, CD1 5, jasná CD33),
časté
jsou aberantní exprese CD2
u variantního M3 podtypu) a CD19. Typická je negativita HLA DR a CD34.
Tento podtyp AML je dnes nejlépe léčitelný
122 .
AMLIMDS u pacientů s konstitutivní trizómií 21 (m. Down)
1.2.1.2.2.
Zvláštní kapitolu
představuje
s Downovým syndromem. U AML a MDS
123 .
AML u
pacientů
těchto pacientů
s konstitutivní trizómií 21. chromozómu -
je známa vyšší incidence leukémií zejména
AMLlMDS u pacientů s Downovým syndromem mladších 4 let
charakteristický fenotyp (AML M7)
124 .
Pacienti s touto formou leukémie mají výbornou
prognózu, častěji jsou ale ohroženi toxicitou léčby, je tedy snaha u nich léčbu redukovat 126.
má
125.
I přes redukci intenzity léčby je prognóza těchto pacientů velmi dobrá. Tyto leukémie
mají specifický imunofenotyp s kombinovanou diferenciací do megakaryocytární a erytroidní linie
127.
Typicky prokazujeme aberantní expresi CD7 a CD56, typicky je dále pozitivní
molekula CD4 na CD33 pozitivních blastech. je pozitivita trombocytárních
antigen ů
Součástí
diferenciace do megakaryocytární linie
CD41, CD42 a CD61. Pro diferenciaci do erytroidní
E. Mejstříková, strana 26 svědčí
linie
pozitivita molekuly CD36 a CD71 (zpravid la s vyšší intenzitou exprese CD71 než
mají proliferující buňky obecně) . Nemaligní CD33 PoZ myeloidní buňky pacientů s m. Down v regenerující fázi kostní pacientů
dřeně významně
s m. Down versus 4%±1 % u
exprimují aberantní molekulu CD56 (71 %±6% u
pacientů
bez Downova syndromu a
léčených
pro AML
/28).
Vysoká exprese CD56 O\l"CAM) korelovala v této studii s vysokou expresí genu RUNXl
128,
který je lokalizovaný na dlouhém raménku 21. chromozómu (21 q22.3). Tuto odlišnost
těchto pacientů paměti při
v regeneraci kostní
hodnocení MRN u
dřeně
pacientů
v podmínkách stresové hematopoézy je nutné mít na s m. Down. Biologie AML u
starších 4 let se již podobá více sporadické formě tohoto onemocnění
pacientů
s m.Down
129.
AML s přestavbami genu MLL (l1q23)
1.2.1.2.3.
AML s přestavbami 11 q23 jsou časté spíše u menších dětí či
klasifikovány jako podtypy M4 monocytární diferenciaci. Tomu
M5 dle FAB klasifikace,
většinou
Morfologicky jsou zpravidla
/3 0.
čili
mají
naznačenou či
odpovídá i imunofenotyp. Navíc
úplnou
často aberantně
exprimují NG2 (molekulu chondroitin sulfátu korelující s přestavbami MLL genu). Vztah k prognóze je v
l i teratuře
rozporný,
současný
rizika jen na podkladě přestavby 11q23 v,
, /32
Iepsl prognozu
názor je nestratifikovat tyto pacienty do horšího
131 , V
některých studiích dokonce tato skupina měla
.
AML s FL T3 interní tandemovou duplikací (ITD) nebo
1.2.1.2.4.
aktivační
mutací
FLT3(D835) AML s FLT3 mutacemi odpovídaj í horší prognóze a s horší prognózou jsou spojeny signalizace spuštěné přes FLT3
především
/ 33. /34 .
II
dětí
jsou
přítomny
asi v 15%
případů,
mutace vedoucí ke konstitutivní aktivaci
V protokolu AML BFM 2004 jsou tito pacienti řazeni
do vyššího rizika.
1.2.1.3.
Kli nické a biologické p rojevy AML,
Mezi projevy nemoci nacházíme hematopoézy, tak i projevy velmi agresivní intenzivní
onemocnění
léčby.
příznaky zmíněné
způsobené
primárně
extramedulární leukémie
u ALL, vyplývající jak z útlaku nemaligní
zmnožením leukemických
buněk. Obecně
je AML
nejčasnější
zahájení
vyžadující rychlé stanovení diagnózy a co
Mezi faktory
ovlivňující časnou
diseminovaná intravaskulární koagulopatie, která leukostáza. Mezi faktory spojené s koagulopatiÍ
prognózu může
být
patří
koagulopatie, zejména
poměrně
patří především
rychle fatální a
M3 , M4, M5 podtyp,
E. Mejstříková, strana 27 hyperleukocytóza a infekce
98, 99 .
(kůže ,
leukemické infiltráty
Při diagnóze onemocnění někdy nacházíme extramedulární Vzácně
CNS apod .).
j sou di agnostikovány AML s primární
extramedulární infiltrací bez infiltrace kostní dřeně - tzv. myelosarkom kůži,
infiltrace postihuje tkáně,
prakticky ale
být infil trováno jakékoli místo v těle
CNS, orbita atd .). Pacienti mohou být na ro zdíl od dřeň
kostní
může
být
někdy
dětí
se
s AML
primárně
se
průtoková
obtížná,
vést rychle ke správné diagnóze. U
pacientů
často
zpravidla v lepším klinickém stavu a
nádoru. Interpretace biopsie může
může
často
primárně
nejčastější
mohla být
se nachází myelomonocytární pacientů
bylo
většina pacientů
Pro vzácnost myelosarkomu je
/ 35, / 36.
o prognóze, z retrospektivních dat se ukazovala horší prognóza, která
způsobena
odkladem
léčby
nebo nižší intenzitou
špatné diagnózy (např. za záměnou za NHL) s myelosarkomem být zcela 1.2.1.4.
původ
cytometrie v těchto situacích
identifikovanou abnormitou byl nález translokace t(9; 11) a (8 ;21),
dělat závěry
infiltrující
Cytogeneticky, resp. molekulárně geneticky
135, /36.
neměla ale zachycenou žádnou odchylku v genotypu
obtížné
(měkké
myslí na jiný
diferenciace, ze studie profesora Reinhardta a profesorky Creutzigové 73% podle FAB klasifikováno jako MS podtyp
Nejčastěji
135.
léčeni
v rámci AML
/35.
léčby nejčastěji
na
podkladě
V současné době by měli pacienti
protokolů.
Farmakogenetika AML
Podobně
jako u ALL nacházíme individuální odchylky v citlivosti leukemických
toxických
účincích
tkáně
na normální
u jednotlivých
cytostatik.
buněk
Podkladem jsou
polymorfismy nejrůznějších genů podílejících se mimo jiné na metabolizmu těchto látek Studie CCG prokázaly horší výsledky pravděpodobným
léčby
u
dětí
glutathion S-transferázy theta 1 (GSTT 1) a
např .
častějších
Princip
léčby
AML, perspektivní nová
byl publikován vztah ztráty funkce
s použitím cytosin arabinosidu a
léčebné
AML v
Evropě
skupiny
antracyklinů
neindikuje v iniciální
léčiva
např.
počtu bloků
Většina
chemoterapie
skupin zabývajících se
alogenní transplantaci kostní
dřeně,
jiné
COG (Children 's Oncology Group, USA) indikují alogenní
transplantaci v první remisi u všech genem AMLlIETO a
(obrázek 8).
léčbě
přežitím
/ 38-140.
AML je velmi intenzivní a je postavena na limitovaném
l éčbou
černošského původu ,
toxických komplikací a horším
v porovnání s pacienty s alespoň jednou funkční alelou
Léčba
/ 37.
podkladem je rozdílná farmakogenomika. Dosud byly publikovány spíše
jednotlivé polymorfismy a jejich vztah k prognóze,
1.2.1.5.
hispánského a
a v
pacientů kromě pacientů
CBF~/MYHl1.
Velkým problémem
s nejlepší prognózou s fúznÍm
léčby
AML je velký
počet
smrtí
E. Mejstříková, strana 28 v souvislosti s
léčbou
(tzv. "treatment related deaths ') a nabízí se hledání specifické
biologické léčby s menším počtem nežádoucích účinků v
současné době
zahrnující pro
dospělé
celkově
léčba
malý
Velký počet nových léků je
/3 7.
především
vyvíjen a testován, ale jejich testování probíhá pacienty.
Rozšíření těchto
počet pacientů. Podobně
léku do
léčeb ných protokolů
jako u ALL hraje v
l éčbě
část lé čeb ných
subklinického postižení CNS chemoterapií,
pro
jako u ALL je
zřetelný
1.2.1.5.1.
Léčba
Příkladem
již používané
buněk.
ozařování
studií u
pacientů
léčby
V
se specifi ckým
pacientů
účinkem
na myeloidní
97, 141.
Nicméně
buňky
je použití ATRy
s fúznÍm genem PMLlRARa, které vede k diferenciaci
současné době
se u tohoto podtypu AML
s indukční léčbou postavenou pouze na A Tře a na AS20} podtypu v budoucnu dosáhnout dlouhodobého současné době řeší
indikuje i
CNS .
začíná
dospělých používat oxid arsenitý (As 2 0 }) v kombinaci s ATRou
v
je obtížné
AML M3 s fúzním genem PMLIRARa
(aH trans retinoic acid) u leukemických
ústup od preventi vního
děti
AML významnou roli
radioterapii CNS (např. protokoly BFM s výjimkou pacientů mladších 1 roku) podobně
ve studiích
144.
vyléčení
v rámci randomizovaných studií. U
zatím
/42, /43.
především
u
Probíhají studie
Otázka, zda je možné u tohoto
bez intenzivní chemoterapie, se dětí
jsou zkušenosti s As 2 0 } ještě
více limitované, v ČR zatím byl léčen pouze jeden dětský pacient s relapsem AML M3, který na této l éčen
léčbě
nedosáhl molekulární remise a vyvinul dalŠÍ hematologický relaps, který byl
intenzivní chemoterapií a alogenní transplantací.
Imunoterapie AML
1.2.1.5.2.
Molekula CD33 je exprimována prakticky na všech blastech u AML. Anti- CD33
léčba
s konjugovaným cytostatikem calicheamicinem gemtuzumab ozogamicin (GO) byla původně
designována pro
chemoterapii.
Postupně
léčbu
starších
pacientů
je tento lék používán již v iniciální
cytostatiky, zatím především u dospělých pacientů s tímto lékem u dětí jak jednotlivě/ 46.
J
s AML, netolerujících standardní
147,
145.
léčbě
v kombinaci s ostatními
Jsou už publikovány i zkušenosti /48.
U dětí
selhání standardních
bloků
tak i v kombinaci s ostatními cytostatiky
zatím 00 spíše využíván jako záchranná
léčba
při
chemoterapie. Nevýhodou tohoto léku je možná komplikace ve formě venookluzivní nemoci (VOD)
/49, /50
a opožděné rekonstituce počtu trombocytů
/5/.
Za VOD pravděpodobně stojí
interakce 00 s Kupfferovými buňkami jater, které jsou rovněž CD33
poz
.
Při použití
analogického konjugátu calicheamicinu s jinou protilátkou (anti-CD22) se sice rekonstituce
trombocytů,
ale VOD zatím nebyla pozorována.
Pravděpodobné
opožďuje
jsou i
určité
E. Mejstříková, strana 29 imunopatologické mechanismy, vzhledem k publikované GO intravenózními imunoglobuliny odpověď 153.
/52 .
úspěšné léčbě
trombocytopénie po
U AML M3 je rovněž dokumentována i excelentní
na imunoterapii pomocí jiné anti-CD33 protilátky i bez konjugovaného cytostatika
Na AML blastech se dále nachází i řada dalších vhodných cílů pro imunoterapii (např.
anti-CD52, anti-CD45 apod.) leukemické kmenové
buňky,
je
Důležitou otázkou, pokud lze akceptovat hypotézu
/54-/ 56.
přítomnost
cílové struktury
právě
na této populaci
buněk.
se, že molekula CD33 je přítomna i na populaci leukemických kmenových buněk
/57, /58.
Jiným vhodným cílem na leukemických kmenových buňkách se zdá být molekula CD44
Zvýšená kinázová aktivita hraje významnou roli v patogenezi AML či
f1t3). Tuto aktivitu lze
například
/6/.
Ph Poz ALL
současné době
přes
c
/60 )
(viz léčba BCRJABUoZ
zapojovat do standardních
/62, 163.
Snahou
léčebných protokolů , podobně
jako u
je důl ežité i vyřešit vhodné načasování této léčby v kontextu ostatní chemoterapie
jeden lék
nemusí být rovněž
signalizace
Jako perspektivní se jeví rovněž inhibice fit3 pomocí např. PKC412
je tyto léky v
(např.
(např.
blokovat imatinibem nebo tzv. duálním inhibitorem
dasatinibem Oe prokázána i účinnost u pacientů s mutací C-KIT ALL)
/59.
Inhibitory kináz
1.2.1.5.3.
kit
Zdá
může
fungovat jako senzitizátor pro lék druhý, v
vůbec účinná, někdy m ůže
být
vhodnější
léky dávat
opačném případě současně).
nutné zm ínit blokátory farnesyl transferázy, které vedou k
kombinace
Na tomto
blokádě
RAS
místě
je
signalizační
dráhy.
epigenetických Jak již bylo deacetylace
ovlivnění
Inhibitory /tiston deacetyláz, demetylující látky, terapeutické
1.2.1.5.4.
změn
u leukemickÝc/1
zmíněno
histonů.
buněk
změn
regulujících expresI genu Je acetylace resp.
zmíněných
tzv."core binding leukémií" (AML lIETO nebo
výše, jedna ze
Typicky u již
C B F ~/MYH l 1 ) může léčba
inh ibitory histon deacetyláz (HDAC)
PMLlRARa pozitivní AML indukovat diferenciaci a apoptózu je kyselina valproová, používaná v humánní
medicíně
usnadňuje zařaz ní tohoto léku do léčby leukémií
/65.
/64.
podobně
jako A TRA u
Nejznámějším inhibitorem
jako antiepileptikum, což
významně
Dalším známým inhibitorem HDAC je
trichostatin nebo depsipeptide /66, /6 7, postupně jsou vyvíjeny i nové inhibitory
/68.
je již
zmíněná
mety lace, resp. demetyla ce cytozinových bazí v promotorových CpG oblastech. V
současné
Dalším mechanismem, kterým jsou regulovány exprese jednotlivých
době
genů,
jsou dostupné léky, které jsou schopné demetylovat metylové skupiny na cytozinech.
Hypermetylace nebo aberantní metylace jsou častou změnou u AMLlMDS
/69- 171 .
Tyto léky
E. Mejstříková, strana 30 jsou
primárně
určené
deoxycytidine (decitabine) dřeni,
snižuje
1. 2.1. 5.5.
potřebu
léčbu
pro
MDS v
dospělém
veku a
patří
sem
např.
S-aza-2'-
Léčba vede k lepšímu vyzrávání jednotlivých linií v kostní
1 72.
počet infekčních
transfúzí a redukuje
komplikací.
Nová cytostatika
Rezistence na cytosin arabinosid je jedním z proto je snaha vyvíjet nové analogy
nukleotid ů
2-chlorodeoxyadenosin (c1adribine)
ak
buněčné
selhání
léčby
AML, a
nukleových kyselin.
podobně
trifosfát a je tak inkorporován do DNA buněčného dělení
nejd ůležitějších faktorů
jako cytosin arabinosid je fosforylován na
během
postupně
její syntézy,
zástavě
vede k
smrti. Dokáže indukovat apoptózu u neproliferujících
buněk
je rezistentní vůči inaktivaci deaminací. Je prokázána většÍ senzitivita podtypu F AB MS zařazen
Y rámci protokolu AML BFM 2004 je cladribine kterého jsou
řazeni
Clofarabine je
pro
léčbu
173.
AML vysokého rizika, do
i pacienti s AML MS podtypem.
podobně
jako cladribine analog
vlastnosti t1udarabinu a cladribinu. U
pacientů
jen částečná odpověď na toto cytostatikum
nukleosidů
a kombinuje farmakokinetické
s relapsem/refrakterní AML byla prokázána
174.
1.2.1.5.6.
Odpověď
na léčbu A ML, terapeutické ovlivnění leukemické kmenové buňky
1.2.1.5.6.1.
Odpověď
na
léčbu
definovaná morfologicky
Na rozdíl od ALL jsou pacienti s AML
léčeni podstatně agresivnější
toxicitou až úmrtností v souvislosti s terapií. S
agresivnější
léčebných protokolů
blastů
chemoterapie. Y protokolech BFM skupiny je pacient s
po 1.
indukčním
četností blastů
je prognóza
obecně
kteří
stratifikace blocích
>5% po 1. bloku
remise v klasickém smyslu
dobrá.
genetické a cytogenetické cíle detekce MRN
Detekce minimální reziduální nemoci nemá na rozdíl od ALL standard. Jen u malé
Léčebná
Podobně jako u ALL většina pacientů dosáhne kompletní remise
přesto
Molekulárně
rychle upadá
po jednom až dvou
bloku, výjimkou jsou pacienti s AML M3 ,
slova často nedosáhnou, 1.2.1.5.6.2.
175.
většina pacientů
BFM98 a 2004 je na obrázku 8).
v jednotlivých protokolech zahrnuje hodnocení
zařazen do vyššího rizika
chemoterapií s vyšší
chemoterapií souvisí i nemožnost
sledování bližší dynamiky poklesu leukémie, protože naprostá do aplázie (schéma
a
části pacientů
lze použít
molekulárně
obecně přijímaný
genetický cíl
(např.
zlatý
AML 1/ETO,
CBF~/MYH11, PMLlRARa, přestavby 11 q23 s identifikovaným fúznÍm partnerem)
/76-/78
E. Mejstříková, strana 31 Kromě pacientů
remise ve smyslu úpravy pozitivitou je indikována pacientů
s několikaletou
v kompletní remisi časovém
přesně
s fúzním genem PMLlRARa není
179.
léčebné
léčba
180.
pacientů
se znovuobjevenou PMLlRARa
pomocí A TRy, resp. AS203 . Naopak jsou publikovány
pozitivitou
fúzních
genů
AMLlIETO
nebo
případy
CBF~/ MYHl1
Nejen z našich zkušeností spíše tato pozitivita vedla v různém
intervalu k relapsu
podmínkou vyléčení
strategie. U
definován význam molekulární
onemocnění
a dosažení molekulární remise je zpravidla
Publikované práce s perzistující pozitivitou AMLl /ETO ne vždy
dostatečně dlouho sledovaly pozitivní pacienty, jej ichž MRN může přetrvávat i řadu let
Zprvu slibný marker AML, gen WT -1 , lze použít ke sledování jen s obtížemi poměrně slabě exprimován ve srovnání s nemaligním pozadím
181-183.
179.
často
je
Mutace v transkripčním
faktoru GA T A-I jsou typické pro myeloproliferace charakteristické pro Downův syndrom (TMO a AML-M7) a lze je využít jako případné cíle pro sledování MRN rovněž
MRN lze
u
některých
pacientů
1.2.1.5.6.3.
Princip detekce MRN cytometricky
v ývoj fyziologické myeloidní především
antigeny
řady
asynchronně
je
změnou
s vhodnou genetickou
cytogenetické zejména fluorescenční in si tu hybridizaci (FISH) II
podstatně méně
184-186.
Pro detekci
použít metody
187
AML imunologicky definovaný. Zajímají nás
exprimované (antigeny asociované s progenitory a antigeny
charakteristicky exprimované na zralejších myeloidních
buňkách, např.
exprese myeloidního
antigenu C0 15 na CD34 PoZ blastech), antigeny aberantně exprimované (např. C019, C056, CD7, C02 nebo N G2) nebo hyperexprese buňkám.
či
snížená exprese
Pro AML je typická imunofenotypová heterogenita
detekovat více kombinacemi protilátek srovnáním imunofenotypu AML u
188, 189
dětí při
potřeba
diagnóze a 190.
a je zpravidla nutné MRN
při
relapsu, že
změnu
v expresi
alespoň
Je tedy zřejmé, že v rámci detekce AML
hledat univerzální odchylky leukemických
ohledu na iniciální imunofenotyp pacienta.
blastů
oproti fyziologickým
(příloha 4). Langebrake et al. rovněž prokázala
jednoho antigenu má 88% sledovaných pacientů je
antigenů
Průtoková
buněk
od nemaligní kostní
cytometrie se na rozdíl od
dřeně
bez
molekulárně
genetických metod nabízí jako metodika dostupná pro prakticky všechny pacienty. Jak u dospělých
191,
tak u dětí v regenerující KO nacházíme nOlmální prekurzorové buňky splňující
kritéria leukemického imunofenotypu v některých časových bodech až v řádu procent
/92.
V rámci mezinárodní studie, kterou koordinovali kolegové z Německa, jsme prokázali prognostický význam reziduální nemoci u AML pouze v univariantních analýzách, nikoli v rámci multivariantní analýzy zahrnující všechna kritéria použitá pro stratifikaci pacientů Dosud publikované studie se zabývaly sledováním MRN
průtokovou
l89
.
cytometrií u AML
E. Mejstříková, strana 32 především u dospělých pacientů
studie u
dětí
cytometrie
ukazující na horší prognózu
/97- /99.
I přes vzácnost AML u dětí byly publikovány dětí
s prokazatelnou MRN podle
buněk
a "nestandardní chování" regenerujících nemaligních
které mohou exprimovat i molekuly považované za asynchronní
Proto je vhodné definovat normy bud' v jednotliv)fch časových bodech
192,
' ., , h pre k urzorec h exprese antigenu, k tere' neprok azuJeme na regeneruJlclc •
průtokové
Interpretaci MRN u AML komplikují dvě okolností: již zmíněná nestabilita
imunofenotypu leukemických prekurzorů,
/78. /9/ . /93-/96.
o
či
aberantní.
nebo nalézt vzorce 200-202 .
Stu d'le van
Rhenen et a!. se zabývala především detekcí imunofenotypových odlišností CD34pozCD381leg leukemických buněk a CD34pozCD381leg prekurzorů. V rámci fyziologické kostní dřeně jsou CD34pozCD381lcg Lin lleg prekurzory považované za nejvíce kmenové a s největším potenciálem repopulace kmenovou
203 . 204. 205.
Analogicky u AML je tato subpopulace blastů považována za nejvíce Tato subpopulace může z celé populace blastů tvořit variabilní část, častěji
tento imunofenotyp je přítomen u nediferencovan)fch AML (AML MO)
206.
E. Mejstříková, strana 33
AMLBFM98 Inlen
Inducbon
I I
IDIl
EJ B EJ
HAM
AM L-BFM 2004 InduIdIon 1
haM1
ISR I 0
I
1
B
}
I lIRil HAM I~ BB KlIP
. PJ.
.
..
Tag 1 MAD
16 liRO
42 ~
21-28 MRO
li RO
Doku
!)(mJ
88 MRO
-140
Doku
DoIaJ
Ddw
RiIIDon'i~ ARA.c ; ~n ~ kSaIůIId n
t E:
2.cOA: HA:
HocI'Ido!IIs CytaIabjn
-.._.,..-M:
Mllolcarůon
mltteIhocl'do!lertes ~rabin
Obrázek 8.
Léčebný
Doku
• bel PaIIaúIn Dit malelu"an MIRer
HocIrislkogl\ClP8 Mlrímaf residua! diseaS8 KMP:
DoIur
ha:
MRl>"
Sratill2lerurv HodOos1s CyQr.IbI~ E~I Stlndanfr1sltogn.ope
DatIlln.tlldn; L.(JNR
EIDpasId-Phospha 2.QDo.2-deaoxy;iderD5ln
Erhaltungstheraple 1 Jatlr Cytarabln I.th. 1..2
-112 liRO"
fOr KIn:Ier mll ANI.. bel Down Syndn!mlAMI.. FAR M3 beachEn! 21xJWoche 4 WochBn IGArĎI1I der D~
Ox:
18.Jni1_ J
~~/I-------
1~
® A:.
12 GY' I
~
KnochBrmar1q)1DIon obIgatlnscll! DoIamer'Qtion des The!1Ipleelemenls
protokol AML BFM 98 a 2004 (AML BFM 98 schéma
převzato
z publikace 141, AML BFM 2004 převzato z protokolu). Je patrné, že léčba ANIL je již od začátku velmi intenzivní, takže většina pacientů velmi rychle upadá do aplázie. Časné
hodnocení odpovědi v den 15 slouží jako přídatné kritérium pro stanovení rizika t:-5% vysoké riziko).
1.2.2.
Přístup
k
pacientům
s rel apsem AML a k
primárně
vysoce rizikové
AML Jak již bylo pacien tů .
zmíněno
Obecně
výše,
různé
pracovní skupiny
zařazují
alogenní SCT pro rozdílnou
panuje shoda neindikovat SCT v první kompletní remisi u
část
pacientů
s příznivou cytogenetikou (AMLlIETOPoz, CBF~/MYHll POZ). Přístup BFM skupiny, podle jejíchž protokolů se léčí i děti v ČR, je velmi rezervovaný k SCT v první kompletní remisi u
E. Mejstříková, strana 34 de novo AML a modalitou.
obecně
Obecně
není indikována. V
případě
prognóza relapsu závisí na
relapsu AML je SCT
něko lika
faktorech, jedním z
délka první remise «1 rok, > 1 rok). Další významnou podskupinou k SCT jsou pacienti se sekundární AML, jak na tak v souvislosti s pro jiné nádorové
léčbou
pro jiné nádorové
onemocnění
lze
rozdělit
pod kl adě
onemocnění.
pro
předcházet
něž
jsou typické
změny
léčbě
dospělými obecně hůře
je
indikovaných dřeně, léčbě
inhibitory topoizomeráz,
(1 až 3 roky) a typicky nesou přestavby MLL později
(4 až 6 let po primární
na 7. a 5. chromozómu, fáze sekundárního MDS
sekundární AML. Nová
v porovnání s
pacientů
Pacienti s AML vzniklou po
genu, a na AML po alkylujících cytostaticích, které vznikají léčbě) ,
nejdůležitějších
kongenitálního selhání kostní
na AML vzni kl é po
léčby
které mají krátký odstup vzniku od této
důležitou léčebnou
léčiva včetně
dostupná. Již
může
monoklonálních protilátek jsou u
zmíněná
konjugovaná s calicheamicinem) a chemoterapie je u
dětí
dětí
kombinovaná imuno (anti-CD33, indikována v
případě
rezistentní
nemoci nebo relapsu. V Č R byla tato léčba použita dosud u 5 dětí s AML (2 pacienti s rezistentním relapsem, 3 pacienti s relapsem po SCT). umožnila SCT (ve 2
případech
Léčba
byla
úspěšná
pacientů
u 3
a
šlo o retransplantaci) .
2. Cíl e 1)
Je mezi myeloidními antigeny exprimovanými u ALL souvislost? Dá se předpokládat
obecná
příčina
exprese myeloidních
antigenů
u ALL? Má
některý
myeloidní antigen prognostický význam u ALL? Pokud ano, je tato exprese d ů l ež itější antigenů
2)
než exprese jakéhokoli myeloidního antigenu? Je exprese myeloidních
stabilní?
Jaký je prognostický význam
diferenciačního
antigenu CD10 u B prekurzorové
AL L? 3)
Jaký je význam nových kritérií definovaných podle MRN? Význam cytometrie v monitorování
4)
časné léčebné odpovědi .
Význam MRN u transplantovaných B
řady
specificitu MRN podle četnost ,
průtokové
dětí
přestaveb
s ALL.
Ovlivňuje
rekonstituce nemaligní
imunoglobulinových
genů ?
5)
Zj istit
biologický charakter a prognózu sekundární ALL po
6)
Prognostický význam 4barevné cytometrie v monitorování MRN u
léčbě
dětí
ALL.
s AML.
E. Mejstříková, strana 35
3. Výsledky a Diskuse 3. 1. Imunologická diagnostika ALL 3.1.1. ALL z
prekurzo rů
B
I ym focytů
(BCP ALL)
B prekurzorové ALL imunologicky členíme na proS, common a praeB
23
podle exprese
antigenu CD I 0 a intracelulárního IgM. Funkce molekuly CDI O na leukemických
a na
fyziologických prekurzorech není známa. Prognosticky nepříznivý význam má především podtyp proB ALL (CDI0 negativní a intra-IgM negativní), který koreluje s častý
MLL (llq23) a je
u
kojenců.
přestavbami
genu
Naopak v podstatě žádný prognostický význam nemá
klasifikace podle pozitivity intracelulámího IgM na podskupinu praeB a cALL (Obrázek 9). Data jak z dospělých , tak i
dětských
ALL navíc ukazují, že ani pro korelaci s přestavbami
genu MLL není intracelulámí expres IgM důležitá 207, 208 .
CuIllUlaIIl.' e
P f~tion
SurvUII \9
C Ql1\plel u
r
( ~M~ )
c ....~t.'1:I
1."" r - - -- - - -- - ----,
~: ~ 0 , 00
~ o...
-
~
-:f.. ofo: J
cALL
o." O. l ~
0.70
'---:':--:' - . - ,-
o... '-:.. 0-
'---=7:--': ---'
--''o
R""v
Obrázek 9. Obrázek ukazuje stejné pl~ežití u pacientů diagnostikovaných jako cALL nebo praeB ALL. V ana(vze jsou zahrnuti všichni pacienti s námi určenou diagnosou cALL nebo praeB ALL v letech 1999 až 2006, n=356, p =ns.
Na obrázku lOje exprese v
antigenu
pravděpodobnosti
znázorněna
na
CD10 PoZ
CP ALL klasifikovaných do dvou skupin podle
(2:20%)
a
CDlOneg,
v přež ití. Exprese antigenu CD I0 i u
pozitivní, se u jednotlivých blastů
analýza
pacientů
mohou
významně
ukazující pacientů ,
signifikantní
rozdíl
které hodnotíme jako
lišit (od pozitivity
s většino u bla tLl negativních až po typickou hyperexpresi všech
hraničního počtu
buněk,
obrázek ll).
Proto jsme se rozhodli testovat hypotézu, že i pacienti s cALLlpraeB ALL s významným podílem CD lOneg blastů mají horší prognózu.
E. Mejstříková, strana 36
IatJ\l. A oportlOn Swvtt.r rl'iJ { ~"--Mttor )
CanpIaI III
+
Cenaot ed
CD1 0ne g (n=41) 0,5
p = 0.0086
n
4~----
____________________
~
10
Signtfikantně
Obrázek 10.
,~~l."
12
horší přežití pacientů s negativitou CDi 0«20%).
'OOOO J=========;;====~
cAll l proS 8ubklon (60%)
proB ALL
..:":'~.,
. . ' .( t : '"
l!"':.'~
f 21
>~
10 .J---,-----~.,.;----4
'00
o
100
' 000
10000
10.01
'0
'~ . C020FlTC
'00
' 000
Lot•. C02ll-FrlC
10000
Obrázek 11. Různá exprese antigenu CDiO při diagnóze BCP ALL, obrázek uprostřed ukazuje typického pacienta s proB subklonem a vpravo je pacientka s typickou hyperexpresí CDIO. ~.
A"0P0!1KJJ1
,o 0.9
O,.
"' :ti
~L
C4..mAtu... ~s...vMťQ fKliplan-""" Cc"llet. t em.urod
Sl.LJ"'J I\I'-'; (K.apWI.~ )
~.
r
CMl&Gred
p(3 skupiny)=O.0025
~
'-4' ,
o.e
.~~ .
12\~1".:~~
(11
tI I i";--'~i'" I I I 1 I ,I I
l o.
"'
~
!.li
I
o
•
4
• Ro<
•
'0
-
MLUAF.
-
O' 0.4
-
zcel. CD1 0+
12
0.7
0.6
proS ni Cll
,""oB 0, 4
L..-._ __ _ _ _ _ __ _--I pt'Q8 (nebo LMtatrM!i proSl
Obrázek 12. Vlevo: Rozdělení pacientů podle hladiny exprese CDiO (zcela CDiO+ jsou pacienti s méně než 20% CDi(feR blasty, proB ALL exprimují CDi O na méně než 20% blaSltt, (~ástečně proB exprimují CDiO (?20%), ale současně obsahují významný podí! (?20%) CDi017eg blastů, všichni MLL AF4po= pacienti měli expresi CDiO nižší než 20% a nejsou zahrnuti do křivky proB. Vpravo: Stejná analýza se spojenou skupinou proB a částečné proB. Signi fi kantně
horší prognóza klasicky definované proB ALL, ukazovaná na obrázku 10, je do
značné míry dána nepřízni vou MLLlAF4
poZ
ALL; oddělíme-Ii MLLlAF4 poZ ALL, přestane být
rozdíl mezi proB a ostatními ALL významný. Námi definovaná podskupina
"částečně
proB"
E. Mejstříková, strana 37 má ale podobnou prognózu jako proB ALL (obrázek 12)
Populační
analýza tedy ukazuje, že
významná přítomnost CDlOneg blastů určuje špatnou prognózu - bez ohledu na to, zda jsou přítomny
i C D 1Opoz
buňky.
3.1.2. Aberantní exprese
antigenů
Jak již bylo zmíněno v Úvodu, leukémie napodobují svým imunofenotypem své nemaligní protějšky
v kostní
leukemické
dřeni,
resp. v thymu. Expresi
buňky, označujeme
AML neznáme. V a) aberantní
zásadě
varianty AML)
z jiné linie,než z které vycházejí
Přesnou příčinu
jako aberantní.
aberantní exprese u ALL a
exi stuj í 2 základní hypotézy:
exprese jsou (např.
leukemogeneze
antigenů
důsledkem
deregulované
v
důsledku
mechanismus aberantní exprese CD2 u mikrogranulámí
209.
b) aberantní exprese Je dllsledkem nezralosti leukemických odpovídat stádiu fyziologických linií
genů
exprese
prekurzorů
buněk
nebo
s potenciálem diferencovat do
může
různých
2/0.
Exprese myeloidních horší prognózy.
antigenů
Zároveň
u ALL byla považována na
základě
prvních studií za marker
známou podskupinou exprimující myeloidní antigeny u
jsou pacienti s fúzním genem TEL/AMLl
2// .
2/2
dětí
s ALL
asociovaným s dobrou prognózou.
Prognostický význam exprese
těchto antigenů
k protichůdným závěrům
2/2-2/ 7.
Slabé místo většiny těchto studií bylo, že všechny myeloidní
antigeny by ly brány jako
sobě
si rovné a zpravidla pozitivita kteréhokoli antigenu pacienta
v jednotlivých studiích vedl v
průběhu
let
klasifikovala j ako tzv. MyAg pozitivního. V situaci, kdy neznáme biologický podklad aberantní exprese některých
antigenů,
není správné je v analýzách hodnotit
studií byla analýza B a T ALL
společně
společně.
Další slabinou
nebo krátká doba sledování, která
nezachytila pacienty s pozdním relapsem A LL. Souhrn charakteristi k jednotlivých studií je uveden v tabulce,
předložené
v rámci recenzního
V rámci protokolu BFM 95 Jsme prokázali s aberantní expresí molekuly CD33 v v multivariantní analýze v
příloze
příloze
řízení
k
příloze
signifikantně
1 (tabulka na
straně
81).
horší prognózu BCP ALL
1. Tuto horší prognózu jsme potvrdili i
1. Horší prognózu jsme potvrdili i u
pacientů léčených
v další studii (ALL IC BFM 2002) s kratší dobou sledování (BCP ALL: medián 2,9 let, minimum 0,09, maximum 4,99 let) (obrázek 13).
E. Mejstříková, strana 38
Cumulallve Proportion Su",ving (Kaplan-Meier) Complete
T Censored
1,00
i '--ť " \ . '--t, "
0,95
-:-"1:
+--
.~
,.tT-i(
,f::
> 0,85
~., ,'>-:t,l--'t,, 1----t-11+1
'~
~ ~
,1:::1:: I ::::1 :,;,:1:
'1
10
li
J-;.;f7t
li
0,90
J~I+:-+-:t-"
0,80
~
Cl.
4i'" 0,75 et:
-CD33>10% - , CD33< 10%
0,70
0,65
0,60
°
2
Obrázek J3. než 10%
Zřetelně
horší přežití bez relapsu u pacientLI s BCP ALL a expresí CD33 na více
blastů léčených v
neg
rámci protokolu ALL JC BFM 2002
(Cox-Mantelův
test, p=0,0204,
D33 neg 90%±3,4%, pro CD33po~ 77%±7,2%, pEFS ve 4 letech pro
pRFS ve 4 letech pro CD33
6
4
roky
88%±3,5% pro CD33po~ 74%±7,2%). Výsledek nezávisle potvrzuje naše zjištění
v příloze 1.
3.1.3. Akutní hybridn í leu kémie Leukémie se signifikantním vzhledem hybridní. Podle typu je lze koexpresí
antigenů
členit
blastů,
jak z lymfoidní, tak i myeloidní linie, nazýváme
na leukémie s
primárně
zjiné linie, dále na leukémie s
definovanou linií a významnou
dvěma liniově různými
definované imunologicky a/nebo morfologicky, a nakonec na leukémie, které změní svůj
fenotyp do jiné linie
před
CD3, lehké
řetězce
zároveň
alespoň
dvou ze
antigenů:
CD19,
imunoglobulinu kapa a lambda. T ALL je definována pozitivitou splněním alespoň
kritérií: intraMPO Poz , CD 13 Poz , CD33 P0 Z, CD6S PoZ a/nebo CDl17 příp,
tří
musí být negativní všechny tyto molekuly: intra CD3,
(intra)CD3 a pozitivitou CD7. AML je definována
(intracelulární,
signifikantně
dosažením kompletní remise. BCP ALL je definována
podle adaptované EGrL klasifikace pozitivitou (intra)CD79a a/nebo, CD22,
populacemi
PoZ
dvou následujících a současně úplnou
i povrchovou) negativitou CD3, CD79a a CD22,
Imunologicky jsou AHL s primárně definovatelnou linií a signifikantní koexpresí antigenů z j iné linie
nejčastějším
klasifikaci
/6,
podtypem AHL a pro její definici lze
např.
použít již
zmíněnou
EGlL
ve které jednotliv)/m antigenům je přiřazeno skóre podle jejich příslušnosti
E. Mejstříková, strana 39 k lymfoidní, resp. myeloidní linii (tabulka 5, viz též
příJoba 12).
V
případě překročení
skóre
nad 2 je nález klasifikován jako AHL.
skóre
Břada
2 I 0.5
(intra)CD79a, intra CD22 CD 19, COlO, CD20 intraTdT, CD24
T řada
myeloidní linie
IgM, (intra)CD3, TCRap, TCRy8 intraMPO CD2, CD5, CD8, CDI0 CDl3, CD33 CD65, CDl17 CDl4, CD15, CD64 intraTdT, CD7, CD la
Tabulka 5. Skóre jednotlivých antigenů pro diagnostiku AHL. Splní-li blasty definici jedné z řad ALL a současně překročí myeloidní skóre 2 nebo splní-li definici AML a překročí-li skóre B nebo T řady 2, je nález klasifikován jako AHL. Pro porovnání uvádím alternativní klasifikaci AL a AHL, která je používána v St. lude's Hospital v Memphisu (USA) a definuje jednotlivé podtypy takto
/0:
B-lineage My+ALL splňují všechna následující kritéria: poZ pOZ a) Blasty jsou CD19 a (CD22 nebo intra-CD79a POl. nebo intra-IgM (fl řetězec» neg b) Blasty jsou intra-CD3 c) Blasty jsou intra-MPOneg d) Blasty exprimuj í j eden nebo více myeloidních
antigenů
(CDI3 CDI5, CD33 nebo
CD65)
T-Iineage My+ALL splňují všechna následující kritéria: poz a) Blasty jsou CD7 PoZ a intra-CD3 neg b) Blasty jsou CD22 c) Blasty jsou intra-MPOneg d) Blasty exprimují jeden nebo více myeloidních
antigenů
(CDI3 , CD15, CD33 nebo
CD65)
LyPOZ AML
splňují
všechna následující kritéria:
a) Blasty jsou intra-MPOPoz (nebo je pozitivní nespecifická esteráza u AML M5) neg b) Blasty jsou intra-CD3
c) Blasty j sou intra-IgM(fl řetězect
eg
a neexprimují současně CD22 a intra-CD79a
d) Blasty exprimuj í jeden a více z následujících antigenů: C 02, CD5, CD7, CD19, CD22 , CD56 nebo intra-CD79a.
Bifen otypické leukémie
splňuj í jedno
z následuj ících kritérií
a) Myeloid/B - Eneage bifenotypové leukémie koexprimují intra-MPO a CD22 spolu s CD 19 nebo intra-CD79a. b) Myeloid/T - lineage bifenotypové leukémie koexprimují intra-MPO a intra-CD3
E. Mejstříková, strana 40 c) Mixed BIT lineage bifenotypové leukémie koexprimují intra-CD3 a intra-IgM(1l řetězec)
nebo intra-CD3 spolu s intra-CD79a.
V rámci diagnostiky považujeme za vyjádřila
nejsou v
důležité,
aby osoba hodnotící cytometrický nález se
k primární linii a které znaky hodnotí jako aberantní. EGIL ani jiné klasifikace současné době
zcela vyhovující a v druhé
polovině
WHO klasifikace hematologických malignit, kde bude
roku 2008 je
očekávána
část věnována
nová
i hybridním
leukémiím. Většina doposud publikovaných studií hodnotila bud' dospělé pacienty nebo kohorty dětí dohromady s dospělými zohledněn
pokud
léčby
typ
striktně
nejčastější
Poměrně málo byl v těchto studiích
219-221
a její výsledky na celkovou prognózu. Z našich dat je čistě
aplikujeme EGIL klasifikaci na
podskupinou definovanou
218,
zřejmé ,
že
pediatrickou kohortu, zdaleka
imuno-genotypově
jsou pacienti s fúzním genem
TEL!AMLl (viz tabulka 8) vzhledem k častému výskytu tohoto podtypu, ale není rozdíl signifikantní oproti
skupině
klasifikované jako non AHL. TEL! AML 1 genotyp je známý
svoji afinitou exprimovat myeloidní antigeny k překročení AHL skóre,
přestože
211.
212a
tato koexprese může vést až průběhu
se jinak
jedná o typické ALL (i když je jej ich prognóza signifikantně horší - obrázek 14).
Podobně
z pohledu diagnostiky a klinického
jako v předchozích studiích jsme potvrdili vyšší incidenci u AHL genotypu BCRlABL a MLL! AF4 (tabulka 6) a s tím související celkovou horší prognózu
dětí
klasifikovaných
jako AHL. V analýze v rámci jednotlivých imunolgenotypových podskupin (obrázek 14) prokazujeme
signifikantně
horší
přežití
u
dětí
klasifikovaných jako AHL v rámci
TEL! AML 1 a MLL! AF4 podskupiny. Jen pro úplnost retrospektivního hodnocení AHL u dětí,
v uvedeném
AHL
primárně
lymfocytárních
rozpětí (září
1996 - srpen 2006): 4 další
AML s lymfoidními znaky. Všechny 4 znaků,
děti
jsme klasifikovali jako
děti přesáhly
prognóza této podskupiny byla
extrémně
AHL skóre expresí T špatná (v 1. roce od
diagnózy žil a dosud žije pouze 1 pacient po alogenní transplantaci od HLA identického sourozence), u žádného z
těchto dětí
jsme neprokázali klonální
přestavby
Ig/TCR.
E. Mejstříková, strana 41
AHL 18% 32% 4% 9,4%
TA LL TEL/AML1 Hyperdiploidní BCRlA BL MLL/AF4 Ostatní BCP ALL
NonAHL 14% 24 % 21% 2,6% 1,3% 35%
10,7%
21%
p hodnota n.s. n.s. p=O,026 p=O,044 p=OOO98 n.s
Tabulka 6. Cetnost jednotlivých imunologických a molekulárně genetických podskupin u pacientů s AHL podle skóre EGIL a u nehybridních ALL. Statistická významnost rozdilu v četnosti jednotlivých podskupin je hodnocena Fisherovým exaktním testem. Červeně znázorněny signifikantní rozdily v zastoupení AL klasifikovaných jako AHL.
~.~ ~7 =~1~\
10
~1)=28
0.6 0.5
0 .9
-non AHL (AM L)
0 .8
-NI.. (AMLJ\..y+)
0.6 0.7
;.....;;;;.;:;....----
0.5 04
1::=88
0 .4
03
n=4
0.3
0 .2
p = 0.00339
02
0.1 10
P = n.s.
0. 1
OO~--~--------------~
0. 0~0:-~--"""----;;---;:-~10---'!"2
12
roky
"
Ol
..... Ol
1
=71 t'I.
,.;t;l
n=~
07
"
" ""
- Ilon AHL (T ALL) - AHL (T AI..1../t.\'+) p =n.s.
~2 n=9
- non AHL (hyr>erdlp'IOld ALL)
AtL (hypériIIpIoIMA)
n=1
- non AHL (TEUAML1' AlL)
- AHl (1EJAM..1"'JMy+) p = 0.00492
non AHL
MLUAF' ~
p = not relevant
ALL)
- AHL (MUIAF4"'"IMy+)
Nf
I
t:~tOd&~ .
~AH
AR._~._"",,)
n=J
p=O.OOO8O 10
p=n.s.
1
,1
(J
10
12
Obrázek 14. Nahoře vlevo: Křivky přežití ukazují signifikantní rozdí! (Cox-Manlelův test) mezi skupinou AHL 0=28 a non AHL 0=557. Nahoře vpravo: V rámci AML nevidíme signifikantní rozdíl. Dole: Pokud hodnotíme přežití v jednotlivých imuno-genotypových podskupinách, signifikantní rozdí! v přežití nacházíme v podskupinách MLLlAF4 a TELlAMLI.
E. Mejstříková, strana 42
3.1.4. leukémie s ná lezem časné
léč by před
fáze
blastů
různých
z
linií a liniový
přesmyk během
dosažením kompletní remise.
Přesmyk
z jedné linie do druhé vyžaduj ící reklasifikaci leukémie nebo nález dvou
populací
při
jednoznačně zařadit
diagnóze leukémie s nemožností nález obecně
respektive myeloidní leukémii, jsou
lze najít spíše jednotlivá kasuistická
léčbě
v
sdělení
oddělených
jako lymfoidní,
leukémií velmi vzácnou situací.
Literárně
než epidemiologická data. Leukémie s nálezem
více různých populací blastů například shrnul Weir et al
222.
V jednotlivých kasuistikách je
přesmyk zjedné linie do druhé popisován např. u ALLlAML s přestavbou MLL genu
akutních leukém ií s monozómii 7. chromozómu BCR/ABL
228.
do myelomonocytární leukémie a v relapsu
stejnými klonálními
přestavbami
19-TCR bez prokázané blastů
plasticita fenotypu leukemických
na
počátku
u
nebo u pacientů fúzním genem
226. 227
Bierings et al. popsala případ pacientky s přesmykem
indukční léčby
223-225,
zpět
229
do BCP ALL se stále
přestavby
léčby
z BCP ALL během
MLL genu.
by mohla
zásadně
Případná
ovlivnit
spolehlivost diagnostiky, zejména imunologické. Proto jsme si položili otázku, zda k přesmyku nedochází hraje
častost
častěj i
předpokláda lo.
část pacientů
leukémie
expresního profilování
230,
řady genů,
transkripci
před
přesmyku
linie
dosažením kompletní
nakonec dosáhne kompletní remise, a fenotypový
proběhne během indukční léčby
obecně mění během léčby
bez povšimnutí. Leukemické což lze identifikovat jak na
tak i na úrovni imunofenotypových změn
pravděpodobně přímo způsobují podané léky např.
Velkou roli v identifikaci
a precisnost sledování dynamiky nádorové nálože
remise. Je totiž možné, že přesmyk
než se
232.
23/.
buňky základě
Část těchto změn
Typicky tyto změny u BCP ALL postihují
tyto antigeny: pokles intenzity až vymizení pozitivity CD10, TdT (lze pozorovat i u T
ALL) , vzestup exprese - CD20, CD45 .
Zm ěny
vyzrávání během nemaligního vývoje B řady
v expresi
233-235.
těchto antigenů částečně
kopírují
Tyto změny zpravidla nevedou ke změně
klasifikace vlastní leukémie, jsou však velmi důležité především pro vlastní detekci a správnou interpretaci MRN. V rámci protokolu ALL IC BFM 2002 bylo jedním z zodpovědět
otázku, zda
časná odpověď
nejdůležitějších
v den 8 a den 15 ukazuje prognózu
výzkumných
cHů
pacientů (přílohy
7 a 11). Součástí protokolu tak byla i centrální evaluace dne 8 a dne 15 morfologicky, imunologicky i
molekulárně
v Praze a v Olomouci.
geneticky. Morfologické
elkem u 4
dětí
vyšetření
bylo
s BCP ALL (charakteristika
prováděno centrálně
pacientů
je v tabulce 3)
s BCP ALL (3x uniformní infiltrace KD blasty, u pacienta DH signifikantní koexistence BCP ALL
blastů
(57%) a myelomonocytárních
blastů
(15%) jsme identifikovali v den 8
E. Mejstříková, strana 43 signifikantní
změnu
lymfocytárních
fenotypu (pacienti PM, LD a DK)
antigenů,
myeloperoxidázy). V natolik, že v V těchto
tyto
CD33 , CD 14 a intracelulární buňky změnily svůj
při
diagnóze. Artefakt četností
způsobený příměsí
leukemických
přestaveb
neobjasněným
dosud
mechanismem
dokonce i mikroskopický vzhled natolik, že buňky.
Tuto
skutečnost
jsme pozorovali
Incidence mezi pacienty se špatnou
připomínají
hlavně
odpovědí
dětí
u
přestavby
s původním
nechtěných buněk
v "monocytech", jednak
kontrolou (nemaligní T lymfocyty), která byla negativní. Na buňky
počátku léčby
změnit
rozporných
výsledků
nemaligní myeloidní se špatnou
či
odpovědí
monocytámí
na prednison.
na prednison a BCP ALL na protokolu ALL
stanovení
blastů
procenta
mezI
může
aberantní antigeny.
ukazuje dosud očekáváme
nedoceněnou
P řesmyk
pozdějším
z lymfoidní linie do myeloidní linie u
plasticitu leukemických
buněk
shodné dosud nepoznané genetické pozadí.
a
alespoň
u
Je být
molekulární přesmykem
spojovala již při diagnose BCP ALL signifikantní exprese molekuly CD2, která jinak časté
vnitřní
nejen imunofenotyp, ale
genetikou, imunofenotypizací a morfologií. Všechny pacienty s
mezi
jsme
mohou tedy
BFM 2002 (1112002 - 10/2007) byla 19%. Ukázali jsme, že tento biologický jev příčinou
fenotyp
imunologicky a morfologicky vypadaly jako normální monocyty.
jednak vysokou
1eukemické
k myeloidní linii (pokles B
jsme ale sortováním prokázali identické klonální
Iymfoidním klonem vyloučili
antigenů včetně
průběhu indukční léčby postupně
podstatě
buňkách
vzestup myeloidních
směrem
nepatří
části pacientů
části těchto pacientů
E. Mejstříková, strana 44
Iniciály
rmuno fenotypl Morfologie
Ivěk(r)
OH 12
PM 115 LO 117 OK 15 IS /1 6
Identické klon. přestavby
cALL (60%) a myelomono (40%) pro8 L I , aberantní CO2 cALL Ll aberantni CO2 cA LL L I aberantnl CO2 cALL LI abcrantní CO2
Tabulka 7.
Přehled
Leukoc yty 3 (10 1 /lL)
19-TCR ve srovnání s leukemick ými blasty mye lomono: ano ( s ouča s ně i FL T3ITO+)
Genotypové zm ěn y
Prednisono vá odpověď
Změna léčby po standardním protokolu ALL
Prognóza
léčby
190
FLT3 IT O delece ll q23 včetně MLL
špatná
lnterfant, SCT
Relaps ALL po SCT
mono ano
15
nena lezeny
špatná
Interfant, SCT
CRI 2,5r
mono ano
6,9
nenalezeny
špatná
lnterfant, SCT
CR 1,2r
mono ano
45
nenalezeny
dobrá
O
CRl8m
"sek"AML ano
138
nenalezeny
špatná
smrt v relapsu
sek AML 6m po dg, smrt v progresi
základních charakteristik pacientlí s přesmykem
směrem
k myeloidní linii
v rámci protokolu ALL Je BFM 2002, pro úplnost uvedena i pacientka JS z retrospektivní
analýzy protokolu ALL BFM 95.
3.1.5. Imunologická b uň c e.
Anti-CD33
protilátek v
rovněž
eventuální specifickou
léčených
léčba ,
léč bě dětské
lmunofenotypizace
prognóze
diagnostika
pacientů
na
leukemické
ALL informace o
případných
molekulárních cílech pro
- zejména monoklonálními protilátkami. Vzhledem k horší
s aberantní expresí CD33, kterou jsme prokázali na
kohortě pacientů
v rámci protokolu ALL BFM 95, se nabízí otázka, zda by se u této podskupiny
nedala využít cílená anti-CD33 terapie protilátkou
anti-CD33
(klon
(Příloha
hp67.6)
většina
leukemických
testováno v monoterapii u
buněk
pacientů
1). Terapie humanizovanou monoklonální
konjugovanou
(gemtuzumab ozogamicin - 00) je používána naprostá
cíllJ
perspektiva využití dalších monoklonálních
při náší
léčbu
terapeutických
především
s cytostatikem u
dospělých pacientů
exprimuje molekulu CD33.
netolerujících
calicheamicinem
Původně
konvenční indukční léčbu
s AML, kde bylo
léčivo
AML (zejména
pacienti starší 65 let s významnými komorbiditami znemožňujícími klasickou intenzivní chemoterapii). V současné době již byly publikovány a probíhají další studie kombinující 00 s další chemoterapií u dospělých pacientů s AML. U dětí byly publikovány zkušenosti s 00 v terapii jak u AML
/46,
tak i u ALL
236-238.
Studie Zwaana et al. prokázala signifikantně nižší
E. Mejstříková, strana 45 hodnotu LC so samotného calicheamicinu in vitro u pacientských
vzo rk ů
s ALL oproti AML, u
kterých Ooemansová et a!. prokázala významno u jnter-individ uální variabilitu v citlivosti na calicheamicin
239.
Zdá se, že určitou roli v odpovědi na tento lék u AML (u ALL nebylo
dosud zkoumáno) hrají i polymorfismy molekuly CD33
240.
Širšímu použití u ALL brání
v současnosti omezení na CD33 poZ ALL a literárně počet publikovaných pacientů nepřesahuje maximálně několik
Naše zkušenosti s s refrakterním
málo desítek
léčbou
onemocněním ,
hladině nejméně
pacientů léčen
zpravidla s relapsem, resp. refrakterní nemocí.
monoklonálními proti látkami u ALL jsou omezené na pacienty
monoklonální protilátky byla na
pacientů,
resp . relapsem po SCT. U všech ověřena
pozitivita
příslušné
pacientů před
podáním
molekuly (u všech byla pozitivita léčbě
ALL podána u 6
částečnou
expresí CD33 byl
40%). Celkem by la monoklonální protilátka v
s relapsem ALL. Pacient s 2. relapsem BCP ALL a s
kombinací anti-CD33 (00) s chemoterapi í bez efektu, pacient s refrakterním velmi
časným
1. relapsem praeT ALL byl
léčen
anti-CD33 (00) bez efektu , pacient s 1. relapsem
sekundární AHL (proB/My+) po alogenní transplantaci dosáhl kompletní remise a kompletní dárcovské krvetvorby po dvou dávkách samotné anti-CD33, dva pacienti s 1. relapsem BCP ALL byli
léčeni
kombinací anti-CD52 (Campath) a chemoterapií pouze s
částečným
efektem
bez dosažení molekulární remise a pacient s 1. relapsem BCP ALL s vysokou pozitivitou CD20 na reziduálních leukemických blastech byl
léčen
anti-CD20 (rituximab) spolu
s chemoterapií a dosáhl molekulární remise.
I cALL lécba ca; ného relapsu I
lCOf-oJ
. ,.
I cALL lécba
pozdního relapsu
I
t------+-~___::::...;I
-'---_
_
_
___
.....
--L-~
Obrázek 15. Ukázka monitorování léčebného efektu kombinace chemoterapie a anti-CD52 (Campath TM). MRN byla detekována pomocí kvantitativního sledování klonálních přestaveb 19-TCR a pomocí 8-barevné průtokové cytometrie (CD58 FITC/CD66c PE/CDIO ECD/CD45PerCp/CD34 APC/CD20 PB/CD38 A 700). U pacienta s čas ným relapsem jsme při pozitivitě MRN podle průtokové cytometrie před alogenní transplantací prokázali selekci CD52 negativních blastů (ověřeno klonem s jiným cílovým epitopem než má anti-CD52 v Campathu). DH2-5, Db2, Vd2, Vbl- 19/TCRjsou cíle pro sledování MRN
E. Mejstříková, strana 46 V současné
době
se již do klinické praxe dostává humanizovaná anti-CD22 protilátka
konjugovaná s calicheamicinem - inotuzumab ozogamicin. První perspektivní výsledky již byly ukázány u dospělých pacientů s Iymfomy a na zv ířecím modelu ALL
24/,242.
Perspektiva
tohoto léku u ALL je veliká, prakticky všechny BCP ALL tuto molekulu exprimují. Podobný je i mechanismus fungování jako u gemtuzumab ozogam icinu, po navázání protilátky dochází k internalizaci komplexu, v lysozymu dojde k cílové
buňky
oddělení
podobným mechanismem jako ionizující
bez redukce a/nebo
indukční
do
léčebných
léčby
změny
ostatních
léků
a nelze tak
243,
očekávat
do indukční léčby relapsu ALL
244.
protokolu pro recidivy před evším
z
léčby
polyrezistentní leukemické
Léčba
době
zapojení
samostatně
bez
léčebných protokolů přidávat snadněji
přidává
anti-CD22 protilátky (epratuzumab)
přímo
Podobná strategie se v současné době zvažuje i v novém
dětské
Převažující
ALL v rámci BFM skupiny.
zkušenosti
refrakterní ALL, kde již zpravidla testujeme efekt na
buňky.
Za zmínku stojí
ještě občasné
protilátky OKT-3 (anti-CD3) v léčbě refrakterního relapsu T ALL
3. 2.
v dohledné
anti-CD20 rituximab, anti-CD52 campath). V současné
protokol COG pro relaps BCP ALL
jsou tedy
navíc ke standardní protokolární
protokolech je vyšší než 95%. Monoklonální protilátky fungující
(např. anti-CD22 epratuzumab
léčebném
Toxicita jak gemtuzumab tak i
ALL, když procento dosažených kompletních remisí v uznávaných
navázaného cytostatika lze zpravidla do stávajících
době
záření. přidat
inotuzumab ozogamicinu není zanedbatelná a nelze je léčbě
cytostatika, které poškozuje DNA
použití monoklonální
245.
leukémie, monitorování účinnosti léčby
3.2.1. Hodnocení minimální rezid uální nemoci Základními metodikami pro sledování MRN je kvantitativní sledování a
průtoková
postupně
zapojuje do
současnosti
léčebných protokolů
RQ PCR pro Ig/TCR. Reziduální nemoc
nejen
dětské
pomocí PCR je tak definován nový pojem - molekulární remise. kteří
dosáhnou v rámci
protokolů
léčbu
ALL. Díky sledování MRN Obecně
BFM molekulární remise po prvním
pacienti s velmi dobrou prognózou odpovídají na ., 79 nere Ia bUJI .
Ig/TCR
cytometrie (detailní charakteristika metodik je rozvedena dále). Zlatým
standardem pro monitorování ALL je v se
přestaveb genů
246.
pacienti s BCP ALL,
indukčním
bloku, jsou
Pacienti sT ALL mají jiný charakter odpovědi, obecně
pomaleji, platí ale, že pacienti negativní
tři měsíce
od diagnózy prakticky
E. Mejstříková, strana 47 V
současné době
méně
vyspělých
rozšířená
řadu
je sledování MRN pomocí RQ PC R Ig/TCR pro
průtoková
zemí metodikou nedostupnou. Naproti tomu
v zemích
standardizaci cytometrie,
zázemím.
zejména ekonomicky cytometrie je
Vzhledem
k
nedostatečné
s horším
ekonomickým
především
její interpretace, jsou všechny dosavadní cytometrické
studie MRN omezené bud' na jednu instituci nebo laboratoř 39. 69. 247-249.
3.2.1.1.
Imunofenotypová detekce MRN u ALL
lmunofenotyp ALL prekurzorů.
blastů
odpovídá více
či méně
protějšky prekurzorů
Nemaligní
imunofenotypu
Iymfocytů
T
normálně
se fyziologicky vyskytují pouze dřeni
v thymu, pro sledování reziduální nemoci v periferní krvi a kostní teoreticky
mělo stačit
se vyvíjejících
detekovat blasty podle imunofenotypu nezralých T
u T ALL by
prekurzorů
intracelulárního TdT, negativita povrchové C03 v rámci C07pozC05poZ T lymfocytů je známá typická hyperexprese molekuly C099 u významné části T ALL nemaligní B prekurzory jsou ve variabilním 13.
počtu
25/.
250.
Dále
Naproti tomu
dřeni
prakticky vždy v kostní
(exprese
zastoupeny
Známý je tzv. rebound fenomén, kdy nacházíme v regenerující kostní dřeni B prekurzory
např.
po chemoterapii až v desítkách procent. Hledání
rozdílů
v imunofenotypu
nemaligních B prekurzorů je zásadní problém reziduální nemoci u BCP ALL Pro odlišení leukemických B
prekurzorů
typických vlastností leukemických
od nemaligních
protějšků
blastů
od
250.
využíváme
několik
blastů:
snížená nebo naopak zvýšená exprese molekuly C038, hyperexprese CO lO,
některých
molekul
(např.
snížená exprese až negativita
0 58)
exprese aberantních molekul zjiné linie než Iymfoidní
(např.
exprese C066c, C033, C013,
C0 15) exprese molekul v rámci nOlmální hematopoézy se
vůbec
nevyskytující (NG2 - molekula
chondroitin sulfátu, která se exprimuje typicky u ALL i AML s
přestavbami
asynchronní exprese molekul (např. intracelulární exprese TdT na C034 C02l na C034 PoZ blastech) Je nutné zmínit i
změny
kdy významnou
část
molekul, které
při
imunofenotypu u
terapie
tvoří
dětské
23/ . 232.
253.
při
diagnóze jako
MLL genu)
blastech, pozitivita
V rámci indukce ALL,
časté potlačení
diagnóze hodnotíme jako hyperexprimované
(např . sníženě
exprese
některých
COlO, C034) nebo exprimované
(např.
Mezi diagnózou a relapsem často nacházíme nestabilitu exprese aberantních
molekul, jako je např. C0 33, C013 a CD 15 stabilitu
během léčby.
kortikosteroidy, je
naopak upregulace molekul, které hodnotíme C045)
ALL
0eg
252,
naopak molekula C066c ukazuje významnou
Prokázali jsme ale stabilitu myeloidního antigenu C033 v časné fázi léčby
252.
E. Mejstříková, strana 48 V
současné době
se
umožňující
16),
přístupem
postupně
stává dostupná 9 a více-barevná cytometrie
analýzu více než I I
parametrů
zvýšení senzitivity a specificity
na j ednotlivé
průtokové
(příklad
na obrázku
buňce. Očekáváme
s tímto
časových
bodech
cytometrie zejména v
s regenerací v KD.
10000
lrukem"k~
...
c.. <'> '00
8U 10
., 5;;
.
,.
O>
u
U,"',
C058 FITC
'''''''''
' 000
>-
g
C?
.....
lť , ..
o
o
o
'.' "
10000. , - - - - - - - - - ,
1.0 1000
o
"~~.~
I. ...
,r..x'O
o
C? lť ,..
p-
BPt'rrorr .. ~.
,""'" o
cr> ~DOO
.~
"f
' W
bb.
U
" 10
.
,
CD20 PB
Ul'"
'OOM
M
o
U to
'" HJO
W OQ
C058 FITC
, OO\X)
10
I DO
' 000
1001!0
C058 FITC
Obrázek 16a. MRN podle pr/'Itokové cytometrie u pacienta s vysoce rizikovou B prekurzorovou ALL 3 měsíce od zahájení léčby. Pacient s iniciální hyperleukocytózou až 200000 WBC/ uL, dobrou odpovědí na prednisom a více než 25% blastů (M3) v KD v den i5, což pacienta kval(fikovalo do vysokého rizika. Anal)/za kombinace CD58/CD66c/CDiO/CDi9/CD45/CD34/CD38/CD2G. Všechny obrázky jsou z gatu CDiC/o: l1eg CD58 po:(be: buněk. V kostní dřeni jsou zřetelné jak nemaligní B prekurzory (CD66c po hyperexprese) CD38 : ), Lak i leukemické buňky (aberantní exprese CD66c, hyperexprese CD58, nízká až negativní CD38, vyšší intenzita CDiO oproti nemaligním B prekurzorům). Celková hladina MR N je i, 2% ze všech jaderných buněk a tato hodnota by pacienta podle nejnovějších BFM kritérií kvalifikovala pro maximální terapii vče tně alogenní SCT i od nepříbuzného dárce (pacient žije v i . kompletní remisi 22 měsíců od iniciální diagnózy na udržovací terapii dle protokolu ALL JC BFM 2002).
E. Mejstříková, strana 49
IZ
CD19PoS buněk
I
,o' , - - - - - - - - - - ,
10' , - - - - - - - - - - ,
pacient lť o
;; u
.
10"
~ 10'
'~~~r-~~~~~
"f
o
'o'
,o'
,r/'
10'
10'
B·530f30-A: C020 FITC B-53013Q-A
10 1
ul 10' FL1-H: CD20 FITC
..... C
'o' , - - - - - - - - ,
U
III všechny CD19PoS ,o'
,r/'
10'
sortované CD 19pos34++2 one g , IgH/TCR=2,1 (zcela leukemlcké)
10'
B-530f30-A: CD20 riTe B-530!JO·A
CD20 Obrázek 16b. Vysoká hladina MRN (modře) na hladině 5,8% s koexistencí nemaligní B řady (červeně) u pacienta s Pfr= ALL 180 dní po SCT Po neúspěšné indukční terapii bylo remise dosaženo až po přidání imatinibu. Pacient byl transplantován v molekulární remisi podle přestaveb 19/TCR (pacient 6 sl0 36 z přílohy 5), přesto velmi časně narostla MRN a následně nemoc zrelabovala úedná se o jediného pacienta s negativní MRN před SCT, který zrelaboval}. Pacient zemřel v léčbě relapsu na .\ystémovou mykotickou infekci. Leukemický původ blastlJ znázorněných na tomto obrázku jsme ověřili přímým sortováním a následným ověřením přítomnosti klonálních přestaveb 19/ TCR, 3.2.1.1.1.
Cytometrická reziduální nemoc v mezinárodní studii Mini-Mini
Jak již bylo
zmíněno,
všechny dosud publikované i
aktuálně
probíhající studie využívající
cytometrií jsou zpravidla omezeny na jednu laboratoř, respektive jednu . . . 39' 69, 247,-?54, -155 . Je zrejme, ze d ete k ce MRN ve stu d"II to h oto rozsah InStItUCI u 'je mozna jen
MRN
průtokovou
v'
multicentricky.
Průtoková
•
-
_o
systém, bud' z přestaveb,
materiálu pacientů
d ůvodu
nebo z při
nenalezení vhodných
důvod ů
diagnóze
obecně dostupnější
cytometrie je považována za metodiku
v zemích s horším ekonomickým zázemím. Asi u 5 až 10% dostatečně
pacientů
se
nezaslání rozhodujících
vzorků
do
logicky vzniká otázka použitelnosti klasifikace MRN podle
Dosud ale nebyla publikována
jednoznačná
nepodaří
i
zavést peR
specifických a senzitivních klonálních
logisticko - biologických (nemožnost získat
onemocnění,
•
dostatečné
množstvÍ
laboratoře).
U
průtokové
těchto
cytometrie.
kritéria pro hodnocení cytometrické MRN.
E. Mejstříková, strana 50 V rámci studie Mini Mini Jsme navrhli jednotný panel monoklonálních protilátek s použitelností prakticky pro všechny pacienty s ALL. V rámci hodnocení MRN jsme definovali široké spektrum subpopu lací rámci B
řady
iT
řady
imunofenotypové odchylky vzniklé jak v časné fázi léč by definovali podle
pacientů
z pilotní fáze
PCR a pacienti s regeneruj ící kostní a v rámci T Maďarska
řady
(děti
dřeni
V rámci B řady jsme
23/. 256-258.
a po SCT) celkem 29 subpopulací
celkem 5 subpopulací. Do studie se zapojily
laboratoře
z Chorvatska, Izraele,
a Hong Kongu. Navrhli jsme jednotné templáty pro interpretaci dat v softwaru
Cellquest (BD San Jose) a
lowJo (TreeStar, Oregon).
dřeni
situaci, kdy v kostní
dochází
postupně
Léčba
ALL
představuje
průtoková
obsáhnout. Vzhledem k tomu, že podle protokolu ALL IC BFM-2002 se medián sledování celé kohorty stále není najít v první fázi jiný zmíněno
způsob,
jak
ověři t
definovali pozadí v kostní subpopulací, které
přesáhli
RQ PCR negativních
dřeni
( příloha
platformě
hodnocení cytometrických dat. Jak již
současné době vzorků
3.2.1.2. O becně
bod a podle tzv. "cross
kombinacemi proB
specifické pro každý
časový
řadu
a naopak) jsme
bod
zvlášť.
Hodnoty
označili
nejvyšší z nich.
cytometrie bylo tedy vztaženo k
odnocení senzitivity a hladině
MRN podle RQ
definované subpopulace mohou poskytnout relevantní informaci i
rozdělí
v den
příloze
11,
standardně
význam pro plánování dalších
Molekulárně
definované
15 pacienty s BCP ALL (ale nikoliv sT ALL) podle
MRN podle citlivé PCR v den 33 a v týden 12. Tyto
závěry
mají
léčebných protokolů.
genetická detekce MRN u ALL
u ALL je možné nalézt molekulárně genetické cíle u většiny pacientů (minimálně u
80%). Jednak využíváme jak (např.
časový
jsme definovali jako MRN, pokud bylo více hodnot subpopulací
průtokové
předem
pravděpo dobnosti bezprostředn í
je detekce MRN podle RQ PCR Ig/
v daný
4barevné cytometrie. Jak rozvádím v
subpopulace
2002 a
ll ).
Zjisti li jsme, že v
října
způsob
změření
hodnoceno jako MRN, jako MRN jsme
PCR
od
trojnásobek 98. percentilu subpopulací "cross lineage" kontrol a
vzorků,
specificity MRN podle
léčí
pro korelaci s prognózou, bylo nutné
TCR. Podl e náhodné poloviny negativních lineage" kontrol (pacienti s T ALL
cytometrie schopná
dostatečný
tento
výše, zlatým standardem v
komplexní
k redukci nádorového klonu a znovuobjevení
nemaligní hematopoézy. Oba tyto dynamické aspekty je
bylo
případné
s relapsem ALL a sledovanou MRN podle RQ
průběhu l éčby
v
s cílem postihnout i
přestavby genů
Ig/TCR, u
pacientů
s
přítomným
fúzním genem
TEL/AMLl nebo BCRlABL) lze použít i sledování jejich transkriptů. Většina
E. Mejstříková, strana 51 léčebných
stratifikací v
současné době
především
vychází u ALL
z kvantitativního sledování
přestaveb Ig/TCR nebo průtokové cytometrie 247, 249, 259. přestaveb
Sledování MRN u ALL pomocí
3.2.1.2.1.
genů
imunoreceptorových
(lg/TCR)
Detekce MRN pomocí kvantifikace klonálních buněčné
receptory (TCR) vychází z
nezralých lymfoidních buněk
přestaveb genů
předpokladu,
buňky
že
pro imunoglobuliny (lg) a Tprotějšky
ALL jsou maligními
Pokud maligní zásah postihne lymfoidní prekurzor, který již
2/0.
zahájil proces V -(D)-J rekombinace, mají všechny jeho
dceřiné buňky
přechodových
specifické pro leukemický klon
oblastí podjednotek antigenových
daného pacienta. Mezi
nejčastěji
řady. Většinu
nekompletní O-J
z nich
tvoří
přestavby.
vyšetření
rutinně
přestavby
ALL (TCRG, TCRB a nekompletní přestaveb IgH u T ALL
určení
buněčného
receptoru gama
přílišnou
složitost
Iymfocytů
u B prekurzorové
TCRD) a naopak výskyt nekompletních
laboratoří)
27 PCR reakcí pro jednotlivé rodiny
26/
řetězců 26 /.
použitím DNA z diagnostického vzorku kostní dostatečnou
nejčastěji
a pro
Monoklonální produkty PCR ověřené
těchto primerů 46, 262, 263.
dřeně
a
fluorescenčně značených
MRN je kvantifikována s
jako standardu. Takto je možné v
zkušeností zavést systém pro sledování MRN až u 95%
je vždy celková
(nedostatečné
imunoglobulinů
jsou sekvenovány a na základě přechodových V -(D)-J
sond je pak zavedeno pacient-specifické RQ-PCR
i logistický
20% lze nalézt
podjednotky alfa se pro
ekvencí jsou navrženy specifické primery. Pomocí
skutečnosti
přibližně
pomocí screeningového panelu, který
(lGH) a lehkých (lGK)
pomocí analýzy heteroduplexů
s ALL, ve
ALL
přestaveb
klonality
přestavované segmenty TCRB, TCRG a TCRD
s
T
pro receptory T
přestavby
zahrnuje (podle postupu používaného naší
laboratořích
přestavby
dětských
260 .
Prvním krokem metodiky je
těžkých
V-D-J, v
řady patří
Zvláštním fenoménem u leukémií je tzv. liniová
způsobuje přestavování genů
promiskuita, která
segmentů
přestavby
U T-ALL lze sledovat
nevyšetřují.
u leukémií z B
které lze detekovat u více než 95%
kompletní
(TCRG), delta (TCRD), a beta (TCRB), tohoto
přestavby
detekované klonální
přestavby těžkých řetězců imunoglobuli nů,
zB
receptorů
stejné sekvence
úspěšnost
univerzálnost je však vykoupena velkou
finanční, časovou
nebezpečím
laboratoře
a metodickou
metody. Dalším negativem je možnost nespecifického nasedání sekvence Ig/TCR ve zdravých lymfocytech s
často
nakonec nižší, vstupuje do toho
množství materiálu, nedodání vzorku do
primerů
falešné pozitivity
pacientů
faktor
apod.). Tato
náročností
na
této
přestavěné
vyšetření.
Proto
byla Evropskou skupinou pro reziduální nemoc u ALL (ESG-MRD-ALL) navržena pravidla
E. Mejstříková , strana 52 pro vyhodnocování MRN pomocí této metody, která zahrnují mimo jiné použití multiplikátu DNA z l ymfocytů zdravých dárců jako negativní kontroly
264 .
řad y
s fa lešnou
po
transplantaci
imunoreceptorových řetězce, přestože
regenerace
dřeně
kostní
genů
( příloha
přestaveb těžkého
i lehkého
podání
kortikoidů
je spojen
významně
falešných pozitivit v systémech pro detekci MRN podle IglTCR
přestaveb
souběžně
závisí na
podávané imunosupresi:
prekurzorů
až jej ich absenci , s
při
čímž
Sledování MRN u ALL pomocífúzních genů
části pacien tů
s ALL jsme schopni identifikovat tzv. fúzní geny. Tyto fúzní geny lze
současn ými
metodikam i detekovat v ně kterých
úrovni DNA
( například
d ů vo dů
typ y
přestaveb
6).
3.2.1.2.2. U
některé
pozitivitou
vzorky byly vyhodnoceny j ako pozitivní podle kritérií ESO. Míra této
především
počet
spojená
v cílech zahrnující
nacházíme významnou redukci B nižší
je
Prokázali jsme, že regenerace B
případech
fúzní geny zahrnující gen MLL),
jak na úrovni mRNA, tak i na
některé
fúzní geny se z technických
detekuj í pouze na RNA úrovni (BCR/AB L). Prokázána je dobrá korelace hladiny
transkriptu TEL! AMLl a přestaveb Ig/TCR nemožnosti sledování specifických
cí l ů
265.
TEL! AMLl se tedy nabízí jako vhodný cíl při přestaveb
podle
Ig/TCR. Dosud není publikována
práce která by technicky a prognosticky srovnala detekci MRN podle
přestaveb
pomocí fúzního genu BCR/ABL. U korelace fúzního genu BCR/ABL s
přestavbami
je
třeba
mít na
pam ě ti ,
že fúzní gen BCR/ABL
může
být u ALL
blastického zvratu CML) i v jiných liniích než v Iymfoidní pomocí fúzních dále
c e lk ově
3.2.1.3.
Ig/TCR
přítomen (podobně
jako u
Nevýhodou sledování MRN
na RNA úrovni je možnost kontaminace s rizikem falešné pozitivity a
vyšší požadavky na kvalitu a Nepří mé
Pro úplnost je obecně
genů
266-268.
Ig/TCR a
stáří
vzorku.
sledování M RN pomocí chimérismu po alogenní SeT
třeba ještě
zmínit sledování
případných
reziduálních leukemických
buněk
u hematologických malignit, jak akutních, tak i chronických, pomocí chimérismu
s přibližnou citlivostí 10-2 . Zlatým standardem v současné době je sledování pacienta a dárce pomocí specifických STR (short tandem repeats) sekvencí nemusí vždy znamenat
nutně
269-272.
Samozřejmě signál příjemce
relaps malignity. Populaci , ve které detekujeme genotyp
příjemce, lze ozřejm it sortováním
269 .
E. Mejstříková, strana 53
3.2.2.
Léčebný
protokol ALL IC-BFM 2002 pro primární
léčbu
de novo
ALL Studie ALL IC-BFM 2002 je léčebný protokol, který byl koordinován Českou republikou a jeho vznik byl moti vován snahou zlepšit léčebné výsledky v zemích s dostatečnou zkušeností s protokoly BFM skupiny, ale bylo srovnat stratifikaci
nedostatečně
pacientů
zkušených v monitorování MRN. Jedním z cílů
podle kli nických a biologických vlastností se stratifikací
protokolu ALL-BFM 2000, ve kterém byli pacienti stratifikováni protokolů
MRN. Základní schéma obou změnila
stratifikace,
některá
předevš ím
je na obrázku 17. V ALL-BFM 2000 se
klasická kritéria, která se používala v
zásadně
předcházejících léčebných
protokolech, jako věk a leukocytóza, nebyla zahrnuta do stratifikace pacientů
7J
podle hladin
259, 273
(příloha
Tato kritéria byla nahrazena parametrem MRN podle přestaveb Ig/TCR ve dvou časových
bodech - v den 33 a týden 12 od diagnózy ALL. Tyto
časové
body pro detekci MRN byly
podloženy retrospektivní analýzou MRN publikovanou v roce 1998 246 . Pacienti s negativní reziduální nemocí v den 33 a v týdnu 12, s dobrou
BeRl ABL a MLLI AF4 jsou
řazen i
odpovědí
na prednison a bez fúzního genu
do standardního rizika. Pacienti se špatnou
odpovědí
na
prednison, s prokázaným fú znÍm genem BCRlABL nebo MLLlAF4, nedosažením remise v den 33 nebo MRN vyšší než 10-3 v týden 12 j sou řazeni do vysokého rizika. Ostatní pacienti,
včetně
těch ,
imunoreceptorových 2002 zahrnuje 2002
zařazeno
země
u kterých není možné zavést systém na sledování
genů ,
jsou
ze všech
řazeni
do
kontinentů
více j ak 4000
dětí
s
protokolu je porovnání klasických
stře dn í ho
av
nově
rizika.
současné době
Léčebný
přestaveb
protokol ALL IC-BFM
bylo do tohoto protokolu od
října
diagnostikovanou ALL. Základní otázkou tohoto
stratifikačních
kritérií s novou stratifikací
zohledňující
MRN v rámci ALL-BFM 2000. Pacienti standardního rizika v protokolu ALL IC-BFM 2002 splň uj í
následujícÍ kritéria: jsou mladší 6 let, iniciální
počet
bílých krvinek je nižší než
20000/flL, v den 33 je dosaženo kompletní remise, v den 15 v kostní blastů ,
mají dobrou
odpověď
splňují alespoň
méně
než 25 %
jednu podmínku z následujících: špatná
na prednison, nedosažení kompletní remise v den 33 , prokázaný fúzní gen
BCRlABL nebo MLLlAF4, více než 25% standardní riziko. Ostatní pacienti jsou dvě
je
na prednison a nemají prokázaný fúzní gen BCRlABL ani
MLLlAF4. Paclc:nti vysokého rizika odpověď
dřeni
řazeni
blastů
do
v den 15 u
středního
pacientů
jinak
nesplňujících
rizika. V rámci protokolu probíhají
výzkumné studie srovnávající výslednou stratifikaci v obou protokolech: studie Mini Risk
sleduje, zda pacienti standardního rizika by splnili kritéria pro zařazení do nízkého rizika i vprotokolu ALL BFM 2000, studie Mini Mini (viz kapitolu 3.2.1.1.1 , str. 4) řeší otázky
E.
Mejstříková,
detekce MRN
strana 54
průtokovou
cytometrií. Prokázali jsme, že skupina standardního rizika se liší
v obou protokolech 273(příloha 7).
ALL IC-BF~12002 : TREATl\tIENT
anall lrrdY leT d 15
dx ~ !
d33
,., (" " .pl' ,·o"d I" A.lI l1ou ,· ,,23.02.2002
SR -
w12
hIIJI~~'-1 ~}=:::2 I~ ~1flJ> I~'1 : ~
~
~nB ' !Ai
I_I__~I
r
m)
III
ic
t
I
t 6-MP I WX
t ___ i_~ _.__~_1U_._t______~
LJ_ _ _
[fl' **
.
L . . . I_ _ _ _
~.;;í::~:'I"~[~J
L..I_ MP_: _MIX --I
Do ·C
1 '~;F';-Il.!: 1=II~II--I
II
f. IIL~I
II
_ ú,_:\iP_ . MT_X
---II
I
6·M P.M'!X
§ §
tJ I':.r·
~I~'~'~I~I----~'~'--------~--------------------------~),~),____~.~ 10
(j
"' nM~
~
!'I_
12
't rr ~m: ~,
I D!\1Ut1'.....' ,z.11Il $<,," l( p;aIioal
'lot 8CP-ALL. \ fIX 1" ""'...ll.., . rOl
r
wiůollll'-All
: \ITX $;Im'1.Jln
.J
'
~2 lO~ '"
t, tr~ J ~!lC' rrtDaJ Lm!.'bnDn ' <#Irntd lb. .,~ ...'i( Tm aU ""'" " , HR I ~ _ c f AlEOP ,... /lf M bul Ú>OIC< bor ~ . ccordm!; '" 1"""_
,ndIi::l._
""",,""'" . ,I11_1IC"1IIo: h'U loKh -' .... >IUJqIb IU O"" PI
AIEOP LLA 2000 - ALL-BFNI 2000
MRD
(3)
TůnepoinlS
la
11]
l
2
lb
(4) (5)
~
strl fikace Pl entu ~--,l
"
10
I
11
10
- I KD Ptoto«>ll. PII.J:..e ,\. ,., DE.' ú\ J....p Prun>col l. PJw. .,lh PRED
Obrázek 1 7 Schéma léčebného protokolu ALL JC-BFM 2002 a AJEOP BFM 2000. Zd~lrazněny jsou časové body, v kterých byla monitorována MRN v protokolu ALL Je BFM 2002 a v kterých časových bodech se stratifikuje v léčebném protokolu ALL BFM 2000.
E.
Mejstříková,
strana 55
3.2.3. Detekce MRN po relapsu ALL a u
pacientů
indikovaných k alogenní
SeT Pacient i, kteří jsou zahrnuti republice sledována pacienti
představovali
průtokové
příloze 5
léčebná odpověď
první kohortu
jsou vůbec prvními, u kterých byla v České
274,
přestave b
podle
pacientů
pro
ověření
před
SCT (vzorek odebraný
těsně před
zahájením
je podle našich dat významným prediktorem
laboratoři
přestaveb
přípravného
přežití
Tito
senzitivity a specificity MRN podle
cytometrie, která byla v pilotní studii používána v naší
V rámci studie jsme prokázali, že negativita MRN podle
genů.
imunoreceptorových
již od roku 1996.
imunoreceptorových
genů
režimu - tzv. "conditioningu")
bez události jak v univariantní, tak i
multivariantní analýze. Do studie byli zahrnuti pacienti jak s relapsem ALL, tak i pacienti s vysoce rizikovou formou ALL, u kterých byla indikována SCT v I. kompletní remisi. Dosud jedinou publikovanou studií o prognostickém významu MRN podle cytometrie u relapsu ALL je práce Coustan - Smith et al. všichni pacienti, MRN po
vč etně pacientů
skončení
indukce a
275.
s extramedulárním relapsem,
měli
průtokové
V této studii byli hodnoceni kteří měli vyšetřenou
vhodný imunofenotyp pro sledování. Tato studie v rámci
multivariantní analýzy ukázala jako negativní prognostický faktor relaps vzniklý v léčby
a MRN
~0,01 %
pracovních skupin na
na konci indukce. Tato publikace
světě,
že
hladinu
časný
relaps v kostní
dřeni
průběhu
potvrzuje zkušenost i jiných
na
léčbě
je pouhou chemoterapií
prakticky nevyléčitelný 276. U pacientů zahrnutých v článku Šrámkové et al. byla sledována MRN rovněž průtokovou cytometrií, v letech 1996 až 2002 det kce byla kombinací protilátek, které reflektovaly
prováděna měřením
nejvhodnější
jedné nebo dvou
imunofenotypové aberace (tzv. LAIP -
Leukemia Associated Immunophenotype) odlišující maligní od nemaligních strategie je v detekci imunofenotypových
velmi
MRN
shiftů
imunofenotypové aberace
279.
se
častá
doporučuje
254,
použít
277,
278 .
pro
Pro
detekci
pokrytí MRN
je postaveno na expertním hodnocení cytometristy. Tento
komplexní
přístup
k reziduální nemoci
průtokovou
definovat, jaké populace jsou v daný
procentuálního zastoupení již
Tato
eventuálních alespoň
dvě
Rozlišení leukemických buněk od případných regenerujících
prekurzorů
přístupem
buněk.
svědčí
cytometrií,
časový
pro MRN. Od
např.
přístup neumožňuje
není možné tímto
bod normou a jaká hranice jejich
září
2002 jsme zahájili detekci MRN
v rámci studie Mini Mini v protokolu ALL IC-BFM 2002 podle jednotného panelu čtyřbarevných
kombinací
zvlášť
pro
BCP
ALL
(SYTOI6/CDI9/CD45.
CD20/CDIO/CD I9/CD34, CD IO/CD66c/CD19/CD45, CD58/CDlO/CD19/CD34) a pro T
E. Mejstříková, strana 56 ALL
(SYTO 16/CD 19/CD45 ,
( příloha pacientů
CD99/CD7/CD5/CD3
ll). Tyto kombinace jsme ve stejné sledovaných po relapsu
u možňují
definovat
či
intra-T dT/CD7 /CD5/intra-CD3)
podobě začali
po transplantaci . V
dostatečný počet
a
používat pro detekci MRN i u
pod statě
nám výše uvedené kombinace
imunofenotypových aberací vhodných pro sledování
MRN.
pac ientů
3.2.4. Detekce MRN u
s AML
Detekce MRN u AML byla první aplikace, kde jsme cytometrii a v rámci této studie j sme definovali
začali
řadu principů
používat 4-barevnou
průtokovou
použitelných i pro detekci MRJ\[
u ALL. Celá studie probíhala v rámci protokolu AML BFM 98, který byl koordinována kolegy z Milnsteru. V naší studii jsme definovali jako základní 2 antigeny CD34 a CD33 ( "páteř
(backbone)"), na nichž se detekovaly jednotlivé imunofenotypové aberace. Pro
kvantifikaci MRN z jaderných které
volně
prochází
okolo 1 roku
/02 , 280,
buněčno u
buněk
k odlišení jaderných
buněk
jsme zavedli
membránou a
měření fluorescenční barvičky značí
SYTO-16,
nukleové kyseliny a tím pomáhá
od nejaderné drti . Medián vzniku
relapsů
u
dětských
AML je
což umožňuje rychlé hodnocení vlivu MRN na výsledky léčby. Detekce
MRN probíhala ve dvou
ča sově oddě l ených
periodách,
přičemž
na pacientech z první kohorty
(2000 až 2001 ) se stanovily prahové hodnoty ("cut offy") jednotlivých subpopulací v definovaných
onemocnění
časových
192 .
imunologickou
bodech (den 15, 21 až 28, 42 - 56, 70 až 84) pro predikci relapsu
Na společných datech jsme prokázali, že pacienti s prokazatelnou
přítomností blastů
bez události (příloha 4) hodnotu MRN podle
/89.
ve 3 a íce
časových
bodech
měli signifikantně
horší
přežití
V rámci multivariantní analýzy jsme ale neprokázali přídatnou
průtokové
podl e protoko lu AML BFM 98
cytometrie ke kritériím, která se používala ve stratifikaci
/4/ ,175.
4 . Závěr
4. 1. Mye/oidní antigeny u ALL Obj evili j sme nové souvislosti mezi aberantními myeloidními antigeny a jejich vztah k prognóze a ke genotypu CD33 u ALL
celkově
iu
( přílohy důležitých
], 3). Objevili jsme silný prognostický význam exprese podskupin
pacientů (příloha
I). Naše výsledky ukazují,
rozdílně
a že tedy dosavadní praxe
že jednotlivé myeloidní antigeny mohou být regulovány jejich
směšování
byla nesprávná. Demonstrovali jsme nestabilitu exprese myeloidních
E. Mejstříková , strana 57 antigenů
zejména
při
primární při
mají srovnatelnou expresi
4.2.
Význam
léčbě
ALL
( přílohy
relapsu ALL, jiné
diferenciačního
Naše data ukazují , že exprese jednoho z
1, 3),
(např.
přičemž některé
antigeny (CD66c)
CD33) nikoliv.
antigen u CD10 nej dů ležitěj š ích anti genů
u leukémií, CD 10,
v rámci klasifikace B prekursorové ALL použita k lepšímu prognostickému
může
rozdělení
být
ALL.
4.3. Minimální reziduální nemoc Průtoková
cytometrie
při
primární
léčbě rozdělí
pacienty s B prekurzorovou ALL podle
hladiny MRN. Spolehlivé použití 4 barevné cytometrie s předdefinovanými gaty je limitováno na prvních 15 zařazení
dnů
protokolu ALL-IC BFM2002,
do skupin M RN
měřených
není možné pro jinou kinetiku U
nepřízni
( příloha
B
řady
S),
přičemž
PCR. Analogické využití
léčebné o dpovědi ( přílohy
tato hodnota
časné odpovědi
díky menší fenotypové
na
léčbu
T ALL
7, ll ).
ých forem ALL lze cytometricky detekovat MRN i v pravděpodobně
spolehlivě určí
plasticitě
pozdějších
maligních
blastů.
je potenciálním zdrojem falešných pozitivit i při detekci MRN pomocí
fázích
léčby
Rekonstituce
přestaveb genů
19 ( příloha 6) Cytometrie detek uje MRN u AML, ale není schopna prognosticky existujících rizikových skupin
( příloha
rozdělit
pacienty v rámci
4).
5. Seznam zkratek ABL AEIOP AF10
AHL ALL AML AMLM4eo AML1
ara-C ARA-G ATRA
BCP BCR BFM
gen V -abl Abelson murine leukemia viral oncogene homolog, lokalizace 9q34.1 Associazonie ltaliana Ematologia Oncologia Pediatrica (pracovní skupina) gen ALL 1 fused gene from chromosome 10, lokalizace 1Op 12 akutní hybridní leukémie akutní lymfoblastická leukémie akutní myeloidní leukémie akutní myeloidní leukémie varianta M4 s eosinofilii gen Acute myeloid leukemia 1 (synonymum RUNX1 (runt-related transcription factor 1), CBF A2 (core binding factor A2) , lokalizace 2 1q22.3 cytosin arabinosid arabinosyl guanin all-trans retinová kyselina (retionoic acid) B prekurzorová (B cell precursor) Breakpoint cluster region, lokalizace 22q11.2 (součást fúzního genu BCRlABL) Berlín-Frankfurt-Munster (pracovní skupina)
E. Mejstříková, strana 58 bHLHB I CALM CBF CBFp CCG CD CEBPa C-KIT CNS COG DAPI DNA EFS EGIL ESG-MRD-ALL EsPhALL ETO
FAB FLT3 FLT3/1TD GMALL GO GSTTI HDAC HLA HOXII HOXllL2
19 ITD JMML L LAIP LMOI LM02 LYLI MDS MLL MRN MYC MYHll NCI NG2
gen Basic helix-Ioop-helix domain containing, class B, 2, lokalizace 21q22.11 gen calmodulin I (phosphorylase kinase, delta), lokalizace 14q24-q31 core binding factor gen podjednotky ~ CBF, lokalizace 16q22 Childhood Cancer Group (pracovní skupina) cluster of differentiation (nomenklaturní systém antigenů) gen pro CCAA T enhancer binding protein alp ha, lokalizace 19q 13.1 gen receptoru pro stem cell faktor, lokalizace 4q 12 centrální nervový systém Childhood Oncology Group (pracovní skupina) 4",6 - diamidino - 2 - phenylindole, di hydrochloride (fluorescenční barvivo značící stechiometricky DNA) deoxyribonykleová kyselina (deoxyribonucleic acid) přežití bez události ("událostí" je smrt, relaps nebo sekundární malignita, event free survival) European Group for the Immunological Characterization of Acute Leukemias European Study Group on MRD detection in ALL (pracovní skupina) European lntergroup Study on Post Induction Treatment ofPhiladelphia Positive Acute Lymphoblastic Leukaemia with lmatinib gen Eigth twenty one (synonymum RUNX1 Tl (runt-related transcription factor 1; translocated to, 1 (cyclin D-related)), lokalizace 8q22 F rench-American-British (klasifikace leukémií) gen FMS-like tyrosine kinase 3, lokalizace 13q 12.2 interní tandemová duplikace FLT3 German Multicenter study group on Adult acute Lymphoblastic Leukemia (pracovní skupina a protokol) gemtuzumab ozogamicin, Mylotarg enzym glutathion S-transferáza theta 1 hi ston deacetyláza (histon deacetylase) human leucocyte antigen rodina genů Homeboxll (TLXl,2,3) gen Homebox 11 L2, lokalizace 7p 15-7p 14.2 imunoglobulin interní tandemová duplikace juvenilní myelomonocytární leukémie litr leukemia associated immunophenotype gen (LIM domain only 1 (rhombotin-like 1)), lokalizace l1p15 gen (LIM domain only 2 (rhombotin-like 2)), lokalizace 11 p13 gen Lymphoblastic leukemia derived sequence 1, lokalizace 19p13.2 myelodysplastický syndrom gen Myeloid/lymphoid or mixed lineage leukemia, lokalizce l1q23 minimální reziduální nemoc protooncogen MYC, lokalizace 8q24 gen Myosin heavy chain, lokal izace 16p 13 National Cancer Institute molekula chondroitin sulfátu
E. Mejstříková, strana 59 NHL NOTCHl
NPMl OKT-3 PAXS PCR
pEFS Ph PML PNP POG
pRFS RAEB RAEB-t RARa RC
RFS RNA RQPCR SCT STR SYTO-16
TALl TAL2 TCR TdT TEL TLXl TLX3 TMD TPS3 VOD WTl
Non Hodginský Iymfom gen Notch homolog 1, translocation-associated (Drosophila), lokalizace 9q34.3 gen nucIeophosmin, lokalizace 5q35 klon anti-CD3 protilátky Paired box gene 5, lokal izace 9p13 polymerázová řetězová reakc (pol ymerase chain reaction) pravděpodobnost přežití bez události ("událostí" je smrt, relaps nebo sekundární malignita, event free survival) Filadelfský (Ph chromozóm na podkladě trans lokace BCRI ABL (9;22) gen promyelocytární leukémie purine nucIeoside fosforyláza Pediatric Oncology Group (pracovní skupina) pravděpodobnost přežití bez relapsu refrakterní anémie s excesem blastů (refractory anemia with excess of blasts) refrakterní anémie s excesem blastů v transformaci (refractory anemia with excess of blasts in transformation) gen receptoru pro retinovou kyselinu alfa (retinoic acid receptor alpha) refrakterní cytopénie přežití bez relapsu onemocnění (relapse free survival) ribonukleová kyselina (ribonucIeic acid) kvantitativní polymerázová řetězová rekace transplantace krvetvorných buněk (stem cell transplantation) short tandem repeats zelené fluorescenční barvivo značící nukleové kyseliny a volně procházející buněčnou membránou gen T -cell acute leukemia 1, lokalizace 1p32 gen T-cell acute Iymphoblastic leukemia 2, lokalizace 9q31 T buněčný receptor (T cell receptor) terminální deoxynukleotidyl transferáza gen Translocation ets leukemia, lokalizace 12p 13.1 gen T cellleukemia, gen z rodiny homebox genů, lokalizace 10q24 gen T-cellleukemia, gen z rodiny homebox genů , lokalizace 5q35.1 tranzitorní myeloproliferativní nemoc (transitory myeloproliferative disease) gen proteinu p53 (o velikosti 53 kDa), lokalizace 17p 13 venookluzivní nemoc (venoocIusive disease) tumor supresorový gen Wilmsova nádoru, lokalizace 11 p 13
E. Mejstříková, strana 60
6. Literatura 6. 1. Použitá literatura 1. Young NS, Scheinberg P, Calado RT. Aplastic anemia. Curr Opin Hematol 2008; 15(3): 162-8. 2. Hrusak O, Trka J, Zuna J, Polouckova A, Kalina T, Stary 1. Acute lymphoblastic I ukemia incidence during socioeconomic transition: selective increase in children from l to 4 years. Leukemia 2002; 16(4):720-5. 3. Pui CH, Robison LL, Look A T. Acute Iymphoblastic leukaemia. Lancet 2008;371 (9617): 1030-43. 4. Strahm B, Malkin D. Hereditary cancer predisposition in children: genetic basis and clinical implications. Int J Cancer 2006; 119(9):2001-6. 5. Zuna J, Cave H, Eckert C, Szczepanski T, Meyer C, Mejstrikova E, et al. Childhood secondary ALL after ALL treatment. Leukemia 2007;21 (7): 1431-5. 6. Shivakumar R, Tan W, Wilding GE, Wang ES, Wetzler M. Biologic features and treatment outcome of secondary acute Iymphoblastic leukemia--a review of 101 cases. Ann Onco12008. 7. Jaffe ES, Diebold .1, Harris NL, Muller-Helmelink HK, Flandrin G, Vardiman JW. Burkitťs Iymphoma: a single disease with multiple variants. The World Health Organization c1assification of neoplastic diseases of the hematopoietic and Iymphoid tissues. Blood 1999;93(3): 1124. 8. Burrneister T, Schwartz S, Horst HA, Rieder H, Gokbuget N, Hoelzer D, et al. Molecular heterogeneity of sporadic adult Burkitt-type leukemia/lymphoma as revealed by PCR and cytogenetics: correlation with morphology, immunology and clinical features. Leukemia 2005; 19(8): 1391-8. 9. Attarbaschi A, Mann G, Schumich A, Konig M, Pickl WF, Haas OA, et al. CD44 deficiency is a consistent finding in childhood Burkitťs lymphoma and leukemia. Leukemia 2007;21(5):1110-3 . 10. Campana D, Behm FG. Immunophenotyping of leukemia. J Immunol Methods 2000;243(1-2):59-75. ll. Kansal R, Deeb G Barcos M, Wetzler M, Brecher ML, Block A W, et aJ. Precursor B lymphoblastic leukemia with surface light chain immunoglobulin restriction: a report of 15 patients. Am J Clin Pathol 2004; 121 (4):512-25. 12. VasefMA, Brynes RK, Murata-Collins JL, Arber DA, Medeiros Ll. Surface immunoglobulin light chain-positive acute Iymphoblastic leukemia of F AB Ll or L2 type: a report of 6 cases in adults . Am .1 Clin Pathol 1998; 11 0(2): 143-9. 13. McKenna R W, Washington LT, Aquino DB , Picker Ll, Kroft SH. Immunophenotypic analysis ofhematogones (B-Iymphocyte precursors) in 662 consecutive bone marrow specimens by 4-color tlow cytometry. Blood 2001 ;98(8):2498-507. 14. Kroft SH, Asplund SL, McKenna RW, Karandikar N1. Haematogones in the peripheral blood of adults: a four-colour tlow cytometry study of 102 patients. Br J Haematol 2004; 126(2):209-12. 15. Sims GP, Ettinger R, Shirota Y, Yarboro CH, Illei GG, Lipsky PE. Identification and characterization of circulating human transitional B cells. Blood 2005; 105(11):4390-8. 16. Bene MC, Castoldi G, Knapp W, Ludwig WD, Matutes E, Orfao A, et aJ. Proposals for the immunological classification of acute leukemias. European Group for the Immunological Characterization of Leukemias (EGIL). Leukemia 1995 ;9( 1O): 1783-6.
E.
Mejstříková,
strana 61
17. Bene MC, Bernier M, Casasnovas RO, Castoldi G, Knapp W, Lanza F, et al. The reliability and specificity of c-kit for the diagnosis of acute myeloid leukemias and undifferentiated leukemias. The European Group for the Jmmunological Classification of Leukemias (EGIL). Blood 1998;92(2):596-9. 18. Szczepanski T, van der Velden VH, van Dongen]J. Flow-cytometric immunophenotyping of normal and malignant lymphocyt s. C lin Chem Lab Med 2006;44(7):775-96. 19. Attarbaschi A, Mann G, Dworzak M, Wiesbauer P, Schrappe M, Gadner H. Mediastinal mas s in childhood T -cell acute lymphoblastic leukemia: significance and therapy response. Med Pediatr Oncol 2002;39(6): 558-65. 20. Reiter A, Schrappe M, Ludwig WD, Tiemann M, Parwaresch R, Zimmermann M, et al. Intensive ALL-type therapy without local radiotherapy provides a 90% event-free survival for children with T-celllymphoblastic lymphoma: a BFM group report. Blood 2000;95(2):416-21 . 21. Schrauder A, Reiter A, Gadner H, Niethammer D, Klingebiel T, Kremens B, et a1. Superiority of allogeneic hematopoietic stem-cell transplantation compared with ehemotherapy alone in high-risk ehildhood T-cell aeute lymphoblastie leukemia: results from ALL-BFM 90 and 95. J Clin Oncol 2006;24(36):5742-9. 22. Bene MC, Castoldi G, Knapp W, Ludwig WD, Matutes E, Orfao A, et al. Classification of acute leukemias: reply from EG lL to Dr van Dongen. Leukemia 1996; 10(8): 1363-4. 23. Kalina T, Mejstrikova E, Vaskova M, Hrusak O. Imunofenotypizace dětských leukémií. Transfúze a hematologie dnes 2006; 12(1 ):20 - 25. 24. Graux C, Cools J, Miehaux L, Vandenberghe P, Hagemeijer A. Cytogeneties and moleeular genetics of T-cell acute Iymphoblastic leukemia: from thymocyte to lymphoblast. Leukemia 2006;20(9): 1496-51 O. 25. van Grotel M, Meijerink JP, Beverloo HB, Langerak A W, Buys-Gladdines lG, Sclmeider P, et a!. The outcome of molecular-cytogenetic subgroups in pediatrie T -cell acute lymphoblastic leukemia: a retrospective study of patients treated aceording to DCOG or COALL protocols. Haematologica 2006;91(9):1212-21. 26. van Grotel M, Meijerink JP, van Wering ER, Langerak A W, Beverloo HB, BuijsGladdines lG, et a!. Prognostic signifieance of molecular-cytogenetic abnormalities in pediatrie T-ALL is not explained by immunophenotypic differences. Leukemia 2008;22(1): 124-31. 27. Niehues T, Kapaun P, Harms DO, Burdach S, Kramm C, Korholz D, et a!. A c1assification based on T cell selection-related phenotypes identifies a subgroup of childhood T-ALL with favorable outcome in the COALL studies. Leukemia 1999;13(4):614-7. 28. Baak U, Gokbuget N, Orawa H, Schwartz S, Hoelzer D, Thiel E, et a!. Thymic adult T-cell acute lymphoblastie leukemia stratified in standard- and high-risk group by aberrant HOX11 L2 expression: experienee ofthe German multicenter ALL study group. Leukemia 2008. 29. Crist WM, Shuster]J, Falletta l, Pullen DJ, Berard CW, Vietti TJ, et a!. Clinical features and outcome in childhood T -cellleukemia-lymphoma according to stage of thymocyte differentiation: a Pediatrie Oncology Group Study. Blood 1988;72(6): 1891-7. 30. Raff T, Gokbuget N, Luschen S, Reutzel R, Ritgen M, Irrner S, et a!. Molecular relapse in adult standard-risk ALL patients detected by prospective MRD monitoring during and after maintenanee treatment: data from the GMALL 06/99 and 07/03 trials. Blood 2007;109(3):910-5. 31. Asnafi V, Be1djord K, Boulanger E, Comba B, Le Tutour P, Estienne MH, et a!. Analysis of TCR, pT alpha, and RAG- l in T -acute lymphoblastic leukemias improves
E. Mejstříková, strana 62 understanding ofearly human T-Iymphoid lineage commitment. Blood 2003;101(7):2693703. 32. Asnafi V, Beldjord K, Oarand R, Millien C, Bahloul M, LeTutour P, et aJ. IgH Dl rearrangements within T -ALL correlate with cC079a expression, an immaturelTCRgammadelta phenotype and absence of lL 7Ralpha/CO 127 expression. Leukemia 2004; 18(12): 1997 -2001. 33. Asnafi V, Beldjord K, Libura M, Vi llarese P, M illien C, Ballerini P, et al. Age-related phenotypic and oncogenic differences in T -cell acute Iymphoblastic leukemias may reflect thymic atrophy. Blood 2004;104(13):4173-80. 34. Asnafi V, Buzyn A, Thomas X, Huguet F, Vey N, Boiron 1M, et al. Impact ofTCR status and genotype on outcome in adult T-cell acute Iymphoblastic leukemia: a LALA-94 study. Blood 2005; 105(8):3072-8. 35. Pui CH, Relling MV, Oowning lR. Acute Iymphoblastic leukemia. N Engl J Med 2004;350(15): 1535-48. 36. Trueworthy R, Shuster J, Look T, Crist W, Borowitz M, Carroll A, et aJ. Ploidy of Iymphoblasts is the strongest predictor oftreatment outcome in B-progenitor cell acute Iymphoblastic leukemia of childhood: a Pediatrie Oncology Oroup study. J Clin Oncol I 992; I 0(4): 606-13. 37. Zuna J, Hrusak O, Kalinova M, Muzikova K, Stary J, Trka J. Significantly lower relapse rate for TEL/AML 1-positive ALL. Leukemia 1999; 13( 1O): 1633. 38. Zuna J, Hrusak O, Kalinova M, Muzikova K, Stary J, Trka J. TEL/AMU positivity in childhood ALL: average or better prognosis? Czech Paediatric Haematology Working Oroup. Leukemia 1999; 13(1 ):22-4. 39. Schultz KR, Pullen OJ, Sather HN, Shuster JJ, Oevidas M, Borowitz MJ, et aJ. Riskand response-based classification of childhood B-precursor acute Iymphoblastic leukemia: a combined analysis of prognostic markers from the Pediatrie Oncology Oroup (PO O) and Children's Cancer Oroup (CCO). Blood 2007; 109(3):926-35. 40. Arico M, Valsecchi MO Rizzari C, Barisone E, Biondi A, Casa1e F, et aJ. Long-term results ofthe AIEOP-ALL-95 Trial for Childhood Acute Lymphoblastic Leukemia: insight on the prognostic value ofDNA index in the framework ofBerlin-Frankfurt-Muenster based chemotherapy. J Clin Oncol 2008;26(2):283-9. 41. Nachman .lB, Heerema NA, Sather H, Camitta B, Forestier E, Harrison CJ, et aJ. Outcome oftreatment in children with hypodiploid acute lymphoblastic leukemia. Blood 2007;110(4):1112-5. 42. Stark B, Jeison M, Oobuzov R, Krug H, Olaser-Oabay L, Luria O, et al. Near haploid childhood acute lymphoblastic leukemia masked by hyperdiploid line: detection by fluorescence in situ hybridization. Cancer Oenet Cytogenet 2001; 128(2): 108-13. 43. Attarbaschi A, Mann O, Konig M, Steiner M, Oworzak MN, Oadner H, et aJ. Neartetraploidy in childhood B-cell precursor acute lymphoblastic leukemia is a highly specific feature of ETV6/RUNX I-positive leukemie cases. Oenes Chromosomes Cancer 2006;45(6):608-11. 44. Forestier E, Andersen MK, Autio K, Blennow E, Borgstrom O, Oolovleva I, et al. Cytogenetic pattems in ETV6/RUNXl-positive pediatrie B-cell precursor acute lymphoblastic leukemia: A Nordic series of245 cases and review ofthe literature. Oenes Chromosomes Cancer 2007;46(5):440-50. 45. Raimondi SC, Zhou Y, Shurtleff SA, Rubnitz JE, Pui CH, Behm FO. Near-triploidy and near-tetraploidy in childhood acute lymphoblastic leukemia: association with B-lineage blast cells carrying the ETV6-RUNX1 fusion, T-lineage immunophenotype, and favorable outcome. Cancer Oenet Cytogenet 2006; 169(1 ):50-7.
E. Mejstříková, strana 63 46. Pieters R, Schrappe M, De Lorenzo P, Hann I, De Rossi G, Felice M, et a!. A treatment protocol for infants younger than 1 year wi th acute lymphoblastic leukaemia (lnterfant-99): an observational study and a mul ti centre randomised tria!. Lancet 2007;370(9583):240-50. 47. Mann G, Cazzaniga G, van der Velden VH, Flohr T, Csinady E, Paganin M, et a!. Acute lymphoblastic leukemia with t( 4; ll) in children 1 year and ol der: The 'big sister' of the infant disease? Leukemia 2007. 48. Arico M, Valsecchi MG, Camitta B, Schrappe M, Chessells 1, Baruchel A, et a!. Outcome oftreatment in children with Philadelphia chromosome-positive acute lymphoblastic leukemia. N Engl 1 Med 2000;342(14):998-1006. 49. Schrappe M, Reiter A, Ludwig WD, Harbott 1, Zimmermann M, Hiddemann W, et a!. Improved outcome in childhood acute lymphoblastic leukemia despite reduced use of anthracycIines and cranial radiotherapy: results oftrial ALL-BFM 90. German-AustrianSwiss ALL-BFM Study Group. Blood 2000;95( 11 ):3310-22. 50. De Keersmaecker K, Marynen P, Cools 1. Genetic insights in the pathogenesis ofTcell acute lymphoblastic leukemia. Haematologica 2005 ;90(8): 1116-27. 51. Asnafi V, Radford-Weiss I, Dastugue N, Bayle C, LeboeufD, Charrin C, et a!. CALM-AFI0 is a common fusion transcript in T-ALL and is specific to the TCRgammadelta lineage. Blood 2003;102(3):1000-6. 52. Caudell D, Aplan PD. The role of CALM-AF 10 gene fusion in acute leukemia. Leukemia 2008;22(4):678-85. 53. Sulis ML, Williams O, Palomero T, Tosello V, Pallikuppam S, Real Pl, et a!. NOTCHl extracellular juxtamembrane expansi on mutations in T-ALL. Blood 2008. 54. Suwanjunee S, Wongchana W, Palaga T. Inhibition of gamma-secretase affects proliferation of leukemia and hepatoma celllines through Notch signaling. Anticancer Drugs 2008; 19(5):4 77-86. 55. Smith M, Arthur D, Camitta B, Carroll AJ, Crist W, Gaynon P, et a!. Uniform approach to risk classification and treatment assignment for children with acute lymphoblastic leukemia. 1 Clin Oncol 1996; 14(1): 18-24. 56. Moricke A, Zimmermann M, Reiter A, Gadner H, Odenwald E, Harbott 1, et a!. Prognostic impact of age in children and adolescents with acute lymphoblastic leukemia: data from the trials ALL-BFM 86, 90, and 95. Klin Padiatr 2005;217(6):310-20. 57. Stanulla M, Schaeffeler E, Flohr T, Cario G, Schrauder A, Zimmermann M, et a!. Thiopurine methyltransferase (TPMT) genotype and early treatment response to mercaptopurine in childhood acute lymphoblastic leukemia. lama 2005 ;293(12): 1485-9. 58. Boissel N, Auclerc MF, Lheritier V, Perel Y, Thomas X, Leblanc T, et a!. Should adolescents with acute lymphoblastic leukemia be treated as old children or young adults? Comparison ofthe French FRALLE-93 and LALA-94 trials. 1 Clin Oncol 2003 ;21 (5):774-80. 59. Ramanujachar R, Richards S, Hann I, Goldstone A, Mitchell C, Vora A, et a!. Adolescents with acute lymphoblastic leukaemia: outcome on UK national paediatric (ALL97) and adult (UKALLXIIIE2993) trials. Pediatr Blood Cancer 2007 ;48(3) :254-61. 60. Ramanujachar R, Richards S, Hann I, Webb D. Adolescents with acute lymphoblastic leukaemia: emerging from the shadow of paediatric and adult treatment protocols. Pediatr Blood Cancer 2006;47(6):748-56. 61. Conter V, Valsecchi MG, Silvestri D, Campbell M, Dibar E, Magyarosy E, et a1. Pulses of vincristine and dexamethasone in addition to intensive chemotherapy for children with intermediate-risk acute lymphoblastic leukaemia: a multicentre randomised tria!. Lancet 2007;369(9556): 123-31. 62. Moricke A, Reiter A, Zimmermann M, Gadner H, Stanulla M, Dordelmann M, et al. Risk-adjusted therapy of acute lymphoblastic leukemia can decrease treatment burden and
E. Mejstříková, strana 64 improve survival : treatment results of 2169 unselected pediatric and adolescent patients enrolled in the trial ALL-BFM 95. Blood 2008. 63. Pui CH , Pei D, Sandlund JT, Campana D, Ribeiro RC, Razzouk BI, et al. Risk of adverse events after completion oftherapy for childhood acute lymphoblastic leukemia. J Clin Oncol 2005 ;23(31 ):7936-4l. 64. Schrappe M, Reiter A, Zimmermann M, Harbott J, Ludwig WD, Henze G, et al. Longterm results of four consecutive trials in childhood ALL performed by the ALL-BFM study group from 1981 to 1995. Berlin-Frankfurt-Munster. Leukemia 2000; 14(12):2205-22. 65. Riehm H, Reiter A, Schrappe M, Berthold F, Dopfer R, Gerein V, et al. [Corticosteroid-dependent reduction of leukocyte count in blood as a prognostic factor in acute Iymphoblastic leukemia in childhood (therapy study ALL-BFM 83)]. Klin Padiatr 1987; 199(3): 151-60. Dordelmann M, Reiter A , Borkhardt A, Ludwig WD, Gotz N, Viehmann S, et al. 66. Prednisone response is the strongest predictor oftreatment outcome in infant acute lymphoblastic leukemia. Blood 1999;94(4): 1209-17. Schrappe M, Arico M , Harbott J, Biondi A, Zimmermann M , Conter V , et aJ. 67. Philadelphia chromosome-positive (Ph+) childhood acute lymphoblastic leukemia: good initial steroid response allows early prediction of a favorable treatment outcome. Blood 1998;92(8):2730-41. Gajjar A, Ribeiro R, Hancock ML, Rivera GK, Mahmoud H, Sandlund JT, et aJ. 68. Persistence of circulating blasts after 1 week of mul ti agent chemotherapy confers a poor prognosis in childhood acute lymphoblastic leukemia. Blood 1995;86(4):1292-5. 69. Borowitz MJ, Pullen DJ , Shuster]J, Viswanatha D, Montgomery K, Willman CL, et al. Minimal residual disease detection in childhood precursor-B-cell acute lymphoblastic leukemia: relation to other risk factors. A Children's Oncology Group study. Leukemia 2003 ; 17(8): 1566-72. Steinherz PG, Gaynon PS, Breneman JC , Cherlow JM, Grossman NJ, Kersey m, et al. 70. Cytoreduction and prognosis in acute Iymphoblastic leukemia--the importance of early manow response: report from the Childrens Cancer Group. J Clin Oncol 1996; 14(2):389-98. 71. Balduzzi A, Valsecchi MG, Uderzo C, De Lorenzo P, Klingebiel T, Peters C, et aJ. Chemotherapy versus allogeneic transplantation for very-high-risk childhood acute Iymphoblastic leukaemia in first complete remission: comparison by genetic randomisation in an international prospective study. Lancet 2005;366(9486):635-42. 72. Wassmann B, Pfeifer H, Goekbuget N, Beelen DW, Beck J, Stelljes M , et aJ. Alternating versus concunent schedules ofimatinib and chemotherapy as front-line therapy for Philadelphia-positive acute lymphoblastic leukemia (Ph+ ALL) . Blood 2006; 108(5): 146977. 73. Ottmann OG, Wassmann B, Pfeifer H, Giagounidis A, Stelljes M, Duhrsen U, et aJ. Imatinib compared with chemotherapy as front-line treatment of elderly patients with Philadelphia chromosome-positive acute lymphoblastic leukemia (Ph+ALL). Cancer 2007; 109( 10):2068-76. 74. Ottmann OG, Wassmann B. Treatment ofPhiladelphia chromosome-positive acute Iymphoblastic leukemia. Hematology Am Soc Hematol Educ Program 2005: 118-22. 75. Pfeifer H, Wassmann B, Pavlova A, Wunderle L, Oldenburg J, Binckebanck A, et aJ. Kinase domain mutations of BCR-ABL frequently precede imatinib-based therapy and give rise to relapse in patients with de novo Philadelphia-positive acute lymphoblastic leukemia (Ph+ ALL). Blood 2007;110(2):727-34. 76. Ramirez P, Dipersio JF. Therapy options in imatinib failures. Oncologist 2008; 13(4):424-34.
E. Mejstříková, strana 65 77. Quintas-Cardama A, Cortes 1. Management of patients with resistant or refractory chronic myelogenous leukemia. Oncology (Williston Park) 2008;22(4):430-7; discussion 437, 442, 446 passim. 78. Porkka K, Koskenvesa P, Lundan T, Rimpilainen J, Mustjoki S, Smykla R, et a!. Dasatinib crosses the blood-brain barrier and is an effic ient therapy for central nervous system Philadelphia chromosome-positive leukemia. Blood 2008. 79. Willemse MJ, Seriu , Hettinger K, ďAnie llo E, Hop WC , Panzer-Grumayer ER, et a!. Detection of minimal residual disease identifies differences in treatment response between TALL and precursor B-ALL. Blood 2002;99(12):4386-93. 80. Cheson BD. Clinical management ofT-cell malignancies: current perspectives, key issues, and emerging therapies. Semin Oncol 2007;34(6 Suppl5):S3-7. 8 1. Holowiecki J. NOTCH 1 pathway: a molecular target in T -cell cancers? Lancet 2005;365(9455):197-9. 82. De Keersmaecker K, Lahortiga I, Mentens N, Folens C, Van Neste L, Bekaert S, et a!. ln vitro validation of gamma-secretase inhibitors alone or in combination with other anticancer drugs for the treatment ofT-cell acute lymphoblastic leukemia. Haematologica 2008;93(4):533-42. 83. Einsiedel HG, von Stackelberg A, Hartmann R, Fengler R, Schrappe M , Janka-Schaub G, et a!. Long-term outcome in children with relapsed ALL by risk-stratified salvage therapy: results of trial acute lymphoblastic leukemia-relapse study of the Berlin-Frankfurt-Munster Group 87. J Clin Oncol 2005;23(31 ):7942-50. 84. Eckert C, Biondi A, Seeger K, Cazzaniga G, Hartmann R, Beyermann B, et a!. Prognostic value of minimal residual disease in relapsed childhood acute lymphoblastic leukaemia. Lancet 2001 ;358(9289): 1239-41 . 85 . Langmuir PB , Aplenc R, Lange Bl. Acute myeloid leukaemia in children. Best Pract Res Clin HaematoI2001;14(l):77-93. 86. Biondi A, Rovelli A, Cantu-Rajnoldi A, Fenu S, Basso G, Luciano A, et a!. Acute promyelocytic leukemia in children: experience of the Italian Pediatric Hematology and Oncology Group (AIEOP). Leukemia 1994;8(8): 1264-8. 87. Maule MM, Dama E, Mosso ML, Magnani C, Pastore G, Merletti F. High incidence of acute promyelocytic leukemia in children in northwest Italy, 1980-2003: a report from the Childhood Cancer Registry ofPiedmont. Leukemia 2008;22(2):439-41. 88 . Niemeyer CM, Baumann I. Myelodysplastic syndrome in children and adolescents. Semin Hematol 2008;45(1 ):60-70. 89. Yeasmin S, Nakayama K, Ishibashi M , Oride A, Katagiri A, Purwana IN, et a!. Therapy-related myelodysplasia and acute myeloid leukemia following paclitaxel- and carboplatin-based chemotherapy in an ovarian cancer patient: a case report and literature review. lnt J Gynecol Cancer 2008. 90. Schaapveld M, Visser O, Louwman MJ, de Vries EG, Willemse PH, Otter R, et a!. Risk of new primary nonbreast cancers after breast cancer treatment: a Dutch populationbased study. J Clin Oncol 2008;26(8): 1239-46. 91. Flaig TW, Tangen CM, Hussain MH, Stadler WM, Raghavan D, Crawford ED, et a!. Randomization reveals unexpected acute leukemias in Southwest Oncology Group prostate cancer tria!. J Clin Oncol 2008;26(9): 1532-6. 92. Navid F, Billups C, Liu T, Krasin MJ, Rodriguez-Galindo C. Second cancers in patients with the Ewing sarcoma family oftumours. Eur J Cancer 2008;44(7):983-991. 93. Howard R, Gilbert Lynch CF, Hall P, Storm H, Holowaty E, et a!. Risk of leukemia among survivors of testicular cancer: a population-based study of 42,722 patients. Ann Epidemiol2008;18(5):416-21.
E. Mejstříková, strana 66 94. Bamard DR, Lange B, Alonzo TA, Buekley J, Kobrinsky JN, Gold S, et al. Aeute myeloid leukemia and myelodysplastic syndrome in chi ldren treated for cancer: comparison with primary presentation. Blood 2002; 100(2):427-34. 95. Bamard DR, Woods WG . Treatment-related myelodysplastic syndrome/acute myeloid leukemia in survivors of childhood cancer--an update. Leuk Lymphoma 2005;46(5):651-63. 96. Alter BP. Diagnosis, genetics , and management of inherited bone marrow failure syndromes. Hematology Am Soc Hematol Educ Program 2007;2007:29-39. 97. Creutzig U, Zimmermann M, Ritter J, Reinhardt D, Hermann J, Henze G, et al. Treatment strategies and long-term results in paediatrie patients treated in four conseeutive AML-BFM trials. Leukemia 2005;19(12):2030-42. 98. Creutzig U, Zimmermann M, Reinhardt D, Dworzak M, Stary J, Lehmbecher T. Early deaths and treatment-related mortality in children undergoing therapy for aeute myeloid leukemia: analysis ofthe multicenter cIinical trials AML-BFM 93 and AML-BFM 98. J Clin OncoI2004;22(21):4384-93. 99. Lehrnbeeher T, Varwig D, Kaiser J, Reinhardt D, Klingebiel T, Creutzig U. Infectious eomplications in pediatrie acute myeloid leukemia: analysis ofthe prospective multiinstitutional cIinical trial AML-BFM 93. Leukemia 2004; 18(1):72-7. 100. Creutzig U, Bender-Gotze C, Ritter J, Zimmermann M, Stollmarm-Gibbels B, Korholz D, et al. The role of intensive AML-specific therapy in treatment of children with RAEB and RAEB-t. Leukemia 1998; 12(5):652-9. 101. Hasle H. Myelodysplastic and myeloproliferative disorders in children. Curr Opin Pediatr 2007; 19(1): 1-8. 102. Abrahamsson J, Clausen N, Gustafsson G, Hovi L, Jonmundsson G, Zeller B, et al. Improved outcome after relapse in chi1dren with acute myeloid leukaemia. Br J Haematol 2007; 136(2):229-236. 103. Bacher U, Kem W, Schnittger S, Hiddemann W, Schoch C, Haferlach T. Further eorrelations of morphology according to FAB and WHO cIassification to cytogenetics in de novo acute myeloid leukemia: a study on 2,235 patients. Arm Hemato12005;84(12):785-91. 104. Bermett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR, et al. Proposals for the cIassifieation ofthe acute leukaemias. French-American-British (FAB) cooperative group. Br J HaematoI1976;33(4):451-8. 105. Reinhardt D, Diekamp S, Langebrake C, Ritter J, Stary J, Dworzak M, et al. Acute megakaryoblastic leukemia in ehildren and adolescents, excIuding Down's syndrome: improved outcome with intensified induction treatment. Leukemia 2005; 19(8): 1495-6. 106. Bamard DR, Alonzo TA, Gerbing RB, Lange B, Woods WG. Comparison of childhood myelodysplastic syndrome, AML F AB M6 or M7, CCG 2891: report from the Children's Oncology Group. Pediatr Blood Cancer 2007;49(1): 17-22. 107. Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR, et al. Proposed revised criteria for the cIassification of acute myeloid leukemia. A report ofthe French-Ameriean-British Cooperative Group. Ann lotem Med 1985; 103(4):620-5. 108. Harris NL, Jaffe ES , Diebold J, Flandrin G, Muller-Hermelink HK, Vardiman J, et al. World HeaIth Organization cIassification of neoplastic diseases of the hematopoietic and lymphoid tissues: report ofthe Clinieal Advisory Committee meeting-Airlie House, Virginia, November 1997. J Cl in Oncol 1999; 17(12):3835-49. 109. Creutzig U, Harbott J, Sperling C, Ritter J, Zimmermann M, Lomer H, et al. Clinical significance of surfaee antigen expression in ehildren with acute myeloid leukemia: results of study AML-BFM-87. Blood 1995;86(8):3097-108. 110. Hrusak O, Porwit-MacDonald A. Antigen expression patterns reflecting genotype of acute leukemias. Leukemia 2002;16(7):1233-58.
E.
Mejstříková,
strana 67
111. Renneville A, Roumier C, Biggio V, Nibourel 0, Boissel N, Fenaux P, et a1. Cooperating gene mutations in acute myeloid leukemia: a review of the literature. Leukemia 2008;22(5):915 -31. 112. Goemans BF, Zwaan CM, Miller M, Zimmennann M Harlow A, Meshinchi S, et a1. Mutations in KIT and RAS are frequent events in pediatric core-binding factor acute myeloid leukemia. Leukemia 2005; 19(9): 1536-42. 113. Rubnitz JE, Raimondi SC, Halbert AR, Tong X, Sri vastava DK, Razzouk BI, et al. Characteristics and outcome of t(8;21 )-positive childhood acute myeloid leukemia: a single institution's experience. Leukemia 2002; 16(10):2072-7. 114. Yamagata T, Maki K, Mitani K. RunxllAMLl in nonnal and abnonnal hematopoiesis. IntJ Hematol 2005;82(1): 1-8. 115. Mrozek K, Bloomfield CD. Chromosome aberrations, gene mutations and expression changes, and prognosis in adult acute myeloid leukemia. Hematology Am Soc Hematol Educ Program 2006:169-77. 116. He G, Wu D, Sun A, Xue Y, Jin Z, Qiu H, et a1. B-Lymphoid and myeloid lineages biphenotypic acute leukemia with t(8;21)(q22;q22). Int J HematoI2008;87(2):132-136. 117. He G, Wu D, Sun A, Xue Y, Jin Z, Qiu H, et a1. CytCD79a expression in acute leukemia with t(8;21): biphenotypic or myeloid leukemia? Cancer Genet Cytogenet 2007; 174(1 ):76-7. 118. Gibson SE, Dong HY, Advani AS, Hsi ED. Expression ofthe B cell-associated transcription factors PAX5, OCT-2, and BOR 1 in acute myeloid leukemia: associations with B-cell antigen expression and myelomonocytic maturation. Am J Cl in Pathol 2006; 126(6):916-24. 119. Valbuena JR, Medeiros U, Rassidakis GZ, Hao S, Wu CD, Chen L, et a1. Expression of B cell-specific activator protein/PAX5 in acute myeloid leukemia with t(8;21)( q22;q22). Am J Clin Pathol 2006; 126(2):23 5-40. 120. Tiacci E, Pileri S, Orleth A, Pacini R, Tabarrini A, Frenguelli F, et a1. PAX5 expression in acute leukemias: higher B-lineage specificity than CD79a and selective association with t(8;21)-acute myelogenous leukemia. Cancer Res 2004;64(20):7399-404. 121. Fenaux P, Wang ZZ, Degos L. Treatment ofacute promyelocytic leukemia by retinoids. Curr Top Microbiol Tmmunol 2007;313: 101-28. 122. Lo-Coco F, Ammatuna E. The biology of acute promyelocytic leukemia and its impact on diagnosis and treatment. Hematology Am Soc Hematol Educ Program 2006:156-61, 514. 123. Hasle H, Clemmensen IH, Mikkelsen M. Risks of leukaemia and solid tumours in individuals with Down's syndrome. Lancet 2000;355(9199): 165-9. 124. Bourquin JP, Subramanian A, Langebrake C, Reinhardt D, Bernard O, Ballerini P, et al. Identification of distinct molecular phenotypes in acute megakaryoblastic leukemia by gene expression profiling. Proc Natl Acad Sci U S A 2006; 103(9):3339-44. 125. Abildgaard L, Ellebaek E, Gustafsson G, Abrahamsson J, Hovi L, Jonmundsson G, et al. Optimal treatment intensity in children with Down syndrome and myeloid leukaemia: data from 56 children treated on NOPHO-AML protocols and a review ofthe literature. Arm Hematol 2006;85(5):275-80. 126. Creutzig U, Reinhardt D, Diekamp S, Dworzak M, Stary J, Zimmermann M. AML patients with Down syndrome have a high cure rate with AML-BFM therapy with reduced dose intensity. Leukemia 2005; 19(8): 1355-60. 127. Langebrake C, Creutzig U, Reinhardt D. Immunophenotype of Down syndrome acute myeloid leukemia and transient myeloproliferative disease differs significantly from other diseases with morphologically identical or similar blasts. Klin Padiatr 2005;217(3): 126-34.
E. Mejstříková, strana 68 128. Langebrake C, Klusmann JH, Wortmann K, Kolar M, Puhlmann U, Reinhardt D. Coneomitant aberrant overexpression of RUNXl and N CAM in regenerating bone marrow of myeloid leukemia ofDown's syndrome. Haematologiea 2006;91(11):1473-80. 129. Hasle H, Abrahamsson J, Arola M, Karow A, O'Marcaigh A, Reinhardt D, et al. Myeloid leukemia in ehildren 4 years or older with Down syndrome often laeks GA TAl mutation and eytogeneties and risk of relapse are more akin to sporadie AML. Leukemia 2008;22(7): 1428-30. 130. Raimondi SC, Chang MN, Ravindranath Y, Behm FG, Gresik MV, Steuber CP, et al. Chromosomal abnormalities in 478 ehildren with aeute myeloid leukemia: clinieal eharaeteristies and treatment outcome in a eooperative pediatrie oneology group study-POG 8821. Blood 1999;94(11 ):3707-16. 131. Betts DR, Ammann RA, Hirt A, Hengartner H, Beek-Popovie M, Kuhne T, et al. The prognostie signifieanee of eytogenetie aberrations in ehildhood aeute myeloid leukaemia. A study ofthe Swiss Paediatrie Oneology Group (SPOG). Eur J HaematoI2007;78(6):468-76. 132. Lie SO, Abrahamsson J, Clausen N, Forestier E, Hasle H, Hovi L, et al. Treatment stratifieation based on initial in vivo response in aeute myeloid leukaemia in ehildren without Down's syndrome: results ofNOPHO-AML trials. Br J Haematol 2003; 122(2):217-25. 133. Small D. FLT3 mutations: biology and treatment. Hematology Am Soe Hematol Edue Program 2006: 178-84. 134. Meshinehi S, Alonzo TA, Stirewalt DL, Zwaan M, Zimmerman M, Reinhardt O, et al. Clinieal implieations of FL T3 mutations in pediatrie AML. Blood 2006; 108(12):3654-61. 135. Reinhardt O, Pekrun A, Lakomek M, Zimmermann M, Ritter J, Creutzig U. Primary myelosareomas are assoeiated with a high rate of relapse: report on 34 ehildren from the aeute myeloid leukaemia-Berlin-Frankfurt-Munster studies. Br J Haematol 2000; 11 0(4):863-6. 136. Re inhardt D, Creutzig U. Isolated myelosareoma in ehildren--update and review. Leuk Lymphoma 2002;43(3):565-74. 137. Kaspers GJ, Zwaan CM. Pediatrie aeute myeJoid leukemia: towards high-quaJity eure of all patients. Haematologiea 2007;92(11): 1519-32. 138. Weiss JR, Kopeeky KJ, Godwin J, Anderson J, Willman CL, Moysieh KB, et aJ. Glutathione S-transferase (GSTMl, GSTT l and GSTAl) polymorphisms and outeomes after treatment for aeute myeloid leukemia: pharrnaeogeneties in Southwest Oneology Group (SWOG) clinieal trials. Leukemia 2006;20(12):2169-71. 139. Bolufer P, Barragan E, Collado M, Cervera J, Lopez JA, Sanz MA. Inf1uenee of genetic polymorphisms on the risk of developing leukemia and on disease progression. Leuk Res 2006;30(12): 1471-91. 140. Eyada TK, El Ghonemy EG, EI Ghoroury EA, EI Bassyouni SO, El Masry MK. Study of genetie polymorphism of xenobiotie enzymes in aeute leukemia. Blood Coagul Fibrinolysis 2007; 18(5):489-95. 141. Creutzig U, Zimmermann M, Lehmbeeher T, GrafN, Hem1ann J, Niemeyer CM, et al. Less toxicity by optimizing ehemotherapy, but not by addition of granulocyte eolonystimulating faetor in ehildren and adoleseents with aeute myeloid leukemia: results of AMLBFM 98. J Clin Oneol 2006;24(27):4499-506. 142. Bonati A, Rizzoli V, Lunghi P. Arsenie trioxide in hematologieal malignaneies: the new diseovery of an aneient drug. Curr Pham1 Bioteehnol 2006;7(6):397-405. 143. Wetzler M, Brady MT, Traey E, Li ZR, Donohue KA, O'Loughlin KL, et al. Arsenie trioxide affeets signal transdueer and aetivator of transeription proteins through alteration of protein tyrosine kinase phosphorylation. Clín Caneer Res 2006; 12(22):6817 -25. 144. Quezada G, Kopp L, Estey E, Wells Rl. AI1-trans-retinoie aeid and arsenie trioxide as initial therapy for aeute promyeloeytie leukemia. Pediatr Blood Caneer 2008;51 (1): 133-5.
E. Mejstříková, strana 69 145. Oleissner B, Sehlenk R, Bornhauser M, Berdel W . Oemtuzumab ozogamiein (mylotarg) for the treatment of aeute myeloid leukemia--ongoi ng trials. Onkologie 2007;30( 12):65 7-62. 146. Reinhardt D, Diekamp S, Fleisehhaek G, Corbaeioglu C, Jurgens H, Dworzak M, et a!. Oemtuzumab ozogamiein (Mylotarg) in ehildren with refraetory or relapsed aeute myeloid leukemia. Onkologie 2004;27(3):269-72. 147. Zwaan CM, Reinhardt D, Corbaeioglu S, van Wering ER, Bokkerink JP, Tissing WJ, et a!. Oemtuzumab ozogamiein: first clinical experienees in ehildren with relapsed/refraetory aeute myeloid leukemia treated on eompassionate-use basis. Blood 2003;101(1O):3868-7l. 148. Aplene R, Alonzo TA, Oerbing RB, Lange BJ, Hurwitz CA, Wells RJ, et a!. Safety and effieaey of gemtuzumab ozogamiein in eombination with ehemotherapy for pediatrie aeute myeloid leukemia: a report from the Children's Oneology Oroup. J Clin Oneol 2008;26( 14):2390-3295. 149. MeKoy JM, Angelotta C, Bennett CL, Tallman MS, Wadleigh M, Evens AM, et a!. Gemtuzumab ozogamiein-associated sinusoidal obstruetive syndrome (SOS): an overview from the researeh on adverse drug events and reports (RADAR) projeet. Leuk Res 2007;31 (5):599-604. 150. Nabhan C, Rundhaugen L, Jatoi M, Riley MB, Boehlke L, Peterson LC, et a!. Oemtuzumab ozogamiein (MylotargTM) is infrequently assoeiated with sinusoidal obstruetive syndrome/veno-oec1usive disease. Ann Oneol 2004; 15(8): 1231-6. 151. Larson RA, Sievers EL, Stadtmauer EA, Lowenberg B, Estey EH, Dombret H, et a!. Final report of the effieaey and safety of gemtuzumab ozogamiein (Mylotarg) in patients with CD33-positive aeute myeloid leukemia in first reeurrenee. Caneer 2005; 104(7): 1442-52. 152. Leung A Y, Liang R. Thromboeytopenia after gemtuzumab is reversible by intravenous immunoglobulin. Leukemia 2005; 19(6): 1077-8. 153. Sievers EL. Native antibody and antibody-targeted ehemotherapy for aeute myeloid leukemia. Adv Pharmaeol 2004;51: 169-83. 154. Mulford D. Antibody therapy for aeute myeloid leukemia. Semin Hematol 2008;45(2): 104-9. 155. Maslak PG, Jureie JO, Seheinberg DA. Monoc1onal antibody therapy of APL. Curr Top Mierobiol Immunol 2007;313 :205-19. 156. Lin Y, Pagel JM, Axworthy D, Pantelias A, Hedin N, Press OW. A genetieally engineered anti-CD45 single-ehain antibody-streptavidin fusion protein for pretargeted radioimmunotherapy of hematologie malignaneies. Caneer Res 2006;66(7):3884-92. 157. Krause DS, Van Etten RA. Right on target: eradieating leukemie stem eells. Trends Mol Med 2007;13(11):470-8l. 158. Hauswirth AW, Florian S, Printz D, Sotlar K, Krauth MT, Fritseh O, et al. Expression ofthe target reeeptor CD33 in CD34+/CD38-/CDI23+ AML stem eells. Eur J Clin Invest 2007;37(1 ):73-82. 159. Jin L, Hope KJ, Zhai Q, Smadja-Joffe F, Diek JE. Targeting ofCD44 eradieates human aeute myeloid leukemie stem eells. Nat Med 2006; 12( 10): 1167-74. 160. Sehittenhelm MM, Shiraga S, Sehroeder A, Corbin AS, Oriffith D, Lee FY, et al. Dasatinib (BMS-354825), a dual SRC/ ABL kinase inhibitor, inhibits the kinase aetivity of wild-type, juxtamembrane, and aetivation loop mutant KIT isoforn1s assoeiated with human malignaneies. Caneer Res 2006;66( 1):473-81. 161. Cairoli R, Beghini A, Morello E, Orillo O, Montillo M, Larizza L, et a!. Imatinib mesylate in the treatment of Core Binding Faetor leukemias with KIT mutations. A report of three eases. Leuk Res 2005;29(4):397-400. 162. Pratz K, Levis M. Ineorporating FL T3 inhibitors into aeute myeloid leukemia treatment regimens. Leuk Lymphoma 2008;49(5):852-63.
E. Mejstříková , strana 70 163 . Knapper S, Mills Kl, Oilkes AF, Austin Sl , Walsh V, Bumett AK. The effects of lestaurtinib (CEP70 1) and PKC412 on primary AML blasts: the induction of cytotoxicity varies with dependence on FLT3 signaling in both FLT3-mutated and wild-type cases. Blood 2006; 108(10):3494-503. 164. Serrano E, Camicer M l , Lasa A, Orantes V, Pena J, Brunet S, et a!. Epigenetic-based treatments emphasize the biologic differences of core-binding factor acute myeloid leukemias. Leuk Res 2008;32(6):944-53. 165 . Kuendgen A, Schmid M, Schlenk R, Knipp S, Hildebrandt B, Steidl C, et al. The histone deacetylase (HDAC) inhibitor valproic acid as monotherapy or in combination with all-trans retinoic acid in patients with acute myeloid leukemia. Cancer 2006; 106(1): 11 2-9. 166. Yu WP, Scott SA, Dong WF. lnduction oflDl expression and apoptosis by the hi stone deacetylase inhibitor (trichostatin A) in human acute myeloid leukaemic cells. Cell Prolif2008;41 (1 ):86-97. 167. Klisovic MI, Maghraby EA, Parthun MR, Ouimond M, Sklenar AR, Whitman SP, et al. Depsipeptide (FR 901228) promotes histone acetylation, gene transcription, apoptosis and its activity is enhanced by DNA methyl transferase inhibitors in AMLl /ETO-positive leukemic cell s. Leukemia 2003 ;17(2):350-8 . 168. Oarcia-Manero O, Assouline S, Cortes J, Estrov Z, Kantarjian H, Yang H, et a!. Phase I study of the oral isotype specific histone deacetylase inhibitor MOCDO 103 in leukemia. Blood 2008. 169. Koschmieder S, Agrawal S, Radomska HS, Huettner CS, Tenen DO, Ottmann 00, et a!. Oecitabine and vitamin D3 differentially affect hematopoietic transcription factors to induce monocytic differentiation. lnt J Oncol 2007;30(2):349-55. 170. Muller CI, Ruter B, Koemer HP Lubbert M. DNA hypermethylation ofmyeloid cells, a novel therapeutic target in MDS and AML. Curr Pharm Biotechnol 2006;7(5):315-21. 171 . Oalm O, Wilop S, Luders C, Jost E, Oehbauer O, Herman JO , et al. Clinical implications of aberrant DNA methylation pattems in acute myelogenous leukemia. Ann Hemato12005;84 Suppl 1:39-46. 172. Scott SA, Dong WF, Ichinohasama R, Hirsch C, Sheridan D, Sanche SE, et al. 5-Aza2'-deoxycytidine (decitabine) can relieve p21 W AF I repression in human acute myeloid leukemia by a mechanism involving release of histone deacetylase 1 (HDAC 1) without requiring p21 WAFl promoter demethylation. Leuk Res 2006;30(1):69-76. 173. Hubeek I, Peters OJ , Broekhuizen R, Zwaan CM, Kaaijk P, van Wering ES , et al. In vitro sensitivity and cross-resistance to deoxynucleoside analogs in childhood acute leukemia. Haematologica 2006;91(1) : 17-23. 174. Kantarjian H, Oandhi V, Cortes J, Verstovsek S, Du M, Oarcia-Manero O, et a!. Phase 2 clinical and pharmacologic study of c10farabine in patients with refractory or relapsed acute leukemia. Blood 2003; 102(7):23 79-86. 175. Creutzig U, Reinhardt D, Zimmermann M. Prognostic relevance ofrisk groups in the pediatrie AML-BFM trials 93 and 98. Ann Hematol 2004;83 Suppl I :S 112-6. 176. SchoU C, Breitinger H, Schlenk RF, Dohner H, Frohling S, Dohner K. Development of a real-time RT -PCR assay for the quantification of the most frequent MLL! AF9 fusion types resulting from translocation t(9;11)(p22;q23) in acute myeloid leukemia. Oenes Chromosomes Cancer 2003 ;38(3):274-80. 177. SchoU C, Schlenk RF, Eiwen K, Dohner H, Frohling S, Dohner K. The prognostic value of MLL-AF9 detection in patients with t(9 ; II )(p22;q23)-positive acute myeloid leukemia. Haematologica 2005 ;90( 12): 1626-34. 178. Kem W, Schoch C, Haferlach T, Schnittger S. Monitoring ofminimal residual disease in acute myeloid leukemia. Crit Rev Oncol Hematol 2005;56(2):283-309.
E. Mejstříková, strana 71 179. Kusec R, Laczika K, Knobl P, Friedl J, Oreinix H, Kahls P, et al. AMLlIETO fusion mRNA can be detected in remission blood samples oť aU patients with t(8;21) acute myeloid leukemia after chemotherapy or autologous bone marrow transplantation. Leukemia 1994;8(5):735-9. 180. Mitterbauer M, Mitterbauer-Hohendanner O, Sperr WR, Kalhs P, Oreinix HT, Fonatsch C, et al. Molecular disease eradication is a prerequi site ťor long-term remission in patients with t(8;21) positive acute myeloid leukemia: a single center study. Leuk Lymphoma 2004;45(5):971-7. 181. Trka J, Kalinova M , Hrusak O, Zuna J, Krejci O, Madzo J, et al. Real-time quantitative PCR detection oť WT1 gene expression in children with AML: prognostic significance, correlation with disease status and residual disease detection by flow cytometry. Leukemia 2002; 16(7): 1381-9. 182. Ostergaard M, Olesen LH, Hasle H, Kj eldsen E, Hokland P. WT1 gene expression: an excellent tool for monitoring minimal residual disease in 70% oť acute myeloid leukaemia patients - results ťrom a single-centre study. Br J Haematol 2004; 125(5):590-600. 183. Hosen N, Sonoda Y, Oji Y, Kimura T Minamiguchi H, Tamaki H, et al. Very low ťrequencies oť human normal CD34+ haematopoietic progenitor cells express the Wilms' tumour gene WT1 at levels similar to those in leukaemia cells. Br J Haematol 2002;116(2):409-20. 184. Pine SR, Ouo Q, Yin C, Jayabose S, Levendoglu-Tugal O, Ozkaynak MF, et al. OA TA 1 as a new target to detect minimal residual disease in both transient leukemia and megakaryoblastic leukemia ofDown syndrome. Leuk Res 2005;29(11):1353-6. 185. Hitzler J, Zipursky A. OATA 1 mutations as clonal markers oťminimal residual disease in acute megakaryoblastic leukemia of Down syndrome--a new tool with significant potential applications. Leuk Res 2005 ;29(11):1239-40. 186. Hasle H, Lund B, Nyvold CO, Hokland P, Ostergaard M . WTl gene expression in children with Down syndrome and transient myeloproliťerative disorder. Leuk Res 2006;30(5):543-6. 187. Bacher U, Kem W, Schoch C, Schnittger S, Hiddemann W, Haferlach T. Evaluation oť complete disease remission in acute myeloid leukemia: a prospective study based on cytomorphology, interphase fluorescence in situ hybridization, and immunophenotyping during ťollow-up in patients with acute myeloid leukemia. Cancer 2006; 106(4):839-47. 188. Kem W, Danhauser-Riedl S, Ratei R, Schnittger S, Schoch C, Kolb HJ, et al. Detection oť minimal residual disease in unselected patients with acute myeloid leukemia using multiparameter fl ow cytometry for definition of leukemia-associated immunophenotypes and determination of their frequencies in normal bone marrow. Haematologica 2003;88(6):646-53. 189. Langebrake C, Creutzig U, Dworzak M, Hrusak O, Mejstrikova E, Oriesinger F, et al. Residual disease monitoring in childhood acute myeloid leukemia by multiparameter flow cytometry: the MRD-AML-BFM Study Oroup. J Clin Oncol 2006;24(22):3686-92. 190. Langebrake C, Brinkmann I, Teigler-Schlegel A, Creutzig U, Oriesinger F, Puhlmann U, et al. Immunophenotypic difťerences between diagnosis and relapse in childhood AML: Implications for MRD monitoring. Cytometry B Clin Cytom 2005 ;63(1): 1-9. 191. Kem W, Voskova D, Schoch C, Hiddemann W, Schnittger S, Haferlach T. Determination of relapse risk based on assessment of minimal residual disease during complete remission by multiparameter flow cytometry in unselected patients with acute myeloid leukemia. Blood 2004; 104(1 0):3078-85. 192. Reinhardt D, Langebrake C, Creutzig U, Vormoor J, Brune C, Thorwesten M, et al. [Minimal residual disease in acute myeloid leukemia in children--standardization and
E. Mejstříková, strana 72 evaluation of immunophenotyping in the AML-BFM-98 study]. Klin Padiatr 2002;214(4): 179-87. 193. Kern W, Voskova O, Sehnittger S, Sehoeh C, Hiddemann W, Haferlaeh T. Four-fold staining including C045 gating improves the sensitivity of m ultiparameter f10w eytometrie assessment of minimal residual disease in patients with acute myeloid leukemia. Hematol J 2004;5(5):410-8. 194. Kern W, Haferlach C, Haferlach T, Sehnittger S. Monitoring of minimal residual disease in aeute myeloid leukemia. Cancer 2008; 112(1 ):4- 16. 195. San-Miguel JF, Vidriales MB, Orfao A. lmmunological evaluation of minimal residual disease (MRO) in acute myeloid leukaemia (AML). Best Pract Res CI in Haematol 2002;15(1):105-18. 196. Vidriales MB, San-Miguel JF, Orfao A Coustan-Smith E, Campana O. Minimal residual disease monitoring by f10w cytometry. Best Pract Res Clin Haematol 2003; 16(4):599-612. 197. Sievers EL, Loken MR. Oetection of minimal residual disease in acute myelogenous leukemia. J Pediatr Hematol Oncol 1995; 17(2): 123-33. 198. Sievers EL, Lange BJ, Alonzo TA, Gerbing RB, Bernstein ID, Smith FO, et al. lmmunophenotypic evidence of leukemia after induction therapy prediets relapse: results from a prospective Children's Cancer Group study of252 patients with acute myeloid leukemia. Blood 2003; 10 I (9):3398-406. 199. Coustan-Smith E, Ribeiro RC, Rubnitz JE, Razzouk BI, Pui CH, Pounds S, et al. Clinieal significanee of residual disease during treatment in childhood acute myeloid leukaemia. Br J Haematol 2003; 123(2):243-52. 200. van Rhenen A, Feller N, Kelder A, Westra AH, Rombouts E, Zweegman S, et al. High stem cell frequency in acute myeloid leukemia at diagnosis predicts high minimal residual disease and poor survival. Clin Cancer Res 2005;11(18):6520-7. 201. van Rhenen A, Moshaver B, Kelder A, Feller N, Nieuwint A W, Zweegman S, et al. Aberrant marker expression patterns on the C034+C038- stem cell compartment in acute myeloid leukemia allows to distinguish the malignant from the normal stem cell eompartment both at diagnosis and in remission. Leukemia 2007;21(8): 1700-7. 202. van Rhenen A, van Oongen GA, Kelder A, Rombouts EJ, Feller N, Moshaver B, et al. The novel AML stem cell assoeiated antigen CLL-1 aids in discrimination between normal and leukemie stem cells. Blood 2007; 110(7):2659-66. 203. Wood BL. Myeloid malignancies: myelodysplastic syndromes, myeloproliferative disorders, and aeute myeloid leukemia. Clin Lab Med 2007;27(3):551-75, vii. 204. Rubnitz JE. Childhood acute myeloid leukemia. Curr Treat Options Oncol 2008;9(1):95-105. 205. Florian S, Sonneek K, Hauswirth A W, Krauth MT, Sehernthaner GH, Sperr WR, et al. Oetection of moleeular targets on the surface of C034+/C038-- stem cells in various myeloid malignancies. Leuk Lymphoma 2006;47(2):207-22. 206. Costello R, Mallet F, Chambost H, Sainty O, Amoulet C, Gastaut JA, et al. The immunophenotype of mini maIly differentiated aeute myeloid leukemia (AML-MO): reduced immunogenieity and high frequeney ofC034+/C038- leukemie progenitors. Leukemia 1999; 13(10): 1513-8. 207. Attarbaschi A, Mann G, Konig M, Steiner M, Strehl S, Sehreiberhuber A, et at Mixed lineage leukemia-rearranged ehildhood pro-B and COl O-negative pre-B acute lymphoblastic leukemia eonstitute a distinet clinical entity. Cl in Cancer Res 2006; 12( 10):2988-94. 208. Gleissner B, Goekbuget N, Rieder H, Arnold R, Sehwartz S, Oiedrieh H, et al. CD 10pre-B acute Iymphoblastic leukemia (ALL) is a distinet high-risk subgroup of adult ALL
E. Mejstříková, strana 73 assoeiated with a high frequeney of MLL aberrations: results of the Oerman Multieenter Trials for Adult ALL (OMALL). Blood 2005; 106(13) :4054-6. 209. Orimwade D, Outram SV, Flora R, Ings SJ, Pizzey AR, Morilla R, et a!. The Tlineage-affiliated CD2 gene lies within an open ehromatin environment in aeute promyeloeytie leukemia eells. Caneer Res 2002;62(16) :4730-5 . 2 10. Oreaves MF. Differentiation-linked leukemogenesis in lymphoeytes. Science 1986;234( 4777):697-704. 21 1. Baruehel A, Cayuela JM, Ballerini P, Landman-Parker J, Cezard V, Firat H, et al. The majority of myeloid-antigen-positive (My+) ehildhood B-eell preeursor aeute lymphoblastie leukaemias expres s TEL-AMLl fusion transeripts. Br 1 HaematoI1997;99(1):101-6. 212. Pui CH, Rubnitz lE, Haneoek ML, Downing lR, Raimondi SC, Rivera OK, et al. Reappraisal of the elinieal and biologie signifieanee of myeloid-assoeiated antigen expression in ehildhood aeute lymphoblastie leukemia. 1 Clin OneoI1998;16(12):3768-73. 213. Wiersma SR, Ortega 1, Sobel E, Weinberg Kl. Clinieal importanee ofmyeloid-antigen expression in aeute lymphoblastie leukemia of ehildhood. N Engl 1 Med 1991 ;324( 12):800-8. 214. Fink FM, Koller U, Mayer H, Haas OA, Orumayer-Panzer ER, Urban C, et al. Prognostie signifieanee of myeloid-assoeiated antigen expression on blast eells in ehildren with aeute lymphoblastie leukemia. The Austrian Pediatrie Oneology Oroup. Med Pediatr OneoI1993;21(5):340-6. 215. Sobol RE, Miek R , Royston I, Davey FR, ElIison RR, Newman R, et al. Clinieal importanee of myeloid antigen expression in adult aeute lymphoblastie leukemia. N Engl 1 Med 1987;316(18): 11 11-7. 216. Putti MC, RondeLli R, Coeito MO, Arieo M, Sainati L, Conter V, et a!. Expression of myeloid markers laeks prognostie impaet in ehildren treated for aeute lymphoblastie leukemia: Italian experienee in AIEOP-ALL 88-91 studies. Blood 1998;92(3):795-801. 217. Uekun FM, Sather HN, Oaynon PS, Arthur DC, Trigg ME, Tubergen DO, et al. Clinieal features and treatment outeome of ehildren with myeloid antigen positive aeute lymphoblastie leukemia: a repol1 from the Children's Caneer Oroup. Blood 1997;90(1 ):28-35. 218. Legrand 0, Perrot lY, Simonin 0, Baudard M, Cadiou M, Blanc C, et al. Adult biphenotypie aeute leukaemia: an entity with poor prognosis whieh is related to uníavourable eytogeneties and P-glyeoprotein over-expression. Br 1 Haematol 1998; 100(1): 147-55. 219. Matutes E, Morilla R , Farahat N, Carbonell F, Swansbury 1, Dyer M, et al. Definition of aeute biphenotypie leukcmia. Haematologiea 1997;82(1 ):64-6. 220. Owaidah TM, AI Beihany A, Iqbal MA, Elkum N, Roberts OT. Cytogeneties, moleeular and ultrastruetural eharaeteristies of biphenotypie aeute leukemia identified by the EOIL seoring system. Leukemia 2006;20(4):620-6. 22l. Killiek S, Matutes E, Powles RL, Hamblin M, Swansbury 1, Treleaven 10, et al. Outeome ofbiphenotypie aeute leukemia. Haematologiea 1999;84(8):699-706. 222. Weir EO, Ali Ansari-Lari M, Batista DA, Oriffin CA, Fuller S, Smith BD, et al. Aeute bilinealleukemia: a rare disease with poor outeome. Leukemia 2007;21(11):2264-70. 223. Krawezuk-Rybak M, Zak 1, Jaworowska B. A lineage switeh from AML to ALL with persistent transloeation t(4;11) in eongenitalleukemia. Med Pediatr Oneol 2003;41(1):95-6. 224. Louniei A, Cony-Makhoul P, Dubus P, Laeombe F, Merlio lP, Reiffers 1. Lineage switeh from aeute myeloid leukemia to aeute lymphoblastie leukemia: report of an adult ease and review of the literature. Am 1 Hematol 2000;65(4):319-2l. 225. Stasik C, Oanguly S, Cunningham MT, Hagemeister S, Persons DL. Infant aeute lymphoblastie leukemia with t(11;16)(q23;pI3.3) and lineage switeh into aeute monoblastie leukemia. Caneer Oenet Cytogenet 2006; 168(2): 146-9.
E. Mejstříková, strana 74 226. Fujisaki H, Hara J, Takai K, Nakanishi K, Matsuda Y, Ohta H, et a!. Lineage switch in childhood leukemia with monosomy 7 and reverse of lineage switch in severe combined immunodeficient mice. Exp Hematol 1999;27(5):826-33 . 227. Cutting R, Taj M, Vora A. Childhood precursor B-cell acute lymphoblastic leukaemia and monosomy 7 with phenotypic shift at relapse ; evidence for the stem cell origins of monosomy 7. Br J Haematol 2008; 140(5):584-6. 228 . Reardon DA, Hanson CA, Roth MS , Castle VP. Lineage switch in Philadelphia chromosome-positive acute lymphoblastic leukemia. Cancer 1994;73(5): 1526-32. 229. Bierings M, Szczepanski T, van Wering ER, Willemse MJ, Langerak A W, Revesz T, et aJ. Two consecutive immunophenotypic switches in a child with immunogenotypically stable acute leukaemia. Br J HaematoI200 1~1 13(3):757-62. 230. Rhein P, Scheid S, Ratei R, Hagemeier C, Seeger K, Kirschner-Schwabe R, et al. Gene expression shift towards nomlal B cells, decreased proliferative capacity and distinct surface receptors characterize leukemic blasts persisting during induction therapy in childhood acute lymphoblastic leukemia. Leukemia 2007;21(5):897-905 . 231. Gaipa G, Basso G, Maglia O, Leoni V, Faini A, Cazzaniga G, et a!. Drug-induced immunophenotypic modulation in childhood ALL: implications for minimal residual disease detection. Leukemia 2005 ; 19( 1):49-56. 232. Gaipa G, Basso G, Aliprandi S, Migliavacca M, Vallinoto C, Maglia O, et a!. Prednisone induces immunophenotypic modulation of CD 1O and CD34 in nonapoptotic Bcell precursor acute lymphoblastic leukemia cells. Cytometry B Clin Cytom 2008. 233. Loken MR, Shah va, Dattilio KL, Civin CI. Flow cytometric analysis of human bone marrow. II. Normal B lymphocyte development. Blood 1987;70(5): 1316-24. 234. Loken MR, Shah va, Hollander Z, Civin CI. Flow cytometric analysis of normal B Iymphoid development. Pathol Immunopathol Res 1988;7(5):357-70. 235 . Shah va, Civin CI, Loken MR. Flow cytometric analysis of human bone marrow. IV. Differential quantitative expression ofT-200 common leukocyte antigen during normal hemopoiesis. J ImmunoI1988;140(6):1861-7. 236. Balduzzi A, Rossi V, Corral L, Bonanomi S, Longoni D, Rovelli A, et a!. Molecular remission induced by gemtuzumab ozogamicin associated with donor lymphocyte infusions in t( 4; ll) acute lymphoblastic leukemia relapsed after transplantation. Leukemia 2003;17(11):2247-8 . 237. Cotter M, Rooney S, O'Marcaigh A, Smith OP. Successful use of gemtuzumab ozogamicin in a child with relapsed CD33-positive acute lymphoblastic leukaemia. Br J HaematoI2003;122(4):687-8. 238. Zwaan CM, Reinhardt D Jurgens H, Huismans DR, Hahlen K, Smith OP, et a!. Gemtuzumab ozogamicin in pediatric CD33-positive acute lymphoblastic leukemia: first clinical experiences and relation with cellular sensitivity to single agent calicheamicin. Leukemia 2003; 17(2):468-70. 239. Goemans BF, Zwaan CM, Vijverberg SJ, Loonen AH, Creutzig U, Hahlen K, et a!. Large interindividual differenees in eellular sensitivity to ealicheamicin may intluence gemtuzumab ozogamiein response in aeute myeloid leukemia. Leukemia 2008 . 240. Lamba JK, Pounds S, Cao X, Downing JR, Campana D, Ribeiro RC, et al. Coding polymorphisms in CD33 and response to gemtuzumab ozogamiein in pediatrie patients with AML: a pilot study. Leukemia 2008. 241. Boghaert ER, Khandke KM, Sridharan L, Dougher M, Dijoseph JF, Kunz A, et a!. Determination of pharmaeokinetic values of calicheamiein-antibody eonjugates in mice by plasmon resonance analysis of small (5 mul) blood samples. Caneer Chemother Pharmaeol 2008;61 (6): 1027-35.
E. Mejstříková, strana 75 242. DiJoseph JF, Armellino DC, Boghaert ER, Khandke K, Dougher MM, Sridharan L, et a1. Antibody-targeted chemotherapy with CMC-544: a CD22-targeted immunoconjugate of calicheamicin for the treatment of B-Iymphoid malignancies. Blood 2004; 103(5): 1807-14. 243. Carnahan J, Wang P, Kendall R, Chen C, Hu S, Boone T, et a1. Epratuzumab, a humanized monoclonal antibody targeting CD22: characterization of in vitro properties. Clin Cancer Res 2003;9(10 Pt 2):3982S-90S. 244. Raetz EA, Cairo MS, Borowitz MJ, Blaney SM, Krailo MD, Leil TA, et a1. Chemoimmunotherapy reinduction with epratuzumab in children with acute lymphoblastic leukemia in marrow relapse: a Children's Oncology Oroup Pilot Study. J Clin Oncol 2008;26(22):3756-62 . 245. Oramatzki M, Burger R, Strobel 0 , Trautmann U, Bartram CR, Helm 0, et a1. Therapy with OKT3 monoclonal antibody in refractory T cell acute Iymphoblastic leukemia induces interleukin-2 responsiveness. Leukemia 1995;9(3):382-90. 246. van Dongen JJ, Seriu T, Panzer-Orumayer ER, Biondi A, Pongers-Willemse MJ , Corral L, et a1. Prognostic value of minimal residual disease in acute Iymphoblastic leukaemia in childhood. Lancet 1998;352(9142): 1731-8. 247. Pui CH, Sandlund JT, Pei D, Campana D, Rivera OK, Ribeiro RC, et a1. Jmproved outcome for children with acute Iymphoblastic leukemia: results ofTotal Therapy Study XmB at St Jude Children's Research Hospita1. Blood 2004; 104(9):2690-6. 248. Coustan-Smith E, Ribeiro RC, Stow P, Zhou Y, Pui CH, Rivera OK, et a1. A simplified flow cytometric assay identifies children with acute lymphoblastic leukemia who have a superior c1inical outcome. Blood 2006; 108(1 ):97 -102. 249. Borowitz MJ , Devidas M, Hunger SP, Bowman WP, Carroll AJ, Carroll WL, et al. Clinical significance of minimal residual disease in childhood acute Iymphoblastic leukemia and its relationship to other prognostic factors: A Children's Oncology Oroup study. Blood 2008. 250. Campana D, Coustan-Smith E. Advances in the immunological monitoring of childhood acute Iymphoblastic leukaemia. Best PractRes CI in HaematoI2002;15(1):1-19. 251. Dworzak MN, Froschl 0, Printz D, Zen LD, Oaipa 0, Ratei R, et a1. CD99 expression in T -lineage ALL: implications for flow cytometric detection of minimal residual disease. Leukemia 2004; 18(4):703-8. 252. Mejstrikova E, Kalina T, Trka J, Stary J, Hrusak O. Correlation of CD33 with poorer prognosis in childhood ALL implicates a potential of anti-CD33 frontline therapy. Leukemia 2005; 19(6): 1092-4. 253. Kalina T, Vaskova M, Mejstrikova E, Madzo J, Trka J, Stary J, et al. Myeloid antigens in childhood Iymphoblastic leukemia: cIinical data point to regulation of CD66c distinct from other myeloid antigens. BMC Cancer 2005;5(1):38. 254. Coustan-Smith E, Behm FO, Sanchez J, Boyett JM, Hancock ML, Raimondi SC, et al. Immunological detection of minimal residual disease in children with acute lymphoblastic leukaemia. Lancet 1998·351(9102):550-4. 255. Coustan-Smith E, Sancho J, Hancock ML, Boyett JM, Behm FO, Raimondi Se, et al. Clinical importance of minimal residual disease in childhood acute lymphoblastic leukemia. Blood 2000;96(8):2691-6. 256. Oaipa 0 , Basso 0, Ratei R, Biondi A, Dworzak M. Reply to van der Sluijs-Oelling et al. Leukemia 2005; 19(12):2351-2. 257. van der Sluijs-Oelling AJ, van der Velden VH, Roeffen ET, Veerman AJ, van Wering ER. lmmunophenotypic modulation in childhood precursor-B-ALL can be mimicked in vitro and is related to the induction of cell death. Leukemia 2005; 19(10): 1845-7. 258. van der Sluijs-Oelling AJ, van der Vel den VH, Roeffen ET, Veerrnan AJ, van Wering ER. Reply to Oaipa et AI. Leukemia 2005 ;1 9(10):1858.
E. Mejstříková, strana 76 259. Flohr T, Schrauder A, Cazzaniga G, Panzer-Grumayer R, van der Velden V, Fischer S, et a!. Minimal residual disease-directed risk stratification using real-time quantitative PCR analysis of immunoglobulin and T -cell receptor gene rearrangements in the international multicenter trial AIEOP-BFM ALL 2000 for childhood acute lymphoblastic leukemia. Leukemia 2008. 260. Szczepanski T, Pongers- Willemse MJ, Langerak A W, Harts W A, Wijkhuijs AJ, van Wering ER, et a!. 19 heavy chain gene rearrangements in T-cell acute lymphoblastic leukemia exhibit predominant DH6-19 and DH7-27 gene usage, can result in complete V-D-J rearrangements, and are rare in T -cell receptor alpha beta lineage. Blood 1999;93(12):407985. 261. van Dongen Jl , Langerak AW, Bruggemann M, Evans PA, Hummel M, Lavender FL, et al. Design and standardization of PCR primers and protocols for detection of clonal immunoglobulin and T -cell receptor gene recombinations in suspect lymphoproliferations: report ofthe BlOMED-2 Concerted Action BMH4-CT98-3936. Leukemia 2003; 17(12):2257317. 262. Szczepanski T, Flohr T, van der Velden VH , Bartram CR, van Dongen Jl. Molecular monitoring of residual disease using antigen receptor genes in childhood acute lymphoblastic leukaemia. Best Pract Res Clin Haematol 2002; 15(1 ):37-57. 263. Beillard E, Pallisgaard N, Van Der Velden VH, Bi W, Dee R, Van Der Schoot E, et a!. Evaluation of candidate control genes for diagnosis and residual disease detection in leukemic patients using 'real-time' quantitative reverse-transcriptase polymerase chain reaction (RQPCR) - a EUl·ope against cancer program. Leukemia 2003. 264. van der Velden VH, Cazzaniga G, Schrauder A, Hancock J, Bader P, PanzerGrumayer ER, et a!. Analysis of minimal residual disease by Ig/TCR gene rearrangements: guidelines for interpretation ofreal-time quantitative PCR data. Leukemia 2007. 265 . Fronkova E, Madzo J, Zuna J, Reznickova L, Muzikova K, Hrusak O, et a!. TELlAML 1 real-time quantitative reverse transcriptase PCR can complement minimal residual disease assessment in childhood ALL. Leukemia 2005; 19(7): 1296-7. 266. Primo D, Sanchez ML, Espinosa AB, Tabemero MD, Rasillo A, Sayagues JM, et a!. Lineage involvement in chronic myeloid leukaemia: comparison between MBCRI ABL and mBCRlABL cases. Br J HaematoI2006;132(6):736-9. 267. Raanani P, Trakhtenbrot L, Rechavi G, Rosenthal E, Avigdor A, Brok-Simoni F, et a!. Philadelphia-chromosome-positive T-lyrnphoblastic leukemia: acute leukemia or chronic myelogenous leukemia blastic crisis. Acta Haematol 2005; 113(3): 181-9. 268. Anastasi J, Feng J, Dickstein JI, Le Beau MM, Rubin CM, Larson RA, et a!. Lineage involvement by BCRI ABL in Ph+ lymphoblastic leukemias: chronic myelogenous leukemia presenting in lymphoid blast vs Ph+ acute lymphoblastic leukemia. Leukemia 1996; 10(5):795-802. 269. Lion T. Detection of impending graft rejection and relapse by lineage-specific chimerism analysis. Methods Mol Med 2007; 134: 197-216. 270. Kristt D, Israeli M, Narinski R, Or H, Yaniv T, Stein J, et a!. Hematopoietic chimerism monitoring based on STRs: quantitative platform performance on sequential samples. J Biomol Tech 2005J6(4):380-91. 271. Kristt D, Klein T. Reliability of quantitative chimerism results : assessment of sample performance using novel parameters. Leukemia 2006;20(6): 1169-72. 272. Kristt D, Stein J, Yaniv I, Klein T. Assessing quantitative chimerism longitudinally: technical considerations, clinical applications and routine feasibility. Bone Marrow Transplant 2007;39(5):2 55-68.
E. Mejstříková, strana 77 273. Fronkova E, Mejstrikova E, Avigad S, Chi k KW, Castillo L, Manor S, et al. Minimal residual disease (MRD) analysis in the non-MRD-based ALL IC-BFM 2002 proto col for childhood ALL: is it possible to avoid MRD testing? Leukemia 2008 . 274. Sramkova L, Muzikova K, Fronkova E, Krejci 0 , Sedlacek P, Formankova R, et al. Detectable minimal residual disease before all ogeneic hematopoietic stem cell transplantation predicts extremely poor prognosis in children with acute lymphoblastic leukemia. Pediatr Blood Cancer 2006. 275. Coustan-Smith E, Gajjar A, Hijiya N, Razzouk BI, Ribeiro RC , Rivera GK, et al. Clinical significance of minimal residual disease in chi!dhood acute lymphoblastic leukemia after first relapse. Leukemia 2004; 18(3 ):499-504. 276. von Stackelberg A, Seeger K, Henze G, Eckert C. Cli nica! significance of minima! residual disease in childhood acute lymphoblastic leukemia after first relapse. Leukemia 2004;18(10): 1727-8 ; author reply 1728-9. 277. Kerst G, Kreyenberg H, Roth C, Well C, Dietz K, Coustan-Smith E, et al. ConculTent detection of minimal residual d isease (MRD) in childhood acute lymphoblastic leukaemia by f10w cytometry and real-time P R. Br J HaematoI2005 ;128(6):774-82. 278. Malec M, van der Velden VH, Bjorklund E, Wijkhuijs 1M, Soderhall S, Mazur 1, et al. Analysis of minimal residual disease in childhood acute lymphoblastic leukemia: comparison between RQ-PCR analysis ofIg/TcR gene rearrangements and multicolor flow cytometric immunophenotyping. Leukemia 2004; 18(1 O): 1630-6. 279. Szczepanski T. Why and how to quantify minimal residual disease in acute Iymphoblastic leukemia? Leukemia 2007;21 (4 ):622-6. 280. Rubnitz lE, Razzouk BI, Lensing S, Pounds S, Pui CH , Ribeiro RC. Prognostic factors and outcome of recurrence in childhood acute myeloid leukemia. Cancer 2007; 109(1): 157-63.
6.2.
Přiložené
Příl o ha
publikace s impact faktorem
1
Correlation of CD33 with poorer prognosis in childhood ALL implicates a potential of anti-CD33 frontline therapy E
Mejstříková,
T Kalina, 1 Trka, 1 Starý, O Hrušák
Leukemia 2005 (fF 6, 924)
E. Mejstříkov á, strana 78
Correspondence
Correlation of C033 with poorer prognosis m childhood ALL implicates a potential of anti-C033 frontline therapy
Leukemla (2005) 19, 1092-1094. doi:10 .1038/sj .leu.2403737 Published online 14 April 2005 TO TH E EDITOR
There has been a co ntroversy about the prognosti c impact of myeloid antigens (MyAgs) in All . The issue has now regained sigllificallce since an anti-CD33 monoclonal antibody (mAb) is avai lable for clinica l treatment, w hich is considered for AMl as we ll as for CD33 P o, Al l .' - J We evalu ated the prognostic impact of MyAgs in uniform ly treated palients. From 9/1996 to 10/2002, 343 children with All were enroll d in a Czec h nationwide sludy All-BFM95. In 327 paticnts (96 0/., ) immunophenotyping and molecular gelletics were performed centra lly in our reference labs . Predn isone gooo responders (PCR) with WBC at diagnosis < 20000/pl and age 1- 5 ycars were assigned inlo standard risk group (SRC). High RC (HRC) corresponded to prednisone poor response (PPR) Ol' no remission at dav 33 or BC R/ABl or MLUAF4 fusions. The remaining patients were in the intermediate RC (IRC). Full stJt istics delails are in the Supplementary Information. Immunophenotyping was períormed on bone marrow (B MI samples at diagnosis on flow cyto metry (FO using FACSCalihur (BD San Jose, CA, USA). Informed conse nt was obtained from the pati ents and/or their guardians . A standard panel of mAbs wa s used for FC immunodiagnostics, including: CD 13 (Sll Dl Imm ullotech, only in B-cell precursor (Bep) Al l ), CD 15 (MMA BDI, CD33 (D3Hl60.251 Immunotech) and CD65 (88 H 7 Immunotech), unvaryingly throughout the study . ln a mu ltivar iate analysis of BCP All patients (Table 1I, CD33 was the only MyAg with an independent prognostic im pact. A simi lar analysis on T AlL (n - 45) did not show , l prognostic impact of li sted M yAgs, therefore, we excluded I All cases from further analyscs. Three cases (0.92%) were excl uded due to the absence of a representative leukemic popul ation by FC and one case was excluded si nce CD33 was not assessed. RFS was signifl ca ntly worse in BCP ALl (tola l = 278) pati ents with higher CD33 ros percE'ntage (patients with higher and lower CD33 percentage: 5-yea r RFS 57 ± 6 vs 87 ± 3%, med ian fo llow -up 3.3 and 4.0 years, n 98 and n 180, respectively; culoff level: 10%; Fi gure 1) . A similar analysis with a cutoff level of 20% would reveal significant differcncc (P = 0.0073, not shown ). When analyzed separately by RC , this difference remai ns apparent and remains significanl for SRC and IRC (5-year RFS in cases with lower and higher CD33 percentage, cutoff 10%: SRC, 88 ± 5 vs 59± 11 %, n = 70 and n = 36, P- O.Ol ; IRC , 92±3 v 60 ± 9°;'" n = 88 <:Ind 1l - 49, P= 0.00073; HRC, 58±13 vs 38 ± 16%, n=22 and n=13 , P=0 .076). The difference in IRC was significant at any cutoff level from lOto 70% (always P<0 .02 , data not shown). Thu s, CD33 is prognostically important especially in IRC, which is the most
=
=
Correspondence: Dr O Hrusak, ClIP - DepartmL'nt or Immunology, V uva lu 84, 150 06 Praha 5, Cz ch Republic; Fax: + 4202 2443 5962; E-mail : Ondrej .Hrusak @lrmOIOI.cuni.cz Received 1 Novcmber 2004; accepted 1 March 2005; Published online 14 April 2005
abundant RC among relapsed cases but paticnts al risk are ot herwise difficu lt to discern The BCP All ca ses with an intermediate percentage of C033 PoS blasts (10- S0'!!o) had a poorer outcome (5-)'ear RFS: 61 ± 7.9%) than the CD 33,wg cases (P=0 .00043 1, but not different than the c" ses with high C0 33 PoS percent age (;;. 50°,{, blasts; P= 0.32). Thc diffcrence in the EFS of cases with higher CD33 P'" cell frequenC)' was less profound (BCP All: P= 0.0022 and P> 0.05, for cutoff 10 ,md 20%, respectively; dala nol shown) co mpJren to RFS (P= 0.00002 and P= 0.0073, Fi gure 1 a) A signifi("ant differencp was noted in sepMate ana lyses for PCR (total = 25 21, TEU AML 1pas (total = 82) and hyperdiploid (total = 62) (P < 0.00001, P= 0.0051 and P= 0.019, respectively; Table 2). In PPR, we failed to prove prognostic significance of C0 33 expression (P > 0.05 , data not shown). The striki ng prognostic importance of C033 contrasts with several studies from the 19905,4-6 which failed to find J diffcrcllt prognosis of M yAgPOS Al l. These papers followed a period of co ntroversy between studies showing significantly worse prog · nosis of MyAgPOS A l L7-~ or no difference. 10 However, reccnt in vitro data showed a higher resistan ce for AlL blasts with MyAgs. '1 Allhough the aberrant MyAg express ion has hccn known for al least 40 years, little is known about its regu lation and ce ll biologica l re levance. The mech anisms that lead to thlo' coexpression of several MyAgs in some cases are obsCllre; it is also not proven whether the mechani sms ca using MyAg expre ss ion are comparable in T-Al l and in BCP All. Therefore, it is inappropriate to combine several MyAgs into compos ite criteria or to ana lyze the sign ifica nce of M yAgs in Bcr Al l together with T-Al l cases. A shorter follow-up is among possible reason s for a nonsignificant outcome; analysi s of Dur data by the criteria used by Uckun et ať (CD3 3 and/or CD13 at a 30% cutoff value) does reveal a significant di ffe rence in outcome (5-yea r EFS 52 ±9.2 and 77 ± 3.4%, 1' = 0 .0052, 4 compa red to 4-year EFS 77 ± 4 and 76 ± 1.8%, n.s. in MyAgP'" and MyAgneg BCP Al l, respe ctive ly). The median fo ll ow-up was 4 3.6 and 2.7 years in this and the previous study, respectivel)' . Since toxic deaths are less likely to correlate significantly wi th the immunophenot)' pe, impact on RFS should alwa)'s be analyzed . The difference in EFS may become insignificanl whe n smaller coho rts of patients are analyzed. Another obscuring
Table 1 Cox multivariate analysis in B-cell precursor All: corre lation of variab les with RFS Varlab/es CD65% CD33% CD13% CD15% WBC at diagnosis Response to prednisone Age at diagnosis bcr/abl or mll/af4
P 0.85 0.0020 0. 44 0.51 0.84 0.079 0.088 0.000013
The independent prognostic significance of myeloid antigens (MyAgs) together with the protocol- defined risk crilena is assessed by the Cox's proportion al hazard regression model. AII patients in whom all listed MyAgs were investigated, n = 255, are included.
E. Mejstříková, strana 79
Correspondence
• 1093
a 0.9
~
0.8 0.7 0.6
p=0.00002
0.5
o
2
3
4
5
6
8
years
b lRG 0.9
SRG
O.A (f)
12
0.7
IRG
~HRG
0.6
0.5 0.4
HRG
0.3
o
2
3
4
5
6
7
8
years C
100 90
l:l VI
c :O'" 0=
80 70
f o'"
60
,., oa.
40
u'"
30
;;
Q. III 'Jt .~ CI ,t.=
'"o U t
50
20 10
o diagnosis figure 1
day 15
day33
OUl come significance of CD JJ in B-cell precursor (BCP)
~ Ll.
(al Cases with high ' r perccntage of CD3J P oS cclls (bold black nesl are compared to those with lower C033 1'" percentag (thin gray
nes). Cutoff 10% is represented in th ese plots, analyses on other IJlues are described in text; all BCP ALL cases (n = 278); (b) RFS in Bep ALL split by risk groups (P-valu es are 0.011, 0.000 73 and 0.076 řvr SRC, IRC and HRG, respectively). The cutoff value and the line ,olorl are identical to the panel (a). (e) Percentage of leukemie cell s ť~prcss ing CD33 during the induction therapy. C ses with cytomeIrica lly detectable minimill residual di sea e were evaluated for CD33 expression. Cells were gJted according to optical scatter properties Jnd by the expression of COlO Jnd CD19 in three-color flow l)'tomelric measur ments.
·,ariable is thc difference in mAb c10nes used in the study w ious mAb c10nes differ in the sensitivity towards the 12 respective MyAgs. This may have influenced especially multicentric analyses (Baruchel, personal communication) Therefore, it is appropriate to analyze RF5 of BCP and T-ALL stparately, the MyAgs should be assessed by identical mAb dones preferably in a single laboratory and the cohort should be
treated by a single chemotherapy protocol. The presenled population-baseo analysis contains all these features. Anti gen shift represents one of the possibl e reasons of the immunotherapy failure. 5ince the anti-CD33 immunotherapy is a potential extension of the presented analysis, we checked for the stability of CD33 expression at relapse. In 39 of 51 relapsed patients, we could compare the levels of CD33 expression at diagnosis and at relapse. In line with previous studies,1 3 the expression among relapsed patients was ralher unstable (17 of 39 switched between positivity and negativily, cutoff 20'Yo, data not shown), and included also cases whose blasts completely lost CD33 expression at relapse. The expression af CD13 and CD15 was also unstable, contrasting with the stability of cl MyAg CD66c. 14 We next asked whether a substantial shift in CD3 3 expression could be observed already during the induction treatment. lNe selected patients whose BM specimens containecl at least 50'10 CD33 PoS blasts at cliagnosls. Of these, specimens were selected with a clear minimal re sidua I disease at days 15 and/or 33 by Fe. CD33 expression was ascertained with a thrce-color combination of CD10 (ALB2), CD33 (D3HL60.251) and CD19 U4.119, all from Immunotech). Percentage of CD33 1" " cells was assesseo in gated cells according to optical scatler ano CD10 and CD19 expression. The data show that despite flu ctuations (Figure 1cl, most patients retain the high CD33 percentage cluring the induction treatment (9/ 12 at day 15 and 7/8 at clay 33). Loss of CD33 expression thus usually does not occur at an early phase of trealment. 50 far, anti -CD33 treatment in ALL has been considered only for patients with high percentage of CD33'"'' blasts and no other realistic hope for cure. I- 3 The minimum percentage of CDDfl"" blasts suitable for anti-CD33 immunotherapy is unknown. In the presented cohort, there are 18 patients (5.5%) who meet the arbitrary criteria of having BCP ALL with CD33 detectable on malority (greater th an 50%) of blasts and being in IRG or HRG. The 5-year RFS of these patients was 40 ± 1SOf, . (thc 5-year RF5 of the remaining patients from the same subset was 80±4%, p= 0.0011 ). In total, eight of these patients relapsed and one developed a secondary AML. Three of them underwent allogeneic BM transplantation in second or thircl compl ete remission . Economic reckoning of anti-CD33 immunotherapy should consider that a proportion of the patients might be rescued from an intensive relapse treatment including BM transplantation . 5etting the anti-CD33 immunotherapy to the induction phase of treatment seems to overcome the issue of Jntigen shift (Figure 1cl. In addition, peri pheral nonmal igllant CD33!'oS cells are reduced by the intensive initial chemotherapy. This limits the risk of therapy failure due to peripheral consumption of th e anti-CD33 mAb 15 and the effective dose may be lower. Clinical experience is cumulating with single anti-CD33 immunotherapy as well as in combination with chemotherapy. If the presented results are to be taken into c1inical trials, a dose escalation study early during treatment is preferable to select a safe dose that does not cause a significant delay in chemotherapy but significantly reduces the leukemic c1one. Thc fact that CD33 is not expressed on all blasts in some cases Inay appear Cl problem for the immunotherapy since CD33 neg cells may remain unaffected. However, this is not different from the contemporary treatment philosophy; most single agenls can kill just a portion of the leukemic cells by themselves and the final success is achieved by combining them. INhereas standard chemotherapy targets mostly dividing cells, the suggested immunotherapy would aim at cells based on partly diffcrent characteristics.
Leukemia
E. Mejstříková, strana 80
Correspondence
Table 2 subset)
Incidence or C033"o, cases in ALL subtypeS usi ng two cutoff va lues for positivity (percent C033 Po" number of CD3J Po, cases/tut, I per
C0 33 at 10%
P·value
C033 at 20%
P-value
BCPALL T ALL
35% (98/278) 13% (6/45)
0.0003
23% (65/278) 8.9% (4/45)
0.007
PGR (BCP ALL) PPR (BCP ALL)
35% (89/252) 35% (9/26)
NS
23% (57/252) 31% (8/26)
NS
TEUAML1 + hyperdiploid BCRlABL+ Nonhyperdiploid, BCRlABL, MLUAF4, TEUAML1 negative (BCP ALL)
59% 26% 50% 27%
0.0003
40% (33/82) 9.7% (6/62) 38% (3/8) 22% (23/105)
0.0001
(48/82) (16/62) (4/8) (28/105)
Statistics show differences in respective subgroups (chi-square).
Acknowledgements We thank J Ridoskova, K Pospisilova, L Condorcinova, P Hanusova and N Ho losova for technical assistanee. Collabora ti o n of Czech Pediat rie Hematology (CPH) centers (data manager A Vrzalova, leaders: B BIJZC'k (Ostrava), Z CC'rna (Plze nl, Y Jabali (Ceske Budej oviee), V MihJI (O lomouc), f) Proehazkova (U st i nad Labem), J Sterba (Brnol, J H<1k Jnd K Tousovska (Hradec Kralove)) and stat isti ca l help of P. MJrtinkova is Jppreciated. The work was supported by gra nt s: CA UK 80 and 65 (2004), MZ CR 7430-3 and MSM0021620813
E Mejstř íko vá 1 ,2 T Ka li na 1 ,2 J Trka' " J Starý) O Hrušá kl
'CLlP (Childhood Leukemia Investigation Prague), 2nd M edica l 5chool, Charles University, 2 Prague, Czeeh Republic; Department af Immunology, 2nd MedKa l 5ehool, Charles University, Prague, Czeeh Republie; and JDepartment af Pediatrie H ematology and Oneology, 2 nd Medical 5chool, Charles University, Prague, Czech Republic
Supplementary Information Supplementary Information aceompanies the paper on the Leukemia webs ite (http://www .nature.eom/leu).
References Zwaan CM, Re inhardt O, Jurgens H, Hui smans DR, Hahlen K, Smith OP et al. Gemtuzumab ozogami cin in pediatric C03 3-positive acute Iymphob lasti c leukemi a: first cl inical experi ences and relation w ith cellul ar sensitivi ty to single ag(~ nt ca li cheam icin. Leukemia 2003; 17 : 468-470. Balduzzí A, Rossi V, Co rral L, Bonanomi S, Longoni D, Rovelli A et al. Mol eeular remission induced by gemtuzumab olOgamic in associated with donor Iymphocyte infusion s in t(4; 11) acute Iymphoblasti c leukem ia relapsed after transplantation. Leukemia 2003; 17: 2247- 2248. Corter M, Rooney S, O'Marcd lgh A, Smith OP. Successfu l use of gemtuzumab ozogam iein in ,I child w ith relapsed C033-positive acute Iymphoblasti c leukaemia . Br } Haematol 2003; 122: 687-688. 4 Uckun FM, Sather HN, Gaynon PS, Arthur O ,Trigg ME, Tubergen DG et al. Clinical features and treatment outcom e of ehi ldren with
myeloid antigen positive acute Iymphoblastic leukem ia: a report from the Ch ildren 's Cancer Group. Blood 1997; 90: 28-35. Putti MC, Ronde"i R, Coe ito MG, Arico M, Sainati L, Conter V er al. Expression of myeloid markers lac ks prognostic imp, Cl in children treated for acute Iymphoblastic leukemia: Italian experie nce in AIEOP·A LL 88-9 1 studies. Bload 1998; 92 : 795 - 801. 6 Czuczman MS, Dodge RK, Stewart ce, Fra nkel SR, Oavey FR, Powell BL et al. Value of immunophenotype in intensively treated adult acute Iymphoblastic leukemi a: ca ncer and leukemi a Group B study 8364. Blood 1999; 93: 393 1- 3939. 7 Wi ersma SR, Ortega j, Sobel E, Weinherg Kl. Clinical importance of myeloid -an ti gen expression in ac ute Iymphoblasti c leukemia of childhood. N EnglJ Med 199 1; 324: 800-808. 8 Fink FM, Ko ll er U, Mayer H, Haas OA, Grumilyer-Pa nzer ER, Urban C el al. Prognosti c significa nce ol' myeloid-assoc iated antigen express ion on bl ast ce ll s in children with acute Iymphohl astic leukemi a. The Austri an Pedi atri e O ncology Group. Med Pediatr OncoJ 1993; 21 : 340- 346. 9 Sobol RE, Mick R, Royston I, Davey FR, Elli son RR, Newman R et al. Clinica l importJn ce of myeloid anti gen exp ressl on in adult acute Iymphoblastic leukemi a. N Eng l J Med 1987; 316: 1111-111 7. 10 Pui CH, Behm FG, Singh B, Rivera GK, Sche/l Mj, Roberts WM et al. Myeloi d-assoc iated anti gen expressio n laeks prognostie value in childhood acute Iymphoh lastic leukemia treated with intensive multi agent chemotherapy. Blood 1990; 75: 198- 202. 11 Den Boer M L, Kapaun P, Pieters R, Kazem ier KM, Janka-Schaub GE, Veerman AJ. M yeloid anti gen co-expression in childhood acute Iymphohlasti e leu kaemia: relationship with in vitro drug resista nce. Br J Haemaro /1999; 105: 876-882. 12 Fi rat H, Favier R, Adam M, Leverger G, La ndm an· Parker j, Cayre Y et al. Oeterrni nation of myeloid antigen exp ression on childhood acute Iymphob lastic leukaemia cells: discrepancies using di ffe rent mon oclona l antibody clones. Leuk I. ympho ma 200 1; 42: 675-682. 13 van Weri ng ER, Beishuizen A, Roeffe n ET, va n der Linden-Schrever BE, Verhoeven MA, Hahl en K el al. Immunop henotypic ch;,mges between diagnosis and relapse in childhood acute Iymphoblasti c Icukemi a. Leukemiil1995 ; 9: 15 23 -1 533. 14 Ka linil T, Vaskova M , Mejstrikova E, MadlO j, Trka j, Stary J et al. Myel oid anti gens in ch ildh ood Iymphoblastic leukemi a: clinical data point to regulation of C066 distinct from other myeloid antigens. BMC Cancer 2005 (in press). 15 van der Vel den VH, Boeck x N, Jedema I, te Marvelde JG, HoogevE:-en PG, Boogaerts rVI et al. Hi gh CD33 -antigen load s in peripheral blood limit the díicacy of gemtuzumab ozogam iein (Mylotarg) treatment in acute myeloid leukem ia patients. Leuke· mia 2004; 18: 983-988.
E. Mej s tříková, strana 82
Příloha
2
Transfer of genomics information to flow cytometry: expression of CD27 aod CD44 discriminates subtypes of acute lymphoblastic leukemia
M Vášková, E
Mejstříková,
Leukemia 2005 (fF 6,924)
T Kalina, P Martínková, M Omelka, J Trka, ] Starý, O Hrušák
-
--
-~-
-
- - - - - - - ----
7
-
-
-
-
- - -.- -
--
-
-
- - - J
-
I lab diagno sis
age jlroup
single lab
Childhoo d Childhoo d Childhoo d Childhoo d
single lab single lab single lab
Childhoo d
single lab
Reference number
Definition of MyAg
%01 posilive Cutoff lor No~ ol patients positivity patien (B and T) ts 11%1
1(1)
CD11b, 13, 33, 36,14,15, w12. Grouped as: 0,1 and >1 MvAgs
25
1(2)
CD13 or CD33 orCD14
30
(3)
CD65,CD13,CD 15 CD33
20
-
- --- -
- - - - - 71
-
--
% CD33 above %01 cutoff positive (BCP BCP and T) ALL
~
% CD33 above cutoff (BCP multivariate univariate ALL) analysis analysis
267
16 (out of 372)
nd
3.2
nd
nd
236
22
24
6.6
9.7
significant
206
12
1.9
13
2.3
significant
My9
microscopy or St. Jude 1FC/2FC I protocol
not shown
nd
nd
significant My9
1FC/2 FC
nd
nd
significant My9
ALL-A-84 and not microscopy ALL-BFM-86 shown FC
CCG
Fe
n. s.
30
1557
nd
14
nd
nd
od
n.s.
(5)
One of: CD33, CD11b, 13, 14 , 15 and 65
20
906
32
33
14
15
ns
n.s.
20
562
9.3
8.9
7.3
6
nd
n.s.
I
single la b
Adult
single lab
Childhoo d
(7) this analysis, Cox's multivariate
CD33 and/or CD13 CD33 and/or CD13 CD33, CD13, CD15, CD65
10
259
30
33
N/A
2.7
3.0
nd CD33 signilicant
N/A
MyAgIrealmenl resulls (EFS ol entire cohort) 1[%1
method
CD33 or CD13
multi-
MyAg+ Irealmenl resulls (EFSol entire cohort) 11%1
CD33 clone
(4)
~ - centric (6)
Trealmenl resulls (EFS ol entire cohort) 11%1
n.s. N/A
not shown not shown not shown , varied not shown D3 HL60 .251
I protocols
MyAg+ Ireatmen Treatmenl t results results (EFSol (EFS of BCP) BCP) 1%1 1%1
MyAgIrealme ni results (EFS of BCP) 11%1
Med ian 10110 w up [yea EFS at rsl [years]
I nd
Graph approx .: 80
Graph approx .. 75
2.5
4
nd
nd
39+-13
78+-5
nd
3.3
nd
nd
nd
38+ -13
75+-4
3.3
5
nd
nd
nd
nd
77+-4
76+-1.8
2.7
4
AIEOPALL 88 AND 91
66+-1.6
69+-2 .3
65+-2 .3
nd
nd
nd
4
6
usually by FC
LALA87
nd
nd
nd
20
nd
nd
28
3
FC
CALBG (8811,9111,9 311,9511)
nd
39+-8
Graph approx .. 45
Graph approx.: 45
3.8
3
2FC/3FC
ALL BFM 95
70+-3
3.6
5
CCG
nd
nd N/A
72+-3
N/A
---
255 ~.
CD33 single lab
Childhoo d
this analysis, univariate
10
32
35
32
35
significant
CD33
20
21
24
21
23
CD33
30
18
20
18
20
significanl
130
24
25
18
20
significant
CD33 or CD13
324
N/A
significant
55+-6 D3HL60 2FC/3FC .251
ALL BFM 95
70+-3
78+-3
54+-6 .5
82+-3
53+-8
76+-3
51+-9
78+-3
53+-8.8
74+-3.2
52+-9
77+-3.3
53+- 8.7
74+-3.2
52+-9 .2
77+-3.4
72+-3
3.6
Abbreviation sa.'5 lisled in thc pap\:r, nd ::o nol dene (nOt provi ded). n s = nOI sagrllficanl. (1 _1.3)Ff' ''' t 1.2,3)-color now cy1ometry, Thl.!' data demonsrrate the poin lS mmtlonoo ln Ihl! paper, includin g· Cnlcna ror MyAg poslU \lJ ty di fTer, malang comparisons cumbcr$Ome. Vari ab le percenl~e of MyAg- or ofCDJ 3.... d ones frequently unspcc:lfil.'d. ;\~ nc\ ..i )' added to th e paper. shon follo\Y up pos.5ibl y c allSCS some ofth e dl fference'S (compare ref 4 :U1d the lasl line) Imponlnl pJl1S of SU ....'1vaJ Jnalyses are often missmg
References (numbered separately just for this Table) Pui CH. Dchm f O, Smsh 8 . R.m.!TB GK . Sché'l l M J. Robe rts WM o('rist \.VM , Mirro J. Jr Mycloid-assocl atc
I.
5
I
E. Mejstříková , strana 82
P říl o ha
2
Transfer of genomics information to flow cytometry : expression of CD27 and CD44 discriminates subtypes of acute lymphoblastic leukemia M Vášková, E
M ejstříková,
Leukemia 2005 (fF 6,146)
T Kalina, P Martínková, M Omelka, J Trka, J Starý, O Hrušák
E. Mejstříková, strana 83 Leukemia (2005). 1-3 1::1 2005 Nat re Publishing Group AI! rights reserved 0887-6924/05 $30.00
www.nature .com/leu
CORRESPON DE NCE Transfer of genomics information to flow cytometry: expression of CD27 and CD44 discriminates subtypes of acute Iymphoblastic leukemia
Leukemia advance online publication. 10 March 2005; lO:10.1038/sj.leu.2403706
oTf IE EDITOR \ pression profiling studies have provided data on an unprece.lťn ted number of genes that are expressed in malignant cells. 1--4
The critical number of genes which can predict a particular ,énotype is a matter of discussion. Downing poi nted out that as ;e..y as 20 genes may be necessary for an acc urate prediction of Jbsets and prognos is; thes genes should be spec ifically "udied, perhaps by methods like reverse transcription poly'llerase chain reaction (RT-PCR) or flow cytometry (FO (Down'lG in Carroll cl ;J / ). Fe. which shows the expression of -lOlecules in mutual context o n individual ce ll s, appe ars uotimal for such analysis. The present study (Microarray-guided FO is designed to Itematically screen for genes within the existing expression lJ rofilill§ studies on childhood acute Iymphoblast ic leukemia ~LLJ. I . The genes, which are identified as best co rrelating with :Iic pediatric ALL subgroups (E2NPBXl , MLL, TEL!AML 1, BCRI ~B L , 'novel' and hyperdiploid genotypes; pati ents who later rclarscd and those who developed therapy- induced acute eloblaslic leukernia), are selected. Afler reca lcul alion just kJr the B precursor A L cases, we select the gen e~ in which the diiíerence in express ion is likC'l y to be observed at thc protein Il!\'el. Next, we select molecules with suitable cellular localiza'ion (nonsecreted proteins) and with an avail able mAb. Reaclivity and specificity of mAbs are tested in healthy peripheral blood ell s and/or in cell lines . The respective molecules are investigated by four-color FC in diagnostic bone 'IIolrrow (BMI samp les. Five molecules have proceeded into this ep (CD44, CD2 7, D49f, CD247 and CDI03) . We present nefe thc resu lts of CD44 and CD27 expression, which are .nvestigated in the largest cohort. A tota l of 66 patients w ith B-cell precursor ALL and 14 atients with T lineage ALL were considered to enter CD44 and CD2? investigation. These patients repre sent all Czech child ren age lower than 18 yea rs) diagnosed with ALL between 03/ 2003 lnd 02/2004 . Five patients (four B-cell precursor and one T neage ALl) were excluded due to low sample volume. CD44 xpression was investigated in 62 B-cell precursor ALL ratients 11 TEUAMLI Po" 18 hyperdiploid, four SCRlABLPo", two ~IL LR'"'' and 17 with no ne of the above-menti oned genotype) jnd (D27 express ion in 56 S-cell precursor AL L patients (2 1 iEUAML 1ros, 15 hyperdiploid, four BCRlASL Po" on . MLLRPo, md 15 with none of thC' above-mentioned genotype). Informed onsent was obta ined from patients and/or their guardians . Patients were trcated according to ALLlC SFM 2002 or Interfant
i orrespondence:
Dr O Hrusak. Childhood Lcukaemia Investigation t'riJgue, Department of Immunology, Charles University 2nd Medical ~ hool, V uvalu 84, 150 06 Praha 5, Pragu e, Czech Republic; Fax : 42022 44 3 5962; E-mail : Ondrej.Hrusak Ifmotol.cuni .cz .eceived 25 Augusl 2004; accepted 12 January 2005
99 protocols. The presence of TEl!AML 1, BCRlABL fu sion genes and ML L gene rearrangement (M LLR ) was detected by tworound nested PCR; hyperdiploidy was assessed using DNA index flow-cytometric measurement. CD44 and CD27 antigens were stained with an anti-CD44 FITC and ant i-CD27 FITC (BD, San Jose, CA, USA) in four-color co mbinations with antigen s f1'Om a standard panel (a nti -CDI0 PE, anti-C019 PCS or PC7, anti CD34 APe, Immunotech, Marseill e, France). Data were acquired using a FACS Ca libur flow cytometer (BD). Antigen positiv ity was ana lyzed on gated malignant ce lls according to the iso type control . To eva luate differences oi express ion between th e subgroups, a nonparametric Mann- Whitney test was performed using StatView software (SAS Institute, Cary, NC, USA) Expression profiling found the gene for CD44 (Hermes, Pgp-l) to be one of the best correlating with the MLL genotype and with the subg1'Oup of T-ALL patients who later developed hematological relapse. u Although MLLRPuSblasts in ou r co hort did show CD4 4 positivity, 50 f,Jr we have not observed a higher CD44 expression compared to other C044 1" " B-cell precursor ALL cases (Figures 1a, 2a). In addition, D44 express ion significantly correlated with higher risk T-ALL (P = 0.032 ) (Figure 2b). This also indi cates that the c urrent risk stratification of the T-ALI_ patients w ithin the ALL-IC BFM2002 protocol (based o n age, leukoc yte count at presentation, ea rly treatment response and unfavorable molecular genetics, as in other major frontline therapy protocols) corresponds to the true biologi ca l risk. Furthermore, C044 expression was significantly lower in TEU AML 11'°' ALL (P < 0.0001). which is in line with thc observation of one of the two express ion profiling studies. 1 Thc assoc iation with TEUAML 1 genotype was also found in CD27 (TN FRSF7) gene expressioll,' but another expression profiling study] showecl only a correlatioll with BCRlABL genotype. We found a strong correlatioll of CD27 with TEl! AML 1 positivity. CD2 7P'" blasts above 30'Yo were detectcd in 20/21 and 2/35 patients with TEUAMLl IJOS and TElIAML 1"0 1: ALL, respectively (P < O.OOO1) (Fi gure 1a). Since the oppositc correlations with TEUAML 1 were observed for CD27 and CD44, we ana lyzed the composite picture of the expression of th ese two molecules simultaneoLlsly (Fi gure 1a, b). Most CJses in both expression profiling studi es and in our cytometric study can be co nsidered either CD44 PoS or C027 Po,. Dual C027posCD44PoS blasts are typi ca lly seen in BCR/ABLI"" ALL and a subset of TEL! AML 1po, patients exists with CD44po'CD271l0S bl asts. ln our cohort, th e TEl!AML 1pos case with Sl 0(., CD44"'" ce lls prese ntecl with an unusually high blast count (peripheral leukocytosis 109 x 10 6 per ml) and higher percentage CD66 pos blasts (25%) compared to thc typical TEUAML 1po, patients G - other presentation parameters corresponcled we ll with the genotype. (D27 has heen considered to be a general marker for memor)' S-ce ll s in humans and so far its expression in human B-ce ll precursors has not been reported. Here we show th at this antigen is indeed expressed in malignant S precursors at a protein leve!. Moreover, an experim ent with seven -color FC on nonmalignant BM (FACS Aria, BD, data not shown) did confirm
•
E. Mejstříková, strana 84
Correspondence
a
Flow cytomelry
80 Q)
cr>
jg
c
60
• • •
Q)
40
N
o
U
20
•
50000
b-
•
.TEUAML1 a Hyperdiploid 6BCRlABL xMLL OOther
40000 c
.Q Ul Ul
30000
~
"x
~ 20000
•• ••
t'
Q)
"-
r--
b
100
c
Q)
cr>
b-
r-N
o
10000
u
O
O
O
Expression profiling
•
.TEUAML1
•
6.
. ... ... .0 t
~
tJ· ~~
.;t~ '0.&.
•
60
O
-MLL
6.
6. OOther
6.
6.
•
x o o o o
-1QOOO 40
•
. ~~
O 20
6.
O
O O
a Hyperdiploid ABCR/ABL
Ul
o o o
100
80
CD44 gene expression
CD44 %
C 100
Q)
:o
~80
""So u
Q)
o>
-g80 Ul Q) Ul
Ol U
'"m ~
-' 40
::<
§ UJ f-
t' Q) 20 t' Q) Cl..
o
o
20
40
80
80
100
Percent TEUAML 1pos cases above cutoff value
Figure 1 Clu ste ring of ALL gcnotyp ic subgroups according to C044 and C027 expression. (a) FC data. (bl Express ion profil ing data from Yeoh er a/.' Probe set number 404<J 3_
C0 27 POS cells in all four specimens (12 ±2.2% among [)APlneg D 19 P' ''C010 l''''C020,,,,g cells, among these C027 PoS cells 45 ± 13% were C034 pn ' l . The dynamics of C044 expression on developing B- ce lls during hematopoiesis was already reported . We are going to analyze the recomb inati on status of 19 genes in subpopu lalions of precursor B-ce lls with C044 and C0 27 expression corres ponding to Iymphobl asts. To compare the value of C0 44 and CD27 with the other expression profiling data for the predi ctio n of TEUAML 1 status, we used th e sam e plot as described previously in the FC metaanalyses" (F igure 1cl . This format clepicts graphically lhe predictive va lue of eac h probe set or molecule for the TEUAML 1 kemia
status. Each probe se t or molecule is separately co mpared to its optimal cutoff value in all BCP ALL patients ol' thc respectivc cohort. lhe probe-set-speciric optimal c utoff value was determined using a statistical software R (http://www.r- projec t.org ). The optimal cutoff value is thc one that leads to th e besl re sol ution between TEUAML 1pos and TEUAMLl "eg subsets, judging by the distance from a nonc:orrel ating diagonal. Th e perce ntage of TEUAML 1neg patients above the cutoff value is compared to the respecti ve value in TEUAML 1pa< pati ents. Th e overall predictive valu es for C044 and CD27 are 93 and 95% in Fe. 82 and 91% in Yeoh,l an d 90 and 90% in RoSS,l respectively. Th e difference in predictive valucs is in compli-
E. Mejstříková, strana 85
•
Correspondence
3
b 8000
12
6000
a
4000
x
Ol
~
2000
Ol Dl
.,.
či
o
.2000
Expression profiling P=0.11
o
~g
ligure 2
~:
~ 50
Flow cytometry P=0.032
o
c~ ---------_._---~-----------_. __ .. _----_._-_._------_.- ._- §~~ ---------_._-- ------_._------------~-_._------------
·4000 - ' - - - - - --
CD44
BQ
----------
Relapsed
CCR
~ 40 O;
a.
30
-10 - ' - - - - - - - - - - - - - - - - -
HR
IR
Significance of CD44 expression in specific subsets. (a) Intensity of CD44 expression ln B-cell precursor ALL. Histograms represent
·ILLR''''' (thick lines), together with typi cal hyperdiploid (thin lines) and TEL/AML 11'°' (dashed line) cases. Y-axi s shows frequency. (b) CD44 1
'pression in T-ALL pati enls. Exprcss ion profiling data of patienls in clinical complete remission (CCR) and relapsed patients. Prube set number 126_3t is shown (Ieft panel) . FC data of patients in the intermediate (IR) and high-risk (HR ) groups (right pane\).
oce with the fact that FC can investigate the expression of ·Ied molecules on pure ALL cells . We studied the other "Illecules (C049f, C0247 and CD103) in spe c imens of fewer lients and thus it would be too early to establish their edictive va Iues. The princi pie that class-defining genes may be selected within 'Icroarray data has been suggested previously (Downing in Carroll et a15 ). The results o( the presented screening strategy nlOve this principle. Lack o( in(ormation on protein expression ~ ared to be the most limiting factor reducing the number of .•ndidate genes in the final FC testing. Improbability o( the cell1und (orm also excluded molecules during the screening (or ~C - these secreted molecules may be studied by protein ochemistry. Although cytometric studies on molecules that "me from a systematic screening strategy in microarrays have ·'1 )1 been presented yet, one new molecule (C058) has been Ilroduced into FC testing based on expression pro(iling? The mented data not only show that in(ormation (rom microarrays an betransferred to cell-based investigation by Fe, but also that lile composite mi croarray in(ormation can be suceessfully ,tploced by strong predi ctors likc CD44 and C027. The oogoing projeet Microarray-guid('d Fe tests whether other molecules ean be found with comparable or better predictive alues for ALL genotype and prognosis.
~clmowledgements
,'Ie thank the cytometric technicians Jana Ridoskova, Klara
work was supported by Crants Nos. CA UK MSM0021620813, ICA MZ CR 7430-3 and 6929-3. M Vaskova l.2 E Mejstrikova'·2 T Kalina 1 . 2 P Martinkova] M Omelka:l lA JTrka J Stary4 O Hrusak 1.2
'CLlP - Childhood Leukemia Investigation Prague, 2nd Medical School, Charles University, Prague. Czech Republic; 10epartment of Immunolugy, 2nd ME'dical School, Charles University, Prague, Czech Republic; JOeparlment of Probabilily ancl Mathcmatical Stalistics, Faculty of Malhemalics and Physics, Charles University, Prague, Czech Rcpublic; anrl 'Pedialric Hematology and Oncology, 2nd MedicalSchool, Charles Univcrsity, Prague, Czech Republic
References
2
3
4
~pisilova and Lucie Condorcinova, and the molecular genetic ~hnicians Katka Muzikova and Leona Rpznickova and Marketa '.ahnova. MV highly appreciates the practieal and theoretical perience in the laboratory of Z Trajanoski (Institute of Jiomedical Engineering and Christian Ooppler Laboratory (or L>l'nomics and Bioinformatics, Craz University o( Technology, IJJslria), as well as the (rienclly atmosphere. The collaboration o( \I Czech Pediatrie Hematology (CPH) centers (Ieaders: B Blazek Ostrava), Z Cerna (Pl zen), J Hak (Hradec Kralove), Y Jabali (Ceske ;udejovi ce), V Mihal (Olomouc), O Proehazkova (Usti nad :.b€ml, J Stary (Praha) and J Sterba (Brno)) is appreeiated. The
8012004,
5
6
7
Ycoh E), Ross ME, Shurtleff SA, Williams WK, Patel D, M hfouz R CI JI. Classification, subtype discovery, and prcdiction of uutcume in pediatrie acute Iymphoblastic leukemia by gene expression profiling. Cancer Ce1/2002; 1: 133-143. Armstrong SA, Staunton JE, Silverman LB, Pieters R, den Boer ML, Mi nden MD el al. MLL translocations specify a distinct gene expression profile that distinguishes a unique leukemia. Nal Genel 2002; 30: 41-47. Ross ME, Zhou X, Song G, Shurtleff SA, Girtman K, Willi;lms WK el al. Classifjeation of pediatrie acute Iymphoblastic leukemia by genc expression profiling. 810ad 2003; 102: 295 1-2959. Fine BM, Stanulla M, Schrappe M, Ho M, Viehm ann 5, HarbulI J el al. Gene expression patterns associated with reeurrent chromosomal transloeations in acute Iymphoblastic leukemia. 8Io(xI2004; 103: 1043-1049. Carroll WL, Bhojwani D, Min DJ, Raetz E, Relling M, Davies S Cl .11. Pediatrie acute Iymphoblastic leukemia. Hematology lAm Soc Hemalol Educ Program) 2003, 102-131. Hrusak 0 , Porwit-MaeDonald A. Antigen exprcssion palterns reflecting genotype of acute leukemias. Leukemia 2002; 16: 1233-1258. Chen JS, Coustan-Smith E, Suzuki T, Neale GA, Mihara K, Pui CH el al. Identification of novel markers for monitoring minimal residual di sease in acute Iymphoblastic leukemia. B/oocl 2001 ; 97: 2115-2120.
Leukemia
E. Mejstříková, strana 86
Příloha
3
Myeloid antigens in childhood Iymphoblastic leukemia:clinical data point to regulation of CD66c distinct from other myeloid antigens Tomas Kalina, Martina Vaskova, Ester Mejstrikova, JozefMadzo, Jan Trka, Jan Stary and Ondrej Hrusak BMC Cancer (fF 2, 709)
E.
Mejstříkov á,
strana 87
()
BMC Cancer
BioMed Centra l
Open Access
Research article
Myeloid antigen s in childhood Iymphoblastic leu kemia:clinical data point to regulation of CD66c distinct from other myeloid antigens Tomas Kalina 1,3, Martina Vaskova l.3, Ester Mejstrikova1,3, Jozef Madzo 2,3 , Jan Trka 2,3, Jan Stary2 and Ondrej Hrusak* l.3 Add ress: 'D epa nlllC lll o f Im mun o logy, Ch a rl es U ni versily 2nd Medical Sc hoo l, Pragu e, Czec h Re publ ic. 2Dep a nment o f Pediatri e H em alo logy and On co logy, Cha rles U nivers il y 2 nd Med ical S,hool. Pragu e, Czech Re publ ic a nd 3CLl P - Ch ildhood Leu kem ia lnvesliga lion Prague Czech Repub li c Emai l: Tom as Kalin a · lo mas .ka li na @ lfm olO l. cun i.cz; Ma n inaVaskova·m ani na .v inova @l fm olol.cun i.cz; ESler Mejs lrikova · ESler. Me jsl rikova@ lfrno lo l.cu ni.cz; loz ef Mad zo · jou f. m adzo@ lfmo lo l.cuni .cz; Jan Trka . jan .lrka@ lfmo lo l. cun i.cz; Jan Slary· ian .s lary(aJlfm o lol. cuni.cz; Ondre j l-lrllsak' · o ndrej .hrusa k@ lfm o lO l.cun i. cz • Co rres pondi ng aU lho r
Publ is hed: 12 April 2005
BM C Concer 2005 , 5:38
doi : I 0. 1 186/ 1471-2407·5- 38
Received: 16 Novembe r 2004 Accepeed: 12 Ap r íl 2005
This areicle is available fro m: http://www. biomedceneral.com / 1471-2407/5/38
© 2005 Ka lina e e al; licensec Bio Med Centra I Ltd. This is an Ope n Access art icl e dis tr ibu ted under the te r ms of th e C re ative Commo ns Attr ibu tio n Li ce nse (http ·/Icreatjyecommoos w hich permi ts unres tric t ed use, dis tributi on . and rep ro du ction in any med ium, provid ed t he original w ork is properly cited.
Q r~lI ic e o se$!byl2
O),
Abstract Background: Aberrant expression of myeloid antigens (MyAgs) on acute Iymphoblastic leukemia (ALL) cells is a well-documented phenomenon, although its regulating mechanisms are unclear. MyAgs in ALL are interpreted e.g. as hallmarks of early diffe rentiation stage and/or lineage indecisiveness. Granulocytic marker CD66c Carcinoembryonic antigen-related cell adhesion molecule 6 (CEACAM6) is aberrantly expressed on ALL with strong correlatio n to genotype (negative in TEUAMLI and MLUAF4, positive in BCRlABL and hyperdiploid cases) . Methods: ln a cohort of 365 consecutively diagnosed Czech B-precursor ALL patie nts, we analyze distribution of MyAg+ cases and mutual relationship among CD 13, CD 15, CD33 , CD6S and CD66c. The most frequent MyAg (CD66c) is studied further regarding its stability from diagnosis to relapse, prognostic significance and regulation of surface expression, For the latter, flow cytometry. Western biot and quantitative RT-PCR on sorted cells is used. Results: We show CD66c is expressed in 43% patients, which is more frequent than other MyAgs studied. In addition. CD66c expression negatively correlates with CD 13 (p < 0 ,000 I). CD33 (p = 0.002) and/or CD6S (p = 0.029) . Our data show that different myeloid antigens often differ in biological importance. which may be obscured by combining them into "MyAg positive ALL". We show that unlike other MyAgs, CD66c expression is not shifted from the onset of ALL to relapse (n = 39, time to relapse 0.3-5.3 years) . Although opposite has previollsly been suggested. we show that CEACAM6 transcription is invariably followed by surface expression (by quantitative RT-PCR on sorted cells) and that malignant cells containing CD66c in cytoplasm without sllrface expression are not found by flow cytometry nor by Western biot in vivo. We report no prognostic significance of CD66c. globally or se parately in genotype subsets of B-precursor ALL. nor an association with known risk factors (n = 254) . Conclusion: ln contrast to general notion we show that different MyAgs in Iymphoblastic leukemia represent different biological circumstances. We chose the most frequent and tightly genotype-associated MyAg CD66c to show its stabile expression in patients from diagnosis to relapse. which differs from what is known on the other MyAgs. Surface expression of CD66c is regulated at the gene transcription level . in contrast to previous reports.
Page 1 of 11 (page number no/ for ci/a/ion p urposes)
E. Mejstříková, strana 88
BMC Cancer 2005, 5:38
Background Although expression of surface markers in acute lymphoblastic leukemia (ALL) parallels that of normal hematopoietic precursors, several markers of myeloid lineage are found on ALL lymphoblasts. This phenomenon is referred to as "aberrant expression". The issue of the regulatory mechanisms that allow it has been addressed repeatedly throughout the recenl 40 years [L2] . Although several hypotheses stressing either possible lineage ind ecis iveness or genetic misprogramming have been rJised , the phenomenon is slili not fully understood. We and others have shown thal the myeloid antigen CDGGe is very frequently aberrantly expressed in B-precursor ALL, however, a large sludy showing its frequency in the light of other myeloid antigens has been missing. CDGGc expression was found on cases of childhood and adult ALL in strong correlation with nonrandom genetic changes (BCR! ABL positivity [3 J, hyperdiploidy and TEL/AMU negativity [4J, reviewed in
[5]) . CDGGc (CEACAMG, previously caJled Nonspecific crossreacting antigen, NCA 90/50 and KOR-SA3s44 antigen) is a member of the carcinoembryonic antigen family. This heavily glycosylated molecule consists oftwo constant IgIike domains and one variable Ig-like domain and it is anchored to the membrane via its glycosylphosphatidylinositol (CPI) . Within the hem atopoietic system, CDGGc expression is limited lO granulocytes and its precursors [3,GJ, where it serves homotypic and heterotypic adhesion [7J, Ca2+ mediated signaling [8J and is markedly upregulated from intracellular stores after activation {9J . It is also found in epithelia ofvarious organs 17J. Upregulation ofCDGGc is an early molecular evenl in lransformation leading to colorectal tumors [lOl . II was also confirmed to inhibit anoikis (apoptotic res ponse induced in normal cells by inadequate ar inappropriate ad hes ion to substrate) in the in vitro model of carcinoma of colon [ll] and specific silencing of this gene led to decreased metastatic potential in pancreatic adenocarcinoma [12] . Surprisingly, Sugita et al [13J reported intracellular presence ofCDG6c in allleukemic celllines examined, regardless of surface presence or absence, with a different an tigen distribution in cytoplasm that determined surface expression. They speculated that presence of an undisc10sed transporter would target this molecule to granules and for surface expression, whereas surface CDGGc"eg cell lines lack this transporter. This intriguing hypothesis prompted us to test whether transcription of CEACAMG gene and/or intracellular CDGGc expression is always followed by surface expression. llniqueness of aberrant expression of CDGGc on malignant lymphoblast is exploited for diagnosis of ALL and
http ://www.biomedcentral.com/14 71-2407 /5/38
follow-up of a minimal residua I disease (MRD) using f10w cytometry [14,15]. To use a marker for a MRD assessment a critical question must be addressed, whether the aberrant expression is a stable property of the malignant clone or whether it can be subject to immunophenotype shift. ln the present study we set out to address the frequency of C DGGc molecule expression in childhood ALL, the regulation ofCDGGc expression from gene transcription to cytoplasmic and surface expression, and we follow immunophenotype stability from diagnosis to relapse. We also discuss relevance of CDGGc for prognosis prediction. Methods Potients The cohort of all Czech children (<18 years) diagnosed with B-precursor ALL investigated in our reference laboratory from 1.5.1997 to 23.7.2004 was used for current study (n = 381). lnformed consent was obtained from patients and/o r their guardians. The presence of TEL/ AMU , BCR/ ABL and MLL/ AF4 fusion genes was detected by two-round nested PCR, hyperd iploidy was assessed using DNA index f10w cytometric measurement as described previously [4] . Patients' genotype and corresponding surface CDGGc expression is shown in Figure 1 (genotype available in 98% of patients) . For intracellular staining and FACS sorting, only samples with enough material were selected. Celllines Surface CDGGc negative cell lines with typical translocation found in childhood ALL: TEL/AMUpos (REH) was kindly provided by R. Pieters (University Hospital Rotterdam), MLL/AF4 pos (RS4;1l) translocation and with no fusion (NALM-G) were obtained from Cerman Cell Line collection (DSMZ, Braunschweig, Cermany) Flow Cytometry
Flow cytometry immunophenotyping of bone marrow (BM) aspirates was performed in at diagnosis and at relapse. Routine immunophenotypic c1assification using panel of monoclonal antibodies (moAbs) was performed as descri bed previously [4]. Brief1y, BM samples were stained with 2-, 3- and 4-color combinations of moAbs for 15 min in darkness, erythrocytes were Iysed with NII 4 CIcontaining Iysing solution for 15 min, washed and data were acquired using single FACS Calibur instrument throughout the study (BD Biosciences, San Jose, CA, USA) f10w cytometer. Anti-CDGGc (CEACAMG) moAb used in all diagnostic and relapse measurements in this study was clone KOR-SA3s44 directly labeled to FITC (lmmunotech, Marseille, France). Intracellular staining was performed using Fix & Perm kit (Caltag, Burlingame, CA,
Page 2 of 11 (page number no/ for ci/a/ion purposes)
E. Mejstříková, strana 89
BMC Cancer2005, 538
hltp//www.biomedcentral .com/1471-2407/5/38
CD66c% 120
100
BO
o 60
I
~
tg'
~
00
00
~o
o o
0O@O
odP
o
= fl
8
ceP
o 40
20 1
o o
~ <9
",
~
o o
~
o 8
o
~
o
'*'
TEUAML1 BCRJABL Hyperdiploid MLUAF4
I M~an
I* Other
0.55
72
60
0.33
11.5
105
91
15
8
154
Figure I Correlation of ALL genotype categories and percentage of CD66c positivity. Median percentage of CD66c PoS
blasts is listed below each genotype group. Data of consecutive unselected patients with BCP ALL (n = 373) are shown.
USA) according to manufacturer's protocol. Acquired data was analyzed with Cell Qucst (BO Biosciences) or Flow Jo (Tree Star, Ashland, OR, USA) software, Iymphoblast gate was drawn based on optical scatter and C019PoS blast population was selected for further analysi s. Value of 20% was chosen as a threshold of positivity as recommended by ECII.1161 . For robust prognostic significance testing, other threshold values were also tested as indicated in results. Cross-blocking of CD66c moAbs Bone marrow samples of C066c poslllve blasts were stained with anti-C066c moAb clone 9A6 (Cenovac, Freiburg, Germany) moAb for 15 min, erythrocytes were Iysed with NH~CI-containing Iysing solution for 15 min, washed and sample was incubated with anti-C066c moAb KOR-SA3544 PE moAb conjugate. Western biot Samples containing 5 x lOG cells were Iysed for 30 min at 4 o Cin 100111 Iysis buffer containing 20 mM Tris-HCI (pH 8.2), 100 mM NaCI, 50 mM NaF, 10 mM EOTA, 10 mM
pyrophosphate (Na 4 P2 0 7 ) and Complete Mini EOTA-Free (protease inhibitor cocktail tablets, Roche Oiagnostics, Mannheim, Cermany). Oeb ris was sedimented by centrifugation for 3 min at 13000 rpm, O°c. Supernatants were mixed with 100 !ll 2x Laemmli's SOS-polyacrylamide gel electrophoresis (PACE) sample loading buffer, and heated for 5 min at 100 C. Proteins were fractionated by SOS-PACE on 12.5% gels and electrophoretically transferred to PVOF membranes (Bio-Rad, Hercules, CA, USA). Membranes were blocked for 1 h in ['BS (pH 7.4) containing 0 .5% Tween-20 and 5% nonfat dried milk. Blots were then incubated for 1 h at room temperature with antiKOR-SA3544 (Immunotech, Marseill e, France) or antibeta-actin (Sigma-Aldrich, Saint Louis, MO, USA) moAbs and then developed using goat anti-mouse IgC (H+L)HRP conjugate (Bio-Rad). Immunoreactive material was then revealed by enhanced chemiluminescence (ECL, Amersham, Little ChaJfont Buckinghamshire, UK) accord· ing to the manufacturer's instructions. 0
Isolation of RNA and Real-Time Quantitative PCR analysis (RQ-PCR) For RQ- PCR analysis, leukemic blasts were FACS soned using sorting option on FACS Calibur or on FACS Aria instrument (1.1 x 10 4 - 4.7 x lOs cells from one patient). Isolation of RNA from FACS-sorted cells was performed using Trizol-reagent (Cibco BRL, Carlsbad, CA, USA) according to manufacturer's instructions 1171. Complementary DNA was prepared using M-MLV Reverse Transcriptase (Cibco) according to manufacturer instructions. Clycogen (Cibco) 250 !lg /mL was added when initial cell number was lowerthan 10 5 Quality of cONA was verified by PCR on beta-2-microglobulin (B2M) housekeeping gene.
RQ-PCR was performed in the LightCycl erT>' rapid thermal cycler system (Roche Oiagnostic CmbH, Mannheim, Cermany), according to manufacturer's instructions, using SYBR green intercalating dye. CEACAM6 specific primers 3'-CCCCITICTACCACCTCTAA and 5'-CCATCTCCCCTCCAACCA designed by Baranov [181 were used for CEACAM6 amplification and B2M specific primers 3'CATCCTCCITACATCTCTCC 5'-CCACCAGACAATCCAAACTC [19Jwere used for total cONA quantification. PCR amplification was carried out in 1 x reaction buffer (20 mmoljL Tris-HCI , pH 8.4 ; 50 mmol/L KCI); and 2.0 mmol MgC1 2 containing 200 !lmoljL of each dNTP, 0.2 !lmol/ L of each prim er, 5 !lg bovine serum albumin per reaction, and 1 U of Platinum Taq DNA polymerase (all from Cibco) in a final reaction volume of 20 !lL. For each PCR reaction, 2 !lL of cONA template and 2 !ll of SYBR Creen 5 x 10.4 (FMC BioProducts, Rockland, MA, USA) fluorescent dye was included. The cyding conditions were 2.0 minutes at 95 o C followed by 45 cycles of
Page 3 of 11 (page number not for citafion purposes)
E. Mejstříková, strana 90
http ://www.biomedcenlral.com/14 71-2407/5/38
BMC Cancer 2005, 538
Table I: Frequency of CD66c and myeloid antigen expression. Cases with >20% blasts are regarded positive, coexpression of CD66c and other MyAg is tested by Fisher's exact test.
Molecule
No of ca ses (total
CD66c CD33 CDI5 CDI3 CD65 CD66c and CD66c and CD66c and CD66c and
156 85 72 57 14 21 30 9 2
CD33 CD 15 CD 13 CD65
= 365)
Proportion [%]
Coexpression with CD66c
43 23 20 16 3.8 5.8 8.2 2.5 0.55
mutuallyexclusive random mutually exclusive mutually exclusive
denaturation at 94 ° C for 5 seconds, anneal ing at 59 ° C for 30 seconds, and extension at n oc for 15 seconds. CEACAM6 and B2M gene were amplified separately from the same cDNA, and al! experiments were performed in duplicale. Melting curve analysis was performed after each run; in case of peak melting lempe rature shift, PCR prodUCLS were verified on agarose gel electrophoresis. Normalized CEACAM6 Expression (CEACAM6n) Amplification and calibralion curves were generaled by using affi liated software (LightCycler 3 data-anal ysis softwa re; version 3.5.28; Idaho Technology Inc., Sall Lake City, UT, USA). A calibration curve for the B2M and CEACAM6 housekeeping ge ne was generated using the series of lOx diluted cDNA from peripheral blood granulocytes as a standard for both reacti ons. Crossing point (Cp) value was calculated with LightCycler 3 software using second derivative maximum method. CEACAM6n value is relative and represents a ratio of CEACAM6 to B2M (CEACAM6n = CEACAM6/ B2M). Standard cDNA from granulocytes was assigned CEACAt\16n value of 1, lhe same aliquot of granulocytes cDNA was used throughout of study. Statistics Statistical evaluation was do ne with Statview software, (SAS Institute Inc, Ne. USA) . We used Fisher's exact test, regression coefficient, Mann-Whitney test and Logrank (Mantel-Cox) test as described in text.
p = 0.002 NS P < 0.0001 p 0.029
=
ALL diagnosed in the study period. The CD66c molecule was expressed on 43% cases (Table 1, cases with >20% positive blas ts were considered positive). For the fra ction of positive cells and correlation with genotype see [S] , of note, 29% of patients expressed CD 66c on more then 50% blasts. Comparison with other MyAg showed thal CD66c is more frequently expressed. Coexpression of CD6 6c with other MyAg was not a usual finding (Table 1, Figure 2). Expression ofCD13, CD33 and CD6S tended to be non-random (mutualIy exclusive) with CD66c (Table 1). Coexpression of CD66c with any 2 of th e other MyAg was found in fewer than 4 cases in each combination. Interestingly, mutual relationship of other MyAg was random , with the exception of CD 13 and CD33 coexpression (p < 0.0001) and CDlS and CD6S coexpression (p = 0.000 2). The analysis was performed also at different culoffvalues (10, 30 and 50 %; data not shown). The same or less significant correlations were also observed at differe nt cutoffvalues. Cross-blocking of KOR-SA3544 c/one with 9A6 c/one The moAb clone KOR-SA3S44 was not included in Human Leukocyte Differentiation Antigens workshop, but was characterized by Sugita et al [13]. To prevent ambiguous interpretation of our data we extended characterization of KOR-SA3S44 clone of CD66c moAb by blocking experiments on CD66c PoS blasts. Pretreatment of cells with workshop-typed clone 9A6 moAb completely blocked binding of KOR-SA3S 44 clone in all 9 leukemie specimens and in granulocytes (data not shown).
Results Frequency of CD66c and mye/oid antigen (MyAg) expression We selected 365 patienťs samples obtained at diagnosis of B-precursor ALL with available information on the ex:pression of MyAg CDl3, CDI5, CD33, CD6S and CD66c. This subcohort represents 96% of al! B-precursor
Cytoplasmic presence of CD66c in ALL blasts We have studied surface and cytoplasmic expression of CD66c in 20 ALL diagnostic samples by flow cylOmetry. In contrast to findings of Sugita et al 113], we have detected CD66c exclusively in all 8 surface positive cases. None of th e 12 surface negative cases stained in cytoplasm
Page 4 of 11 (page number not for citation purposes)
E. Mejstříková, strana 91
BMC Cancer 2005,538
httpl/www.biomedcentral.com/1471-2407/5/38
100 Vl
O
o a.
u
O
80
t:l
U 60 u .~ 40 Cll
a.
S>.
20
u
'O
O ....., .....r--
?1-
o
10
20
30
40
50
% of surface
Figure 2 Graphical iIIustration of myeloid antigen positivity in childhood B-precursor ALL. For each antigen, positive cases are represented by a colored formo The areas of the forms roughly correspond to the frequency of positive cases (observed numbers of patients are marked in red) while the shapes are constructed to illustrate the respective coexpressions. An arbitrary cutoff value of 20% is used for all antigens. The CD66c positivity correlates with negativity of any of the following: CD33 (p 0.002), CD 13 (p < 0.0001) and CD65 (p = 0.029). There was a signifícant correlation between CD33 and CD 13 positivity (p < 0.000 I) and between CD 15 and CD65 positivity (p 0.0002) whereas the positivity of no other two antigens of the ones shown correlated signifícantly with each other. Total number of B-precursor cases illustrated is 365.
=
=
(Figure 3). The probable cause of the opposite finding in several cases (lower percentage after permeabilization than on surface) is a higher background after permeabilization (isotypic control mean fluorescence intensity was 4.3 ± 2.0 and 9.7 ± 3.7 for surface and permeabilized staining, respectively), which covers borderline events. Transcription of CEACAM6 gene To extend the above findings, we used Real-Time Quantitative Reverse Transcription-PCR (RQ-RT-PCR) to quantitatively assess presence of specific CEACAM6 mRNA. We FACS-sorted CD 19posCD66c'leg or CD 19posCD66cPoS blast cells for RQ-RT-PCR analysis. We didn't find significant amount of CEACAM6 transcript in surface CD66c"eslymphoblasts, whereas CD66cPOS cells contained CEACAM6. When CD66c''''S and positive fraction was
60
70
80
90
CD66c PoS
'F igure 3 'R elationship of surface and cytoplasmic expression of CD66c. Percentage of surface expression of CD66c in ALL blasts is plotted against cytoplasmic expression (after cell membrane permeabilization). Samples of 20 patients at ALL diagnosis are shown, 12 CD66c negative and 8 CD66c positive. Regression coeffjcient R2 0.927
=
FACS-sorted of heterogeneous specimens (lymphoblasts partly positive for CD66c) the level of CEACAA16 was observed higher in CD66c" eg cells and lower in CD66cPos cells as compared to uniform populations (Figure 4). In one specimen (ALL patient with Down syndrome), CEACAM6 wasn't increased in CD66c PoS fraction. Western biot We further question the intracellular CD66c positivity in surface CD66c negative cell lines. We performed Western bIot as described by Sugita et a!. [13 J on REH (TEL/ AMUPos ) and RS4;11 (MLL/ AF4 pos ) celllines and found no CD66c proteín (Fígure 5). Furthermore we found NALM-6 (surface CD66c neg, no translocatíon) cell line negative. Two BCR/ ABL and four hyperdiploid (aH surface CD66c Pos ) diagnostic samples used as positive controls were positive, with the similar1y narrow band contrasting to broad band detected in granulocytes (Fígure 5), suggesting different glycosylation in keeping with report by Sugita. Stability of surface expression from diagnosis to relapse All relapsed patients up tiJ1 12/ 2003 with available information on CD66c expression at diagnosis and at relapse were used to assess stability of CD66c expression . Comparison of CD66c expression in 39 cases of relapsed childhood ALL cases to theír ímmunophenotype at díagnosis
Page 5 of 11 (page number not for citation purposes)
E. Mejstříková, strana 92
BMC Cancer 2005, 538
http ://www.biomedcentral.com/1471-2407/5/38
10-
lO-
lO-
CD66c-neg
CDooe-oog
10 3
103
10'
102
10'
10 '
CD66cneg
CD86C-pOS
CD6Se-pos
CD66cneg from heterogenous
1000------------------~------------~
p
P
o:: 100 U a..1<.0 c J-
0.0001 0.0012
2 10
0:: «
IDU
.E
«
J-W
1
mU
o o
ID
o::
0,1 ~
Figure 4 Transcription of CEACAM6 versus surlace CD66c expression on sorted cells. FACSsorted CD66c surface negative (CD66c neg ) or positive (CD66c Pos ) ALL Iymphoblasts, five patients with heterogeneous CD66c expression were sorted into both CD66c negative and CD66c positive fraction (lines connect sorted fractions from the same specimen) . Mann-Whitney test was used to compare groups (n 32). CEACAM6n value is normalized to beta-2-microglobulin (see Methods).
=
Page 6 of 11 (page number not for citation purposes)
E. Mejstříková, strana 93
BMC Cancer 2005 , 5:38
fl)
(/J
Q)
Q)
~ "5
~ "5
aJ
co ....
U
c
"O
Cl
~
-
Ci
aJ
'E !>.
~
O
co
aJ
Ol
'0
...J
...J
c
~
127 kO
httpl/www.biomedcentraLcom/1471-2407/5/38
~
"O
:§ Cl.
...
i5
~
>.
~
"O
'0
Ci
:o... 8. >. ~
"O
'0
Ci
'E
~
>.
~
::I:
UJ
a:::
'9
....-:r
:E
a:::
z
Cf)
...oJ
<:
--- - -
-
CD66c
~~
(3-actin
Figure 5 Western biot of granulocytes, ALL samples of CD66c positive cases and surface CD66c neg cell lines with TEUAMLI pos (REH), MLUAF4pos (RS4; 11) translocation and with no fusion (NALM-6),
100
00
O
O
O
80 Q) (/J
g.60
O
Q)
O 00
O
o::
O
Ví 40 O 20
- - - - - - - -
O
O
20
40
60
80
100
Diagnosis Figure 6 Stability o( CD66c (rom diagnosis to relapse. Each circle represents one patient (n = 39). Percentage of CD66cPoS blasts at diagnosis is plotted against percentage of CD66c PoS blasts at relapse. Regression line with 95% confidence R2 =
0.755
revea led that both negativity and positivity of this antigen was retained from diagnosis to relapse (Figure 6; median time to relapse 2.5y min 0.3y, max 5.3y). Although the q uantitative levels of CD66c expression differed in some patients (median difference 0.0%, standard d eviation 21 %), no case of CD66c comp lete loss or gain was found in our cohort. Prognostic significance of CD66c expression Only B-p recu rsor ALL patients treated on the same ALL BFM 95 treatment protocol [20] (n = 254 ) were eva luated for prognostic impact. The prognosis did not differ for cases with either CD66cPOS blasts exceed ing either 20% (Figure 7) or any other cutoff value tested (5%, 10% and 50%, d ata not shown).
Next, we asked whether CD66c expression correlated with the risk fac tors used in ALL BFM-95 protocol for stratiflcation into risk groups 12l J. No difference in relapse free survival (RFS) was noted when analyzed separately for each risk group or higher and lower initial leukocytosis (cutoffvalue: 2 x 10 4 cells per ml), age group or response to prednisone (gro ups as in Ta ble 2) . When analyzed with respect to a genotype, we found no prognostic va lue of CD66c in any defined group (BeRl ABlYos, TEL/ AMLlPos, hyperdiploid ALL and no ne of the above-menti oned genetic changes, Figure 7 and Table 2).
Page 7 of 11 (p age number no/ for ci/a/ion purposes)
E. Mejstříková, strana 94
BMC Cancer 2005, 538
http://www.biomedcentral.com/1471-2407 /5/38
1,0
1,0
'.
0,8
0,6 0,4
.+ .-- •
0,6 0,4
Relapse fr. . survlval
Relapse free survival
p=O,77
OVERAll C068c posltive
0,2
TEUAML1 0,2
n=109
p=0 .95
CD66c positive
CD66c negatíve n=145
0,0
n=2
CD66c negative n=77 0,0
o 1,0
4 3 Time (years)
2
+ •• ~ +
0,8
,
6
5
•
". ~"'
7
2
O
4 3 Time (years)
5
6
7
5
6
7
1,0
'
0,8
.. ' _ ' f l
0,6 0,4
~
0,8
0,6 0,4
Relapse free survival Hyperdiploid
0,2
CD66c positive
Relapse free survival no (TEL/AML1. BCRlABL. MLUAF4 or hyperdiploidy)
p=0 .35 0,2
n=55
CD66c positive
CD66c negative n=7
0,0
2
3
4
5
6
7
Time (years)
n=45
CD66c negative n=59
0,0
°
p=O .15
o
2
4 3 Time (years)
Figure 7 Relapse free survival of cases with CD66c pos (blue line) or CD66c neg (red line) B-precursor ALL. Unselected consecutive patients treated on ALL BFM95 protocol (median follow up 3.64 years). Since surface CD66c associates with genotype, separate analyses for distinct genotype subgroups are shown.
ln contrast to the study by I-Ianenberg et al [221 , there was no correlation between initial leukocytosis and CD66c in our cohort (Table 2). Discussion aur data on childhood B-precursor ALL show that CD6 6c is more frequently expressed than the myeloid antigens included in the standard immunophenotyping panels for ALL. To our knowledge, CD66c is the most frequent myeloid ma rker in childhood ALt. This, together with the tight correlations between CD66c and genotype [5], makes CD66c a perlinent object of research on aberra nt expression regulation.
ln line with the data from Sugita, we confirm the specificity ofKOR-SA3S44 clone moAb for C D66c by CEACAM6
mRNA detection and by cross-blocking of KOR-SA3S44 binding by representative 9A6 clone, that suggests a spatial proximity of the two epitopes recognized. Furthermore we show that all CD66c PoS ALL specimens show a similar extent of glycosylation as cell lines analyzed by Sugita, which differs from the extent of glycosylation in granulocytes. Since there is a strong correlation of ALL genotype and CD66c expression, we hypothesized lhat surface CD66c expression would be controlled by gene transcription rather than by targeting to surface from intracellular stores as proposed by Sugita [13J. In accordance with this, both intracellular staining and Western hlot failed to identify cytoplasmic CD66c protein in any surface CD66c,1ťg cells. Down the same line, no CEACM16 transcript was
page 8 of 11 (page number no/ for ci/a/ion purposes)
E. Mejstříková, strana 95
BMC Cancer 2005, 5:38
http ://www.biomedcentral.com/1471-2407/5/38
Table 2: Correlation between risk factors and CD66c expression. The distribution o f CD66cPoS and CD66c neg cases (cutoff 20%) is shown. In addition, no difference was observed in the RFS of the risk-defined subsets based on the CD66c expression (Iog-nnk test pvalue> 0.05 in all analyses). Only patients treated by a single ALL BFM-95 protocol are shown here (n 254).
=
CD66cPoS cases
CD66c neg cases
p.value (chi·square)
AII patients
109
145
N/A
Prednisone poor responder Prednisone good responder
9 100
12 133
n.s.
Initial leukocytosis = > 20 x 109/L Initialleukocytosis < 20 x 109/L
28
44 101
n,S.
81 2
77
P < 0.0001
TEUAMLI BCRlABL MLUAF4 Hyperdiploid Other genotype (not TEUAMLI, BCRlABL, MLUAF4 or hyperdiploidy)
7 O
55 45
7 59
Age I-S Age >5
59
88
50
57
Standard risk group Intermediate risk group High risk group
40 54 15
58 72
detected in surface CD66c ne g lymphoblasts. Overall our data suggest that transcription is the checkpoint that leads to surface expression, rather th en the former model, which proposed that all malignant lymphoblasts generate the CD66c molecule but only some of them target it for the ce ll membrane. lnterestingly, importance of this molecule was shown in a model of co lorectal carcinoma where transfection with CEACAM6 inhibited anoikis (l0), high CEACAM6 predicted high risk patients with resectable colorectal cancer (9) and CEACAM6 gene silencing decreased resistance to anoik is in vitro leading to inhibition of metas tatic ability in mouse model (ll). Although the function of CEACAM6 in ALL blasts is stili unknown, this molecule's function has been recently associated with pathogenesis of other types of cancer in man [10-12,23,24 ]. Study of anti-CEACAM6 immunotoxin-based therapy in mouse model of pancreatic carcinoma was published (ecently [25].
50 far, prognostic significance of expression of myeloid antigens CDl3, CDI4. CD33, CD65w, COl lb and CD15 has been studied with conflicting results (summarized in [26]). As determined in our large cohort of patients treated on ALL BFM 95 protoco!, no prognostic significance of CD66c could be revealed in genera!, nor when we analyzed separate risk groups or TEL/ AMLl pOS, BCRj ABL-
n,s.
n.s.
15
pos, hyperdiploid and other B-precursor ALL cases separately. Furthermore, instability of aberrant expression was reported for most myel o id markers (CDl3, CD14, CD15, CD3 3 and CD65). Stabil ity of expression is a major concem of flow cytometric studies of MRD . In present, use of multiple CD markers is widely recommended to prevent MRD underestimation due to the immunophenotype shift (discussed in [15 ,27J). In current study we show for the first time that CD66c expression stays quaJitativel y stable from diagnosis to relapse in all relapsed cases studied. This finding, together with high frequency of CD66cPoS cases, supports inclusion of CD66c into a moAbs panels for MRD detection in patients positive for this CD marker at diagnosis. However, anecdotal downregulation ofCD66c expression during chemotherapy has been observed [15 J, but has not been methodically studied ye L Any temporary downregulation might lead to falsely lower values of MRD measurement - thus, it would be worthwhil e to disclose whether this phenomenon occurs (egularly at certain point.s of chemotherapy. Mutual excl usiveness of MyAg expression as wel! as differmt stability of CD66c compared to other MyAgs [28J chall enges the general practice of prognostic evaluation of MyAgpos ALL cases as a group [26] and favors individual
Page 9 of 11 (page number not for citation purposes)
E.
Mejstříková,
strana 96
http//www.blomedcenlral.com/1471-2407/5/38
BMC Cancer2005, 538
evaluation of contributionjregulation o f each MyAg for blast cell .
Conclusion CDGGc presents some of the tightest associations wilh ALL genotype. Although our findings indicate that CDGGc is unlikely to gain a practical importance as a prognosis predictor, there are several reasons to focus on it in diagnostic and MRD studies. CDGGc, apparently the most frequently expressed aberrant antigen in childhood ALL, is very useful in discriminating leukemic blasts from nonmalignant cells. Aberrant ex:pression remains a puzzling phenomenon that warrants further investigation . If it is confirmed by tec hniques sensitive enough that the so called "aberrant markers" are truly not expressed on any subtle population of lymphoid precursors, there will be an opportunity to find new targets for specific ALL therapy (e.g. monoclonal antibodics against differently glycosylated form of CD66c) thal will spare the nonleukemic precursors, lhus reducing th e treatment toxicity.
4.
5.
6.
7.
8.
9.
10.
II.
Competing interests The author(s) declare that they have no competing interesls.
12.
Autho rs' co ntributions
13.
TK performed flow cytometry, cell sorting, RQ-RT-PCR study and drafted the manuscript, MV ca rried out the Western bIot study, EM acquired and analyzed patients flow cytometry data and performed the statistical analysis, JM design ed and assisted to the RQ-RT-PCR study, JT designed RT-PCR, did the genotype detection and critically discussed the manuscript, JS contributed to the study design and organization and OH conceived of the study, analyzed data and drafted the manuscript. All authors read and approved the final manuscript.
14.
15.
16.
Acknowledgements This work was supported by the Grant Agency of Charles University #44/ 2001 and #65/2004, IGA #7430·3 and MSM0021620813 . Superb technical assistance of j. Ridoskova, K. Pospisilova, L. Gondorcinova, P. Hanusova, K. Muzikova and M. Kalinova as well as the collaboration of all Czech Pediatrie Hematology (CPH) centers (data manager A. Vrulova, leaders: B. Blazek (Ostrava), Z. Cerna (Plzen), Y. jabali (Ceske Budejovice), V. Mihal (Olomouc), D. Prochazkova (Usti nad Labem), j. Scary (Praha), j. Sterba (Brno), j. Hak and K. Tousovska (Hradec Kralove)) is highly appreciated. V. Horejsi is acknowledged for consulting in molecular immunology. We th.nk to F. Grunert for providing us with a sample of 9A6 clone of CD66c.
17.
18.
antigen in human normal and cancerOU5 colon mucosa..: com-
19.
References 2. 3.
Markert CL: Neoplasia: a disease of cell differentiation, Concer Res 1968, 28: 1908-1 914 . Greaves MF: Differentiation-linked leukemogenesis in Iymphocytes, Science 1986, 234:697-704. Mori T, Sugita K, Suzuki T, Okazaki T, Manabe A, Hosoya R, Mizut.ni S, Kinoshica A, Nakazawa S: A novel monoclonal antibody, KORSA3544 which reacts to Philadelphia chromosome-positive acute Iymphoblastic leukemia cell s with high sensitivity, Leukemio 1995,9: 1233-1239.
Hrusak O, Trkaj, Zunaj, Houskovaj, Bartunkovaj, Stary j: Aberrant expression of KOR-SA3544 antigen in childhood acute Iymphoblastic leukemia predicts TEL-AMLt negativity. The Pediatrie Hematology Working Group in the Czech Republic. Leukemio 1998, t 2: 1064-1 070. Hrusak O, Porwit-MacDonald A: Antigen expression patterns rellecting genotype of acute leukemias. Leukemio 2002, 16: 1233-1258. Boccuni p, Di Noto R, Lo Pardo C, Villa MR, Ferrara F, Rotoli B, Dei Vecchio L: CD66c antigen expression is myeloid restricted in normal bone marrow but is a common feature of CD 'IO+ early-B-cell malignancies. Tissue Antigens 1998, 52: 1·8. Kishimoto T, Kikucani H, von dem Borne AEGK, Goyert SM, Mason OY, Miyasaka M, Moretca L, Okumura K, Shaw S, Springer TA, Sugamura K, Zola H: Leukocyte Typing VI. New York, London, Garland Publishing Inc: 1997: 1342. Klein ML, McGhee SA, Baranian j, Stevens L, Herca SA: Role of nonspecific cross-reacting antigen, a CD66 cluster antigen, in activation of human granulocytes. In(eC[ Immun 1996, 64:4574-4579. Skubitz KM , Campbell KD , Skubitz AP: CD66a, CD66b, CD66c, and CD66d each independently stimulate neutrophils. ) Leukoe Biol 1996, 60: I06-1 17. janescheff P, Terracciano L, Lowy A. Glatz-Krieger K, Grunert F, Micheel B, Brummer j, Laffer U, Metzger U, Herrmann R, Rochlin C: Expression of CEACAM6 in resectable colorectal cancer: a factor of independent prognostic significance.) Clin Onco/2003, 21 :3638-3646. Ordonez C, Screaton RA, lIamzis C, Stanners CP: Human carcinoembryonic antigen functions as a general inhibitor of anoikis. Cancer Res 2000, 60:3419-3424. Ouxbury MS, lto H, Zinner MJ, Ashley SW, Whang EE: CEACAM6 gene silencing impairs anoikis resistance and in vivo meta· static ability of pancreatic adenocarcinoma cells. Oneogene 2004,23:465-473. Sugita K, Mori T, Yokota S, Kuroki M, Koyama TO , Inukai T, lijima K, Goi K, Tezuka T, Kojika S, Shiraishi K, Nakamura M, Miyamoto N, Karakida N, Kagami K, Nakazawa S: The :KOR·SA3S44 antigen predominantly expressed on the surlace of Philadelphia chromosome·positive acute Iymphoblastic leukemia cells is nonspecific cross-reacting antigen.50/90 (CD66c) and invariably expressed in cytoplasm of human leukemia cells. Leukem;a 1999, 13:779-785. Campana O, Couscan-Smith E: Detection of minima' residua' dis· ease in acute leukemia by Ilow cytometry, Cytometry 1999, 38: 139-152. Cam pana O, Coustan-Smith E: Advances in the immunological monitoring of childhood acute Iymphoblastic leukaemia. Sest Pract Res Clin Hoemotol ~002, 15: 1-19. Bene MC, Castoldi G, Knapp W, Ludwig WO, Matutes E, Orbo A van't Veer MB: Proposals for the immunological classification of acute leukemias. European Group for the 'mmunological Characterization of Leukemias (EG lL). Leukemio 1995, 9: 1783-1786. Chomczynski p, Sacchi N: Single.step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction, Anal Bioehem 1987, 1.62: 156-159. Baranov V, Yeung MM , Hammarstrom S: Expression of carci· noembryonic antigen and nonspecific cross.reacting 50-kDa
20.
21.
parative ultrastructural study with monoclonal antibodies. Concer Res 19'14, 54:3305-3314. Madzo j, Zuna j, Muzikova K, Kalinova M, Krejci O, Hrusak O, Otova B, Stary j, T rka j: Slower molecular response to treatment predicts poor outcome in patients with TEUAMLI positive acute Iymphoblastic leukemia: prospective real·time quantitative reverse transcriptase.polymerase chain reaction study. Concer 2003 , 97: I 05-11 J. Muller Hj , Beier R, t oning L, :Bluners-Sawanki R, Oorffel W, Maass E, Muller·Weihrich S, Scheel-Walter HG, Scherer F, St.hnke K, Schrappe M, Horn A Lumkemann K, Boos j: Pharmacokinetics of native Escherichia coli asparaginase (Asparaginase medac) and hypersensitivity reactions in ALL-BFM 95 reinduction treatment. Br) Hoemotol 200 I, 114:794-799. Oworzak MN, Froschl G, Primz D, Mann G, Poeschger U, Muhlegger N , Friesch G, Gadner H: Prognostlc significance and modalities
Page 10 of 11 (page number not for cilahon purposes)
E. Mejstříková, strana 97
BMC Cancer 2005 , 5:38
http{/www.biamedcenlral.cam/1471-2407/5/38
of flow cytometric minimal residual disease detection in childhood acute Iymphoblastic leukemia. Blood 2002, 99 : 1952-1958. 22. Hanenberg H, Baumann M, Quentin I. Nagel G. Grosse-Wilde H, von Kleist S, Gobel U. Burdach S. Grunert F: Expression of the CEA gene family members NCA-5 0/90 and NCA-160 (CD 66 ) in childhood acute Iymphoblastic le u kemias (ALLs) and ln cell lines of B-cell origin. Leukemio 1994, 8:2 127-2133 . 21 Scholzel S. Zimmermann W, Schwarzkopf G. Grunert F, Rogaczewski B, Th om pson j: Carcinoembryonic Antigen Family Members CEACAM6 and CEACAM7 Are Differentially Expressed in N ormal Tissues and Oppositely Deregulated in Hyperplastic Colorectal Po ly ps and Early Adenomas. Am j Pothol 2000, 156:595-605 . 24 Duxbury MS, Ito H, Benoit E. Zinner Mj, Ashley SW. Whang EE: Overexpression of CEACAM6 promotes insulin-like growth factor I-induced pancreatic adenocarcinoma cellular invasiveness. Oncogene 2004, 23:5834-5842. 25. Duxbury MS. Ito H. Ashley SW , Whang EE: CEACAM6 as a novel target for indirect type I immunotoxin-based therapy in pancreatic adenocarcinoma. Biochem Biophys Res (ommun 2004. 317:837-843 . 26. Putti MC. Rondelli R. Cocito MG. Arico M. Sainati L. Conter V, Guglielmi C, Cantu-Rajnoldi A, Consolini R, Pession A, Zanesco L. Masera G, Biondi A, Basso G: Expression of Myeloid Markers Lacks Prognostic Impact in Children Treated for Acute Lymphoblastic Leukemia: Italian Experience in AIEOP-ALL 88-91 Studies. Blood 1998. 92:795-80 I. 27. San Miguel jF, Ciudad j. Vidriales MB, Orfao A. Lucio P. Porwit-MacDonald A. Gaipa G, van Wering E. van Dongen]J: Immunophenotypical detection of minimal residual disease in acute leukemia. (rit Rev Oncol Hemo!Ol 1999, 32: 175-185. 28. Mejstrikova E, Kalina T, Trka j, Stary j, Hrusak O : Correlation of CD]] w ith poorer prognosis in childhood ALL implicates a potent ial of anti-CD]] frontline the r a py. Leukemio 2005. in press:.
Pre-publicatíon history The pre- publicati on hislory for lhis paper can be accessed here : bttp-l!www bjQmedceo !ra! com/1471-24Q7/5138/prepllb
Publishwith Bio MedCentral andevery scientist can read your work free of charge "BioMed Central will b e the most significant development for disseminating the results of biomedical research in our lifetime. "
Sir Paul Nurse, Cancer Research UK
Your research papers will be: • available free of charge to the entire biomedical community • peer reviewed and publishedimmediately upon acceptance • cited in PubMed and archived on PubMed Centra I • yours - you keep the copyright Submil your manuscrlpl here: http://wv.!w.biomedcentral.wm/infotpublishing adv.asp
O
I
BioMedcentra!
Page 11 af 11 (page number not for citation purposes)
E. Mejstříková, strana 98
Př íloha
4
Residual disease monitoring in childhood acute myeloid leukemia by multiparameter tlow cytometry: the MRD-AML-BFM study group Claudia Langebrake, Ursula Creutzig, Michael Dworzak, Ondrej Hrusak, Ester Mejstrikova, Frank Griesinger, Martin Zimmermann, and Dirk Reinhardt
Journal ofClinical Oncology (fF 13,598)
E.
Mejstříková,
VO L U M E
24
strana 99
NUM B E R
22
A UGUST
1
2006
JOURNAL OF CLINICAL ONCOLOGY
ORIGINAL
REPORT
Residual Disease Monitoring in Childhood Acute Myeloid Leukemia by Multiparameter Flow Cytometry: The MRD-AML-BFM Study Group CIc/lidi" La/lKťbrake, UrSIII.l Creutzig, Michael DW/lTza'" Ondrej Hrusak, Ester Mejstrikora. Frn/lk Griť5Í11ger, MlllTill Z iml11enllllllll, tlnd Dirk Rcinhllrdt
A B STR
A C T
Purpose M onito ri ng of residual disease (RD) by flow cytometry in childh ood acute myeloid leukemia (AMU m ay predict outcome However, the opt imal time poi nts for investigation, the best antibody comb inat ions. and most im porta ntly, the clinica l impact of RD analysis rema in unclear.
Patients and Methods Five hundred f orty-two specimens ol 150 chi ldren enrolled in the AML-Ber lin-Frankfurt-Mu enster (8FM) 98 study w ere analyzed by l ou r-co lor im munophenotyping at up to lour predefined tlme pomts du rmg treatmem, For each of the 12 leukemia-associated immunophenotypes and tlme po ints, a threshold level based on a previous retrospectl ve analysis of another cohort of ch il dren w ith AML an d on control bone marrows was determined.
Results
-rs' dcsdosuros of polenlldl con ·
~ess
lepfln t HK1UftSIS to C~udra
ake, PIlD. Dep,ortl11en1 ol Padlf'orn3,nlogv dnd Onc;ology ,
g-Str3SS6 1. [)·30625 Hanno..,er, lny, &-m3d.
Rega rding all four t lme po ints, there is a statistica lly sig nificant differen ce in the 3-year event-free survival (EFS) in thos e children presenting with immu nolog ically detec table blasts at 3 or more time poin ts . The levels at bone m arrow punc ture (B M P) 1 and BMP2 turned aut to have the most sign iflcant predictive value for 3-year-EFS: 71 % ::':: 6% versus 48 % ::':: 9%, P LOg.Ra nk ..- .029 and 70 % ::': 6 % ve rsu s 50% ::': 7%, PLOQ Ra n k = .033), resulting in a more than two-fold risk of re lapse , ln a m ultiva nate ana lysis, using a comb ined risk classificat ion based on morphologically dete rm ined blas ts at BM P1 and BMP2, French-American-B ritish class ification, and cytogenetics, the influence of immunologically determined RD w as no longer statistically significant.
Conclusion RD monitori ng before second induction has the same predictive value as examining levels at four diHerent tim e poi nt s dUrlng intensive chemotherapy. Cornpared with commonly defined nsk factors in th e AML-BFM stuo les, flow cy tometry does not provide add itional information for outcome prediction, but may be helpful to evaluate the remission status at dav 28.
filnot!b",kf' _ cIaLJd~~
J Clin Oncol 24:3686-3692. © 2006 by American Society of Clinical Oncology OOOgy
,J.IIl3X1OOI2422 ·36861S20 00
o1200/JCO 2005 05 4312
INTRODUCTION Minima! residua! disease (MRD) monitoring In childhood ;Ind adulL acute myeloid 1e1lkemiJ (AML) using flow cytometry is stiLl ullder discussion in terms of the progllo~tic impacr, tbe optímal time poillťs for ,rn,J1ysis, and the best antibody combina· rions. AML blast cell s do not express specific antigens lhal' could snvť as single and unambiguous murkers for RD in regeneratillg bone marTOW, Ir is thcrdim: necessary to carefully charJcterize combinatiullS of anrigells Ihat are "ble ro sensitively (letecr residu;ú blasr cdls among llormal hematopoietic ccUs during treat.ment. Anot!Je\" obstacle for MRD mo nitoring in AML is the insrability ofrhe b!ast cell antigen cxpression partern. As previously sbown by
us' and others,2.'1 thl' vast majority of AML cases undergo J shift of antigen expression pattem between diagnosis and eventual relapse, II is therefore indispensJble for MRD evalualion to monitor a wide range of leukemia-associated imll1unophcno · types (li\IP). Until now. there are only a few rcp0rls about thc prognostic relevancc of RD monitoring in pcdiatric S,6 and aduit 7 - III /\M L Thcseinvesligatiolls wen: based on different therapy reg-imens and hav.: employed divergent rechnical 'Ipproaches in terms oť utilized lATP, rime poinrs of Jn;llysis during thcrap)', and rhe posirivity thresho!Js lIsed for OutCOIllC predkt.ion, The objectivc of lhis st udl' was to determine t he following assessment criterÍ;i lol' the i\ML-Berlin Frankfult-l'v!uenstcr (I3FM ) tTeatment strJtegy: lhe
Information downloaded from www.jco.org and provided by BIBLlOTHEK DER on July 31,2006 from . Copyright © 2006 by the American Society of Clinical Oncology. AII rights reserved.
E, Mejstříková, strana 100 Residual Disease Monitoring in Chi ldhood AML
(!1sitivity of specitic Li\IP b'ISed lln '1 [et[ospective :1l1alysis of chilťn wirh AML; thé' appropriat cness of Rl) a: 'essment by lnultidiens ional How cytometry for outcome prcd iction in ch ildren with \M L; the most predictive time point cluring rher:1py; Jne! the ditional valu t: of tlow yto met ric assays for o ut come precliction 1cu mparison to known risk ťactors. II
Study Design Thc AML-BFM MIU)stlldyiscolllpo,ed oirwo phdses. Pha,c A includes eSlablishment
r Jc novll A lL bctwccn J ~ llu a r)' I. 2()(L and Iuly 31,21104 wcrc "Iigiblc for prospcctivc cvaluiltiun. cluJcd WlTé ehildrCll with acu le promydocytie :ukcmia (AML French-AJIIl'riean· Rrirish [fAB ] MJ). witll t( 15;17)/P LL-
Tabl. 1. Pal lem CharaClenstlcs al Diagnosis fo r Both Pal'1s ol Ihs Study Phase A IretrospectIVel
CharaclerlsllC ~
al
dl~gn o s os ,
No.
%
Rang
7.59
9. 98
02-17.7
0 06-200
15.400
12, 800
200·500.000
500-550,000
marrow blasls, %
Medlan ~ange
34(31
e
69
62
18-98
15-99 52/ 8
7Sn5
Diagnosis Diagnosis anJ cbssincation werc ~stablishcd accorJing 10 tlll' critcl13 of thl' FAB IH " group by thc refcrcn.:e IaboratolY ni th~ AivlL-llFrvl studic's in Muenster a nd w,'re rl"\~cweJ by all ce'CpClt group of indcpc'ndcnl' hcmato]ogists. In addilion 10 thc Pappcn.heim stainru bone lIlilrr"W anJ bluod Slllcars th e fo 1J0wilig cytoehelllical stainillgs 'VlTC pCrf()ľlllcd: pcriod.ic "cid Schiťf. rTlydoperoxidasc. a1pha-IlJphtyl-acetatc cstcrasc. "Ild aei" phos ph" tasc. Tht' diagn(,ses llfMO and M7 sllbtypcs \Vcrc always cOlllÍľl ncd by illlmunulogical mcthods. C}'1ogcnetic and lllolcculd rgellctic data IVcrenhtainl'd lnlllllhc rcfi.'lTll ce laboratory oflhe AML-13FM stlld y (J. Harbort, Gic.,scn. GmnaIlY)· Multiparameter Flow Cytometry Fom-color Ilow cytomťtI)'-a ccordin g to thc COIlSCllSlIS p,lllcl nf I~hc AML-BFM MRD study gmup ~s prc\~()llsly dcsnibcd in d~taiľ o__ Wd' perfonncd at !hli immunology labomtorics at the University ChilJrcn 's Hospil,lI. MUCllstcr, G~rmany. the St Anna Childrl'lú Hospit',,1. ViCllll". AuslrÍ;l. tlte
IIp.L
Range 'Mti
0/0
No.
year,
Meolan eukocytes, Medlan
PIKlse B (prospec Ivel
.RARa and childrell ,vítl\ Iriso Oly 21 heC3 USl' they cxhibit biolllgically diffcrcnt ICtLkclllias Jml reccivc slightly ditfercnt chclllothcml'Y. Altugcthcr. 542 sa mplcs from 150 children (Tahle II \'\Iere "vailablc. This (ohorl is rcprl's~ nf;ltive '''' comp~rcd wil h the overaJJ stuJy coUcctive III terlllS nf 3gC. sex, WBC, f'AB cla. sifica tion. and risk-gr0l1p allocation. From these 150 childrcn. five childrcn "ilh cktectablc hlaSl cclls hy tlow cytoll\clry, hu t an dntigen eA-pression patrcm that IVas not cClvcrcd by the LA]P.' used fo r RD monitoring (CD33/ 7/117 / ~6. eD3J/56. CO?/3.' ) \'\Icre cxdllded . Thc standard risk gmup (SR) defUlition thal' is gCllcrall uscd for slrati · hcation in the AML-BFM studic, cOlllpriscs FAB subtype' (MIIM2 + ucr rod.>. M3. M4co). favorahlc cytogcnelics, alld lllorphologic~ lI y dctcnnincd hone lll'lITnW hlast.> oťless than 5'Yo at day 15 (not rl'quir~d lór FAB M3).o~ For this a.Ll<1lysis. the morphologic cvaluation at cby 2~ "flrcatmcnt « 5% hlasl~ in bone lnilrrow ) \VilS induded as dn additional par'llnctcr for tllL' SR groul' (herdn rCIl:rrcd to as cxtcnded AML-RFM ri_,k). Thc datl' uf Ci!ch bone lnrc third thcmpy course; BMP4: bcfore fOlllth tllcrapy (oursc) wcre indudcd in Ihc study ( T~bk 2) . Bone marrow specimclls were obrai nťd aťter inrnrmcd mnscnl ťmm each patient or eadl pal1 c nťs guardian. 1\11 chilJrcn wc r~ Irciltcd acwrding to thc GC rIIldll AlVIL-BFM YH study (~S to tll .. Ireatlllcnt schcdubscc C r~ul2ig ct aILl ) . Al! illVL's tigatious pcrforlllťd had bccll :tppnlved hy Ihc 10(,11 cthic.s committccs und were in 'lCcord'lllce ",ith ~n ,lss umncc hlcd \\~ th Jnd approvcd hy thl' Depdltmcnt of Health dnd HUITlJ.l\ Serviccs.
50/50 Inlens\tlCatlon
6
20
M7
5 B
31 34 6 12
Other/nOl classifiable
4 20
22
G 61 62
4 41 41
4
3
12
8
6
5
3
31
24 22
16
16
11 15 32/68
Malntenance t ycar
Ar.·e olh
~a'Volype
Normal 118.211
5
Invll6i
lIq23 • -, group (SR/HRI
15 16/49
65
8 3 23 33/67
22 48/ 102
~
15
:t
@-CsF
±
I'G-CSF
T=-----~T~=T~--~~T~--~O~------~T~---=TH'----~o~---Day
1
15 21-28 BMP l BMP2
42-56 BMP3
70·84
112
140
360
BMP4
150
AllorevlStions: FAB, French-Amencan-S"llsh; SR. stan aod nsk; HR. hlgh nS. 1hree chlldren presenterj w olh "" 20% tllasls al dlagno :s: 'o chlldren 11 7% I1i 18%1 with M2 and 118:2 11 and one chlld wlth M7 115% ln me smear. no marrow punch boopsy aVRllableJ ar. d m yeloflbrosls
Fig 1. Acufe M elold Leu em la· Berlin·Franklun -M u Alste, study 98 Ireatm cm schedule wllh bone ll) rrow punclUre
3687
Inlormation downloaded lrom www .jCO .org and provided by BIBLlOTHEK DER on July 31 . 2006 lrom . Copyright © 2006 by the American Society ol Clinical Oncology. AII rights reserved.
E.
Mejstříková,
strana 101 langebrake et al
Table 2, Olstflbu llOI1 of Eliglble Specimens 811d Tlfl 1e POlnlS
lime POInt
Days From Stan of Chemotherapy
No,of Speclfllens
Medlďn
Range
PrOlocol
SMP l
97
15
13-2 7
15
BMP2
123
29
14·56
21 ·28
BMP3
120
60
31·101
42·56
BMP4
62
96
5 l a6
70-84
punnure,
l"r1l'" Uni\'.:r, ity, " r
on:migh t mail.
wide antibody pand bascd Oll a CD33/CD3·j backhone, indepcn · dent nf thl' illitial illlln unop henl't)'p" íllcluding tluur,'scwce colljug,lted myduid nl,lrk,'rs OU- PE (Sf! Dl; Iml111InLltech, Krddd, Cermany), COl J-HT'C (MM A; [ketun Diduns(ln, Hcidelherg, Cermany), CD3J· P(~ (mHlc,O.25I ; llTllllullOtcch). CD33-APC (D3HLC,O,25 l; ImmunoIcch), and HLA·DR-FlTC ([.143, BeCI II DidullSon), l)'mphoiJ 1l1Jrker~ Cm·PE (8H8.1 ; Im rnunolcch ), CD 10· FI'J'C (Al.B2; lmlllullotech), U)19· FITC (/4. 119; lrn(llllllOt '.:h), CD56 -P E (",CAM ItU; Hecto ll DiL'kroson), tllL' act ivatioLl "ne! prnliteration marker CD38 ·PE (!-I137; Bedon Dickin50n) os wcll oS th~ pmgenitor·associ'ltcd rnorkers CD34-APC 18{;12; Becton Dickinso ll ), CD3 4- P '7 (58 J; Imlllunot<.'ch)
Data Ana/ysis/Gating Strategy Thc spccimclts \wre "naJ)'lcd using tlrc Paint' J-ga tc PRO·Software Dickinsonl ar Cytornic, RXP, (lftw
Be..1011
~ erc rCl'iewed cťlltrally in the íll1l111U10logy IaboratolY ofthc UnivCI'Sl ty ChilJrcn's Hospital Muenster, in ordcr to warmnt homogcneous datl m aJ.ysis and mlcrpl'etarioll, Rc, idual ll1a1igua.llt cdb aOlllng normal h"lTlatopoictic cdb 11m idcntiticc\ hy clu,rer onalysl' lIsing six paramcter (tórward ond sidc
Table 3, Consensus Ant.body P~ll el Used fo r RO MonitOring Withln lha AML·BFM MR D Stuoy Group An ubody FtTC
PE
Statistics Fwnl·!'rec survival (EFS) was caJculated froll1 Jatc of diagnosis to last ar hrsl evcnt (tililure to achieve rcmiss ion, rcsistant kukeUHJ, rl" bpse, scco nd Illaligllanc)', ar dCo th of any causc), F
(rilelti n~t"n,
Tube No
scaltcr propcrties as well os tour antigcns simull<me usly), An tigm positivity was inferrcd if thc fluorcscenc'c intemity could he ckarl)' sep
PC5 or APC
APC or PC7
SyTO 16
CD7
CD45
C034
2
HLA-OR
CD38
CD33
CD34
3
CD15
C01 3
C033
CD34
4
CD2
C07
CD33
CD34
CD 15
C01 17
CD33
CD34
C0 19
C056
C033
CD34
Abbrev,atlOns, RO , res,du, I d.se8se; AML·BFM, Acute Myelold l eukem ta Berl,n-Frankl urJ·Mue n ter; MAO. nllOlmal restdual dlsease, FITC , fluorescern tSO!h'ocyanate: PE. phycooryth ruo; APC, alloplJycocy nlO, pe5. phyco .rytI1t111· cyanne 5: PC7, pl1ycoerythnn-cvanlne 7,
RESULTS Definition of Cut Off Levels by Retrospective Analysis The dclcrrnination of lATP specificity has ueen Jescribc'd in detail previously, I I Tn brief, bone marrow specimt'm of 39 chiJdrcn with ,KUle lymphobhlstic leukemia, Ewing sarcoma, non -Hodgkin's lymphoma, Ol' without a malign<1nt Jisease werc evaluMed ICH the presence ,mJ amount ofdifferent LAIP, Three groups of specificity could be defined according to thc' median perce ntage of LAIP in regenerating bone marrow: low spcóficiť)' ",ith 1.0°;', Ol' more, medium specificity from 0, I tu 1,0%, and high specificit)' with Iess than 0.1%. B,lscd on these results, cJini.:aU y prognost ic relevant thresholds for each specificity group and time points wcre defined by retrospectively analyzing 25 children with relapse and 40 children without relap,e at ,1 met!ia n follow-up of 1.5 years, Lo\\' specilicily L!\lP are oni)' infol'lmtive at BMP I, whiJe high and ver)' high specificity LAIP can be utilized for discriminatioll at .,11 four rime points. According to these data, the thresholds given ill Figure 2 werec,llclllated thar are ab le to un
Serial Assessment of RD at Three or More Time Points Identifies Chi/dren With Poor Prognosis Children with at le'lst three ~pec imens unti! BMP4 (n = Y5) havť been eva luated to invest igate the impact ol' serial inmuII1010gic mon itoring for outcome preJiclion, ln :,4 chilJren, ;111 mt\lsureJ LAIP leVéI, were below th e threshold al e:lch time point. 111 13 rhilJren, the measured LA IPs Úl at least three specime ns were aboVť the Jťlermi ne,l threshold , ln 48 child ren , Dne or more LATPs were above thc lhresho ld at one or two tjme points (Table 4), Using thi.s appro:lch, it wa5 possible to ideJltify ,1 poor-risk group, chdracterized by positive flow c:yto mdric assays at three or more consecutive time poillts dltring chemotherapy, with all EFS oť3l'Yt , :!: II % (Fig :\), However, EFS wa s Ilot significantly diftt:rent berweell children who were ncgutive by f10w cytometl')' at alJ analyzťd time .J! Il Tlt N.\L 1))-' r U N I, ')\!. O NCl I[)J(;'t'
Information downloaded from www ,jeo,org and provided by BIBLlOTHEK DER on Jufy 31,2006 from , Copyright © 2006 by the Ameriean Society of Clinieal Oneology, AII rights reserved,
E. Mejstříková, strana 102 Residual Disease Monitoring in Childh ood AML
60
A Es o '"'" .g
.,.
=
• CCR (n 161 • Relapse (n = 121
>
5.0 " CCR (n : 19) • Relapse (n = 14)
40
l,j
40
J I
i
3.0
(/) 3. 0
~
~
E
20
2.0
I
:3 10
I
a.
I
3
i
0.0
f
10 0. 0
2
3
BMP Tlmepolnl
t?:-
• CCR (n = 21) • Re lapse (n = 12)
~
1 ?:-
20 .
~
r
a.
3
1.5
3
i
!
0.5
2
25
i
20
'"'" :I'
1,5
~
• CCR (n = 18) • Relapse (n = 13 )
3.0
Ul
c:-
10
00
Fig 2. lAl M edl8n percenlag
Sl>OClhcil y
Low
Wedlum
Hlgh
Very hlgh
bone marrow accordlng 10 speclhclty alld !In,., P(llnL Dala are OOlall18C by a relrospec tM! analvslS ol m!dren Wllh aCutc:> myelold ukemla suf ering ITom relapse v DI Su S conl nuous complete r I11ISSIOO tBI Cutolf leve l tas perce ta e of nU cleated calls) accorci"' O 10 LAJP 5peClf,crty
1.0
D-
!.
:5
f
0.5
t
f
00
3
BMP Timepoi nt
SMPl
C D I 31331:J4 CD34138IHUI· D A C 033l34 C0 33l3411 17
2"..
BMP2
CO l 513311 17 COI 51331:J4 C OI 5I117/33/34 C0 7!.l3134
SMP3
BMP4
0.3%
0 .2",.
01 %
0 .1%
1 .8~o
CD2I33134 CD 19!.l3134 CD56133134
0 6'110
C D 19/56rJ3134
0.4"
t
.
3
Immunophenolype
0.1%
Single Tirne Point RD Assessrnent Before Second Induction Is Most Inforrnative for Outcorne Prediction "Vhen investil:(arin g thc four difterent timl' points sep.tralcly, we found th,tt only at BMPl and BMP2, there is a statistiwlly signitiCJnt difference between RD -positive and RD-negative children in the 3-yectr EFS (Fig 4; 7l % :!: 6% II 48'Yr, ::+:: 9%; P" ,~. 'l ", k = .029; 70% ::t:
points Jnd those with on <:: or two ['os itive tim e points (73% ::+:: 8% v 61% ± 7%, P - .43). Combinin g the good ~nd Che intennediate gruup, the diffen:, nce in contrrtS l to thc poo r gruup was statisticaUy ,igniticant: P"FS65% ::+:: S')!u v 31 'i-h ::+:: 13')/0; P - ,02. Shifting the cut off k'vels, which wcre used for group aJJocation , did nllt result in ct b erter distinction in terms of FS (dnta not shown ).
Table 4. Risk Groups and O utcorn e Accordl ng to Residual Disease Detarmined in Chlldren W it i1 Three o r More EIIglble
R,s Group
'Gme POlnts
No. o Pauants
Rela ose
NR
7 16 6
O
Good
AII negauve
34
25
1·2 abov" Illresi10ld ". 3 abova ,hreshold
48
30
13
4
Aborevlill.ons
T"ne Poin ts Secondarv
CCR
ln erl11ed,ata
poor
ol leukem ta-
assoctated IIl11l1unopl1enOlvpes (LAIP' SEl i"
BMP Timepolnl
B
2.
f
35
2.5
Ul
i
BMP TimepolOl
3.0
t
I
,e
Deil t h ln CCR
Lpukenlia
[)
O I O
1
CCA. COnllnUOIJ complete remlSSlon. NR, nonresponse
3689
WI~w.J(l) .nrg
Information downloaded from www.jco.org and provided by BIBLlOTHEK DER on July 31 , 2006 from . Copyright © 2006 by the American Society of Clinical Oncology. AII rights reserved.
E. Mejstříková, strana 103 Langebrake et al
a.nd AMI.-BFM risk show almost the same rL k r'ltio for FFS with sil11ibr 95% 1$ (2.09; 1.00 to 4.39: 2.06: 0.87 to 4.88). Thl' influenc(' of BM P2 on tlw risk of failure is lťs~ thall thc il11pact ol' the "1\-1 L- BF,'v! risk, however, the differencťs are not signifie'1I1l. Using '111 cxtended risk gruup dassitication including l11orphologic.dly determined blasl'.' al da )' 28 as '1 covari'Ht', this rurned out to have more impacl on FF. with a RR of2.8 for both time points (Table 5). We f1l1ther ,mal)'Zed, whether the inclusion offlow c)'tomdry in the risk group as-,eSSlllent could hdp to detine more prccisely risk gr lUp:; for aJdition~1 treatment stratificalion. Therťlore, the 3-year EFS uccording to illllllunologicall)' deterlllined RD W.15 caklllated sep'lrately for tht' SR and high -risk grollp. Tht' re IVa, HO dilfaenet' in both groups .lt m."IP 1 or BIvlP2 (Table 6).
IO~ lc-. I
.
--,
0.9 1
08 07
-\
'_, \...,
L-,-,ll'-=~(~L_ _ _ _ _-.-!0~7~3~,~SE~ = 0.08 -'.
.,
I
06
' ""1 _._'- ___ _._ .._ __ _
l __1
'. 0 5
06 1. SE=007
-I
L_,
0.4
l ___ _ L
0.3
O . 31 ,SE~0' 3
02 Ol
_ __ __ _
- - Good ln ~ 34. 9 evenl5) •. _ - Inrermedl8te (n = 48. 18 events) P: 1-2 = .43 ' ·3 = .OOS 2-3 = .00 7 _ _ poor (n = 13.9 events) o
2 Years fig 3. Evenl·tree survlVllI 13 ye8rs) of children wilh 81 leaSI three specimens ullli1 . ,!P4 (9ooct aH tIIn8 polnts negatJVe. Inlermed le: 1-2 lime pOlnls po itive, poor. 311me pOlnts ~OSltlv ,SE, slandard rror pro blhlY of ev nt-Iree survlvall .
,oJi' I' 50% :!: 7%: P'.og I ~,nk = .0 3). This is also tnte, ir only high and cr)' high specificity LAlPs are rt!g~rdcd (70°/r, :!: 6% \' 37 '!o~, :!: II ')10; O"'g l<.Iflk = .031; 66% :!: 6%, 1/ 45%, :!: 891,: Pu-, ~- R. nk = .045 1. Similarily, oni)' ~r BMPl and BMP2 th e RD- posirive ch ildren differed signiti-
cmrl)' from lhc RD-negative chi1dren rcg'lrding tbeircumulative 110nrrsponse and relapse inciden ť: 48%, ± 10% \' 25% -'-- 6%; P r ; "", = .0_: 1.1% ± 70/r, v 25% :!: 6%; p(;'JY = .01.. Regarding 3-year overal1 Irvival, oni)' Blv!P2 tmned out to be.l statistically sigllitica ntdisLrimnator (P = .(36).
MRD Monitoring Has No Additional Prognostic Impact Compared With Known Risk Factors Usi.ng the 1111ivari"tc COX regression model for FFS risk assess "Ient, both B~lPl and BMP2 had slatistiGlIly sign iticanl impuct: RR ,
:J5 (95% CI, 1. J3 to 4.89; P = .021) and risk ratio, 2,21 (95 % CI, !.l8 11 4.1 4; P - .013), respt'ctively.
Regarding on ly thosť children, who respolldell to treatment as by morphology at day 15, there is Cl statisticaUy 'ignificant liftlTťtKC in 3-year EFS betwecn M RD-negative "nd MRD-posítive .hildren: 6\'% :!: 6% I ' 40% ± 15% (P < .05), A multivariat e analysis on trollin g for AML-BFM risk cbssificauon, induding FAB su btype, cytogene tics, dnd morphologicuUydeterlIint'd bhst atday 15, was pe rfl'rmed. At BMP I, hoth Oowcytometry ktcťtnincd
DISCUSSION Dlle to the fJct that the antigcn t'xpressioll pattcm of AtvlL bbsts ditfers sígnineantly belween diagnosi:; and relapse I ·1 and thal no ~pť cine an tigens exist which clearly can identify leukemic blasts, we h.we devdoped an antibod}' panci that a110w5 liS to delcct residua1 blasl' cells independently of thl' iniLial immunophenotype. Based on a .D33/ D34 basís, thc LA[Ps uti1i7.ed for MRD .1ssessmenl comprisc the commonly accepted antigen expression patlerns in i\ML. ' " ' 11l We art' the tirsl' grollp appl}'ing tillle-depelHknt prognostically rekvant ClIt off leveb that have been determined by the retrospective 'lnalysb nf children treakd within the AML-flFM studies for thc occurrenceoť 12 different Lt\lPs at detined time points, instead ofelllpirie,illy defining cut ofťlevel. [n our intemationa1 prospcctiVt' stud}', \W wťre able to show th,lt tbe detection of residuul blasl cells by flow cytometry at early time points of follow-lIp (ul1til da)' 84) is a significant predictor uf treatment outcome regarding 3-yea r EFS, Espt!cially in the very early course of therapy-before thc start of lbe secontl induclioll, al day 2H frOIll Ji
1.0
A
1.0
B
Og ' - \ .. ,
08 0.7
u.\
~a...
u 48, SE = 0 .09
04
~,-{ .. _,-
- - BMP2 neg.
02
~
0.1
.. _
_..
.. BMP2 pes.
= 0.07
In = 72, 2 1 even!S)
(n ~ 49.
24
event.,
Log-rank P = .033
4
2
Ye"," t
._--_..
0.50. SE
03
. LOg-ronk P .029 O. l j,--_~ __~_~_~
2
05
04
- - 8MPl neg (n = 64, 18evenIS) .. - .. BMP 1 pos (n = 29, 15 evenlS)
02
0.70. SE = 0.06
'1.
06
"-:---'._. _. _u_...
03
l,_ \ '--o
0.7
't ..,
0.6 ,o.. 05
- ·-....T
0.8
071.SE = 0.06
1
0.9
(AI bon" l11arrow puncture IBMPI I and {B ) BMP2
3
Yea,s
(SE, standard
error probabili ty ol event-t re e survlval)
Information downloaded from www.jco.org and provided by BIBLlOTHEK DER on July 31,2006 from Copyright © 2006 by the American Society of Clinical Oncology. AII rights reserved.
E. Mejstříková, strana 104 Residual Oisease Moniloring in Childhood AMl
Table 5. U,wa"ate and Multlva"ate Cox Models for Failure-Free Survlval Tlll e Po,nt BMPI Cox Model
RR
95 % CI
BMP 2
P
RR
95% Ci
P
Unlvanal e 2.35 11 3 to 4 .89
Flow eVlOmelry r,ate
02 2 2 1 1.18 to 4 . 14 .0 1
Mu!u
flow ev omelry
2 09 100104.39 .05
AML·BFM nsk
2 06 08710488
184 O 7 0351
10 2 24
06
1.0 1 to 4 .97 .05
Mulllvanale Flow cYlOmelrv
1 98 O 95 10 4 16 .07
Extended AMl-BFM tiS"
2 80 I 05 to 7.46
I 79 0 .94 to 3 41
04 2 80
08
I 15 to 6 80 02
AbbrOVJahOJlS: BM P. DOne rTl
I
Our reslIlts are cunsistenr with repo rts on pediatrie und adult \M L regard ing th e prognostic im puct ,)ť immllnological blasl detecIlun after hrst indll(tion ..7 .Y Thc series ofCollstan -$mith et ar; comrrist's the analysis of residual blaSb at the md of remissio n indudion Iherapy b)' defined marker combin.nions dependcnt on the initi<11 immunophenotypc. [n the RD-negative grollp, their reslIlts show less IhJn O.J %, residu'll AML cells: six children (2 1%) rebpsed, and three ,hildren 00%) died in CR, wh ereas in thc RD-posilive gro up the p ropoľlion ofchildren who rdap ed and oftho~e in CR is equal (both relapse and CCR. n = 5: 38%; Jeath in C R. n = 3). T he prob<Jbil.ity IIf 2-ye<Jr over,lll survival is stali . tka lly di fferent. but the assimihl'ion of th e two (Ur'ves ;tfter th<1t time inJicatl's that a stable situalio n h.IS not yet bťt'n 'K hi evť d. Othťr invcs tigator groups to und that oni)' th e monitorin g at later :ime po in t~'·~·lo (after wnsolidatioll therapy) is signi fi c
Table 6. 3-Year EFS Ac o rdlng to R,sy Group Srrallf,catron and Flow Gy omeme Oetenn lned Resldual Olsease Level a TII119 POI I1 IS SMPI and BMP2 NO . of Patlents
~'o of Events-
3-Year EFS
SE
P
28
4
0 .86
0 .07
31
6
2
0.67
0.19
Nega!lve
36
14
O. O
0.08
POSI!IV('
23
13
0.4 3
0. 10
Negatlve
31
8
Posltrve
8
~ 's~. GroLlp
SMPI SR Ne a!lve PositIVe HR 16
MP2 SR 0.84
0.07
0.7G
01 5
66
HR NegatlVe
41
16
0 .58
008
Poslllve
40
22
0.44
0 08
22
.lboreviatrons EFS. evenl·frse survlvai; BMP, I)one nlarrow puneture: SR. :andal d risk: HR. hlgh "sk. An event is lha fallure o achreve remlss on. reslsta nl leukemlB. relapse. acond Illalignancv. or death of any cause .
comť
prediction, which m,IY linut the c1inieal usefulness of these data for risk-adapted th erapy taiJoring. Thc most re(ent reporl by Kern et al 21 ťven revcaled tnut nnly time points longer than I year after diag· nosis were indepcndently related to EFS and overull surviv;tl. During this period of therapy ťollow-up, RD positiviry may rather have repre~e nrcd resurgent le ukemia (occlllt relapse) than a surrogale marker of initial therupy response lIsable for treatment stratification . The indi vidu,ll diffťrcn(c s in the kinetics of leukemie recurrences mJ)' tnerefore impede u prospedive application oť Ihis approach for thc carly diugnosis of relapse. Althollgh we could show thut tlow eytomťtry is a reliab le and objť e live method to delect residu;tl bJ<Jst eell s in regeuerating bone m'lHOW speeimens, md that it is therdore upp rop ri ate for outwme prediction, we further wanted to know whethcr these reslIlts bear additive va lllťs for treatment str<Jtiticatioll us comp'lred with co nven !Íonal risk fJctors. otably, thc AML- BFM risk gwup cl'lssitication ll is based on initi al cytogeneťics, FAB classification, and morphologicJ II )' detectabIe blast c('Hs of more or less rhan 5% at du)' 15. Appl)'ing Ihis binar)' risk group clussi tication to the children anu lyzed, J highly signilicant d iffere nct: in terms o f3-year EFS is <1chieved without using inul1unological inlormation: 78% ::!:: 6% ve rsus 49% :!: 6% ( P = .0025). When inclllding illtormation on morphological blast collnts fr0111 day 2R (BM P2) in udditioll to the conve ntional ri sk classiti(ation (extended AM L-BFM risk clussifi(ution), the difťercnce in 3-)'ear EFS betw ťen th e two ri sk groups eve n in c rc aSťd 10 85% :: SUlc, wrsus 48% 5'110 (P = .0002 ). Thc impact of flow c}'tometry results on FFS at either lime poin t W;tS found equiv'llent to that of the AML-llFM ri sk in tertm of risk ratio and 95% Cls. \A/hen induding the extťnded AML-BFM risk classiticatio n in theCOX -model, it turned outthat immllnologi(al Rl) as a .:ovariate does Ilot contribute to a better risk group separation. It has to he mentioned that none of the recently puhlished stlldies of other groups included informa tion of RD monitoring us wmpared with mor~)hologically derermined blast cdb ut the an'llyzed time points JS part ofthe risk dassificJtion s)'stem. Consiclering our resulrs, this comparison has to be recommended in order to interprer poten tial additional va lues of tlow cytomelric invesligations colTCLťly. Applying th e covuriul es l1Sed by othe[ groups (age al diagnos is, leukoc)'te count at diagnosis. kať}'oty pe) to our dura, we ob tain similar signifi can t results for the inflllellce ofBMP I and BMP2 on outeome. However, in our study the more sophisticntt'd i\ML-BFM risk gro up classification hus been laken us the bJsis to delt:rmllle the true addi tional va lue of MRD monitoring. fn conclusion, for the AML- BFM stud ies, ri sk group strati!i.cation based Oll FAB sublype. cytogem:lics, Jnd morphologicully determined bone marrow bbsts heťore second induction dOťs not bcndi t from inclusion ofRD data,
=
3691
Informalion downloaded from www.jco.org and provided by BIBLlOTHEK DER on July 31, 2006 from . Copyright © 2006 by the America n Society of Clinical Oncology. AI! rights reserved .
E. Mejsrříková, strana 105 Langebrake et al
REFERENCES
-,
'
I. Langebrake C. Bnnhnann I. Telgler-Sclllegel 81 al: Im munophenotyplc differences betweso ,,.gn0515 and relapse ln childhood AML ' Imphcaos lor MAD monltonng CylOmeti'y S C'IO Cytom 1-9. 2005 2. Sil r Mana R. Srewart Carleton C. Dodg", ,,,hard K. BI al: Hlgh IreQuency o Immu nopl1911o'VIle changes ln scute myelold lau emla at relapse: fl'phC3tl ons tor reSIdua I dlsease deteclion (Cancer leukem a Group S Study 836 11 Blood 973574)580. 2001 3. Maeedo A. San M lguel JF. Vidnales MB. st al 'I1enotYPlc cl1anges ,n acule nlyelold IElukaemla. pllCatlons ln 11'0 d leellon ol mlnlmal resldual .'Sease J Clln Pari10l 49 15· 18. 1996 4. Oelschlagel U. Now"k R. Se"a bA. el al Stli!l aberran l antigen expresslon
:003 7. Faller N. van dm Pol MA. van Stiln A, et al.
'RD parameTers uSlnu II'lmunophenOlYplC delecn methods are l1'ghly rellab e ln predlctlng surviva1
11 acule IllYeloid Isukaem la . Leukemie 18.13801390. 2004 8. Kern W . Voskove D. Scl10eh C. 81 al: DeIE"mlnauon of relapse flS based on assessmem of m lnlmal resldual dlsease during c0111ple e ren1lssion by multil:>arameter flow cytomstry ln unse lectsd pa llents w lth acule myslo,d leuKemie. BloOd 104 : 3078-3086 . 2004 9. San Mlguel JF. Vldria 'es MB . Lopez·Berges C. et al : Early wnmunophenorypleal svaluJnon ol mimmal reSidua I dlsease ln acule myslold leu emla ldentilles diH erenl pstient risk groups and may contnbute to posllnoucllon treatmam straTi rca rl on. Blood 98 174 6- 175 1. 2001 10. Vend1ll1 A Tamburi ni A. Buec isano F. al al Chnrcal relevance of minima I resloual dlseasa de lecIlon io adull acute mye oid leuksmla J HemalOther Stem Cell Aas II 349-357. 2002 1, . Relnhardt D. Langebra e C. Creurz,g U. st al Mlnlmal resldual dissase 10 acute rnye oid leukemla ,n chlldren : S nda rdlZ3110n and evaluatlon ol immuoophenolYPlng " ' I he AML·B FM·98 srudy. Klln Padlal r 2 14 179- 187.2002 12. CrsutZlg U. ZIInmem ,ann M, Ritter J. s t al. DeloMlon ol a s anoa rd·ns group ln choldren w lth AML Br J Haematol 104 :630-639. 1999 13. CrsumQ U. Zlmmer Biln M . Relnhardl D. et al: Esrly dealhs and l reatman t·rslated morlallty ln children undergoing therapy for cu te m yaloid lelJeml Analysls ot l ha rnulliCenter d ioieel tnals AMl·BF M 93 and AML·BFM 98. J CIIIl Ol1eol 22; 4384-4393 . 2004 14. Bennen JM. Catovs y D. Daniel MT, er al. Proposals lor the classlllC3t,oo of l ha acu ts leukaa-
rnias' Frencll-All1erican-8ririsl1 (FABI co-operatlve group. Br J Hitem 10133:45 1-458. 1976 15. Bennen JM , Catovs,",y D. Demel M r , al al: Proposed revised cri ler for the classiflcallon of acule myelold leukeml A ,eport ol Ihe FrenchA marica n-Bnll 11 Cooperatlve Group . Ann Inrern Mad I03:620-625. 1985 16. Bennett JM. Catov ky D. D(iI1lel MT. 9t al: Crilena lor lha dl8gno IS of aeule ISlJkemr of m egaka ryocyle lineage (M71: A rspon of tl1 e French· Am encan ·Bntlsh Cooperatlve Grou . Ann hllern Med 103460-462. 1985 17. Cll eson BD . Cassiler h PA Head DR. st al: Report ol l ha Natlonal Cancer Instotuls-sponsore workshop 0 11 delll11110ns ol diagi10sls and response ln acule ", yelold leu'em lll. J Clin OncoI8.8 13-8 19. 1990 18. Maeedo A. Or lDo A, Vldriales MB. el al' Charactefl zaliO n of aberr an t phenorypes '" Je ute r yeloblas le "oll ·em la. AI1i1 HemalOl 70: 189-1 94. 1995 19. Reading CL. Esrey EH . Huh YO. SI al: Expres· Slon of unusual imlll unopheno!ype comblnatlons ln acule myelog9nous leu emla. Blood 8 1:3083-3090.
1993 20. Terstappen LW . Sa Hord M. Konemann S. Sl al: Fl ow C om em e ch rac tenza Ion of a~u le mysloid leukem ia Part II . Phenotypic heterogsnelty al dlagnosls lcorrecte and repubhs hed on 'ele ongl!lally pfl nted ln Leukenlla 5:757-767. 199 I I. Le'Jkemla 6.70-80. 1992 21 . Kam W, Voskova D. Schoch C. el al: ProgllosIIC IInoact of m inimal resldual dlsease as de emllned by mulllpa rameler How cytometry '" pa ti ant wlth "ClIte m yelold leukem la . Onkologie 28:260. 2005
• •• Acknowledgment We thank Carolin Augsburg., Juna Mdtzer, Elisabeth Kurzknabe. Gertralld Frlischl, and i\ngcl a Schumich for their excellen t tťchni ca l iSI,tncť . Weaho th<1nk all thr hospital persol1ncl Jnd clinicians particíputillg ill the AClIlť Myeloid Leukemia -Bedin-Frmkfurt-Muenster ,tud)' oup for provid ing bone m ,IITOW ~ample s a nd clillictl data. Thi.s ,utide i, cledicated to D r GlIenther Schellollg in honor ofhis 80th birthda)'.
Appendix The Appelldix is induded in thc full-t ext vers ion of this article , availJble onlinr at www.jco.org. [I is not included in thr PDF vcr,ion ( via Adob ei!l> Re'lder<»).
Authors ' Disclosures of Potential Conflicts of Interest ll,e aUlhors Ind,eated no potel 118. confllcts of nterest.
Author Contributions Conception and desi gn: Claudla Langel1rake. M ichael Dworzak. Ondrel Hrusak. Ursula Creutz g. D,rk Rei nhardt Administrative support: Ursula CreutZlg, Martil' Zimmermann Provision ol study mat erials Or patients: Ctaudla Langsbra.e. Ichael DWO«8k. Ondrsj Hrusak . Este r Mejstn ,ov' . Frank Gneslnger Dir k Asml1ardl Colfection and 8ssem bly ol data: Claudla Lal1gebrake. Michael Dworza . Ondrej H rllsa~ . Ester MSjstriKova. F, ank Grieslnger. Dirk Reinhardt Data analysis and interpretatio n: Claudla Lar1ge brake, Mart,n ZilnmeflTlMI'. Dlr Relnhardt Manuscript wri tlng : Claudla Langebrake Final approval ol manuscript: Ulsula Creu\Zl9. M ichael Dwo«a~ . Dlfk Reln hardl
J Ol TlH\ AL ll~' C I. L~ I ( ·.·\ l. O :-:CI II.OC)
Information downloaded from www.jco.org and provided by BIBLlOTHEK DER on July 31, 2006 from . Copyright © 2006 by the American Society of Clinical Oncology. AI! rights reserved .
E. Mejstříková, strana 106
Příloh a
5
Detectable minima) residua I disease before aHogeneic hematopoietic stem cell transplantation predicts extremely poor prognosis in children with acute Iymphoblastic leukemia Lucie Šrámková, Kateřina Mužíková, Eva Froňková, Ondřej Krejčí , Petr Sedláček, Renata Formánková, Ester
Mejstříkov á ,
Jan Starý, Jan Trka
Pediatrie Blood and Caneer (fF 2, J 64)
E. Mejstříková, strana 107
Pediatr Blood Cancer
Detectable Minimal Residual Disease Before Allogeneic Hematopoietic Stem Cell Transplantation Predicts Extremely Poor Prognosis in Children With Acute l ymphoblastic leukemia Lucie Sramkova, MO,1,2 Katerina Muzikova, Bc,1 , 2 Eva Fronkova, MO,1,2 Ondrej Krejci, MO, PhO,1,2 Petr Sedla ce k, MO, PhO/ Renata Formankova, MO, PhO/ Ester Mejstrikova, MO,1,3 Jan Stary, MO, PhO/ and Jan Trka, MO, Ph0 1 ,2* , for the Czech Pediatrie Hematology Group (CPH) Background. Thc level of minimal residua I di sease (MRD) prior to allogeneic hematopoietic stem cell transplantation (HSCT) has been shown 10 be an indE'penden t prognostie faetor for outeome of pediatrie patients with high-risk Jeu te Iymphoblélstic leukem ia (ALU. Retrospeetive studies whieh uspd (,emi-) quantitation of clone>peeific immunoglobulinfT-ccll receptor Og!TCR) rea rrangements have document ed the feasibility and practi cality of this teehnique. This approach has also been di sputcd due to the occurrence of clonal evolution and gencrall y high MRD leve ls prior to HSCT. Procedure. ln our prospeetive study, MRD before and after HSCT was monitored using quantitati ve real-time PCR in a eohort of 36 children with ALL eon secutively tran,p lanted in our center between VII 1/2000 and VII/ 2004. Results. tn 25 of 36 pati ents, MRD level prior HSCT was assessed. Seventeen patients w ere classifi ed as MRD-negati ve and eight were MRD-positive up to 9 x 10 1 . In MRD-positive subgroup, Key words:
seven eve nts (six re lapses) occurrcd post -transplant in striking eontrast to only on e relapse in MRD-n egative subgroup (event-free survival (EFS) log-rank P < 0.0001). MRD proved to be the on ly significant prognostic factor in a multi va ri ate analysis (P < 0.0001). Adopt ive immunotherapy including donor Iymphocyte infusions in patients with adverse dynamics of MRO after HSCT had only limited and/o r temporary effect. Clonal evolution did not present a problelll precluding MRD monitoring in any of pati ents suffering a posttransplant relapse. Conclusions. We show that MRD quantitati on using elonal IglTCR rearrangements successfully assesses th e ri sk in pediatrie ALL patients undergoing allogpneic HSCT. I\s our ahility to treat deteetable MRD levels after HSCT is very lirnitcd, alternati ve strategies for MRD-positi ve patients prior HSCT are necessary. Pediatr Blood Cancer © 2006 Wi ley- li", tnc.
acute Iymphobla stic leukem ia; childhood ; hematopoietic stem cell transplantation; immunoglobulin and T-cell receptor gene rearrangements; minimal residua I disease
INTRODUCTlON Despite the overall improvement in the chemotherapybased front-line treatment of the childhood acute Iymphoblastic Ieukemia (ALL) , the hematopoietic stem cell transplantation (HSCT) remains an important treatment option for the patients with resistant, very high-risk, and/or relapsed di sease. However, the curative eťťect of allogeneic HSCT is hampered by a relapse occurrence that represents a major cause of the HSCT failure. Already in 1998, Knechtli et al. showed that the level of minima! residual disease (MRD) prior HSCT represents an important prognostic factor [I]. They used a semi-quantitative approach for the detection of immunoreceptor gene-immunoglobulin and T-cell receptor genes (lglTCR) rearrangements. In their cohort, all children entering the pre-transplant conditioning with a high-level MRD suffered a post-transplant event and children with Iow-Ievel MRD had significantly poorer outcome compared to MRD-negative subgroup. Similar data were obtained in the subsequent studies, partly using the new technique of real-time quantitative PCR (RQ-PCR) [2,3]. When this quantitative technique for MRD detection was completely introduced and progressively standardized , the international Pre-BMT M RD Study Group (part of the European Study Group on Minimal Residual Disease in ALL-ESG-MRD-ALL) [4] retrospectively analyzed the pre-transp lant MRD levels and thc post-transplant outcome in a cohort of 140 ped iatric ALL patients [5]. MRD proved to
© 2006 Wiley-Liss, Inc. 001 10.1002/pbc,20794
be a highly significant (P < 0.00 I) independent factor to influence event-free survival (EFS) of this group. High MRD burden, together with a shorter duration of the fiťst co mplete remission (CR) . MLL gene rearrangements, and pro-B immunophenotype proved to be the on ly negative risk factors. This large multicenter, retrospective study was recently di sputed by lm as huku et a!. [6]. Their analysis, based on a group of95 transplanted patients (age < 20 years), showed no correlation between the pre-transplant MRD burden and the post-transplant relapse. lmashuku and colleagues made a couple of rather surprising observations: first, they found 96% of their patients to be MRD-positive prior HSCT.
lCLlP----Z:hildhood Leukem ia Investigation Prague. Prague. Czcch Republic; 20epartment of Pediatric Hem~t(Jlogy and Oncology, Charles University, 2nd MedicJl Schoo'! Jnd University Hospital Motol , Prague, Czech Republic; ~Departmcnt ol' Imll1unology. Charles University, 2nd Medi ea l School and University Hospital Moto\, Prague, Czech Republic Grant spo nsor: FNM; Grant number: 9735; Grant sponsor: MZ; Grant number: 6929-3; Grant sponsor: GAUK; Grant number: 6212004: Grant sponsor: MZ; Grant number: 00064203 ; Grant sponsor: MSM : Grant number: 0021620813 . *Correspondence to: Jan Trka, Department ol' Pediatrie Hematology and Oncology, Charles Uni vers ity, 2nd Medic al Schoo\' V uvalu 84. 150 06--Prague 5, Czech Republic. E-mail : jan.trka @ lťmotoJ. c uni .cz Received 30 August 2005; Accepted II Janumy 2006 '/olWtLEY
In te rSci enc e
E. Mejstříková, strana 108
2
Sramkova et al.
Second , their ability to predict the relapse after HSCT was hampered by the c lonal evolution of IgrrC R rearrangements. Eleven of 16 patients havin g relapse after HSCT showed a totall y diffe re nt rearrangement pattern at relapse when compared to the initial sc reening . This is in strikin g contrast to previously published data that show more than 95% of relapsed patients have preserved rearrangements allowing the MRD follow-up [7]. On the basis of their results, Imashuku et a!. dispute the practicality oť the Ig/TCR-based approach. Here, we prese nt a series of36 pediatrie ALL patients who consecutively underwent allogeneic HSCT using unmanipulated grafts at our institution. In 25 of them, we were abl e to assess MRD leve! prior and after the transpla nt using RQPCR-based detection oť Ig/TCR rearrangements. We show that detectable MRD beťore HSCT is a very strong negative
prognostic factor and that the clonal evolution of the Ig/TCR reanangements does not ham per the relapse prediction. Therefore, our study clearly demonstrates for one thing the feasibi lity of the method and for another a very strong cJinical value of this approach for the identification oť patients ar rhe risk of rel apse after allogeneic HSCT.
MA TERIAlS AND METHODS
Patients and Treatment Between August 2000 and September 2004, 36 consecutive pediatrie patients (age 1. 1-19 years) with ALL indicated to HSCT were enrolled to our study (Table I) . This cohort comprised all such pediatrie patients from the Czech Republic in the given period oť time and all transplants were
TAHLE I. Transplanted Patients With ALL
No.
I 2 3 4
5 6 7 8 9 10 II 12 13 14 15 16 17 18 19 20 21
22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
Agc/sex 111m 121m 101m 8/m 7/m 151m 101m 101m 111m 15/f 121m 191m 161m 101m 8/f 7/f 31m 10lf 21m 8/m 8/m Ilf S/m 10lf 6/m 21m 121m 9/m 51m 8/m 121m 161m 9/m 17/f 7/m 141m
Immunop henotype T-ALL pro-B cA LL T-ALL T-ALL pro- B cALL AHLlcALL cALL pracB/cALL pracB/cALL praeB/cALL cALL cALL cA LL cALL T-ALL cALL pro-B AHLlpro-B cALL T-ALL T-ALL cALUpro-B cALL pro-B cA LL pracB cA LL cALL pro-B/cAI.L praeB cALL AHL eALLlprae B cALL
Fusion gene
Remission
Conditioning
Donor (HLA march)
GVHD grade
Not detect ed ETV6/RUNXI ETV6/RUNXI Not detccted Not ,.!clccted MLLlA F4 ETV6/RUNX I BCR/ABL BCRJABL BCRJABL BCRlABL Not detected Not detected ETV6/RUNXI ETV6/RUNXI Not dctec tcd Not detccted BCRlABL MLL rcarrangement BCR/Al:lL Not detccted Not delec led SIL/fAL I ETV6/RUNXI Not dckcted MLLlAFt.» BCRlABL Not dctected BCR/ABL ETV6/RUNXI Not detected Not detec ted Not dctecteu BC'R/A BL I TV6/RUNXI BCR/ ABL
CRI PR3 CR3 CRI CR2 CRI CR2 CRI CR2 CR2 CRI CR3 CR2 CR2 CR2 CR2 CRI CR2 CR2 CR2 CR2 CRI CRI CR2 PR3 CR2 CRI CR2 CRI CR2 PR3 PR3 CR2 CRI CR3 CRI
TBI 12Gy, VPl6 TBI 12 Gy, VPI6 TBI 14.4 Gy,VPJ6 TBI 12 Gy, VPI6 TBI 14.4, Cy TBI 12 Gy, VPI6 TBI 12 Gy, VPI6 TBI 12 Gy, VPI6 TBI 12 Gy, VPI6 TBI J2 Gy, VPI6 TBI 14.4 G y, Cy TBI 12 Gy, VPI6 TBI 12 Gy, VPI6 TBI 12 Gy, VPI6 TBI 12 Gy, VPI6 TBI 12 Gy, VPI6 TBI12 Gy, VPI6 TBI 12 Gy, VPI6 BuCy, VPI6 TBI 12 Gy, VPI 6 TBI 12 Gy, VPI6 BuCy, VPI6 TBI 12 Gy, VPI6 TBI 12 Gy, VPI6 TBI 12Gy, VPI6 BuCy, Mel TBI 12 Gy, VPI6 TBI 12 Gy, VPI6 TBI 12 Gy, VPI6 TBI12Gy, VPI6 TBI l2 Gy, VPI6 TBI 12 Gy, VP .16 TBI 12 Gy, VPI6 TBI 12 Gy. VPI6 TBI 12 G y, VPl6 TBI 12Gy, VPl6
MSD MSD UD ( 10/10) MSD UCB (5/6) MSD MSD UD (10110) UD (10/10) UD (9/10) UD (9/10) UD (10/10) UD (10/10) UD (7/[0) MSD MSD MSD UD ( 10/10) UD ( 10/10) UD (9/l0) UD (9/10) U D (10/10) MFD (10/10) UD (9/10) UD (10/10) MSD UD (8/10) UD (9/10) UD (10/10) UD (10/10) UD (10/10) UD ( 10/10) MSD UD (7/ 10) UD (9/10) UD (lO/lO)
IIl. cGV HD II
II II II I, cGVHD 1/ 1/
Follow-up (months) 61 15* 55 52 4* 46
44 21 * 24*
6* 4* II II II II 1/
31i 2* 9* 32 32 31 30
O' 1/ II II II II II II II IV II II II II II II
26 26 12* 23 22 19 18 lli 1* 17 17 16 1* 15 14 12
10*
Bu, busulphan; Cy, cyclophosphamide; Mel, mclphalan; VP 16, vepesid ; CR , complete remission; PR, partial remission; TBI, total body irradiation; MSD, matched sibli ng donor; UD, unrel ated donor; UCB, unrelated eord blood; MF D, matched family donor; GVHD, graft-versus-hoSl disease ; cGVHD, chronic graft-vcrs lIs-h oSl disease . ' - time to eve nt. Pediolr Blood COl1cer DOl IO.1002/pbc
E. Mejstříková, strana 109
MRD in Children With ALL Undergoing "SCT performed at our institution. The group of patients consisted of 30 children with B-cell precursor (BCP) ALL and 6 children with T-cell ALL. Twelve children were transplanted in the first CR. all of them due to high-risk ALL (T-ALL and prednisone poor response (4), BCR-ABL-positive (7), MLL rearranged (I 20 children were transplanted in the second or higher remission; and 4 children in partial remission (without increased number of blast cells in the bone marrow but without recovery of hematopoiesis). Donors of hematopoietic stem cells were HLA-identical siblings in 10 cases, unrelated donor from BMT registries with variable rate of HLA match (from 7 to 10/1 O antigen match on high resolution PCR level) in 24 cases; one patient was transplanted using unrelated cord blood and one patient frum phenotypically identical mot her. In the majority 01' transplants (n = 32), we used a similar pre-transplant conditioning regimen based on the total body irradiation (TBl) in the dose 12 or 14.4 Gy and etoposide 2 60 mg/m , in the cord blood transplantation TBl plus cyclophosphamide 2 x 60 mg/kg. In three children under 2 years of age a busulphan-based conditioning (2x busulphan, cyclophosphamide, and melphalan, I x busulphan, cyc\ophosphamide, and etoposide) was used. In the majority of unrelated donor transplants (n = 19) we used rabbit antithymocytic globulin (ATG, Fresenius) at 10 mg/kg for 4 days. Graft-versus-host disease (GVHD) prophylaxis consisted of intravenous (i.v.) cyclosporin A (CsA) in the dose 3 mg/kg/ day in HLA-identical sibling donors and using the combination of CsA 5 mg/kg/day and methotrexate (MTX) administered days +1, +3, and +6 in unrelated donor transplants, always with transition to oral CsA in the adequate dosage. Since February 2003, we sturted reduced GVHD prophylaxis according to ALL SCT-BFM 2002 protocol where only targeted dose of CsA with required serum levels between 80 and 130 ~g/l (Fluorescence Polarization Immunoassay method) was given. Only in one ca se (the cord blood transplant) the combination of CsA and methylprednisolone was used. Incidence of acute GVHD was low in our group with I chi\d developing acute GVHD Grade I, 24 children experiencing acute GVHD grade II, and 7 children having no acute GVHD. We have registered only two cases of acute GVHD grade III-lY. Follow-up of the whole group ranges from 12 to 61 month s with median 26 months after HSCT.
»;
MRD Assessment For the MRD assessment we examined BM samples from both diagnosis and relapse, I week before the start of the pretransplant conditioning and then after HSCT, on a regular basis: days +28, +60, +100, +180, and later 9,12,18, and 24 months after HSCT (or more frequently in MRD-positive patients when adoptive immunotherapy was considered). In children with a very high-risk of relapse, additional Pediarr B/oad Cancer DOll O.lO02/pbc
3
peripheral blood (PB) samples were taken every month during first 6 months after HSCTand every 3 months later on. Mononuclear cells from the diagnostic Oľ relapse BM samples were isolated by Ficoll - Paque (density 1.071 g/ml, Pharmacia, Uppsala, Sweden) density centľifu gation. Follow-up BM or PB samples weľe proces sed by erythrocyte Iysi s. Genomic DNA was isolated by QIAamp '" DNA Blood Mini Kit (QIAGEN GmbH, Hilden, Germany), DNA was stored at -20 D C before processing. Primers and protocols for immunoglobulin heavy chain (IGH), immunoglobulin Iight chain kappa (IGK), T-cell receptor gamma (TCRG), T-cell receptor delta (TCRD) gene rearrangements, and TAL I deletions detection were desc ribed previously [8,9). Clonality of PCR products was confirmed by the heteroduplex analysis [10). Monoclonal PCR products were cut from the gel, reamplified with the same set of primers, and puritied by QIAquick PCR Purification Kit (QIAGEN). Sequencing was performed in the ABI PRISM" 310 Genetic Analyzer with BigDye™ Primer v3.0 Sequencing Kit (Applied Biosystems, Foster City, CA). Variable (V), diversity (D), and joining (1) regions of the immunoreceptor gene were identified by comparison with sequences in GenBank using the lm Muno Gene Tics (IMGT) Database (http:// imgt.cines.fr, IMGT, European Bioinformatics Institute, Montepellier, France) and the IGBlast search (http:// www.ncbi.nlm.nih.gov/igblast/, National Center for Biotechnology Information, Bethesda, MD). Patient-specific forward primers for RQ-PCR were designed using the VECTOR NTI 8 Suite Software (Informax , Bethesda, MD). Family-specific reverse primeľs and probes for IGH, IGK, TCRD, and TCRG were described previously [11-14). Ig/TCR RQ-PCR was performed in the iCycler IQTM Real-Time PCR Detection System (B1ORAD, Hercules, CA) and in the ABI PRISM 1l1 7100 RealTime PCR System (Applied Biosystems). Standard CUl'ves were prepared by diluting the diagnostic (HSCT in CR I) Ol' relapse DNA samples in polyclonal DNA fmm healthy donors. The albumin gene was used to normalize the DNA concentration and quality [15). The ESG-MRD-ALL criteria for RQ-PCR sensitivity and quantitative range (QR) interpretation were used [4). In six patients who suffered from relapse after HSCT, BM samples were re-analyzed for the presence of clonal Ig/TCR rearrangements to evaluate the extent of clona I evolution .
RESUlTS Feasibility of the Approach We were able to evaluate MRD level prior HSCT using 19/ TCR rearrangements in 25 of 36 patients. ln two cases, the pre-transplant sample was not available due to severe bone marrow aplasia, and in nine cases we did not find a target with adequate sensitivity and specificity. AII 25 patients were regularly monitored using IglTCR rearrangements (two
E. Mejstříková, strana 110
4
Sramkova et al.
targets in 14 patients, one target in 9). At leas t one target with 4 sensitivity 10- was available for all but one patient (No. 12), where only one target with sensitivity 10- 3 was found.
Clonal Evolution of Ig/TCR Rearrangements ln our cohort, clonal evolution did not preclude MRD monitoring in any of the patients. Al! post-transplant relapse sam ples when compared to the diagnos is/first relapse spec imen showed at least one stable Ig/TCR rearrangement.
MRD Level Prior HSCT Predicts Post-transplant Outcome According to MRD level in BM prior the conditioning regimen we divided our patients into two groups (Table II) . The first group (MRD-positive) consisted of patients with MRD detectable within the QR of the method. The seco nd group (MRD-negative) included patients with undetectable or very low MRD positivity (below the QR; n 2) prior to the transplant. The MRD-positive subgroup consisted of eight patients (Nos. 2, 8, 9, 10, II, 13 ,22, and 35). One of them died due to post-transplant complications (day + 66-multiorgan faiJure after gram-negative sepsis), six children experienced a hematological relapse, although all of them ach ieved
=
transient post-transplant MRD negativity, and one is alive in continuous hematological remission. However, the latter patient (No. 35) suffered from molecu lar genetic re lapse and was treated with the ad op ti ve immunotherapy with a followup of 12 months. The group of MRD-negative patients included 17 children (Nos. 1,4,12,15, 16,17,18 ,20,2 1,23,24,26,27,29,30,31, and 36), a ll of whom are alive except one and in complete hematological remi ssion with a follow-up of 16-61 months (median 26 months). One patient of this group (No. 36) suffered a relapse 10 months after HSCT. Interestingly, although MRD-negative by Ig/TCR approach, BCR/ABL fusion gene reverse- transcriptase PCR analysis prior HSCT showed borderline positivity (data not shown). EFS analysis clearly supports the hypothesis that MRD positivity prior transpl ant is a significant adverse prognostic factor (Iog-rank P < 0.000 I; Fig. I) . Further division into the subgroups with high MRD positivity (2 10-\ n = 4) and low MRD positivity ( > 10- 4 and < 10-'; n =4) did not show any effect, evidently due to a low number of patients and a high frequency of events (data not shown). Multivariate analysis (i ncluding also sex, age at diagnosis, first CR duration, type of donor, and fusion gene-BCR/ABL, MLLlAF4, and ETV6/RUNXI - presence) identified pre-transplant MRD as the only significant risk factor (P < 0.000 I).
TABLE II. MRD Levels in the Bone Marrow Before Conditioning and Survival of the Patients
Patient no .
ALL subtype
Numhcr of Igrl'CR t~rgcl:;
MRO level 1.47 x 10-
4
Relapse after HSCT
2
ETV 6/RUNX I
8 9
BCRlABL BCRlABL
2 2
5.3 x 10- 2 , 9.2 x 10- 2 Negative, l.3 x 10-4
Ves Ves
10 II
BCRlA BL BCRlABL
2
1.2 x 10- 3 2.3 x 10- 4 ,6.2 X 10- 4
Ves Ves
13
BCP
2.8 x 10- 2
No
22 35
T-ALL ETV6/RUNXI T T BCP ETV6/RlJNX I BCP T-ALL BCRlA BL BCRlABL BCP T-ALL ETV6/R NX I MLLlAF9 BCRlABL BCR/ABL ETV6/RUNXl BCP BCR-ABL
6.4 x 10- 3 ., 4.1 x 10- 4 .3.5 x 10 Negative, negative Positive (below QR) Negative Negative, ncgative Negativc Neg~tive, negative Ncgative, negative Ncgative Positive (bclow QR) Negative, negative Ncgá live Negati ve, negative NegiJtive. ncgative Negative, negat ive Negat.ive, negative Negali ve, negative Negative
Ves No No No No No No No No No No No No No No No No No Ves
4 12 15 16 17 18 20 21
23 24 26 27 29 30 31 36
Pediatr Blood Cancer 001 IO.IOO2lpbc
I
2 2 I 2 2 2 I
2 2 2 2 2 2 2
Ves
Comments Oied in remission (systemic fungal infection, severe induced GVHO after OLl) Oied in remission (system ic fungal infection) Oied in progression of disease (on treatment of relapse) Oied in progression of disease Oied in remissio n (systemic fungal infection, severe induced GVHO after OLl) Oied due to multiorgan failure in gram-negative sepsis day + 66 Alive in CR2; after 2nd HSCT Alive in CC R Alive in CCR Alive in CCR Alive in CCR Alive in CCR Alive in CCR Alive in CCR Alive in CCR Alive in CCR Alive in CCR Alive in CCR Alive in CCR AJive in CCR Alive in CCR Alivc in CCR Alive in CCR Alivc in CCR Oied, system ic fungal infcction
E. Mejstříková, strana 111
MRD in Children With ALL Undergoing HSCT
MRD-negative (n=17; 1 event)
_8 Iii >
.~
::>
.6
~
":- .4 C
'>"
w
.2
MRD-positive (n=8; 7 events)
O
Lo -rank p<0 .0001 O
20
10
40
30
50
60
Follow-up [months)
Fig. 1_ EFS in MRD-positive and MRD-nega ťive patiems. Meuian follow in MRD-negative grou p = 25 months .
a
=
1....--=_---,
-f
U, I
1o
0..0 1
a: ~
O,i)() 1
0 000 '
\
t--\--
c .,
0,0000 '
,
b
li
1 o
II j
,
O'
0.0 '
a: ~
OJII)I
Fig. 2_ Disease course nnd MRD follow-up in Patients 8 and 22. a: Disease progress ion after HSCT in BCR/AI:lL-positi ve patient (No. 8) was only temporari ly retarded by ima tini b trcatmem and Dll. Squarcs = 19 rearrangcmem targct No. I; lriangles = 19 rea rrangement target No. 2 (rull symbols = BM samp les; open symbols = PB samples); black arrows = Dll; 1M = imatinib mesyla te; R = relapse ; T. = allogeneic HSCT. b : Pre-transplant MRD status detemlines posttransplam course oft he discasc in T-ALL int'anl paticnl. Squares = TCR delta incomplete rearrangemem target (full symbo ls = BM samples; open symbols = PB samples). Pediatr 8100d
Cw,,'ť(
001 IO.IOO2/pbc
5
MRD Monitoring and Treatment Strategies After HSCT Post-tran splant MRD dynamics was monitored in all patients with detectable targets. In the group of pre-transplant MRD-positive patients , five displ ayed minimally one MRDpositive sample before the emergence of relapse. In three patients , there was a time-frame for an attempt to ave rt the relapse manifestation, but despite our effort all three patients subsequently relapsed (Fig. 2). Period from the first MRD positivity after transplantation to the diagnos is of hematologieal relapse was 0-486 days (0, 30, 36, 330, 365 , and 486 days). Treatment of post-transplant hematological relapse was heterogeneous and it is summarized in Table III.
DISCUSSION Detee tion of MRD leve ls has already beeome an integral part of treatment of childhood ALL patients including those undergoing HSCT. Ongoing front-line treatment trial s, based on previous retrospec tive studies [16, I 7 J, aim to demonstrate the benefit of MRD-based stratifkation in the prospective setting. Retrospective single- and multicenter analyses of MRD in transpla nted pediatrie ALL ehildren showed cIearly the significant impact ol' pre-transplant MRD on outcome [1-3 ,5]. Imashuku et al. questioned these data , showing II surprisingly high proportion of MRD-positive patients at the start of a conditioning regimen (96% ) :rnd strikingly high frequency of the c10nal evolution hampe ring MRD detection itself [6]. Although they did not use an up-to-date methodology (specifi c probe hybridization was employed instead of RQ-PCR) and a full spectrlll11 of [gITCR rearrangements, their results cast doubt upon the practicality of the whole approach. In this study, we concentrated on IglTCR quantitation only, despite the fact that significant proportion of patients bear the fusion genes (BCR/ABL, ETV6/RUNXl , MLL! AF4, MLLlAF9) as potential targets for MRD deteetion as well . As already mentioned, the clona I IglTCR qu antitation methodology has been inereasingly standardized throughout the last years within the ESG-MRD-ALL. This standardization process and also newly developed interpretation criteria improved significantly its clinical value _ When properly applied , this met hod provides a reliable, clinically lIseful tool, as it was proved by numerous intemational quality controls [4] . However, this is nOI the case for the quantitative ana lys is of the fusion genes expression in pediatrie ALL. We have recently shown a very good correlation between ETV6/RUNX I transcript leve ls and Ig/ TCR quantitation but in the cohort that consisted dominantly from the front-line treated patients [18]. In the current study, we observed a minor but significant discrepaney between BCR/ABL and IglTCR MRD leveJs in Patient 36_ He di splayed borderline positivity in nested qualitative PCR for BCR/ ABL prior to the transplant and he was the on ly patient
E. Mejstříková, strana 112
6
Sramkova et al. the Ig/TC R MRD-negative subgroup who developed relapse after HSCT. Our prospective single-center study clearly documents the feasibility and applicability of the IgrrCR quantitation technique of MRD detec tion in the transplant setting. The efficacy of thi s approach, in terms of the identification of at least one target with adequate specificity and sensitivity per patient , increased steadily throughout the study: it was 61 % in the first half of the cohort a nd 80% in the latter. Our res ults are in agreement with the previously published data [13,5.19] a nd in strikin g contrast to the study by lmashuku et aJ. [6]. Both survival a nd multivariate analyses demonstrate the significance of MRD level before HSCT. Patients who enter the transplant conditioning phase of treatment with MRD 4 level higher than 10- are at high-risk of post-transplant relapse. Treatment ofthese re lapses has been very di sappointing so far. Three differe nt approaches may lead to a potential solution of thi s problem: ( I) reduc tion of the mal ig nant clone prior to the transplant using intensified or additional treatment, (2) modification of the HSCT procedure, and (3) post-transplan t treatment modifications ba sed on the close follow -up of MRD levels. Pre-transplant treatment intensification is complicated by the fact that vast majority of patients have been heavily pre-treated. The idea of employment of so me dru gs that are not no rmally used in the frontline treatment failed to show a significant effec t. Potenti ally, introduc tion of new age nts (e.g ., kinase inhibitors, monoclonal antibodies, new antimetabolites, suc h as clofarabine) might be an option [20]. However, no co nvin cing data are available thus fa r. Modification of tra nsplant procedure based on MRD positivity prior to HSCT aims for the redu c tion of GVHD prophylaxis, thus boosting the graft-versus-Ieukemia effect. Removing ATG, targeted dose of CsA and rapid immunosuppression tapering may lead to this effect. Although partly e ncouraging. results of an ongoing Dutch study have thus far been inconclu sive [19]. In some patients in our cohort, we used targeted dose of CsA. In patients younger th an 16 years of age tra nsplanted from matc hed sibling, we did not use MTX in GVHD prophyl ax is and in so me patients tra nsplanted from well-matched unrel ated donor eve n ATG was omitted. Due to the srna II numbers and heterogeneity of patients with ALL, it is difficult to prove any noticeable effect of thi s strategy in terms of relapse prevention. We should oe cautious as thi s strategy may significantly increase the risk of transpl ant-re lated morbidity and mortality without c lear evidence of efficacy. Adoptive immunotherapy after the transplantatio n was genera lly not successfu l in our co hort . ln our hands, interventions, suc h as early and rap id di scon tinllation uf immunosuppression, infusion of DU in 4-6 weeks interval, and/or use of imatinib mesylate in BCR/ABL-pusitive ALL were not suffic ie nt enollgh to prevent onset oť relapse. Application of these approaches in pati ents with imminenr ln
"
-o
'" 0>0
~
-z
o Z
o
Z
o Z
-o
i:<: " ?: ~
u:.2
o
'"
~
Z
r
u
'"
o
00
""+ o
o ..o ..o
+ o
o
$
o
I"M
++
+ o
00
I I
o Z
I I
I I
o
00
..o M
o
00
+ o
+ o
+
00 M
00
c
"
Pediatr 8/DOd C{//lCer DOr IO.IO02/pbc
o ..o
o
o
o
c;,
+
E.
Mejstříková,
strana 113
MRD in Chndren With ALL Undergoing HSCT relapse have led to extended remission but often at a priee of severe, uneontrolled GVHD, and Jife-threatening invasive fungal infeetions, Moreover, no permanent effeet was seen in our group of patients, Heavily pre-treated patients bearing a chemoresistant leukemia have an extreme morbidity and mortality, Therefore, an early initiation of adoptive therapy might at least postpone, ifnot prevent, relapse, and faeilitate ťurther effieaeious ehemotherapy; if subsequent remission is reaehed, these patients should be indieated for re-transplantation, A second HSCT may be curative in such settings, Our results demonstrate that the possibilities of treatment of the post-transplant relapse are extremely limited, Introduction of new treatment modalities is desirable for the patients with molecular genetic relapse after HSCT. These should include not only those mentioned above (kinase inhibitors, monoclonal antibodies, new antimetabolítes) but also the inhibitors of enzymes controlling epigenetic modifications, specimcally DNA methyltransferases and histone deacetylases [21.), We demonstrate feasibility of MRD quantitation using clona I IglTCR reaITangements as an approach for the pretransplant risk assessment in pediatrie ALL patients undergoing allogeneic HSCT. We show that, although, we are able to identify the patients in an almost certain risk of relapse, our ability to respond and to avert an impending relapse is very limited, ln spíte of the use uf currently available set of treatment approaches after HSCT, we failed to permanently avert a predicted relapse, The change of the approach to MRD-positive patients prior to HSCT is necessary because of very questionable benefit of HSCT in these children, We are confident that all efforts should be ai med to better control pretransplant MRD levels,
Note added in proof Patient No, 35 suffered from hematologieaI relapse 14 months after HSCT and died due to disease progression,
ACKNOWLEDGMENT AH pediatlie hematology eenters in the Czech Republie are acknowledged for their cooperation and reporting the patients to the transplantation center: Brno, Olomouc, Ostrava, Hradec Kralove, Ceske Budejoviee, Plzen, Usti nad Labem, We would like to thank Jozef Madzo, Ph,D, for the help with the data analysis,
REFERENCES I, Knechtli CJ, Gouldcn NJ. Hancock JP. et al. Minimal residua I disease status before allogencic bone marrow transplantation is an important delcnminant of succcssful OUlCOme forchildren and adolescents wilh acute Iymphoblastic leukemia, Blood 1998;92: 4072-4079, 2, Bader P, Hancock J, Kreyenberg H, el al. Minimal res idual diseasc (MRD) Slalu s rnor lO allogene ic stem cell transplantation is a powerful predictor for posl-transplant outcome in children Wilh ALL. Leukemia 2002;161668-1672. Pr-dil/! r Blood COllcer DOl 10, 1002lpbc
7
3, van der Velden VH, Jooslen SA, Willemse MJ, et al. Real -l.ime quantitative PCR for deteclion of minimal residual disease hcforc allogencic stem cell transplantation predicIs oulcoml! in children with acute Iymphoblastic leukemi.!. Lcukemia 2001: I 5: 14851487, 4, van der Velden VH, Hochhaus A, Cazzaniga G, ct al. Detcclion oť minima! res idual disease in hematologie malignancics by rcal-time quantitalive PCR: Plinciples. approaches, and laboratory aspects. Leukem.ia 2003;17:1013-1034. 5, Krejei O, Van der Velden V, Bader P, el al. Level oť minimal residual disease pnor to haemalopoietic transplantation predicls rrognosis in paedialrie patients with acule Iymphoblastic leukaemia: A reporl of the pre-BMT-MRD sludy group. Letter. Bone Marrow Transplant 2003;32:849-851, 6, Imashuku S, Terui K, Matsuyama T, el al. Lack nf clinical utilily ol' minimal residual disease delection in allogeneic Slem cell recipient.~ Wilh ehildhood acute Iymphoblastic leukemia: Muhi-instilulional collaborative study in Japan, Bone Marrow Transrlant 2003;31: 1127-1135, 7, Szczepanski T, Willemse MJ, Bnnkhof B, ct 31. Comparalive analysis of 19 and TCR gene rearrangemenls at diagnosis and at relapse of childhood preeursor-B-ALL pruvides improved stralegies for seIection of stable PCR largets for monitoring of minimal residual disease, Blood 2002;99:23 I 5-2323, 8, Pongers- Willemse MJ, Seriu T, Stolz F, et al. Primers and protocols for standardized detection of minimal residual disease in aeute lymphoblastic leukem.ia using immunoglobulin and T-ccll receptor gene rearrangements and TAL I dclctions as PCR targels: Report of the BIOMED-I concerted aelion: Invcstigation ol' minimal residual disease in acute leukemia, Leukemia 1999: 13: 110-118, 9, Szczepanski T, Pongers Willemse MJ, Langerak AW, Cl al. 19 heavy chain gene rearrangements in T-celJ acute lymrhobJastic lcukemia exhibit predominant DH6- I 9 and DH7-27 gene usage, can result in complete V-D-J rearrangements, and are rare in T-ccll receplor alpha beta 1ineage , B lood 1999;93 :4079-4085. 10, van Dongen JJ, Langerak AW, Bruggcmann M, ct al. Design and standardization of PCR primers and protoeols ťor deteclion ofc1onal immunoglobulin and T-cell receplOr gene recombinations in suspect Iymphoproliferations: Report of lhe BIOMED-2 cOl1certed action BMH4-et98-3936, Leukemia 2003; 17:2257-23 I 7, II. van der Velden VB , Wijkhuijs JM, Jaeobs DC. el ul. T-cell recertor gamma gene rearrangements as target~ for delcelion 01' minimal residual disease in acute Iymphoblastic leukemia by real-timc quanlil3live PCR analysis, Leukemia 2002; I6: 1372-1380, 12, van der Velden VH , Willemse MJ, van der Schoot CE, ct aJ. Immunoglobulin kappa deleling element rcan'wlgements in precursor-B aeute Iymphoblastie leukcmia are stahlc targets for delection of minimal residual disease by real-time quantilalive PCR, Leukemia 2002; 16:928-936, 13, Langerak AW, Wolvers-Tettero lL, van Gastel-Mol EJ , ct aJ. Basic helix-loop-heJix proteins E2A and HEB inducc immaturc T-cell receplor rearrangements in nonlymphoid eeJls, Blood 200 1;98: 2456-2465, 14. Verhagen OJ, Willemse MJ, Breunis WB, ct ul. Application of germline IGH probes in real-time quantitative PCR for the delcclion of minima! residual disease in acute lymphoblastic leukcmia, Leukemia 2000; 14: 1426-1435, 15, Pongers Willemse MJ , Verhagen OJ , Tibbe GJ, ct aJ. Real-timc quantitative PCR for lhe deteclion of minimal residua I disease in acute Iymphobla,tic leukemia using junclional region specinc TAQMAN probe s, Leukemia 1998;12:2006-2014, 16. van Dongen JJ, Seriu T, Panzer-Grumayer ER, ct aJ. Prognostie value of minimaJ residual disease in acute Iymphoblaslic leukacmia in childhood, Lancet 1998;352: 1731-1738,
E. Mejstříková, strana 114
8
Sramkova et al.
17. Couslan·Smilh E, Bchm FG, Sanchez J , et al. Immunol ogical detection of minimal residual disease in c hildren WiLh acuLe lymphoblasLic leukaemia. Lancel 1998;35 1:550-554. 18. Fronkova E, Madzo J, Zuna J , el aJ. TELlAML 1 real-Lime quantitative reverse lranscriplase PCR can compl ement minimal residual disease assessment in childhood ALL. Leukemía 2005; 19: 1296-1 297.
Pediarr Blood Cancer DOl IO. I002/pbc
19. Sch ilham MW, Balduzzi A. Bader P, et al. Is there a role for minimal residua l disease levels in the Lreatment of ALL patie nts who rcccive allogeneic stem cell s? Bone Marrow Transplant 2005;35:S49-S52. 20. Pui CH , Jeha S, Kirkpatrick p, et al. Clofarabine. Na! Rev Drug
Discov 2005;4:369 -370. 2 1. Eggcr G, Liang G, Aparicio A, el al. Epigenetics in human disease and prospects for epigenetic therapy. Nature 2004;429:457-463.
E. Mejstříková, strana 115
Příloha
6
B-cell reconstitution after allogeneic stem cell h"ansplantation impairs minimal residual disease (MRD) monitoring in children with ALL
Frankova va, Muzikova Katerina, Mejstrikova Ester, Kovac Martin, Formankova Renata, Sedlacek Petr ,Hrusak Ondrej, Stary Jan, Trka Jan Bone Marrow Transplantation (fF 3, O)
Bone Marrow Transplantation (2008), 1-1 0 ,, 2008 Nalure Publlshing Group A" rights reserved 0268-3369/08 $30.00
www.nature.com/bml
ORIGINAL ARTICLE
8-cell reconstitution after allogeneic SeT impairs minimal residual disease monitoring in children with ALL EFronkova l , K Muzikova l , E Mejstrikova, M Kovac, R Formankova, P Sedlacek, "nd J Trka
o Hrusak, J Stary
Drpar/men/ oj Paediatrie Haema/%gyIOn c%gy, 2nd Medica/ School, Charles Universi/y {Ind Ulliversi/y HO.l'piwl M%l. Prague, C:ťch Republie
\linimal residual disease (MRD) detection using qua ntiation of clone-specific Igor T CR rearrangements before and after transplantation in chjld ren with high-risk \LL is an important predietor of outcome. The method lod guidelines for its interpretation are very precise to amid both false-negative aod -positive resu lts. 111 a grou p or 21 patients foDowing transplantatioll, we observed detectable MRD positivities in IgjTCR-based real-time quantitative PCR (RQ-PCR) leading to no further progressioo of the disease (ll of 100 (11 %) total samples). We hypothesized that these positivities were mostly the result ol' nonspecific amplifica tion despite the Ipplication of striet internatio na/ly agreed-upoll meaes. We applied two non-self-specific 19 heavy chain ays and received a simiJar number of positivities (20 lod 15%). Nonspccific products amplified in these RQpeR systems differed fro m specific products in length and lequence. Statistkal analysis proved that there was an t ccllent correlation of this phenomenon with B-cell regelleration in BM as measured by ftow cytometry and 19 light chain-K excision circle quantification. We nc\ude that although IgjTCR quantification is a reliable method for post transplant MRD detection, isolated positivities in Ig-based RQ-PCR systems at the time of iotense B-cell regeneration must be viewed with caution to lVoid the wrong indication of treatment. BOlle Martou' Tramplanlalion advance online publication, 19 May 2008; doi: 1O.1038jbmt.2008. 122 Ke,'words: ALL; SCT; min imal residu al disease; KREC; R.~el1 reconstit ution ; donor Iymphocyte infusion
COlTcspondence: Dr E Fra nkova, Departmenl of Paediatric lI aematologyfOncology, 2nd Medic"1 School. Charles Unive rsity, V C valu 84, 15006 Praguc 5, Czcch Rcpublic. E-mail : eva.fronkova (a lf.mo lo Lc·uni.cz These aUlh o rs contribulcd equally to Ihi, work . Recci\'ed 21 December 2007; revised 13 M arch 200~; accc plc-d 20 March :lXlR
lntroduction ouring lhe pasl decade, minimal residual disease (M Ro) moniloring using palienl-specific anligen receplor gene rearrangements has been incorporaled inlo major fronlline and relapse lrealment prolocols for ehildhood ALL. ' " Severa I grou ps, inc1uding ours, reporled lhe unfavorable prognostic signifieance ol' high MRo levels befo rc transplant in children with high-risk ALL. ~" Sludies exploring the significance of post tl'ansplanl MRo were based on lhe detection of mixed chimerisl11,"· II' f10w cylomelry (FC)," th e fusion gene PCR in ca se of Ph + ALL,' 2 peR lIsing c1one-specifie Tg or TC R V- (D)-J seq uences' :' ", and, most recently , real-time quantilalive peR (RQ-PCR) deleclion of c10nal 19/TCR rearra ngemenls.' 7 Consislcnlly, all studies showed that delectable M Ro at any lime after seT represenls a substantial risk of post lranspl ant relapse, both in children and in adults. Several meas ures exisl 10 avert hematologiea l relapse after SCT when molecular relapse is delected. The f'irst oplion is thr reinforcement ol' lhe GVL effeel by immunosuppression reduclion or donor lymphocyte infusion (DLl); lhe second is the applicalion ol' further cytoreductive therapy, including MoAh Ol' specific tyrosine kinase inhibitol's, in the case BCR/ABL-posilive ALL. . 8-26 The benefit from such lherapy has becn described in a small proportion of ALL pa lients . To date, lhcre is no genel'al consenl regal'ding the choice and liming of lherapy in case of post lransplant MRo pOSili vi ty . The ESG-MRo-ALL (European Study Group on Minima! Resid ual Disease in ALL) sel guidelines for thc inlerpreta li on of quantitative IgjTCR-based MRD lhal inc1ude the use of polyclonal DNA frol11 the peripheral blood (PB) of healthy donors in mllltiplicate as a ncgative control and a strict definition of MRD positivily.27 This definition is even stricter when aimed at lhcrapy intensification (fo r examp1e, DLl) 10 prevent false-positive reslIlls. So far, this method ha s provcd to be rcliable, leaving the potential nonspecific amp lification in regenerating nonrnalignant Iyrnphocytes only a theoretica l possibilily. Start in g 140 days and later after SCT, we observed M RD positivities in aceo rd ance with al1 ESG-MRD-ALL criteria in patients who subsequently turned MRD-negative without any anti leukemic treatment and had been in complete rernissio n for several years after SeT. We hypothesizcd that
ar
•
D,.wb.cks of post t,.nsplant MRD in childhood ALL E Frankova el al
the B-cell regeneration after SCT caused an unspecific binding of clone-specific primers, which bypassed the otherwise strict ESG- MRD- ALL criteria for MRD positivity. MRD detection using non-self patient-specific assays revea lcd false MRO positivity in a substan tial portion of post-SCT BM specimens. This false positivity was limited to thc samples con taining high numbers of B-cell progenitors as measured by immunophenotyping a nd by recen tl y described B-cell recombination (r;: deleting) excision circles (KR EC) detection. Moreover, using high-resolution capillary electrophoresis and seq uencing, we observed that the size of such nonspecific RQ-PCR products differed from the specific products. We therefore co nclude that MRD results post-SCT sho uld be approached with extreme caution, and offer recommendations for avoiding MRO misinterpretation.
Patients and methods Patiel1lS A total of 38 children w ith ALL (aged 1-18 years) underwent allogeneic SCT in the Czech Republic from January 2003 to October 2006. Of them 2 1 patients with B-precursor ALL were selected for the stud y based on following criteria: leukemi a-free survi va l with a follow-up of at least 12 months after tra nsplant (median follow-up 46 months, range 17- 62 months) and the availability of M RD results and residua I O A samples from at leas t three (3- 7) of the following time points after S T: days + 30, 1 60, 100, -I 140, + 180, I year, 2 years and 3 years. In total, 100 BM DN A sam ples (Ieftover material a fter MRD detection) were investigated. Washed leftove r material from tubings and the transfusio n bag following the completc administration of the BM graft was used as a control for KRE dctcction. lnforrned consent for the use of rcsidual mate rial afte r protocol-based examinat ion for research purposes was obtained from patients or their guardians. Transplanls The patients were transplanted due to relapsed or high-risk ALL in their first (7), second (ll) Ol' third (3) rem ission. Oono rs of hematopoietic stem cells were HLA-identical siblings in 5 cases, a si bling with 9/ 10 HLA antige n match in I case, and unrelated donors from BMT rcgistries in 15 cases. BM was transplanted in ten cases, PBSC in nine cases, umbilical cord blood (UCB) in one case and both BM and UCB (sibling) in one case. ln the majority of transplants (n = 19), we used a similar pre-transplant conditioning regimen based on TBl at a dose of 12Gy and etoposide 60 mg/ kg, and in two c hildren, a BU-based co nditionin g (I x B ,CY a nd melph alan, I x BU, C Y and etoposide) was used. l n the majority of unrelated donor transpl a nts (n = 13), we used the rabbit an tithymocytic globulin (Fresenius, Bad Homburg, Germ any) at a dose of 10 mg/ kg for 4 days. GVH D prophylaxis consisted of CsA in HLA-identica l sibling transplants a nd a combination of CsA and MTX a dministered on days + I, + 3 an d + 6 in unrel ated donor trans pla nts, respecti ve ly. ln one patient ( C B transplantl, a com binati on of CsA a nd methylpredMarrow Transplantation
nisolone was used. No patient reccived any a ntileukemic trea tment (including adop tive immuno therapy) during the post transplant period.
Deteelion o[ residual disease M ononuclear cells from the diagnostic or relapse BM sam pIes were isolated by Ficoll-Paquc (Pharmacia , Uppsala, Swede n) density centrifugation and stored in Iiquid nitrogen. Follow-up BM samplcs were processed by erythrocyte Iysis a nd stored at -80 ce. Genomic DNA was isolated using a QlAamp DNA Blood Mini Kit (Qiagen GmbH, Hilden , Germany). Primers and protoco ls for the d etection of 19 heavy chain (lGH) rearra ngem en ts, 19 li ght chain-r;: deletions (KDE), TCR-y (TCRG) , TCR- ů (TCRO) gene rearrangements a nd TAL! deletions havc been described previously2U 9 C10nality of PCR products was confirmcd by the heteroduplex analysis.2 ') Sequencing was performed in the ABI Prisl11 310 Genetic Analyzer with Big Oye Primer v3.0 Sequencing Kit (Applied Biosys tems, Foster City, CA, USA) . Variable (V), diversity ( O) and joining (1) regi o ns of the iml11unoreceptor gene were identified by comparison wi th sequcnces in Gen Bank using thc ImMunoGeneTics databasc (http: //imgt.cines.fr, European Bioinformatics Institute, Mo ntepcllier, France) and the IGBlas t sea rch (http: , /www.ncbi.nlm.nih.govjigblast / , N atio nal Center for Biotcchnology lnrormation , Bethesda, MO, US A). P atient-specific forward primers for RQ-PCR were designed using the Vector NTI 8 Suitc Software (InforMax, Bethesda, MO, USA). Family-specific reverse primers and probes for IG H, IGK, TC RO and TCRG have been described previously3U--" Ig/TCR RQ-PCR was performed in the iCycler IQ Rea l-Time PCR De tection System (Bio-Rad, Hercules, CA, USA). Standard curves were prepared by diluting the diagnostic sam ples in pooled polyclonal DNA from the PB of five healthy donors, which was also used as negative control s. MRD using patientspecific as well as non-se lf-specific sys tems was measured in triplicate, with 2.5 ~l of DNA per rea ctioll. The a lbumin gene was used to normalize the DNA concentration and quality J4 The ESG-M RO-ALL criteria for RQ-PC R se nsitivity, quantitative range a nd MRD interpretation were used. 2 7 DNA analysls by on-chip eleclrophoresis RQ-PCR product (I ~I) was a nalyzed in Agilc nt 2 100 Bioanalyzer (Agilent Technologies, Santa Cla ra , CA, USA) lIsing a DNA Series II Kit ( DNA 1000 Assay) and Agilent 2100 Expert software acco rding to thc manuťacturer 's instructions . FlolI' cylOmelry K3 EDTA BM aspirate specimens were stained within 12h from sample collection lIsing a whole-blood Iysi s technique with ammonium chloride . Abollt 50- 100 pl of sa mple (according to cellularity) was incllbated with MoAb at dark for 15 min, and then sampIe was incu bated with Iys ing solution for 15min a t dark and centrifuged (400.1:). Aftcr supernatant remova l, the sample was resuspcnded in 200 ~I of PBS and ana lyzed on flow cytometer (FACSCa libur; RO Biosciences, San Jose, CA, USA ar CyAn ; Oako, Glostrup,
Drawbacks of pcst transplant MRD in childhood ALL E FlOnkova el a/
Denmark). For B- and T-cell regeneralionjFC MRO after SeT, the following MoAb combinations were used: C 020; CDIOjCOl9/CD3 4, SYTOl6jCOl9jC045 and CD3 / CD16-56jC045j 0 19. The following anlibody clones wcre used: C020 FITC (cloneL27; BO Biosciences), CO 10 PE klone SS2/36; Dako), COl9 PC7 jPC5 (clone J4.119; Immunotech, Marseilles, France), C034 APC (clonc L1Q; Immunotech), CDI 9 PE (clone S125CI ; [mmllnotech), CD45 PerC P (clone 20 I; BO Biosciences), C03 FITC (clone Sk7; BD Biosciences), C O 16 PE (c1onc B73.I; BO Biosciences), C D 56 PE (clone B73.1; BO Biosciences) anll CDI9 AP (clone SJ25CI; BD Biosciences). SYTO-16 tgreen tluorescent nucleic acid stain) wa~ purcha sed from Invitrogen-Molecular Prohes (Carlsbad , CA , USA). SYTO-16 green fluorescenl nucleic acid stain exhibits bright green fluorescence upon binding to DNA and RNA. SYTO-16 was used for reporting the percen tage of O 19Po, ;ells out of all nuclea ted cells (SYTO 16'''''). The following lubpopul a tions were reported: C D 191'°', CD 19'''''C 045 dim , COlO " 019 P "" CD IOP<>sCD19 P ''', C D34posC Dl9po, and CD3'x". T he B cells with a lowe r expression of C 045 (dim) correspond to immature cells,15
Detrc/ion
0/ S-cell receptor excision circles
We used the method of KR . C delection,36 slightly modified as follows . We did not employ the sÍmllltaneou s etection of Íntron-Kde rearrangements as sllggesled. Instead, we used a ~c. me thod with a calibrator 'ample- ·the donor D NA extracted from the residuum ,fter BM transplantation. We considered the differences in DNA concentration of lhe samples and used RQ-PCR for [he albumin gene as dcscribed previously.-'4 The final 1IT10unt of KR EC relative to the calibrator was expressed :1S:
2C,calibralor- C.s:.Jl11p lc+ klg2(DNAconc.cJlibra Lor fDNA co nc.$al11p le ) .
SWlistical analyses rhe distribution of freqllen cies between groups wilh posilive md negalive MRD was assessed using Fisher's exact tesl. fhe Mann-Whitney tesl was llsed to estimale the significance .,1' differences concerning continuous M RD va lues. The ,tatistical a nalyscs were pcrformed using StalView version l.O(SlatView Soflware, C a ry, NC, USA). Trend anal ys is in fí gurc 4 was calculated using GraphPad Prism Software i~rsion 5 (GraphPad Software, San Diego , CA, USA). First, l!1eans of individual variables in individual lime points were mrnpared using one-wa y analysis of variancc (ANOVA), Ilten posl-hoc linear lrend analysis was used to lest decrease r increase in nalurally ordered groups. Simullaneollsly, Kruskal-Wallis lesl was used for comparison of means .unong individua l lime points.
Results
Freifuency oj' h)'[!Olhelica/l)' Ilotls[!ecific amplificalion using paticl/ts ' c!one-specijic syslems in lhe course oj' prospective oosl-SCT monitoring \ patient-specific RQ-PCR system with minimal sensitivity
JI' /0- 4 was designed for all patienls, and 17 out of 21 patients were monitored using lwo independent 19/TCR
largets. Figure I shows lhe results of post transplanl MR D monito ri ng. l n total , 16 of 15l (11%) samples from ni ne patients we re M R D positive according to ESG-MRO-ALL criteria in a t least one la rget (the C value of at lcasl one of lhe rhrce rcplica tes was within 4.0 from the highest CI va lllc oť the sensitivily and lhe CI value of at least o nc oť thc three replica tes was ~ 1.0 lower lhan the lowcsl CI of the background) 27 Eighl samples ťulfilled ESG-MRO-ALL criteria for positivity that (lim at therapy intensi!ication (the C. valuc of at least one of the three rcplicates was ~ 3.0 lower than the lowest C. of the background). AII samples were cvalualcd as ' positive, nOl quantifiable'. Fifteen oť sixteen positive samples were examineJ by tw{) largets; one sample was positive in bOlh of them. The RQ-PCR targets with positivc samples llsed IGH (7 x ), KDE (5 x ), TCRD (4 x) and TCRG (1 x ) rearrangemenls for MRD dcteclion. Nine of sixteen positivc samples turned negative in thc following BM examinarions; threc palients were stili positive al the end of follow-llp. Only I ol' 16 posilive samples was taken during corlicosleroid lreatmenl of G H D , compared lo 51 of 135 negalive samples (P = 0.0 I). On the basis oť ťollowing, we hypothesized lhal mosl ar the positive result s wcre false positives and were most likely caused by unspecific hind ing of patienl-specific primers lo similar V- (O)-J sequcnces of nonmalignant Iymphocytes in BM. First, all patients with at least one positive sample remained free of leukemia with a median follow-up o ť 31 months (range 17-61 ; nol significantly different fmm patienls with no posilive samplc, P = 0.17 , Mann-Whitney). Also, a" patienls who entered lhe sludy were MRD nega tive (n = 16) or low positive ( < 5 X 10- 4 , 1/ = 5) hefore the transplant. Finally, simultaneous RQ-RT-PCR detection of the respective fu sion gene (BCR / ABL and TELj AMLl) in 10 relevant samples les led gaye a negative MRD result.
•
Frequency o( amplificalion us ing non-.\'(:I( clone-.I'pecific syslems Because of lhe suspicion of false positivity, we lIsed different RQ-PCR system s specifk ťor unique V- (D)-J sequences of two olher patients (nol specific for thc leukemie cells of the given patients) to test whether rhe V- (D)-J sequences would evcn be amplified in such circumstances. A lotal of 100 post-SCT samplcs with sufficienl leftover DNA were tesled. The frequcncy of MRO-posilive sa mples in the selected cohort dill not diffcr from the whole cohort (ll % ). We used lwo patie nt-specific systems bascd on lhe VH3J H4 rea rrangement. Both the VH3 family and JH4 represent the most freq uently rearranged segments in physiological B-cell development J7 .3" To maintain the reproducibility of the d a ta , we used a ssays with no background amplification of polyclonal DNA ťrom PB ar healthy donors (' bu ffy coat' cells). The assays \vere originally designed for two patients wilh ALL (not included in the inves tigated cohort) using patienl-specific primers situated with their 5' end in the VH3 segment and lhe last seven nucleotides spanning N segments . In t()l a l, 20 out oť 100 (20 % ) and 15 out of 100 (15 % ) samples werc posilive according to the ESG-M RD-ALL criteria using those Bone Marrow Transplantation
D,awbacks ol post t,ansplant MRD in childhood ALL E Fronkova ol af UPNe
d+30
d+60
d+90
d+140
d+180
+1yr
+2yrs
+3yrs
+4yrs
-1/
#82
VH5Nkl
Ij Ij
#321
Ij
#374
#415
Ij
#432
-1/
#450
Ij
Vklll
VH3Nkll BCR/ABL VH3Nkll TEL/AML 1 VH1
BCR/ABL
VH1Nklll TEL/AML1 Vd2NgI
Ij
#497
TEL/AML1
Vgl
VH1NgI
Ij Ij
#375
#429
FG:
VH2ND2 BCR/ABL
#287
VH3NgII
Ij
#639
Vd2NgI VH3Nd2 BCR/ABL
#658
-jj Ij
#710 #718 #719
+5yrs IglTCR assay:
Ij
#4
• •
O
#861
0--0
•
7Ij
VgI/Db2 Vklll/Dd2
Ij Ij
#919
VH3NH4 BCR/ABL VH3NH4 TEL/AML 1
Ij Ij
#906 #914
VH3NgII BCR/ABL
VH4Nd2 DH1 DH3
#923
Figure I POSL Lrrtnsplanl minil11~1 r~sidual di sease (M RD) resulls (lgITCR) [rom 21 palient s included in the sludy AII posilive sump les wcrt cvaluated aS 'rositi"e, nOl quanlifiahle·. O , MRD negati w ; . , MRD positive: UPN, unique ralienl number; FG. fusion gene.
Positive samples: Assay1: 1/18 Assay 2: 2/18
1/21 2/21
3/20 2/20
4 /11 1/11
•
O
40
4/15 1/15
... ... .......... ................ ....
45
ú
6/15 5/15
. . . . . . . . .f.....io . . .~.J {)
.~. . ~
••
·0
35~--~----~----.----.----.----.----_1
d30
d60
d90
d140
d180 1-3 yrs post-SCT
Figure 2 C. valuc~ o r ralsc minimal residual di sť"'c (MRD) positivc sumrle, usin g two no n·sclr-srecilic 19 CR assa )'s. hlled and emrl)' diamo nds rcpresent thc two d irfcrenl ."ays . In lhé Ca se thaL more members of thc cxamincd trip licatc wcrc positivc. thc hi gh('Sl C, value was uscd
assays. Of lhe eleven samples positive in their own patientspecific systems, five and four werc also positive in the two non-self clone-specific assays, respectively . Figure 2 shows C, values of false-positive samples in both assays. As the onset of positive signal was diffcrcnt belween the two assays, their Cl values are not mutually compaľ able. Given the fact that no background amplification was present, aU samples would ful fill the criteria for therapy intensification. 27 ne Marrow Transplantation
Nonspeciflc RQ-PCR producls di/Jer /rom Ih l:' spccific producrs in lenglh and sequence As we observed no difference in the length of most nonspecific RQ-PCR products on 8% polyacrylam.ide gel (data not shown), we employed a morc ~ensitive Agilent DNA analysis by on-chip electrophoľesis. Figurc 3 silow~ an electľophoretogram of the díagnostic sa mple (I: 10 000 dilution in buffy coat) of the patient whose assay was uscd for amplification compared to thľee diffcrcnt false-positive samples from different p ati ents thal differ in size and / or pattern. Sequencing of 10 false-positive RQ-PCR prodllcts revealed the use of different N nuc1eotides down stream from the matching sequence of the patient-specific primer.
Assessmenl of B-cell reconslilulion in lhl! BM liS ing KREC delecrion [n the course of V- (0)-1 recombination, [g (TCR) gene segments are assembled, leaving nonreplicable circular DNA fragments as 'by-products '. T-cell recombinalion circles deteclion has been commonly lIsed as a quantilative marker of thymic output J9 Detcction of B-cell recombination circles has not been widely employcd dlle to the complicated structure of [g genes. Recently, detection excision circles originating in the deletions of [g light chain]( (KREC) was described J (, As IGK delction occurs in all B
or
•
Drawbacks of post transplant MRD in childhood ALL
E FrookGva et a/
120
a
100 80
"
60 40
1
20 O
100
150
15 50
...
300
400
500
700
1SpO
b
80 60 40 .~ c
::J Ol
u
20 O
,.
c Ol
15 50
u
(f)
ll:' o
~ Ol
>
'';::;
cu
100
...
300
i
400
500
700
1500
j
c
!
80
i,
60
I
Ol
cr:
10Ó 150
40
i i
20
~
1
O
, !
100 80
15
...
10Q i
200
300
400
500
700
1500
!
I,
60
d
I
40
i
I
tj ,I
20 O 15
100
200
300 400 Size (bp)
500
700
1500
Figurc 3 Agilenl DNA chip electrophorelog ram of s poeific and unspeci nc rea l-time q uan tilali ve PCR (RQ-PCR ) prod uels. Roth 15 and 1500hp praks represe nl lower "nd upper D NA markers, "nd ... "xi, va lues are cxprcsscd as relalive Auoresccnce units. (a) The product ar patient-spceific RQ-PCR. Thc diag noslie ALL samp lc lIscd for arnr lilicalion was 1' 10 000 diluled in polyclonal DNA rrom heaithy donors. (b-d) N onspecil1c RQ-PC R producl' ampli ljed fmm lhree differcnl palienls' fottow-up ,amples Ilsing lhe non-setf-spccific assay of lhe pa lienl from (a).
Iymphocytes lhat fail to rearrange IGK produ ctively on one or bOlh a lleles, the number of K R ECs in the BM refleets the number of Jcveloping B Iymphocy tes. Figure 4a shows the number of KRECs relative to the DNA from the BM donor sample during t:n e post transplant period in 100 BM samples . With few exceptions, Ihe KREC le ve ls rose continuously, remaining lower than those of the con trol until day 90 and having a median value higher than the control 1-3 years post-SCT. This result was in coneordance wlth the lower intensity of immunosuppression in later time points. Figure 5a shows the impact of immunosuppression taken at the time of sample collection
on the nllmber of KRECs for all time points together. Supplementary Figure I shows the same a nalysis for individual time points. KREC number correlated well with the ťalse positivities observed in non-self-specific 19/TCR assays. Figure 6a shows the difference in the number of K R ECs between the samples with negative and false-po sitive MRD. False MRD positivity was more frequent in samples with higher KR EC number than in the negative ones (P
Orawbacks ol post transplant MRO in childhood ALL E Fronkova e/ a/
Assessmenl uf'lvmphucyle recullslilulion in Ihe BM' samples by immunophenotl'jJing Evaluation of B-cell subsets and T cells was avai lable in 77 out of 100
a
•
8 ~
Ci
7
•
6
c
o 5
TI
•
e eg e
8 4 ID
> 3 m ~
2
(J
w cr: ~
O
b 25 20
~ c
15
uL()
"
10
o(J
5
o ~
"cn
O
+ll~~$ n=18
• •
n=16
35
o ~
15
cn 10
o
(J
5
O
n=15
n=17
n=5
n=12
n=11
g eé Ó 6óO. n=16
n=16
•
n=17
n=5
n=12
d 35 25 ~ o
o ~
"oM
•
20 15 10
(J
5
O
+$~~~
.... ......... -.........
n=14
n=15
n=16
n=5
n=11
n=8
d30
d60
d90
d140
d180
1-3yrs post-SeT
Marrow Transplantation
-
-
MRD monitorin g using antigen receptor gene rearrangement s offers a highly sensitive tool for post transplant management ol' high-risk ALL. On the basis of MRD positivity, Illethods ať preemptive imlllunothera py (DLI , immunosllppression withdrawal) ar chemotherapy ca n be employed earlicr than when using thc lcss sensitive mixcd chime rism Ol' Ilow cytollletric detection. However, our study demo nstrated that even whe n all the criteria for MRD interprelation dcveloped during the past decade are met, there is a considerable risk of false-positivc MRD result s a fter SCT . E ven though a ll such rcsults would be classified only as ' positive, not qllantiflable', thcy wOllld mean at best the necessity ať rcpeated examination together with putting more stress on the patient~ and their parcnts, and at worst the application of preemptive treatment. ln our cohort, 9 of 21 patients were MRD-positive a t 1east o nce during the flrst 3 years post-SCT lIsing clonespecific 19jTCR RQ-PCR assays. No patient has relapsed so far, with a medi an follow-up oť 31 months. This number does not rellect the exact number of positive reslIlts in all tra nsplanted patients, as our cohort was selected based on
n=11
•
30
Discussion
n=15
•
• •
25
*' "
•
n=11
•
30
20
n=20
o éE1Q~ n=16
C
n=21
incrca sc in T-cell numbers during the post transplant period (P = 0.02, one-way ANOVA; Figures 4c an d d). The diffcrence in B-cell progenitor (CD 19 P <"CD45""") numbers was not significant (Figure 4b). The impac t of immunosupp ression on the number of B-cell progenitors, Band T cells is shown in Figures 5b, c and d (for all time points togcthcr) and in Supplementary Figures lb, c a nd d (for individual time points). Figure 6b and c shows that samples with fals e-positive MRO had higher numbers of immature B cells (CD 19 1"-"C045 dim ) and total B cells than negative ones (P < 0.000 I, M an n-Whitney). Also, the numbers of p COIO + ' COI9 po" COlO o'COI9 po' and C034 poSCOI 9 PoS cells were significantly higher in samples with false-positive MR O (P < 0.000 I, P = 0.0002 and 0.000 I, respec tively , data not shown). The number of T cells (C03 pO') was not differcnt between the two grollps (F igure 6d). Supplementary Figures 2b, c and d show the same a na lysis for individual time points.
Figure 4 LymphocYlc rcconslilUtion during Ihc posl-SCT peri od wa s d escribed by Ihe following : (a) K-delelion exc ision circle, (KRE ) numhers expressed relalive (O the donor BM snmpl. (KREC numh"r I).One-wa y ana lysis of varia ncc (A NOVA ), nol significant dilTcrcnce: Kruskal Wallis, P=O .OOl I (significa ntly differenl level bctwcen lime poinls I and 5, I nnd 6,2 and 5). (b) Pcrce ntage ol' B-cell precursors (CU 19''''CD45'''''') out ol' Ihe lOlal BM cell number during lhe posl-ser period . One-way ANOV A, no l significanl diffe rence: Krll skal Wallis, nol significanl diťkrcn cc. (c) Percentage of B cells (CD I 91"" ) aul of Ihe 10 Ial flM cell nurnbcr Juring lhc pOSI-SCT period. On e-way ANOV A, nol significan l di /Terentc: Kr uska l Wallis, nOI significa nl difTerence (P = 0.06) . (d) Percťntagc oť T cells (CD3"''') OUI of Ihe 10lal BM ce ll number during Ihe posl-SCT peri od . One-way ANOV A, signiftcanl diffcrcncc (P = 0.(020) : linear Ircnd, significa.nt increase among time points (slope II. 8, I' d l.OOOI): KruskalWallis, P =0.0015 (si gn iflca nlly differen l level bClween lim" po ints I and 5. I and 6).
Ordwbdcks of post trdnsplant MRO in childhood ALL
E Fronkova el al
a
•
8 7
~
tll
Ci c o 5
r
4 3
~ 2 u w
II:
l<:
b
25
~
E 15 oč 1!1
V
10
"Cl o
5
u
o
I
I
...
n=18
"'\ P
P< 0 .0001
r
•
I
I
1 ~
n=29
c
n=44
P=0 .075
Ťr
'j.
o
=-=
•
.-
.~
•
..
n=37
20
;;
P<0.0001
'I
o .. ....
"' -.
r
P < O.OOO1
"O
B Ol .;;>
P=O.11
r
6
"I.
S
n=34
....... .. -.
n=13
35
P=0 .84
r
30 25
~
20
o" o-
15
Cll
Ci
u 10 5
o
••
~ n=29
d
35 30
,r
P
M
o
u 10 5
o
•
•+ Ó
n=34
n=13
P=0.12
•
r
.r
"',-c-
P=O .054
P=0.57
r
" 'I
•
25
~ 20 "o-o 15
P <0.OO01
'I
d:J~ t • n=24
n=32
n=12
CsA
CsA, MP
nol5
I
Figure 5 The irn pacI of immullosuppr~, ive Ireulrne l11 la ken O'C D45 d ;"" bl, B cells (COI9""" cl and '1' cells (CD 3"''', d) o ul o f Ihe 101a l BM cell numncr . T he Mann Whitney lesl wa s used fo r sla lis!ical anal ys is. MP, me lhylprcdni so lo ne; 1$, immuno,uppressio n, Cs , cyd ospo rin A . Ono pa licnt rcceivcd myco p hc nolalc 111 0 telil pl us mClh ylprcdnisolo ne I yca r pOSl-Ser (no l sh ow n).
ar ' ..
the leukemia-free survival and on the availabili ty of post tra nspla nt ma terial. Kneehtli el a l.'~ obse rved post transplan t M R D positivity in 8 of 36 pa ticn ts who remained in conti nu ing eomplete remissio n. AII M RDpositive samples in their study had been followcd by negative o nes, compared to 9 of J6 positive samples in this study. O n the basis of the interim results, we a lready eo nsidered th e last three positive samples in o ur study to be false positive a nd did not invite patients to un dergo repea ted exa mination out of sched ule, so we have not received the M RD results so fa r. One ean speculate that Lhose positivities were real and retlected GVL effeet preven ting a hematologica l rel apse. AII of the patients se lected for this study were MRD negative or low positive before th e tra nsplant, w hich has been shown to denote a low risk of post transpla nt relapse in several sludies. s H. 17 Moreover, II of 16 pos itive samples coming from paLÍents with fusion genes were negative in sim u ltaneous rusio n tran seript detectio n. We employed two non-sc lf-specific RQ-PC R assa ys a nd observed a substantia l percentage ol' false -posi tive amp lificalions in our co hort of samples. MRD moni to rin g based on anligen receptor gene rearra ngements utilizes 'fingerp rint-like' v- (D)-J seq uc nccs of leukemie clones for primer design. Nonspecific amplifieation in polyclonallymphocytes with simila r rearrangemcnts is usually determined using pooled PB DNA from 5- 10 healthy donors, a nd its oecurrence depends on the type or Ig/TCR targel and th e number of inserted N nllcleo tides (about 30-40% in fGH whereas 90% in TCRG targets) 4 0 Recently, van der Velden el al 41 reported tha l the level of nonspecific amp li ficat ion in IGH largels depends on lhe lime point during the ind uclion treatment and is the hi ghest in the post-maintenance period dlle Lo Ihe preva lence ol' CDIOpo' Tdpeg precursor B eells with com plete VH-JH rearrangemenls in the B-cell eompartmenl. Thc aulhors concluded, however, th a l ESG-MRD-ALL guidelines for interpretation of RQ-PCR data were sufficie nt for fro ntline ALL therapy mon itoring, with less th an 2% falsc-posilive resu lts. As we were limited in sample size, so \Ve co uld o nl y a na lyze two lGH targe ts representing the same type (VH3 and JH4) of rearrangement to minimize assay -specifie variations. Both assays ha d no baekground in PB buffy eoat; s tili , \Ve observed 20 a nd 15% false-posilivc results us ing regenerati ng post-SCT BM samples . Regardi ng immune cell reco nstitutio n, SCT co nstitutcs a unique silu ation . According to the KREC levc ls representing in our setting the output of B eells from BM, thc median Icvel remain$ lower th a n that of the control group unlil I-year post-SCT. This resull is in co ncordance with lhe stud y of K ook el al. : 2 who observed a depression or tot a l B-I ymp hocy te count until 18 months posl-SCT. Howeve r, a fractio n of patients had higher va lues than hca lthy BM donor startin g in a few ca ses as soon as day 60 posl-SCT, and creati ng more than half of the cohort 1- 3 years pos tSCT. Hi gher num bers of lOta l B cells were also observed in a st ud y of patients sur viving 20-30 yea r$ after lranspJa nta ti o n 4J The pro blem or extensive regeneration causing un spccifie primer bindi ng cou ld be theoretically overcol11c by using rege nera ting BM DN A instead of DNA from buffy coats as a negative eontrol; however, the availability oť such
•
Bone Marrow Transptanlalion
Drawbacks ol post transplant MRD in childhood ALL E Fronkova el al
8 7 6
a E
~~ 5
"';:; m (1)
~
0;0
4
3 u o w"O 2 cr: 1 Y: ~
c
o
•
P<0.0001
•
ó
False-positive
35 C 30 25 20 o "- 15 m 10 u 5
* ~
o
o
'6
•• ••
:t n=74
•
P<0.0001
v 10
"-
m
o u
5
o
• P<0 .0001
$ ~False-positive
MRD N e gative MRD
n=18
n=59
d 35
~~ • •
n=18
lI)
:<;
MRD Neg ative MRD
n=26
False-positi ve
b 25 - 20 ~ E 15
MRD Neg a tive MRD n=59
30 25 "i. 20 o "- 15 M o 10 u 5
•
P=0.98
•
~
o
6 False-positive
n=15
I
I
~
MRD Negative MRD n=54
t"igure 6 Lymphocy tc rccOllst it ution in minimal resid ua l di sca sc (MRO)-ncga ti ve vs :vIRO ralsc-positi vc samplcs (a ll time points togcthc r). (a) Expressed as K-d ele tio n cXl'ision cirdes ( KR EC) num bers rela ti ve to the don or BM sa mple. (b) Expresscd as a perccn tage of B-cell precu"ors (C D I9'''''C045'''''') and (c) S-cell percentage (CD 19"''') a ut ol the tOlal BM cell number. (d) The number ofT cells (C D 3"''') in the two groups W
material in practice is very limi ted. Our recommendalion is 10 continue using PB buffy coats while carefully judging cases uf nOl quantifiable posl-SCT MRD positi vilY. As a ll lhe RQ-PCR prod uc ls we s equenc ť d and analyzed by onchip elec lrophoresis differed from the spec ific ones, performing at least o ne of the two techniques would pay ofr in such cases. The impacl of preemplive thera py based on pasL tran sp lant MRD Ol' c him e rism deteclion has nol been pro ve n so far on a larger cohort o f pa ticnts. In our previous sludy, all atlempts used to ave rl posl-SeT relapse showed only a limited or lemporary e ffecl. " On l he basis of lh ese resu lls, we are now workin g on pre-lra ns planl MRD monitoring to enable ad din g lreatme nl to minimize MRD leve l before the transpla nt a nd thu s the ri sk or relap se. Howeve r, po s t lra ns planl MRD m o nitorin g stili remains an exlremely useful tool for thera p y management, if judged with cauti o n.
Acknowledgements
This work was s upp ortcd by MSM00216 208 13, MZO 00064203 , MZdNR8269-3(2005, MZdNR9531-3 and GAUK 754 3/ 2007.
References
Eckert C. Biondi A, Seegcr K , Cazzan iga G, H a rtmann R , Bcyermann B e l al. Prognostic value o l' minima l residual di sease in relapsed childhood ac utc Iymphobla sli c leuk aem ia. Lancel 200 1, 358: 1239- 1241. Marrow Transplantation
2 Floh r T , Schrauder A, Cazza niga G , Pa nzer-Grum ayer R, va n der Veld en V, Fischer S el al. Minimal residual di,easc-d ircclcd risk strat ifica lion using real-time quanlilalive PCR analysis ol' immunogJobu lin and T-ce ll recepto r gene rea rrangements in the internation a l mullicenler lrial AIEOP-BFM t\LL 2000 for ch ildhood acute Jymph oblastic leukemia. Leukemia 2008 (in press). Pui CH , Schrappe M , Ribeiro RC, Niemeyer C M . C hildh ood and adolescent Iym phoid and mycloid Icukcmia . lIenwlolo(JY (A m Soc HemalOl Educ Pro(Jl'am ) 2004; 2004: 118 · 145. 4 van Dongen JJ, Seriu T , Panzer-Grumayer ER, Bi ondi A, Pongers- Willem se MJ , Corral L el al. Prognoslic valuc o l' minim a l residual di scasc in ac ute Iymph oblastic Icukacmia in childh oo d. Lancel 1995; 352: 173 1-1 738. 5 Bader P, Hancock .I, Kreyenberg H, Goulden NJ , Nielhammer O, Oakhill A Cl al. Minim a l res idu a l disease (MRD) slalus prior 10 a llogen eic stem cell lransplantalion is a powerful pred ic to r for post-transplant oulcome in children wilh ALL. 1.RlIkemia 2002; 16 166S 1672. 6 Knech tli CJ, Goulden NJ, Ha ncock JP, Grandage VL, I-Iarris EL Garland RJ el a!. Minima! residual disease sta tus befme allogcncic bone marrow transp la ntalion is an impo rtant determinant of successful o utco me for children a nd adolescents with
Drawbacks of post lransplant MRD in childhood ALL
E FrO/lkova et a/
acul e lyrnp ho blaS lic leuke rni a. Pediolr B/ood Concer 2007; 48: 93- 100 9 Bader P, Bec k J , Frey A, Sch lege l PG, Hebarth H , Hand gre li nge r R cl al. Serial a nd q ua nlitalive analysis of mi xed herna topo ielic chimerisrn by PC R in palients wilh ac ute Icukemia s a ll ows lhc prediction o f relapse a ft er a llogeneic BMT. Bone Ma rro lv Tramplanl 1998; 21 : 487- 49 5. zun el M , Nas rnan-Bjo rk f. 10 Zelterqui st H. M a tlsso n J , Svenberg P, T arnrn ik L el al. Mi xed c himeri sm in the B ce ll linea ge is a ra pid and se ns ili ve indiea tor o f minim a l resi d ua l discase in bo ne rn a rrow transpl a nt recipient> \\' ith pre-B cell acute Jyrnpho bl astic leukerni a. Bone M orro w Ti-ansplan! 2000; 25: 843-851. II Sa nchcz J, Sc rra no J , Gorncz P, M a rlinc z F , Martin C, Madero L cl ol. Clini ca l va lue of immun o logicalmo nito ring o l' minimal residual disease in acule Iyrn pho blas lic leu kaemia after allogeneic tra nsp l.an ta tio n. Br J HaelllulGl 2002; 116: 68 6-694. 12 Radi eh J, Gehl y G, Lec A , A very R , Brya nt E, Edm a nds S cl ol. De tec ti o n o f ber-a bl tra nsc ripts in Phil ade lphia chro moSOllH>positi ve ac ute Iym p hobl a stic leukemi a a ft er ma rrow lran spla nta tioll. Bloort 1997; 89: 2602- 2609. 13 Knec htli CJ , Goulden NJ , H a ncock .I P, Harris E L. G a rl a nd RJ , Jones C 1'1 ul. Minim a l rcs idu a l discase statu s as a p rcdic tor of rela pse aft er allogenei,~ bone ma rrow tra nsp lant a tio n fo r children with aCLlte Iym phoblastic le ukaemia . Br J Hoem otol 1998; 102: 860- 871. 14 Mi glin o M , Berisso G , Gra sso R, Ca nepa L, la vio M , Pierri f CI al. AHoge ne ic bone ma rrow tra nspl anla tion (BMT) fo r adull S wi lh ac ule Iympho bl astie Icukemia (ALL): pred ictive ro le o f minim a l res idual di sease mOnilOI"ing on relap se. Bone MorrolV Tronsplanl 2002; 30: 579- 58 5. 15 Radi ch J, Ladne P, Gooley T. Po lymera se chain reac li onbased detec ti o n of minimal residua l di s use in acute Iymph obla stic leukemi a predicts rela pse a fter a ll ogeneic BMT. Biol Bloort MorrolV TronJp lanl 199 5; I : 24- 3 1. 16 Uzunel M , Ja ksc h M . Mattsso n J, Ringden O . M in ima l re idual disease detccti o n afte r al logeneic stem ce ll tra nspl a matio n is co rrel a ted to relapse in patients with ucule Iym phob lastic leukaemia. Br J Hoem%l 2003; 122: 788- 794. 17 Spindli O, Peruta B, Tosi M. Guerini V, Sa lvi A , Za nott i MC el al. Clearan ce o f minimal residual d iscase a ft er a llogene ic slem ccll tran sp lan la tio n a nd thc prcdict io n o f thc clinica l oulcome of adu lt pa lie nts with hi gh-ris k ac ule Iymph o bla sti c leuk emia . HuemalolOl/ica 2007; 92: 6 12-6 / 8. 18 Bade r P, Klin geb iei T , Seha ud t A , Th eu rer M a inka U, Ha ndgrctin ger R. La ng P el al. Prc ve ntio n of relapse in ped iatrie pa lienb wi th ac ulc leuk cmi as a nd MDS after a llogeneic SCT by ea d y immunotherapy initi a ted o n the basis of in creasing mixed chim erism: a singl e center expe ri ence o f 12 children. Leuke/l/ia 1999; 13: 20 79- 2086 . 19 Co llin s Jr RH, G o ldstcin S, Giralt S, Lev ine J , Po rle r D , Drobyski W Cl ul. Do nor le uk ocy te infusio ns in acute Iymphocy lic le ukemia. Bone MorrOI! ' Tramplanl 2000; 26: 51/ - 5 16. 20 Do mini etl o A , Pozzi S, M igli no M , Albarrac in F , Piaggio G, Bcrt o lo tti F Cl al. Do no r Iymphocyte infusi ons fo r the lrea lme nt o f minimal resid ua l di sease in acute leuk emia. Blood 2007; 109 5063-5064. 21 Keil F , Kalh, P, Haas OA , I;ritsch G , Reite r E, M a nnh a lte r C el al. Relapse of Phil adcl p hi a chro moso mc positi ve acul e Iymphob lasli c leukac mi a a ft e r ma rrow lra nspl a nla li o n: sustai ned mo lec ula r re mission afte r earl y an d dosecsca la ting infusio n o f d o no r IClIcocy lcs. Br J Haemulol 1997; 97: 16 1- 164.
22 Loren AW , Po rte r DL. Donor leukocyle infusions fo r thc treat ment ol' re la psed aCLtte leuk emi a afte r a lloge neic stem cell transplantati o n. Bone M urrolV Tral1 splol1 1 2008; 41: 4l)J 493. 23 Mehta PA , D avics SM . All ogene ic tra nspla nta ti o n fo r childhood ALL. Bone M arrOlv Tramplonl 200S; 41: 133- 139. 24 Pui CH , Jeha S. New thera pe utic stra tegies for lhe treatm ť nt o f acute Iymph o bla stic leukaem ia. N al Reu Drl/q Disell !' 2007; 6: 149-165. 25 Sch il ha m MW , Ba lduzz i A, Bader P. ls lhere a rol e fo r minimal rcsid ua l d isease Icvcls in thc trca tmcnt o l' A LL pa tie nts who receive a ll oge neic stem cell s') BOlle Marro ll' Tronsplunl 2005 ; 35 (Suppl I ): S49-S 52 26 Y aza ki M , And o h M , Ito T , Ohno T , Wad a Y. Succcss ful pn:vc nti on o f hcm a lo logica l rel a psc fo r a pa ticnt with aCUl C lymph o bl as lic Phil ade lphi a c hro mosome- positi ve leukemia a ft er a ll oge neic bone marrow t ra nsp lan ta ti o n by d o no r le uk ocy te infusion. Bone Marrow Tra/lSplalll 1997; 19: 393-394. 27 van de r Velden VH, C t7.z
Drawback, ol po,t transplant MRD in childhood ALL E Fronko'!a el a!
36 va n ZChll MC, Szczepa nski T, van der Burg M, van Dongen Jl. Rcp li cation history of B Iymphocytes rcvcals homcostatic proliferation and extensive antigen-induced B cell expa nsio n. J Exp M ed 2007; 204: 645-655. 37 Cook GP, Tomlinson 1M . Thc human ill1munoglobulin VH repertoire. ImmLln ol Today 1995; 16: 237- 242. 38 Ra :1 phorst FM , Raman CS, Tami J , risc hba ch M, Sa nz r. Human 19 heavy chain CDR3 rcgions in adult bone Illarrow pre-B cells display an adult phenotype of diversity: evide nce for st ructural seleetion ol' DH a mino aeid sequences. 1111 Immlln o! 1997; 9: 1503- 15 15 . 39 H aze nberg M D , Verschuren M C , Hamann D, Miedcma F. van Dongen JJ . T cell receptor excision eircles as Illarkers for recent thymic cmigra11ls: basie aspcets, technica l approach, and guidelines for inLerp retation. J Mol Med 200 I; 79: 631-640 . 40 Van Der Velden VH, Hochha us A, Ca Z7.Átniga G, Szezepa nsk i T, Gabert J. Van D on gen JJ . Deteetion ol' minimal residua l
discase in hematologie malignancies by real-time quantitative PCR: princip les , approaehes, and laborato ry aspects. Leukemia 2003; 17 101 3- 10 34. 41 va n der Velde n VH, Wij k huijs JM, van Donge n JJ. Non-specific amplifieation of patient-speeific Ig ITCR gene rcarrangcments dcpends on the time point during therap y: illlplications ror minimal residu a l disease monitoring. Leukemiu 20013; 22: 641-644. 42 Kook ll . Goldman F , Padley D, Giller R, Rumclh a n S, Holida M el al. ReconstrueLion of thc immullc system artcr unrelated o r partially matched T-cell-depleted bone m~rro\V transplantation in c hildrcn: immunophcnotypic analysi s and factors arrcct in g the speed or recovery. Bloud 1996; 88:
1089- 1097. 43 Storek J, l oseph A, Es pino G , D awson MA , D o ue k DC, Sullivan KM e l al. ImmuniLY of patients surviving 20-30 yedrs after allogencic or synge neic bone marrow transplant 8.lion. B!ood 200 I: 98: 3505- 35 12.
Supplemenlary Informa tion accompanies the paper on Bone M a rrow Transplantation websitc (http://www.nature.com/bmt)
Ma rrow Transplantation
E. Mejstříková, strana 116
Př íloh a
7
Minimal residual disease (MRD) an alysis in the non-MRD-based ALL IC-BFM 2002 protocol for childhood ALL: is it possible to avoid MRD testing? E Fronkova, E Mejstrikova, S Avigad, KWChik, L Castillo, S Manor, L Reznickova, T Valova, K Zdrahalova, O Hrusak, Y Jabali, M Schrappe, V Conter, S Izraeli, CK Li, B Stark,
J Stary and J Trka Leukemia (fF 6.924)
E. Mejstříková, strana 117
leukemia (200S). 1-9 © 2008 ature Publishing Group AII rights reserved 0887·6924/08 $30.00 www.nature.com/leu
ORIGINAL ARTICLE
"inimal residua I disease (MRD) analysis in the non-MR D-based AU IC-BFM 2002 ,rotocol for childhood ALL: is it possible to avoid MRD testing? "onkova 1, E Mejstrikova 1, S Avi ~ad2 , ', KW C hik , L Casti ll0 5 , S M anor ·3 , L Rezn ickova 1, T Va lova 1, K ZdrahJlova \ O Hru sak \ 7 labali6 , M Schrappe , V Conter , S Izraeli 9 . 1o , CK Li 4 , B Stark2.J, J Stary l and J Trka 1 4
iťfJilrtment of Pediatric Hemat%gy and Oncology, CUP, Second Medical Schoo/, Charles University and University Hospita/
ml, Prague, Czech Repub/ic; 2Pediatric Hemat% gy Onc%gy Department, Schn eider Chi/dren 's Medical Center of Israel,
'a h Tikva, /srael; .1 Sack ler Faculty of Medicine, Tel Aviv University, Tel Avi v, Israel; 4 Department of Pediatrics, Prince of Wales 1SfJital, The Chinese Universit:.. of Hong Kong, H ong Kong, China; "Servicio Hemato Oncológico, Centro HospJtalario Pereira '&Sť/l, Montevideo, Uruguay; Department af Pediatrics, Regional Hospital, Ceske Budejovice, Czech Republic; ' Department of '!iiatrics, University H ospita l Schleswig-Holstein, Kiel, Cermany; 8Departm ent of Pediatrics, University of Milano-Bicocca, !pfdale San Cerardo, Monza, Italy; qPediatric Hemato-Oncology, Cancer Research Center, Sheba Medical Center, Tel lShomer, /srael and lOSa ckler Faculty of Medicine, Tel Aviv U niversity, Tel Aviv, /srael
'l1e ALL IC-BFM 2002 prolocol was crealed as an altern alive 10 MRD-based AIEOP-BFM ALL 2000 sludy, 10 Inlegrale early tIpOIlse crileria Inlo risk-group slratlficalion in counlries not lIIorming rouline PCR-based MRO lesling. ALL IC slraliflcaIIXI comprfses Ihe response 10 prednisone, bone marrow (BM) IIIIphology al days 15 and 33, age, WBC and BCRlABL or MLU IF4 presence. Here, we compared Ihis slratificalion 10 Ihe MROcrileria using MRO evalualion in 163 palienls from four tLL le member counlries al days 8, 15 and 33 and week 12. lADnegaUvily al dav 33 was associaled with an age of 1-5 , WBC<20000,11-' , non-T immunophenolype, good ndnisone ,r esponse and non-M3 morphology al dav 15. There no significanl associalions wilh gen der or hyperdiploidy ~ Ihe sludy group, or wilh TEUAML 1 fusion wilhin BCP-ALL. '-Jents wilh M1/2 BM al dav 8 lended 10 be MRO negalive al Ieek 12. Palienls slratlfied inlo Ihe slandard-risk group had a lIIIer response Ihan inlermediale-risk group palienls. HowMr, 34% of Ihem were MRO posltive al dav 33 and/or week 12. Our findings revealed lhal morphology-based ALL IC 'IIi·group slralificalion allows Ihe Idenlificatlon of most MRO h·risk patienls, bul fails 10 discri mlnale Ihe MRO low-risk poop assigned 10 Iherapy reduclion . _',J.~emia advance on line publication. 28 February 2008; :o 10.1038/leu .2008.22 {eywords: acute Iymphoblastic leukemia; minima I residual ase; risk-group strat ification ; immunogl ob ulin and T-cell receptor ~e rearrangements; bone marrow morphology
Introduction ~~ra l retrospective studie s from the la te 1990s ha ve shown the
nical ,ignificance o f minimal resi dual disease (MRO) in "ildhood acute Iymphob lastic leukemia (A LL}.1 -] Based o n resul ts, MRO testing has become a part of the risk-g roup alification procedurc in severa l of th e most progressive ALL :ealment protocols· - 7 ln 2000, the International Berlin'ankiurt- Munster Study Croup (I -BFM-SG) incorpo rated MRO ling into risk-grollp striltifi cati on in the AIEOP-BFM ALL 2000 I. Alrnost all ' classi ca l' risk features (except prednisone 'IpOnse, t(4;11), t(9;22) and indu ction failure) were omitted in 'llTelpondenee: Dr J Trka, Department of Pediatrie Hema tology and l!!cology, Charl es Univ"rsity, 2nd Medieal Sehool, V uvalu 84, 150 - Pragu > 5, zeeh Republic. . ail: [email protected] ':I:eived 2 JJnu ary 2008; aeeepted 17 January 2008
this risk-group stratification. M RO negativity at d ay 33 (d33, end o f indu ction phase 1) and at weekl2 (w12 , before conso lidati on treatmentl stratified patients to the standard- (Iow) ri sk group, w hil e high M RO level at w12 wa s an additiona l contributi on to the cla ss ical high-risk featu res U Mi nima l residual d isease monitoring in the A IEOP-BFM ALL 2000 tri al is based on real-tirne quantitative PCR (RQ-PCR) detection of pat ient- spec ific immunoglobulin (lg) and T-cel l receptor (TCR) gene rearrangements, whi ch is currently consid ered the most relia ble tool for MRO cliagnostics. However, this method is time-consumi ng, logistically dema ndin g and relatively expensive. Moreover, to guarantee the credibility of 9 the result s, it requires regul ar inter-Iaboratory quality co ntrol s For these reaso ns, thi s method is stili inapplicable in many laboratories aroun d the wor ld. ln 2000, many countries within I-BFM-S G were no t able to appl y routine MRO testin g to clinical practice; therefore, a new protocol , AL L IC-BFM 2002, was desi gned in parallel to AI EOPBFM ALL 2000 . Countries that joined the ALL IC co nsortiu m wanted to test the poss ibili ty of (at least partially) gaining additional prognostic inforrnation by non-MRO-based methods. The patients were stratified into standard-risk, intermediate-risk and high-risk grou ps (SR G, IRG and HRG , respecti ve ly), according to age, white blood ce ll (WB C) count, blast proportion in peripheral b lood after 7 days of prednisone and a si ngle intra thec al do se of methotre xate (prednisone response) and the presence of t(9; 22) an d t(4; 11) fu sions (Figu re 1) . As the morphological assessrnent o f bo ne marrow (BM) at day 15 (d15 ) provided additional prognostic information, 10 d15 BM evaluation was newl y added to the stratification sc heme. A 510w response to treatment detected as M3 BM at d15 re-striltificd pa tient s to high er risk groups. More th an 800 chi ldren a year from 12 cou ntries all around the world (inc luding Argentina , Chile, Croatia, Czech Republic, Hong Kong, Hllngary, Isra el, Poland, Serbia, Slovakia, Slovenia and Uruguay) were treJted according to the ALL IC protocol. Apart from the treatmen t questions, o ne of the goa ls of th is stud y was to compa re thi s riskgroup assessment to th e MRO-based criteria used in A IEOP-BFM A LL 200 0. Therefore, this stud y asked the qllc stion wh ether it is possible to avoid MRO testing in some subgro up s of ratients. The study was ca ll ed 'Mini Risk' and cOllsisted of MRO evalu ation in 163 patient s treated in four rn ember countries of the A LL IC co nsortiu m (C zech Republ ic, Isracl , Hong Kon g and Uruguay).
E. Mejstříková, strana 118
MRD and risk group stratification in childhood All treatment prolocol E Frankova et a/
Patients, materials and methods
the analyzed subgroups. Twenty patients we re not analyzed for MRO: 3 because of early death during the induction phase and 17 because of poor quality or missing diagnostic ONA. The BM sampling was performed at up to six fixed time points along the cou rse of the treatment protocol: d8 (Czech Republic only) and d15 of induction, end of induction phase 1 (d33), preconsol idation (wl2), pre-reinduction (w22) and at the end of the first year of treatment (w52). BM was classified as Ml « 5% blasts), M2 (5 - 24% blastsl and M3 (:;,, 25% blasts) by national reference laboratories using standard morphological criteria. AII patients or their parents or guardians gaye informed consent for the treatment study.
Pa tients A total of 207 children with ALL (age 1-18 years) treated in the Czech Republic (November 2002 to February 2004, n = lOS), Israel Uanuary 2004 to Oecember 2004, n = 48), Hong Kong (January 2003 to March 2004, n - 36) and Uruguay (April 2004 to March 2005, n = 18) were included in the Mini Risk study. AII children were treated according to the ALL IC-BFM 2002 protocol. As Ihe Mini Ri sk study was mainly aimed at SRG, sampling of SRG patients in the Czech Republ ic continued to October 2004. Moreover, Czech HRG and T-cell ALL (T-ALU patients consecutively diagnosed up to Oecember 2004 were enrolled in the study to acquire suffjcient numbers of patients in
Treatment Figure 2 shows the treatment sc heme of the ALL IC-BFM 2002 protocol. In brieC after 7 days of steroid prephase and one injection of i.t. MTX (intrathecal methotrexate), all patients rece ived the 8-age nt, 8-week induction therapy (protocol I). Consolidation therapy (protoc ol mM/Ml consisted of four courses of high-dose (HO) MTX 2 g m- 2 (5 g m- 2 for T-ALL) for SRG and IRC patients. Cranial irradiation was applied in non-transplanted HRG patients, T-ALL patients and in patients with CNS status 3. At the beginning of delayed re-induction , patients were randoml)' assigned to receive one protocol II vs two protocol III (using le ss dexamethasone, vincristine, cyclophosphamide and anthracycl ines) treatments for SRC and one protocol II vs three protocol III treatments for IR C. HRC therapy con sisted of three HR-blocks and triple reintensification w ith three protocol III treatments (HR-ll vs six HR-b locks plu s protocolll treatment (HR-2B) vs three HR-blocks
agel·6yand age<10rl!6yrsor 1(9;22), 1(4;11),or WBC<20 000 111-1 and WBC l!20 000 111- 1 and blasls d8l! 1000 111-1 blasls d8<1000 111-1 blasls d8<1000 111-1
dg. Oay 8
Oay 15
Ml/2
I I M3
Oay 33
Risk
group:
Figure 1 The risk-group stratifi cation scheme in the ALL-IC BFM 2002 trial. M1/2/3, BM status according to morphology, Ml «5% blasts), M2 (5- 24% blasts) and M3 (:;" 25% bl asts). BM, bone marrow.
t
t
t
i
6·MP/MTX
8 1~:~ j~~-':~",eWjlh 8 I
12 Gy' only for T-ALL
d15
w12
d33
dx
6-MP /MTX
~ ~
~ III'
~
**
mM S
IR
~
;
t
i
1_ 1_ [I
" .GC~"~~_. 12Gy'
m~ ~
I
i
8
6-MPIMTX
12Gy'
~ 4w'
BFM
I
t
m=[L
6-MPIMTX
[
[
[
[
12
I§
6-MPIMTX
12Gy'
......'I,1i:
10
t 6-MP/MTX
AIIo-SCT#
t
O
I
6-MP/MTX
~8r-~~:"T HR PRED-PR IR. M3d1 5 1(9;22) 1(4;11) NRd33
t
I
~
52 104 W
Figure 2 ALL IC-BFM 2002 trea tment. 1 BM sampling; .u i.t. MTX in maintenance therapy; " Prot. I', DNR 30mgm 2 only for SR pati ents with BCP-ALL; 'BCP-ALL, MTX 2 g m- 2 per 24 h x 4; T-ALL, MTX 5 g m- 2 per 24 h x 4; 'CNS status 1/2, pe RT - 12 Gy; CNS status 3, tCRT:12/18 dosage by age al' treatment del ivery, iniants < 1 year of age had neither pCRT nor tCRT; sno randomization of AIEOP vs BFM but choice by group acco rdin g to previous experience with on of the 1\'\10 high-risk strategies. i.t. MTX, intratheca l methotrexate.
E. Mejstříková, strana 119
MRD and risk group slralificalion in childhood ALL Irealmenl prolocol E FronkOlla el a/
:' l\Vo protocol II treatments (HR-2A). Maintenance therapy a total treatment duration of 24 months consisted of lercaptopurine daily, MTX weekly and extended i .t. MTX children re eiving HO MTX 2 g m- 2 in consolidation. elements of this BFM-type therapy have already been JJ/ished. JJ. J2
concentratron and qualily.22 The E5G-MRO-ALL (European Study Group on Minimal Residual Oisease in ALL) criteria for RQ-PCR sensitivity and quantitative range interpretation were used 2J
•
Feasibi/ity of MRO monitoring
h l' cytometry ~w cytometry immunophenotyping of BM aspirates was fflonned at diagnosis using a panel of mAbs recommended the European Group for th e Immun%gica/ Characterization ·Leukemias. J3
Jelection of fusion genes 'lIe presence of TEUAML 7, BCR/ABL and MLUAF4 fusion genes 'as examined as a part of routine ALL diagnostics by real-time ICR in the Czech Repub/ic and by fluorescence in situ rbridization in Israel, Hong Kong and Uruguay.
A tota/ of 187 patients were investigated for the presence of clonal IglTCR rearrangements. In 179 (96'10) patients, al least one monoclonal marker was found. Patient-specific RQ-PCR assay with minimal sensitivity of 10- 4 was established for 163 (87%) patients. The main reason for inadequate sensitivily was early amplification of contro/ DNA from healthy donors. One hundred and four patients (53%) had two sensilive IgfTCR targets. We have prev ious/y shown a good correlation between two independent IglTCR targets during the ALL IC-BFM 2002 24 induction treatment Higher values were used for MRO analysis in patients with two IglTCR targets.
Statistica/ ana/yses DNA index
:ne
DNA index was assessed routinely at diagnosis using the 'yc/eTestP/u s DNA Reagent kit (BO, San Jose, CA, USA) as scribed previous/y.14 The DNA index w as defined as a ratio of ne mode of fluorescence of cells in the GO/G1 phase and the mde of fluorescence of norma I peripheral blood in GO/ G1. rognostically re/evant hyperdip/oidy was assessed as havi ng a D, A index ~ 1.16 and ";'1.6.
Oetcclion of residua/ disease '.\ononuclear cells from the diagnosti c BM samples were lOlated by Ficoll-Paque (Pharmacia, Uppsa/a, Sweden) density entriíugation and stored in /iquid nitrogen. Follow-up BM or ?eripheral blood sampl es were proce' ed by erythrocyte Iysis ,nd stored at - 80 ve. Genomic DNA was iso/ated by the QIAamp DNA Blood Mini Kit (Qiagen GmbH, Hi/den, Germany). DNA was stored at n -20 C before processing . Primers and protocols for detection of rnmunoglobulin heavy chain, immunoglobulin light chain K, ' -ce ll receptor y, T-cell receptor o gene rearrangements and ' AL 1 deletions were dc'scribed previously . " .11, Clonality of the PeR products was confirmed by heteroduplex anal ysis. J7 lequencing was performed using the ABI PRISM 3 10 Genetic ~na /yzer with the BigOye Primer v3.0 Sequencing Kit (Appl ied 3iosyslems, Foster City, CA, USA). Variable (Vl, diversity (O) Ind joining Ul regions of the immunoreceptor gene were rdentiíied by comparison with sequences in GenBank using the ImMunoGeneTics (IMGT) Oa tabase (http://imgt.cines.fr, IMGT, European Bioinformatics Institute, Montepe/lier, France) Jnd the IGBlast search (http ://www.ncbi.n/m .nih.gov/ igb/as tl, ~a tional Center for Biotechn%gy Information, Bethesda, MD, USA) Patient-specific forward primers for RQ-PCR were designed using the Vector NTI 8 Suite Software (Informax, Bethesda, MO, USA). Famil y-specific reverse primers and probes for immunoglobu/in heavy chain, immunoglobulin light chain K, T-ce/l receptor 8 and T-ce/l receptor y were described previously .J8- 21 IglTCR RQ-PCR was performed using the iCycler IQ'" real -time peR Oetection System (Bio-Rad, Hercu/es, CA, USA) and the ABI Prism 7700 real -time PCR Sys tem (Applied Biosystems, foster City, CA, USA). Standard curves were prepared by diluting the diagnostic samples in pol yclonal DNA from healthy donors. The albumin ge ne was used to normal ize the DNA
Oistribution of variables between groups with positive and negative MRO w as assessed using Fisher' s exact test. The Mann-Whitney and Kruska/-Wa/lis tests were used to cslimatc significance of differences concerning continuous MRO values. The statistical anal yses were performed using StatView version 5.0 (StatView Software, Cary, Ne, USA).
Results For intergroup analys is, a consecutive/y recruited cohort af patients w as anal yzed, with MRO values available in 69 (d8), 87 (d15 J, 133 (d33) and 133 (wI2) samples. For internal SRG, HRG and T-ALL group ana/yse s, augmented consecutively recruited groups were used as described in the Patients section.
MRO and ALL IC-BFM 2002 risk groups Table 1 shows the distribution of MRO positivity and negativity at d33 and w12. Overa/l , 68 of 133 (51 %) consecutively diagnosed patients had, at d33 , a BM Ihat was negative for MRO (d33 MROnc-g ) and 107 of 135 (79%) were MRO negative at w12 (w12 MRO ne~). Twenty-five of 133 (19%) patients had, at d33, MRO levels that were higher than 10- 3 MRO correlated significantly with the ALL IC-BFM 2002 risk groups. Median MRO level at d33 was 1.0 x 10- 3 in HRG patients compared to 1.9 x 10 s in IRG patients and O in SRG patients (Figure 3a). Standard-risk group patients responded better in terms of MRO negativity at d33 than IRG patients (P= 0.0018). Howcver, 21 of 62 (3 3.9%) SRG patients for whom both d33 and w12 fol/ow-up samples were available were d33 and/or w12 MRDPo" (ranging from borderline positivity to '1.5 x 10- 2 at d33 and to 1.2 x 10- 4 at w12). These palients would not qualify as MROba sed SRG in AIEOP-BFM ALL 2000. Figure 3b shows MRO clearanc e during the initial phase of treatment in the ALL IC 5RG, highlighting patients with slow mo/ecu/ar response. They did not differ signiricant/y from those with rapid mo/ ecul ,Jr response with respect to gen der, WBC, BM morph%gy at d15 or presence of hyperdip/oidy or TEUAML 7 fusion. The only difference was in a higher proportion of M3 BM at d8 in MRO slow responders (P = 0.016 , see be/ow). Conversely, 26 of 64 (40.6%) consecutive/y diagnoseo ALL IC IRG patients were MRO negative at both d33 and w12 (15 patients by 2 IglTCR targets, thus fulfil/ing AIEOP-BFM ALL 2000 criterion for SRG). Leukemia
!ukemia
E. Mejstříková, strana 120
MRD and risk group stratification in childhood All treatment protocol E Frankova et al
8a. (])
c
o .'!1
c
"O
(])
o. cr:
8::
MRo and c/inical/biological factors Age. Using a cutoff of 6 years applied in the ALL IC-BFM 2002 stratific:ation, palients older than 6 years were mor0 likely to be d3 3 MRDPo, than patient s 1-5 years of age (P= 0.0007) . Quantitative MRD levels differed significantly at d15 (P=0 .045) and d33 (p=0.002S) between the two groups and showed no significant difference at d8 and w12 (Figure 4a). In w12, neither a 6-year nor 10-year cutoff significantly di stinguished MRD negative patients from those who were MRD positive.
White blood cell count. Patients with a presenting leukocyte count of less than 200001-11 - 1 at diagnosis, corresponding to the cutoff used in the ALL IC-BFM 2002 stratification, were more likely to be d33 MRDneg than those with WBC> 20 000 1-11 - 1 (P= 0.0013). The difference was stili present at w12 (P= 0.0028). Figure 4b shows box-plots representing quantitative MRD levels for WBC lower or higher than 200001-11-1 during the initial phase of therapy, showing a significant difference between the two groups at d33 (P= 0.0009) and w12 (P= 0.018) Cender. There w as no difference in molecular response between boys and girls, neither in achieving d33 or w12 MRD negativity, nor ln MRD level s at respective time points during therapy (Table 1). Immunophenotype. Patients with BCP-ALL were more likely to achieve d3 3 MRD negativity than those with T-ALL (P= 0.0019, Table 1) However, this difference was not distinct at w12 When comparing quantitative MRD levels, patients with BCP-ALL had significantly lower MRD level s at d1S (P= 0023) and d33 (P= 0.002) than T-ALL patients; the difference was not significant at w12 (Figure 4c) A multivariate analysis inciuding age, WBC and B vs T immunophenotype did not show a significant impact of immunophenotype on d33 MRD statu s (P= 0.26), which wa s caused by its correlation with WBC (WBC: P= 0.029; age: P= 0.039).
c
o
'" o§' u
o
:o "D
c
'"
u '" S
u
'O ti
Ol
E
TEUAML 7. Although median MRD level in d33 MRDPoS patients was 2.3 x 1O-~ in TEUAML7-positive compared to 3.0 X lO-l in TEUAMLl-neg ative BCP-ALL patients, there was no significant difference in achieving d33 MRD negativity between patient s with and without the TEUAML 7 fusion (Table 1). Within the consecutively diagnosed TEUAML 7 group, 10 of 31 (32 .3°,{») patients had detectable MRD at the end of induction. There was no impact of the TEUAML I fusion on w12 MRD status. Other chromosomal translocations. As the patients with t(9; 22) have been treated according to the EsPhALL protocol sin ce 2004 in the Czech Republic and Israel, there were onl y four patients with the BeR/ABL fusion in our cohort, thus preciuding statistical analysi s. Similarly, there were no children with t(4; 11) in the cohort. ONA ploidy. The data concerning DNA ploidy were only available in Czech, Hong Kong and Uruguay patients. There was no significant difference in molecular response between patien ts with a hyperdiploid phenotype and those with a norma I karyotype, neither in achieving MRD negJtivity (Tahle 1), nor in MRD levels at respective time points during therapy (data not shown).
E.
Mejstříková,
strana 121 MRD and risk group stratification in childhood ALL treatment protocol E Fronkov3 et aj
a
dB
d15
d33
•
w12
1.00E+Ol 1.00E+OO 1.00E-Ol 1.00E·02 1.00E-03 1.00E-04 1.00E-OS
neg.
fl: U
~
______________- L__
~~~
______
~~
__
~~~i-
____
~~
____
~--U
b
~
C fl:
:E
Figure 3 MRD and ALL IC-BFM 2002 risk-group strati fi cation . (a) Quantitative MRD levels (Iogarithmic scale) at d8, d15, d33 and w 12 accord ing 'v lhe ALL IC-BFM 2002 risk groups. (b ) MRD c1 earance in patients stratified to thl' ALL IC-BFM 2002 sta ndard-ri sk group . Empty diamonds in dicatc ..Iienls with MRD positivity at d3 3 and/ar w 12, who would not qualify as SRG in th e MRD-based AIEOP-BFM ALL 2000 tria I. MRD, minimal residu al di sease; SRG, sta ndard-risk group.
\1RD and early response to therapy Prednisone response N one of the patients with more than 1000 blasts fll
I
in peripheral blood afte r 1 week of predn ison e
mel one i.t. MTX (poo r prcd ni sone respo nse) achieved negativity ,td33 (P = 0.0026, Tabl(' ll. Th e difference rem aine d si gnifica nt II \V12 (P= 003 1) Qua ntitat ive MRO levels differed signi fi-
"ant ly between prednisone poor and good responders at d15 p- 0.037), d33 (P< 0.000 1) and w12 (P= 0.032, Fi gu re 5a)
Day 15 bone marrow morphology (BM d15). The d15 llorphology evaluation was performed cen trall y in the A LL ICBFM 2002 protocol due to BM d1 5 implementation in risk~o up stratificatio n. In the w hole coh ort, on ly two patients were re·stratified from SRC to IRC and two pJtients (ro m IRC to HRC . ·wo IRC patients w ith M3 BM al d15 we re not trea ted as HRC due to treatment intoleran ce. Patients with M3 BM d15 were more likely to be d33 MROPo, than patients with Ml and M2 P= O.03, Tab le 1). Fi gure 5c shows the com pari son of d1 5 BM lt4tus with M RO level s at d 15 and its impact on quantitative d33 md \V12 MRO levels. Th ere was no difference in MRO status ':etween patients with Ml and M2 BM on d15 al ,Jny time point. Figure 6 shows the correlation of morphological- and PCRbased d15 BM assessme nt, w ith 28 of 95 (29%) samples m luated differently by ea ch meth od. Th ree cases we re l.Iscssecl as Ml by morphology and M3 by PCR, and, vice ,~rsa , one case wa s evaluated as M3 by morphology and Ml hy peR. We then evalu ated th e impact of PCR-b ased MRO at d15 on PCR-based MRO at d33 (F igure ll. Using the MRO cu toff
level 5 x 10 - 4 , 15 (1 4%) pat ients had an M RO lowe r th an thi s va lue. N on e of these patients were d33 MROPo" compared 4 to 53 (6 0%) patients with d'15 MRO hi gher th an 5 x 10- . Similar resul ts were observed using an MRO d15 cu toff of 1 x 10- 3 The MRO low-risk group defined by this cutoff compri sed 19 patients (18%) and on ly 1 of these pati ents w as MRO positive at d33, compared to 51 (59%) in the second group.
Day B bone marrow morphology (BM dB) Oay fl BM morphology is assessed as a pa rt of the A LL IC-B FM 2002 in the Czec h Rep ublic. The proportions of Ml, M2 and M3 d8 BM in th e consecut ively recruited grou p were 12, 24 and 64°/.., respectively. BM d8 status had no impact on d33 MRO positivity (Tab le 1). Howe ver, this was main ly due to a higher percentage of Ml BM at d8 (7 o f 21, 33%, P= 0.03 5) in T-A LL patients, wh o tended to have slowe r MRO response during th e indu etion phase. Interestingly, only 1 of 14 SRC patients with d8 Ml or M2 BM w as positive at d33 , compa red to 13 of JO patients with dB M3 BM (P = 0.019 ). The impact of da BM status on MRO status is stili ev ident at w12, even when analyz ing th e w ho le cohort, including T-ALL patients (Table 1). Fi gure Sb shows the quantitative MRO level s accordin g to d8 BM status, with significa ntly higher levels at d15 in patients with M 3 BM at da. lNe obse rved no signi ficant corre lation between dB and d33 MRO level s (data not shown). The overall results of dB BM ev aluation and its impact on survival in a larger eohort will be reported after sufficient follow-up. Leukemia
E. Mejstříková, strana 122
MRD and risk group stratification in childhood ALL treatment protocol E Fronkova el a/
..• . .
a
dB
1.00E+Ol 1.00E+00
~
1.00E-Ol 1.00E-02 1.00E-Q3 1.00E-Q4
·
t
d15
.
$
...
w12
P=O.0025
P=O.62
H" .
P=O.045
P=0.424
1.00E-05
d33
neg. <6 yrs
b ()
1.00E-Ol
~
1.00E-02
Cl
1.00E-Q3
a::
::!:
<6 yrs
~6 yrs
~ 6 yrs
<6 yrs
1.00E+01 1.00E+00
a::
~ 6yrs
T~
1.00E-04
P=O.14
1.00E-05
•.......
E?
.
< 6 yrs
. .
P=O.0009
~ 6 yrs
P=O.01B
•
P=O.50
neg.
I. T~ $ .
<20000 ~t - '
C
1.00E+Ol 1.00E+00 1.00E-Ol 1.00E-02 1.00E-03
~ 20000,,1-'
<20000,,1- 1
~ 20000 ,,1- 1
........... • .
P=O.OO2 .......• ...
P=O.89
1.00E-04
P=O.023
P=O.454
1.00E-05
neg.
BCP-ALL
T-ALL
BCP-ALL
T-ALL
BCP-ALL
T-ALL
BCP-ALL
T-ALL
Figure 4 Relation of ciinical ieatures to MRD at dB, d15, d33 and w12 . Quanlitati ve MRD levels (Iogarithmic scale) were related to (a) age al diagnosis, (b) presenting while blood cell count al diagnosis and (c) B vs T immunophenotype. MRD, minimal residual disease; WBC, white blood cell.
Discussion Most modern treatment protocols for childhood ALL have implemented some sort of MRD assessment into their risk-group stratification, aimed at both treatment intensification and reduction . Using MRD w ith the aim of therapy reduction requires a highly sensitive methocl (at least 10--4 ), which so far has not been commonly accessible for many countries worldwide. The ALL IC proto col was designecl to span the period during whi ch the involved countries cst ablishecl routine MRD testing. One of the goals of the design w.1s to implement d1 5 BM morphology evaluation to re- strJtify earl y slow responders to treatment intensific.1tion. Antigen receptor-ba sed MRD evaluation was used in aur study to determine the correlation of ALL IC risk-group assignment with mo lecular response during the initial phase of therapy. MRD correlated stron gly with ALL IC risk groups. As expected, HRC patients showed slower mol ecular response than IRC and SRC patients. None of the analyzed patients woulcl be stratified into HRC by AIEOP-BFM ALL 2000 criteria, solely based on high MRD level at w12 . AII such patients were .1lso predni sone poor responders . IRC patients had higher MRD levels al dJJ than SRC patients. However, about one-third of ALL-IC SRC patients were positive at the end of induClion and/or at w12. Based on the previous findings, these patients have a higher risk of relapse 3 Consequently, no ukemia
lreatment reduction would be justified in a protocol based on ALL IC stratification criteria . Next to age and WBC as c lassical ri sk factors, we observecl tl significant difference in molecular respon se between BCP-ALL and T-ALL patients. BCP-ALL patients showed a better response at d15 ancl dJJ, while the difference between the two groups was no longer clistin ct at w12. Thi s finding is in agreement with a previou s study comparing BCP-ALL and T-ALL patients treated 25 ancl implies the according to the ALL-BFM 90 protocol potential ben efit available to T-ALL patients from tre.1 tment modalities used later in th e treatment. As most of th e T-ALL patients were positive at d33 and negative al w12 , another time point in between the two mi ght be beneficial for ri sk stratifi ca tion of T-ALL patients in BFM-basecl protocols. BM evaluation at d15 has been an integral part of the ALL IC stratification and MJ BM at d15 showed a negative impact on d33 MRD in our stu cly. Onl y four patients in our cohort w re re-stratified into therapy intensification based o n MJ BM d15 . In all four ca ses, the morphologi ca l evaluation was concordant with the PCR-based evaluation . In general, the correlation between morphology and PCR wa s far from ideal. II ( ould be explained by a difficult BM evaluation at this time point. On the other hand, RQ-PCR also has limitations, mainly in the exact determination of high MRD level s: one PCR cycle clifference means a twofolcl di fference in MRD value. Using a cutoff of
E. Mejstříková, strana 123
MRD and risk group stratification in childhood ALL treatment protocol EFrankova el al
a
d15
w12
d33
•
1.00E+01
I ....
1.00E+OO
C?
1.00E-01 1.00E-02 1.00E-03 1.00E-04 1.00E-05
P=O.032
•
• P=O.037
P
neg.
PGR
b
1.00E+01
1.00E-01
~
1.00E-02
u
C
1.00E-03
:E
1.00E-04
a:
', '
.
P=O.229
" , '
•
P
neg. BMd8
M1
1.00E+01
M2
. ... ...
M3
1.00E+OO
... ... .1 ........Eť
~
1.00E-01
:.:.:::::
~
$
1.00E-02 1.00E-03 1.00E-04 1.00E-05
neg.
PPR
PGR
P=O.122
I
•
1.00E-05
C
PPR
PGR
, 9~
1.00E+OO
a:
PPR
M1
••
• . ;.
P
M3
M2
M2
M1
P=O.01
-:; ' . ' :-
L -______________________
BMd15
M2
0 , "
-L~
____
M3
M1
•
_L~~~~
M2
+
P=O.279
•
I
_ _ _ _ _ _ _ _L __ __ L_ _ _ _ _ _
M3
M3
M1
~
__
M2
~
M3
figure 5 Relation or earl y treatment response in PB (a) and BM (b, c) to MRO at d15, d33 and w 12. Quantitative MRO levels (Iogarithmic scal el were related to (a) blast count in PB aiter 7 days of prednisone and one i.t. MTX (prednisone response) or less than 1000~d-' (PGR) vS more than 1 000~li 1 (P PRI, (b) BM status at d8 (BCI'-ALL only) and (c) BM status at d15. BM, bone marrow; i.t. MTX, intrathecal methotrexate; MRO, minimal residual discase; PB, peripheral blood; PGR, predn isone good responders; PPR, prednisone poor responders.
i X 10.- 4 fo r peR at d15 , we could identify a small group oipatients (14'/'0) who achi eved ,\lIRD negativity at d33. This is n agreement with th c previous stu dy of Panzer-Grumayer et al.,26 who observed an excellent outcome in a similar group of patients treated according to the ALL-BFM 90 protocol. However, the use of IglTCR-based MRD monitoring (or treatment assign ment in such an ea rly time point is limited by the extreme logistical requirements of this method. Flow cytomelry seem s to be a more co nvenient tool for the early identification of patients with excellent prognosis 27 . 28 BM dB evaluation on the BFM-based protocol has not been reporteo so far. The Chi Idren' s Oncology Group slud)' concerning BM dB reflecls a different treatmenl approach (without a prednisone pre-phasel. With regard to this, it is not surprising lhat the proportion of Ml and M2 marrows in the Children ' s Oncology Group study is significantly hi gher. 29 ln our study, BCP-ALL patients with M3 BM d8 were more likel y to be MRD posi tive at both d33 and w12. This trend was even preserved in SRG. However, 67% of SRG patients in our cohort had M 3 BM
d8 ano more than half of I'hem were subsequently d33 MRD negative, which impairs the practical use of this factor. Our results indicate that M'I BM d8 might potentially identify a small group (11 %) of BCP-ALL patients with excellent prognosis. However, it will be necessary to ex tend the co hort and perform a follow-up to confirm this hypothesis. Patients wi th TEUAML 7 translocation did not dif(er from the other BCP-ALL patients in molecular response. About one-third of TEUAMLI-pos itive patients had detectabl e MRD at d33 . According to a previous study, those patients have a higher risk of subsequent relapse 30 ln co ntrast to the well -documented favorable prognostic impact of hyperdiploidy,3 1 about half of the hyperdiploid cascs were MRD positive at d33 . This could potentially mean that those patients have a greater risk of rel apse. Unfortunately, the original BFM MRD stud y 3 lacke d information on DNA ploidy. Borowitz el aj2q reported a surpri singly high percentage of MRD positive ca ses at the end of induction among paticnts with favorable trisomies, despite the excellent prognosis of this group. Leukemia
E. Mejstříková, strana 124
MRD and ri sk group stratification in childhood ALL treatment protocol
E Fronkova et al 10 ,-----~--------------------------------------~
, 2
M3
: M3mo:
, >-
Q.
o E
4
t l ·
01
____________ __ __________ ____ ____ !.. _ _........
o
M2mo/M1 PCR •
(5
.s=
.
0.1
10
.
55.
"C
•
0.01 ,
:
••
•
M1
:
.....
•
_! _____
.t.i i{.- to. -::- - - ..t .:. · . I
- - - - - - -r - - - - - - •• - -+ - -....- - ...;~ - ,~ -
ll)
L
~
__
:. 8 ~ •• ,M2 ~ M2mo/M3PCR 3
•• • •
•
,
:..
tl
~
-; - - - - .
M1mo/M3PCR
: • 9 :
3
,:M1mo/:M2PCR , , Negative (n=1)
1E-05
1E-04
1E-03
0.01
0.1
10
d15 MRD (lgfTCR)
figure 6 Correlation of morphological (mo) and PCR-based BM assessment at d15. Percentage of blasts according to IgfTCR-based RQ-PCR (x axis) vs percentage of bl.Jst5 according to morphology (y axis) in 95 BM sampl es. BM, bone marrow; RQ-PCR, real-time quantitative PCR .
5E-04
10
'"'""C
0.1
U
Ci"
0.01
'a, ::-
1E-03
Cl CI:
:E
I I I I
>::
15 (14%)
I
19 (18%)
I
1E-05
1E-05
,• • • • • • • • •• • •$•• •• • • • # • , •~ .: • • #. • • ~ I• •
>.!
1E-04
Negative
•
1E-04
I I
1E-03
0.01
0.1
10
MRD (lgfTCR) d15
Figure 7 Impact of PCR-based MRD at d15 on PCR-based MRD at d33. Pcrcentage of blasts at d15 (x axis) vs percentage of blasts dt d33 (yaxis) in 104 BM samples. Arrows indicate the respcctive d15 cutoffs of 1 x la-OJ and 5 x 10- 04 thal define the MRD low- ri sk groups. BM, bone marrow; MR O, minimal residual disease.
Our stud)' was aimed at eorrelation of elinieal, morphologieal Jnd biologieal stralifieation criteria used in the ALL IC-BFM 2002 protoeol wi th MRD clea ranee. Generall)', MRO reduetion during the first 3 months of therap)' eorrelated with ALL IC-flFM 2002 risk groups. This was most distinet in the high-risk group, implying that the 'classieal ' criteria used in ALL-IC are able to ielentify most HRG patients. However, there was a large overlap between the intermediatc- and standard-risk group s eoneerning MRO negativity at the end of induetion. Thus, the ALL le criteria dre not able to reliably define the low-ri sk group potentiall y assigned to therapy reduetion, whieh is the ehallenge of modem leukemia treatment. Ouring our stud)', the countries involved implemented the methodology of PCR-based MRO testing and are now prepared to use the MRO-based protoeol. Stili, great effort shou ld be made to the identifieation of sim pler stratifieation criteria (for example, flow eytometry MRD ass('ssment), sinee PCR-based MRO testing is stili unavailable in many eountries. emia
Acknowledgements This work was supported by MSM0021620813, MZOOOO 64203, MZdNR8269-3/2005, Israel Caneer Association and Children 's Caneer Foundation of Hong Kong. We thank the fol lowing participating clinical eenters for elini ea l management and sample colleetion: Czech Republi c: Brno U Sterba), Ceske Budejoviee (Y Jabali), Hradec Králové U Hak), Olomouc (V Miháll, OstravJ (B Blazekl, Plzeň (Z Černá), Prague U Stary), Ústí nad Labem (O Prochazkova); Israel : Afu la (H Gavriel), Beer-Sheva U Kapelushnik), Haifa Bnei-Zion (O Attias), Haifa Rambam (R Elhasidl, Holon (A Ballinl, Jerusalem Hadassa h (M Weintraubl, Jerusalem Shaarei Zede k (H Miskinl, Petah-Tikva (G Avrahamil, Rehovot (O Sthoeger), Tel Aviv (Y Burstein), Tel Hashomer (S Sielorai). 'vVe also tha nk S Bendova for sequeneing and A Vrzalova for data management.
E. Mejstříková, strana 125
MRD and risk group stratification in childhood All treatment protocol
E Fronkova el al
• 9
~eierences
1 Cdve H, van der Werff ten Bosch i, Sueiu S, Guidal C, Waterkeyn C, Otten J et " I. Clinieal signifieanee of minimal res idual disease in ehildhood aeute Iymphoblastie leukemia. European Organi zation for Research and Treatment of CancerChildhood Leukemia Cooperative Group. N Engl) Med 1998; 339 : 591 - 598. 2 Coustan Smith E, San cho i, Hancock ML, Bo)'ett JM, Behm FG, Raimondi SC el al. Clinical import dnce of minimal residual disease in childhood acute Iymphoblastic leukemia. Blood 2000; 96: 2691 - 2696 J van Oongen Ji, Seriu T, Panzer-Grumayer ER, I:liondi A, Pongers-Willemse MJ, Corral L el al. Prognostic value of minimal residual disease in Jcute Iymphoblastic leukaemia in childhood. Lancel 1998; 352:1731-1738. 4 Arieo M , Baruchel A, Bertrand Y, Biondi A, Conter V, Eden T el al. The seventh international childhood acute Iymphoblastie leukemia workshop report: Palermo, Ital y, January 29-30, 2005. Leukemia 2005; 19: 1145-11 52. 5 Pui CH, Relling MV, Sandlund JT, Oowning IR, Campana O, Evans WE. Rationale and dl!Sign of TOI"al Th ~ rapy Study XV for ncwly di agnosed childhood acute I)'mphoblastic leukemia. Ann Hemalo/2004; 83 (Suppl 1): SI24-S126. b Schultz KR, Pullen Di, Sather HN, ShuSler Jj, Oevidas M, Borowitz MJ el al. Risk- and response-ba ed classifi CJtion of childhood B:preeursor acute Iymphoblasti c leukemia : a combined anil lysis ol prognostic markers from the P(!diatric Oncology Group (PO ) and Children 's Can ce r Group (CCG). Blood 2007; 109: 926-935. i Zhou L Goldwasser MA, Li A, Oahlberg SE, Neuberg O, Wang H el al. Quantitativc analysis of minimal residual disease prcdicts relapse in childrcn with B-lineage aeute Iymphoblastie leukemia in OFCI ALL Consortium Protoco l 95 -01. Blood 2007; 110: 1607-1611. B Stanulla M, Schaeffeler E, Flohr T, Cario G, Schrauder A, Zimmermann M el al. Thiopurine methyltransferase (TPMn genOlype and early treatOlent response to Olercaptopurine in childhood acute Iymphoblasti c leukemia. )AMA 2005; 293: 1485-1489 9 van der Velrll 'n VH, Pan zer-Grumayer ER, Cazzaniga G, Flohr T, Sullon R, Schr ..1Uder A et J I. Optimization of PCR -based minimal residual disl'ase diagnosti cs for childhood acute Iymphoblastic leukemia in a multi-centcr setting. Leukemia 2007; 21 : 706-713. 10 Lauten M, Zimmermann M, Reiter A, Beier R, Gadner H, Niemeycr C el al. Bone marrow da y 15 has an additional impact on the prediction of event free survival in children w ith acute Iymphoblastic leukemia characterizcd by th e prednisone response . 131000 2002; 100 (11): 69a, abstract r250 1. I N,1Chman lB, Sather HN, Sensel MG, Trigg ME, Cherlow JM, Lukens IN cl al. Augmented post-induction therapy for children with high-risk acute Iymphoblastic Icukemia and a slow response to initial therapy. N Engl) Med 1998; 338: 1663- 1671 . 12 Schrappe M , Rciter A, Zimmermann M, Harbott J, ludwig WO, Henze G el al. Long-term results of four consecutive trials in childhood ALL performed by th e ALL-BFM study group from /981 to 1995. Berlin-Frankfurt-Munster. Leukemia 2000; 14: 2205-2222. II Bene MC, Castoldi G, Knapp W, Ludwi g WO, Matutes E, Orfao A el al. Proposals for the immunological class ilication of acute If'ukemias. European Group for the Immunologi cal Charac terization cf leukemias (ECll). Leukemia 1995; 9: 1783 - 1786. 14 Hiddemann W, Wormann B, Ritter J, Thiel E, Gohde W, Lahme B el al. Freq uency and clinical significance of DNA dneuploidy in ucut ' leukem ia. Ann N Y Acad Sci 1986; 468: 227-240. 15 Pongers-Willemse MJ, Seriu T, Stol z F, d'Aniello E, Gameiro P, Pisa P el al. Primers and protocols for standardizcd dctection of minimal res idual disease in dcutc Iymphoblastic leukemia using immunoglobul in and T cell receptor gene rearrangements and TAL1 deletions as PCR targets: report of the BIOMEO-l CONCE RTE O ACTION: investi ga tion of minimal residual disease in acute leukemia. Leukemi.1 1999; 13: 110-118 lb Szczepanski T, Pongers Willemse MJ, Langerak AW, Harts WA, Wijkhuijs AJ, v an Wering ER el al. 19 heavy chain gene rearrangements in T-cell acute Iymphoblasti c Icukemia exhibit predominant OH6-19 and OH7 -27 gene usage, can result in
17
e
18
19
20
21
22
23
24
25
26
27
28
29
30
31
comple\e V-O-I rearrangements, and are rare in T-cel\ receptor alpha beta \in eage. S/ood 1999; 93: 4079-4085. van Oongen Jl, Langerak AW, Bruggemann M, Evans PA, Hummel M, Lav'nder FL el al. Design and standardization of PCR primcrs and protocols for detection of clona I immunoglobulin and T-cell receptor gene recombinations in suspect Iymphoproliferations: report of the BIOMED-2 Concerted Action BMH4-CT98-393 6. Leukemia 200.1; 17: 2257- 2317 Langerak AW, Wol vers-Tettero lL, va n Gastel-Mol EJ, Oud ME, van Oongen Jl. Basic heli x-Ioop-helix proteins E2A and HEB induce immature T-ce ll receptor rearrangements in nonlymphoid rI ' lI s. Blood 2001 ; 98: 2456- 2465. Veln der Vel den VH, Wijkhuijs JM, Jacobs DC, van Wering ER, van Oongen JJ. T cell receptor gamma gene rearrangements as targets for detection of minimal residua I disease in acute Iymphobl as tic leukemia by real-time quantitative PCR analysis. Leukcmia 2002; 16: 1372-1380. van der Velden VH, Willemse MJ, van der Schoot CE, Hahlen K, va n Wering ER, va n Oongen JJ. Immunoglobulin kappa deleting element rearrangements in precursor-B acute Iymphoblastic leukemia are stable targcts for dctedion of minimal residua l disease by real-time quantitati vc PCR. Leukemia 2002; 16: 92 8-936. Verhagen OJ, Willemse MJ. Breunis WB, Wijkhuijs AI, Jaeobs DC, Joosten 5A el al. Application of germline IGH probes in rea l-time quantitative PCR for th" detection or minimal residual disease in acute Iymph ob lastic leukemia. Leukemia 2000; 14: 1426-14 35. Pongers Willemse MJ. Verhagen OJ, Tibbe GI, Wijkhuijs Ai, de Haas V, Roovers E el al. Real-time quantitative PCR for the detection of minimal residual disease in acute Iymphoblastic leukemia using junctional region specific TaqMan probes. Leukemia 1998; 12: 2006-2014. va n der Velden VH, Ca zza niga G, Sehrauder A, Hancock J, Bader P, Panzer-Grumayer ER el al. Analysis of minimal resi dua I disease by IgffCR gene rearrangements : guidelines for interpretation of real-time quantitative PCR data. Leukemia 2007; 21: 604-611. Fronkova , Madzo i, Zuna J, Reznicko va l, Muzikova K, Hrusak O lOl al. TEUAML 1 real-time quantitative reverse tran scriptase PCR can complement minimal residua I disease assessment in childhood ALL. Leukemia 2005; 19: 1296-1297. Willemse MJ, Seriu T, Hettinger K, d'Aniello E, Hop WC, Panzer Grumayer ER el al. Oetection of minimal residual disease identifies differences in treatment response between T-ALL and precursor B-ALL. Blood 2002; 99 : 4386-4393 Panzer Grumayer ER, Schneider M, Panzer S, Fasching K, Gad ner H. Rapid molecular response during early indu ction chemotherapy predicts a good outcome in childhood acute Iymphoblastic leukemia. Bload 2000; 95: 790-794. Basso G, Gaipa G, Valsecchi MG, Veltroni M, Oworczak M , Ratei R el al. Early evaluation of bone marrow minimal res idual di scase by flowcytometry on da y 15 is feasible on a multiccntt'r basis and bears strong prognostic valu ' in childhood acutl' Iymphoblastic leukemia. The AIEOP-BFM experience. Blood 2007; 110 (11): 426a, abstr'lCt [1423J. Mejstrikova E, Fronkova E, Batinic O, Oubravcic K, Kiss F, Kappelmayer I el al. Standardized 4 color flow cytometric minimal residual disease detection failed to overcome regenerJtion problems but identiiies early blast clearence predi clive af molecul ar remi ss ion after induction in childhood B lineagc leukemia. Blood 2006; 108 (11): 521 a-522a, abstract [1842J. Borowitz MJ, Pullen Oj, Shuster Jj, Viswanatha O, Montgoml'ry K, Willman CL el al. Minimal residual disease detection in childhood precursor-B-cell acute Iymphoblasti c leukemia: relation to ather risk factors . A Children's Oncology Group study. Leukemia 2003; 17: 1566-1572. Madzo L Zuna J, Muzikova K, Kalinova M, Krejci O, Hrusak O el al. Slower molecular response to treatment predicts poor outcome in patients with TEUAML 1 positive acut I)'mphobl astic leukemia: prospective rcal-time quantil'ative reverse transcriptasepolymerase chain reaction study. Cancer 2003; 97: 105-11 3. Trueworth y R, Shuster J, Look T, Crist W, Borowitz M, Carroll A el al. Ploidy or Iymphoblasts is the strongest predictor of treatment outcome in B-progenitor cell acute Iymphoblastic leukemi a of childhood: a Pediatric Oncology Group study.) Ctin Oncol 1992; 10: 606-613.
Leukemia
E. Mejstříková, strana 126
P říloha
8
CD44 and CD27 delineate B-precursor stages with different recombination status and with an uneven distribution in nonmalignant and malignant hematopoiesis M. Vaskova, E. Frankova, 1. Starkova, T. Kalina, E. Mejstrikova & O. Hrusak
Tissue Antigens, 2007, (fF 2,245)
E. Mejstříková, strana 127
Tissue Antigens ISSN 0001·2815
CD44 and CD27 delineate B-precursor stages with different recombination status and with an uneven distribution in nonmalignant and malignant hematopoiesis M. Vaskova 1,2, E. Fronkova 1,2, J. Starkova 1,2, T. Kalina 1,2, E. Mejstrikova 1,2 & O. Hrusak 1,2 1 Department ol Pediatric Hematology and Oncology, 2nd Medlcal Sc hool. Charles Univers',ty Prague, Prague, Czech Republic 2 Ch ildhood Leukemia Invesllgalion Prague. Prague, Czech Republic
Key words aCute Iymphobla stic leukem ia; B·cell development; CD2 7; CD44 Correspondence
Martina Vas kova Childhood Leukemia Investiga tion Prague V Uvalu B4 150 06 Prague Czec h Republic Tel: + 4202244364 77 Fax: +420 2 2443 64 13 e-mail: martina.vaskova@IImotol .cu ni.cz Recelved 22 March 2007; revlsed 14 August 200 7. 10 October 2007 ; accepted 21 October 2007 dOl 10.111 1/j.1399-0039.200 7 00968 x
Abstract
The expression ofC027 and C044 correlale with lhe genotype of B-precursor acule Iymphoblastic leukemia (ALL). Based on the expression of these antigens, we g identified counterpar ts of TEL/ AML/poS and TEL/AML/"e leukemic cell s in nonmalignant bone marrow. Although C027 is known as a marker of malure memory B cells, we recenlly showed thal C027 is also expressed by malignant and nonmalignant B precursors. Here, we show that C027 and C044 delineate stages of B-precursor development. Well-established dífferentiation markers showed thal Ihe developmental sequence starts from undetermined progenitors, expressing C044 . Upon B-lineage commitment, cell s gain C027 and lose C044. The C027po'C044neg (C027 single positive , 27SP) cells are the ea rliest stage within COIOPosCOl9 PoS B precursors and express RAC-I and TDT. These cells correspond to TEL/AML/pn s ALL (1/4 pediatric B-precursor ALL). The developmenl follows to C027/C044 double-positive (27/440P) stage, 44SP stage and C027/ C044 double-negative (27/440N) stage. Before exit to periphery, C044 is reexpressed. The 27/440P cells are mostly large and profoundly suppress RAG-I. Oespite their presumably high proliferation potential, 27 /440P cells rarely g dominate in leukemia. At 44SP stage, which corresponds to TEL/AML/ne leukemias, RAC-I is reexpressed and 19 light chain gene Slarts 10 be rearranged.
Introduction
By making the lineage decision, human progenitorcells start the B-lineage dcvelopment in the bone marrow (BM) [reviewed in (I , 2»). This early stage is accompanied by the upregulation of B-cell-specific lranscriptional regulators Pax-5 and very likely also early B-cell factor. Since the earliesl stage, the B-lineage-committed cells start to rearrange thc genes for 19 heavy chain (lCH). Upon successful ICH rearrangement, the recombination machinery is suppressed and cells quickly proliferate. In the following stage, 19 light-chain (lCL genes are rearranged. Cells that com plete both ICH and ICL are ready to be functionally competent and develop into na)'ve mature B cells. Such cells leave Ihe BM environment. On cell surface level , the first B-committed cells express C034, which disappears in the course of maturation. Very
CI 200 7 The Au Ihor Joumal compll Ilon ':> 2007 Black well
M un~ s ga ard
ear1y after B-lineage decision , COlO is expressed and continues to be present on the cell surface untillCH and ICL are rearranged . An importa nt surface marker of B lineage, COI9, is expressed since the very early stage when ICH rearra nges. Experimental evid ence shows that CD I O precedes the expression ofC019, perhaps in majority ofcells. The CD IOposC034po,CO19 neg cells are mostly committed to B lineage, although they can develop also to other lineages (3). However, the most immature B-precursor leukemia stage called pro-B is defined as COI9po'C0341""COLOneg (4, 5). Leukemia nomenclature, ref1ecting the current understanding of malignant transformation during Iymphopoiesis, lherefore assumes CD 19 expression before COlO. At COlOposCOl9PoS stages of developmenl, important events occur including ICH rearrangemcnt completion, cell expansion ancl ICL rearrangement. This
E.
Mejstříková,
strana 128
C044 and C027 delineate B-precursor development
CD IOpo'CD 19poS immunophenotype is found in the majority of B-cell-precursor acute lymphoblastic leukemias (BC P AL Ls). Our previous study showed that major genotypic BCr ALL subsets strictly correlate with the surface C027 and C044 expression (6) (updated results are in http: //clip. If2.cuni .cz/ files /archiv /C044_and _ C027 _ updated _results. pdf), thus confirming RNA genomics data on protein leve! (7). [nterestingly, TEL/AMU!"" BCP ALL corresponds to poS C027 phenotype, in the absence of C044. Although C027 was considered to be a general marker for memory B cells in humans, we showed its expression also on COIOPo'COl9 Po S B-cell precursors (6). [ts expression in BCP ALL and in nonmalignant B precursors was then confirmed in another study (8). C027 (TNFRSF7, T-cell activation antigen) is a II O-kOa transmembrane glycoprotein composed of disulfide -linked 55-kOa monomers, and it is a member of the tumor necrosis factor receptor fami ly. In humans, it is expressed on the large subset of peripheral T cells, on most medullary thymocytes (9) and on natural killer cells (10). [n more mature stages of B-lineage development, it is expressed on somatically mutated B cells(ll). C044 (Hermes, Pgp-I) is a cell surface glycoprotein coded by 10 standard exons found in all C044 isoforms; the other 10 are variably spliced (12). C044 was originally identified on hematopoietic cells (13), but it was also found on a wide range of olher tissues (I 16). The C044 medíates cellular adhesíon . It is a receptor for hyaluronate (17), but il binds other compounds of extracellular matrix also. II is involved in the process of Iymphocyte activation (18), and its inleraction with stromal cell s is important during Iymphopoiesis and myeolopoiesís (19, 20). The C044v6 splice variant is often associated with mdast a tic potential of nonhematopoietic neoplasias. for cxample, in gastric adenocarcinoma (21) . lts exprcssion also correlates wíth shorter survival of patients with
2
M. Vas kova et al.
Materials and methods Patients
BM specimens of children without any evidence of malignant or residua I malignant disease (patients I and more years after BM transplant, after the end or the ALL therapy and palients investigated to exclude hematologieal malignancy) were used. Spccimens were collected according to the Czech law and institlltional regulations and with informed consenl. Only leftover material from specimens after completed diagnostic investigations was used.
Flow cytometry
Cell suspension of the unseparated BM or peripheral blood (PB) was stained with four- to eight-color combina tions of monoclonal antibodies (mAbs). The following tluorochrome-Iabeled mAbs were used: anti-C044 fluorescein isothiocyanate (FITC) (reacting with the standard isoform of C044) and anti-C027 phycoerythrin (PE) (BO, San Jose, CA) , anti-CO 10 ECO, anti-COl79a PC5, anti-CO 19 PC7 and anti-C034 allophycocyanin (APC) (Immunotech, Marseille, France) , antí-TdT FITC (Oako , Glostrllp, Oenmark), anti-C020 Alexa405 (Serotec, Kidlington, Oxford, UK) and anti-lgM Oyomics647 (Exbio, Prague, Czech Republic). In addition , OAPI (4,6-díamidino-2phenylindole) (Molecular Probes, Leíden , The Netherlands) and FIX&PERM cel.1 permeabilízation kit (An der Grub, Vienna , Austria) were used. Nonmalignant BM samples were analyzed using FACS Aria (BO) and BO LSRII (BO) tlow cytometers and sorted using FACS Aria tlow cytometer (BO). For polychromatic tlow cytometry experiments, photomultíplier (PMT) voltage was set above electronic noise threshold and automated compensation matrix calculation was performed using single-colors tained tubes (DIVA 4.1.2 or SUMMIT 4.3). Gating strategy of compensated data was determined using Fluorescence Minus One controls (24). Analysis was performed using FLOWJO 8.1.1 software (Treestar, Ashland, ORl using the same strategy of positivity determination. For each sample, 4 x 10 5 to 4 x 10 6 events were recorded. Gating strategy was used for the analysis and sorting of the cells with a given immunophenotype as follows: CO I Opos cells po were selected from the gate of CO 19 'OAPl nc g cells (live B-lineage ceJls). Four subpopulations based on <..'044 and CD27 positivity and negativity were identified according to Fluorescence Minus One conlrols among these cells (Figure I) . The cells falling above or below a threshold of fluorescence intensity set by these controls are called posi/ive or nega/ive subpopula/ions throughout the paper. The experiments with sorted subpopulations were performed in duplicates or triplicates. The purity of sorted subpopulations was always more than 95%. Thcrefore, the
© 2007 The Auth ors Journal co mp'lation 10 2007 Blackwell Munksgaard
E.
Mejstříková,
strana 129
M . Ifaskova et al.
C044 and C027 delineate B-precurs o r development
A
.000
lOoO U
(f)
'o'
'o'
,a'
4 10
o
o ,a' Ol
1000
lL
S'OJ
U 1000
U
,o'
,r:? D
cells we re sel ecled Irom the gate of C0 19""· OAPlnog cells (I .ve B-lineage ce lls) (B) C02 7 and C044 ex pression in COl OOO' C019 Po. cells in BM. in C019"""C0 20PO' C010 noo cells in BM and in C020 00 , cell s in PB ol differen t pal.ents. BM. bone marrow; OAPI. 4. 6-diamidino-2-phenyli ndole; FSC. l orward scaller ; PB. peripheral blood; SSC. s,da scaller.
B
I"('oj
O
4000
o
lOl
,a' 'o' 'o'
SSC
OAPI
C010+
C020+C010-
o
tO'
102
tO'
10"
C010 C020+ in PB
,o'
tO'
,a'
ta'
ta'
,a'
,a'
,a'
,a'
,0'
,0'
to'
()
o 10 2
results of the mRNA and DNA analyses indeed describe the sorted subpopulations.
Real-time quantitative reverse transcriptionpolymerase chain reaction
Complete RNA from sorted cells was isolated by RNeasy Micro Kit (Qiagen GmBH , Hilden, Germany) according to manufactllrer's instructions. RNA was converted into complementary DNA (cDNA) using MoM LV Reverse Transcriptase (Gibco BRL, Carlsbad, TX) . Real-time quantitative reverse transcription- polymerase chain reaction (PCR) analyses were performed lIsing LightCycler™ rapid thermal cycler system (Roche Diagnostics GmbH, Mannheim , Germany). Fluorescent DNA-binding dye SYBR Green (FMC BioProducts, Rockland , MA) was used for quantification of R A G- 1 and TDT gene expression. Control gene beta-2-microglobulin (b2m) was measured using hybridization probes as described previously (25). The primer sequences were as follows - for TDT: forward 5'-gTCgTgCCTTTgCCCTgTI-3 ' , reverse 5' -TCCgCTCATgTgTggCATAg-3' and for RAG-/: forward 5' -TgAgTAATATCAACCAAATTgCAgACA-3 ' , reverse 5' -ggATCTCACCCggAACAgC-3'. The composition of PCR mix was as follows: I U of Platinum Taq DNA polymerase in the PCR buffer provided by the manufacturer (Gibco BRL - Life Technologics Inc., Gaithersburg, MO), MgC I2 3 mM (for RAG-/) or 2 mM (for TDT), deoxynucleotide triphosphate 0.2 mM each, bovine serum albumin 0.25 (lg/(ll, primers 0.5 (lM eaeh; fluorescent signal was generated lIsing 0.2 (ll SYBR Grecn (2 x 10- 4 ofthe stock concentration, dillltcd by dimcthyl sulfoxide; for RAG-I
@
lOoo
1000
'000
Figure 1 C027 and C044 expressi on in nonmal,gnant BM and PB . Gating s Iralegy used in seven-color staining ol one BM spec,men lor analysis ol lhe differentiation markers in parl ,cular subpopulations and lor the ir sorting (A) COl Ope',
'o'
'o· 'a'
a ,o'
,oJ
to'
'o'
o ,o'
,o'
,o'
,o'
C044
and TDT). One microliter of cDNA was added in a final volume of20 (ll. The LlGHTCYCLER program for RAG-I and TDT consisted of the initial denaturing at 94°C for 90 s, followed by 40 PCR cycles: 95°C for 5 s, 62°C for 40 s (single fluorescence measurement) , noc for 12 s. The melting curve analysis was performed to confirm specific amplification and to identify nonspecific templates after each run. RAG- I a nd TdT mRNA expression was shown as a ratio to b2m expression; this va llle was normalized to 1 in the most immature subset investigated.
Real-time quantitative peR analysis of immunoglobulin gene rearrangements
DNA from sorted cells was isolated by QIAamp DNA Blood Micro Kit (Qiagen GmbH). Multiplex real-time quantitative PCR (RQ-PCR) for lGH detecting virtually all com plete lGH rearrangements and RQ-PCR for intron recombin a tion signal seguence to kappa deleting element ( RSS-Kde) was performed using family-specific V segment forward primers and J segment-specific reverse primers and probes as described previollsly (26, 27) in the iCycler IQTM therma l cycler system (Bio-Rad, Hercules, CA). The starting concentration of template was measured against the dilution series of positive-control DNA in germ-line (unrearranged) Hela DNA. REH cell line served as a positive control for intron RSS-Kde recombination . Samples of patients with ALL containing 87%, and 97% of clona l cells with monoallelic rearrangemcnts were used for standard curve preparation in VHI-3- JHI-ó and VH47-JH 1-6 assays, respectively. The cycling conditions wcrc as follows: initial denaturing at 95°C for 10 min , 50 cyclcs of
2007 l he AutholS
Jo urnal complla{,on II:J 2007 BI . ckwell Mun ksgaard
3
E.
Mejstříková,
strana 130
C044 and C027 deline ate B-precurso r development
M. Vaskova el al.
denaturing at 94°C for 15 s, annealing extension at 64°C (lCH) or 62°C (intron RSS-Kde) for I min; DNA at a concentration of 0.01-0.2 ~g/ J.lI was used for cach PCR reaction. AlI the assays reach ed sens itivity of at lea st 1% of rearranged a lleles in the germ-line background. Each sample was run in duplicate, and a mean value was used for further analysis. Albumin gene was used to normalize the D A concentration and quality (28) . IC YCLE R IQ Optical System software version 3.0a was used for quantjfícation, and a fínal value was shown as a percentage of rearranged a lleles relative to the respective clonal control DNA , which was set as 100%.
Results
C027 and C044 expression define phenotypic stages in B precursor development
As CD44 negativity in combination with CD27 positivity is found exclusively in one sub type of leukemia (TEL/ AMU PO') (6), we searc hed for such cells among B precursors in nonmalignant BM . Within CD IO po'CDI9 p o , cells, such 27SP cells were present in all 14 specimens. In addition , 44SP cells as well as 27/44DP and 27/44DN cells were found (Fi gu re 2). Next, markers of B-precursor differentiation were investigated by polychromatic flow cytometry . The expression of CD44, C D27 and a B precursor defíning the combinalion of C D19 and C D IO was studied together with the differenti ati on mark ers C D34, TdT, cytoplasmic IgM and cytoplasmic VpreB (CD 179a). The percentage of CD34 poS cells is the highest in 27SP and decreases gradually in 27/44DP, 44SP and 27/44DN su bpopulations (Figure 3A,B). A similar lrend is found in the percentage of CD lO b",ghl cells , which become virtually missing in CD27 neg B-precurso r stages (Figure 3C,D). This sequence of developmental stages was further supported by
90 80 ~
~ o
70
~ 60
O
Ll C
~ 50
$
cr
~
ID
40
'"
30
<JI
20
.o ::J
O 10
O
o 27/44DP
44SP
27/44DN
Figure 2 Frequencies of subpopula tions de fined by C027 and CD44 expression w ithin CDlO""" CD19PQo B precurso rs. Fourteen spec lmens were used for this analysis. DN, double negative; OP. double positive; SP, single positive .
4
27/440P cells downregulate the key recombination enzymes
We sorted subpopulations ba sed on the CD44 and CD27 expression to compare th eir recombination potential by measuring TdT a nd RAG-I m RNA expression by RQPCR . Similar to the protein level, TdT mRNA expression decreases in concordance with the suggested developmental stages (Figure 5A) . Interestingl y, the 27/44DP cells express the lowest amount of RAG-l transcr ipt s (Figure 58), suggesting that these cell s are in the stage of suppressed RAG - I expression after completed ICH rearrangement. RAG-I is reexpressed during IC L rearrangement, as proven by the reappearance in the 44SP cells. Because the cells with a downregulated RAC-I are known to be frequent among the large proliferating cells, we ana lyzed the percentage of large cells (eslimated by cytometry as the proportion of cells with a higher forward scatter). The 27/ 44DP subpopulation appears to be composed mostly by the large cells, based on the highest percentage of cells with high fonvard scatter (Figure 6A,B). Immunoglobulin gene rearrangement
After a successful rea rrangement of ICH genes in early B precursors, cells proliferate and ICL genes start lo rearrange. In all four subpopula tions, heavy-chain genes (both segme nts VHI-3- JH and VH4-7-JH) were rearranged (data not shown) , proving tha l heavy chains start to rearrange at or before lhe earliest stage oťdifferentiation among lhe analyzed subpopulations. The low quantilalive range ofthe system for VHI-3- JHand VH4-7-JH detection did not enable theexact quantifícation, ma inly because of a limited DNA concentration obtained from so rted cells. Nexl, we investigated the ICL rearrangement. The sys tem detected the intron RSS-Kde rearrangements, whích appear in the late phase of ICL rearrangement. As shown in Figure 7.ICL genes begin to be rearranged al the 44SP stage, whereas intron RSS-Kde rearrangements are virtually missing al earlier stages. Both C027 and C044 reappear at C020poSC010neg stages
O
27SP
a gradualloss ofintracellularTdTand VpreB (Figure 4A,B) and by the increase ofintracellular [gM p ns cells (Figurc 4C). The observed sequence of developmental stagcs o l' 8 precursors is thus: 27SP, 27/44DP, 44SP an d 27í44DN. We fit these su bpopulations into two well-cslablished models of B-cell development (Table I) (2,29).
As reported earlier, mature PB B Iymphocytes are CD44poS (15) . Our data show that majority of CD20 P o' CD I Oneg cells express CD44 already in BM, while PB B Iymphocytcs are almost exclusively CD44 Po, (Figure lB), contrasting with BM CDIO P o, cells, which contain a grealcr fraction of
co 2007 The Authors Journal compllat,on © 2007 Blackwell Munksgaard
E. Mejstříková, strana 131
M . Vaskova er al.
C044 and C027 de line ate B-precursor development
A
B
~
100
4í
80
.o :=J
4i
60
.!!! Qi
40
~
20
a.
~o
9
(J
27/44DP 44SP
o
C")
o
U
27SP
o
27/44DN
o
27SP 27/44DP 44SP 27/44DN
102
103
10<
105
CD34
C
~ Figure 3 Express ion of C034 and COlO w ilhin C027- and CD44-defined subse l s. Tha percenlage of CD 34 ""0 cells (A) and C034 fluore scence inlensity (B). The percenlage of C01 0b<>gh, cells
O 70
4í
60
::>
50
.o
27SP
4i
40
.!!! Qi
30
fo
20
a.
(C) and COlO fluorescence inlensi ty (O) Fourtean
27/44DP
(J
spec im ens w ere used for thiS analysls. The verlica l lines dlslll1gUlSh CD34""" and C034 "0
~
O U
44SP
10
27/440N
O
27SP
OP, double positive ; SP, single posltlve; TdT, lerm inal deoxyribonuc leo lidyllransferase.
0102
27/440P 44SP 27/440N
1()3
10<
105
CD10
CD44 ncg cells. Thus, immature, na'ive B cells reexpress C D44 in BM , before they exit to periphery . lack of C027 expression in stem cell s
Within CD34 P o, cells , Nilsson et al. reported higher proportion of CD27 PoS cel ls in CDI9 ncg subpopula ti on compa red with CD 19''''' subpopul ation , which was in terpreted that B-lineage-committ ed C D 34PoS cells express !ess CD27 (8). Althou gh we used flu o rochrome (PE) with hi gher fluorescence intensity a nd a higher resolution than FITC used in Nilsson 's stud y, we found only 1.9 ± 1.7 % of CD27 po, cells (n = 14) amon g CDI9negC D34pos cells.
As we used a reliable cytometric system acquiring millio ns of events a nd selecting exclu sively viable cells for lhe a na lys is, we could further divide these CD 19ncgCD34pos cells . We used CD 10 for finer division of the CD 19negCD34pos ce lls because it is known that the C D 19neg C D I Opo' CD34 pn , cel ls are biased towa rd s B line age, a lthough they ca n develop also into o ther lineages (30, 31). The CD IO ncg and C D IO poS fra ctions of C D 19ncgC D34pm ce lls contained 0.23 ± 0.21 % and 38.8 ± 26.3% C D27 poS cells (n = 14), respectively. ln com parison, CD 19po'C D34Pos cell s contai n 67 ± 14.1 % C D27 poS cells (11 = 14). ln CD34posCDI9neg su bpopulation , the acqu isition of CD IO is accompa nied by C D44
A
c
B
Figure 4 The I<1 le nsit ies of inl racell ular TdT (AJ,
27SP
27SP
271440P
27/440P
27/44DN
27/44DN
intracell ula r C0 179a (B) and intracellular IgM (C) w ll hlo C027- and C044 -d eflf1 ed subsels. Represen lalive sam pl es of five m easuremenls are shown The verllca l hnes d, s ~ng uish pOSll lve and negatlve cells . ON , dou ble nsgat ive; OP, dou ble posilive; SP, single pos lt ,ve; TdT, lermina l deoxy-
o 102 103
ribonucleo lidyl transferase.
intracellular T dT
© 200 7 The Aulh ors Journa' c o mp'~"on ll!) 2007 Blackwell Munksgaard
C027°09
104 105
010 2 1()3 104
10 ~
intracellular CD179a
O 102 103 10' ID' intracellular 19M
5
E.
Mejstříková,
strana 132
C044 and C027 delineate B-preeu rso r development
M. Vaskova er a/.
Table 1 The nl odels o f B-e II oevelopmen t" LeBien (2) Pro·B (I) Pre-B I (II)
CD19, CD19, CD 19, CD19,
Pre-BI I (II) Immature B (lili (IV)
CD10, CO lO, CD10, CD10,
CD34, CD24, IL-7R, RAG, Vpre B, Ig-(X, VDJ H CD24, pre-B rec eptor, low RAG, IgM HC CD24 , pre-B recepto r, RAG, IgM HC, VJ K CD20, CD21, CD22, CD24, CD40, 19M
Loken et al. (291
CD2 7 and CD44
CDl ot"uh, C034, TdT
27SP 27/44DP 44SP. 27/44DN 27/44DN
CO lO CD1 0+ , lgM IgM
DN, double negative; DP, double posltlve; HC, heavy ehain; SP, single posi tive; TdT, termina l deoxyribonueleotidyl transferase We show CD27- and CD44-defined diffe rentiation stages in lhe cont ext of two publlshed models. The stages in the model by Lo ken et al. (291 and LeBlen (21 are assigned by num erals and the" names, respectively, in the first column.
a
throughout the differentiation. The suppression of RAG-I may be important during the cell proliferation as RAG-l (unlike TdT) ma y cause unwanted DNA recombination during replication. This may explain the differences between TdTand RAG-I expression a t the 27/440P stage, cOnlaining la rge proliferating cells. The knowledge on B-p recurso r development can be combined with the leukemia di agnos tics to discover the cou nterparts of the domin an t leukemia population . The 27SP cell s a re physiological cou nterpa rts of TEL/ A M LlPoS A LL, and the 44SP cells are the principaJ counterparts of aH other B-precursor ALL subtypes. The 44/270P and 44/ 270N imm unophenotype is rarely seen among B-precursor leukemias. This contradicts a logica l expectation th at the proliferating compartment (large pre-B, mostly found in 44( 270P stage) might be more likely to transform into malignancy. Because the nonmalignant counterparts of TEL/ A M LIPO' ALL are less mature than the counterparts of TEL/AM LI" og AL L, it is possible to speculate that the differentiation of TEL/ AMLI PO' leukemic cells is blocked in earlier stage than in other leukemias. After exit to periphery, a subset of B cells reexpresses CD27 during the process ofsomatic hypermutation and it is then constitutively present as a marker ofmemory cells (32). ln line with this, we observed a C027 Po , subpopulation amon gC020posCOlOncgcell s not only in PB butalso in BM. The BM C027po'C020PosCO IOnc g B cells are probably recirculating memory cells. Migration of memory cells to BM was proven by the analysis of BM cells with preferential
downregulation (Figure 8), which is in line with our results that C044 negativ it y together with C 027 positivity is found in the earliest CO 19 poS cells.
Discussion
We previously reported tha t C027 and C 044 define ALL subty pes (6). Thc data presenled here show that the expression of these molecules correlates nonrandomly with several independent markcrs of B-precursor differentiation. Among CD I Opo'CO 19 ''''' cdb.. both maturity of lG rearrangements and the expression of differentiation markers prove th e following developmental sequence: 27SP, 27/440P, 44SP and 27/440N. The data showing heavy chain gene rearrangements in 27SP, 27/440P, 44SP as well as 27/440N cdls indicate that IGH rearrange at stages up to 27SP. Upon transition to the 27 /440P stage, in which the large cells dominate, RAG-I exp ression is suppressed, Afterwards, C027 disappears and cells enler the next stage, 44SP. This stagc is characte rized by res tarting the recombination mach inery (shown by RAG-I reexpression). Only at this stage, cells start to contain completed lG L rearrangements (shown by KDE detection). Cells appare ntly attain smaller size at 44SP stage. This process is fini s hed by loss of CD 44. The resulting 27/440N phenotype dominatcs among the C D JOPos B precursors in human BM. The 27 /440N cells complete th e differentiation proccss before the cells lose CD10, regain C044 and are allowed to exit to periphery. Interestingly, whereas RAG-I is profoundly supprcssed at 27/440P stage, TdT expression decreases gradually
A
E
TdT
1.2
B
N
e
-f2 0.6 I-
~
0
"(J
6
0.6
OA
C2 OA
0.2
0.2
O
Figure 5 Recombinatlon enzyme terminal deoxy· ribonucleotidyl transferas e (AI and RAG-l (BI mRNA expression. Expression levels are shown
E 0.8
0.8
N
I-
RAG-1
1.2
27SP
27/44DP
44SP
27/44DN
O
as ratio to control gene expression; this valu e was normalized to 1 in the most immatu re subset investigated. Exper iments were perfor med in dupl icates . DN, doub le nega tive; DP, dou ble 27SP
27/44DP
44SP
27/44DN
posltlve; SP , sin gle positive; TdT, terminal deoxyribonucleotidyl transferase.
@ 2007 The Au \hors Jovrnal cornpolalion
E.
Mejstříková,
strana 133
M. Vas kova B I al.
CD44 and CD27 delineate B-precursor development
A
B 90 80 27SP
70
~ Figure 6 Frequencies of large ceUs dehn ed as the subpopulation wi th the high FSC (A) and representatlve sa mple of FS C intensity. Fourteen specimen s wer e used for this ana lysis. DN, double negative; DP. double positive; SP, single po sitive ON. double nega ti ve; DP. double positive; FSC, forw ard scalter: SP. single posl tive.
60
'á
27/440P
50 'E 40
~
li.
44SP
30 20 10
o
27/440N
o
O~~~~----------~~~~
27SP
mut a tions in the complementarity dct.:rmining regions (CDR) (33) of th e VeD)] rcgions, which is the pattern that evolves during thc antigen sclcction in gcrminal centers. Interaction of CD27 with its ligand CD70 expressed on activated T and B cclls regulatcs B-cell growth and diffe rentiation to IgG- and IgM-producing cells (34, 35). At the early developmental slages, CD27 expression was reporte:d on munnc hematopoietic stem cells (36), and Nilsson et al. reported a higb C D 27 expression in human CD34 Po> cells even bcfore C D 19, similar to mice (8) . Although we did obscrve sucb ceJls, lhey we re a!most alwa ys CD \Opos and their percentage increases only after Blineage commitment. Because th e C D27 and C D44 are known to play a role in apoptotic Ol' adhesion processes, it can be speculated tbat their expression merely ref1ec!s the functional status of the precursor B cell, regardless of its developmental stage. However, our d a ta clearly show that CD27 and CD44 are expressed in an organized fa shion during the precursor development. Consequently, this
intron RSS-Kde
140
o
120
27/44DP
44SP
27/440N
1000 2000 3000 4000 FSC
forms a basis for a further research into functional similarities between norma! precursor and the Ieukemia cell. A!though the straightforward explanation of the presented data is that the observed subpopulation s ref1ec! consequent developmental stages, alternative scenarios may be provided . We may hypothesize lhat the foJlowing phenomena may not represent regular features in Bprecursor differentiation a t CD \OposCD 19 poS stage: (i) CD27 expression , (ii) the decrease in CD44 exprcssion or (iii) the reappearance of CD44 secn in 27 /44DP and 44SP cells. The scenario (i) is definitely possible in lransgenic a nimals because B cells do develop in CD27 - I - mice (37). However, our data do not support this scenario as a major pathway in healthy humans. The CDIOP
100
C
ll>
E OJ
(J)
CD34+CD19-
80
C
~ Ol !l' "t'
CD34+CD19+
'o'
,o'
'o'
ro'
,o'
,rl
to:!'
,o'
60 o
oo
40 20
o
27SP
27/440P
44SP
27/44DN
o ,02
'o'
Figure 7 Immunoglobulin light-chain ge ne rearrangement.Th e percentage of intron RSS-Kde rearrang emen ts is show n. The experim ent was
perf or med ln triplicate . ON, double negative; DP . double positive: SP. Single positive.
C 2007 lha Authors Journal ~mplla\lOn C 2007 Blackwall Munksgaard
o ,o'
'o'
'o'
,rl
'o'
,o'
CD44 Figure 8 CD44
expression CD19posCD34P<>'cells.
in
CD1 0P<>sCD34posCD19neg
and
in
7
E. Mejstříková, strana 134
M. Vaskova et a/.
C044 and C027 delineate B-precursor development
scenario (ii) , the most immaLure cells would enter the 27 /44DP stage. Therefore, the cells would need to have OIOPo'CDI9 poS stage. completed ICH prior to the Analogically to the scenario (i), we would need to accept that some cells might stili lose CD44 and acquire the 27SP phenotype. Our obscrvation of high RAG-l expression at the 27SP cell s makes this scenario very unlikely as we would assume that the cells with completely rearranged ICH reexpress RAG-l upon losing CD44 before returning back to the RAG-l n"g 27/44DP phenotype. The scenario (iii) also appears unlikely as differentiating from 27SP directly to 27/ 44DN phenotype would circumvent the large pre-B stage with suppressed RAG-l. ff this scenario occurred in a significant fraction of cells, it would contradict the current understanding of B-precursor differentiation . A possible role or CD27 is the regulation of apoptosis because association with apoptotic cascades was reported. Data from different models illustrate that the net effect of CD27 may be either nega tive or positive. Nolte et al. showed that murine CD2T - progenitors proliferated more rapidly, suggesting a regulatory role of C027 in the growth of these cells. They propose that such a regulatory role may be triggered by the interaction of CD27 on B precursors with CD70 on activated T cells. This could be beneficial during infection as the presence of foreign antigen in the BM could induce an unwanted selective tolerance (37). CD27 triggering on primary plasma celll eukemia has antiapoptotic effect (38). Overexpression or CD27-binding protein Siva induces apoptosis in cell lines (39). lt was also shown that CD27 associates with tumo r necrosis factor-receptor-associated factor (TRAF)2 and T RAF5 signal transducers responsible for NF-KB activation (40). Apoptosis plays an important role during B-cell development when cells unsuccessfully rearranging 19 gcnes die by apoptosis. The question is whether CD27 on B precursors could mediate proapoptotic or antiapoptotic signal because the published data concerning CD27 role in apoptosis are diverse in different cell types. Although our data do not answer this question, it is obvious that among B prccursors, CD27 identifies mainly thc earliest stages at or before the start of lG rcarrangement. The specificity of the early-stage detection is strengthened when CD27 positivity is combined with CD44 negativity. Our data show that during B-cell development, CD44 undergoes two waves of downregulation. If the lIncommitp ted CD34 o S cells are most ly CD44 Po" then CD44 expression decreases together with the two-step acqllisition of CDlO (Figure 7). The CD44n,~ cells have been described previously among CD34 Po, BM cells and have been, in line with our results, shown as CDlOpo' CDI9poS B precursors ncg (41). These CD44 cells are 27SP. The following reemergence ofCD44, resulting in 27/44DP stage corresponds with one of its supposed functions, which is regulation of cell proliferation. While suppressing RAG-j expression, cells
8
proliferate at the 27/44DP. The expression of CD44 again ceases only Lwo stages down, at Lhe 27/44DN stage, in which ICL are fully rearranged and the cell s do not proliferate and are not dependent on CD44-mediated contact with stromal cells. The role of CD44 during hematopoiesis was experimentally shown by the addition of anti-CD44 mAb~ that inhibit or enhance stromal ceIJ-dependent hematopoiesis (42). There are more models of B-lymphocyte development, and the nomencla ture of B-cell developmental stages is stili unsettled. Any nomenclature should be based on fítting surface markers with lG gene rearrangement status. CD27 and C 044 extend the B-cell development model becallse their expression also renects not on ly the lG rearrangements status but also a different likelihood to transform into leukemia and / or to block the differentiation in different genetic subsets of A LL.
Acknowledgments
We thank the cytometric technician Pavel Semerak. The collaboration of alJ Czech Pediatrie Hematology centers [Ieaders: B. Blazek (Ostrava), Z. Cerna (Plzen), J. Hak (Hradec Kralove), Y . Jabali (Ceske Budejovice), V. Mihal (Olomouc), D. Prochazkova (Usti nad Labem), J . Stary (Praha) and J. SLerba (Brno)] is appreciated . This work was supported by grant nos GAUK 80/ 2004, IGA MZ CR NR / 8269-3, VZ MSMT MSM002l6208l3 and GACR 301 /06/ P162.
References I. Biom B, Spits H. Oevelopment of human lymphoid cel!s. Annu Rev Imnwno/2006: 24: 287-320. 2. LeBien TW. Fates of human B-cel! precursors. B/ood 2000: 96: 9-23. 3. Gal y A , Travis M, Cen O , Chen B. Human T, B, natural kilJer, and dendritic cells arise from a common bone marrow progenitor cell subset. lmmunily 1995: 3: 459··73. 4 Bene MC, Castoldi G, Knapp W et al. Proposals for the immunological classification of acute leukemias. European Group for the Immunological Characterization ol' Leukemias (EGIL). Leukemia 1995: 9: 1783-6. 5. Hrusak 0 , Porwit-MacOonald A. Antigen expression patlerns renecting genotype of acute leukemias. Review. Leukemio 2002: 16: 1233- 58. 6. Vaskova M, Mejstrikova E. Kalina Tet aJ. Transfer ol' genomics information to Oow cytometr y: exprcssion of C027 and C044 discriminates subt ypes of acute Iymphoblnstic leukemia. Leukemia 2005: 19: 876--8. 7. Yeoh EJ , Ross ME, ShurtleffSA et al. Classification, subtype discovery, and prediction of outcome in pediatri"
~ 2007 l he Authors Journal comp,lation " 2007 Blackwell Munksgaard
E. Mejstříková, strana 135
M. Vaskova el al.
9.
10.
II.
12.
13 .
14.
15.
16.
17.
18.
19.
20.
21.
22.
23 .
implication for diagnosis and therapy of childhood A LL. Exp Hema/O/2005 : 33: 1500-7. van Lier RA, Borst J, Vroom TM et al. Tissue dis tribution and biochemical and functional properties ofTp55 (CD27), a novel Tcell differentiation antigen. J Iml11unol1987: 139: 1589-96. Sugita K, Robertson MJ, Torimoto Y, Ritz J, Schlossman SF, Morimoto C. Participation of the CD27 antigen in the regulation of [L-2-~ctivated human natural killer cells. J Il11muno/1992: 149 1199-203. Klein U, Rajewsky K , Kuppers R. Human immunoglobulin (lg)M + 19D+ periphcral blood B cells expressing the CD27 cell surface antigen carry soma tically mutated variable region genes: CD27 as a general marker for somatically mutated (memory) B cells . J Ex p Med 1998 : 188: 1679- 89. Screaton GR, Bell MV , Jackson DG , Cornelis FB, Gerth U, Bell JL Genomic structure of DNA encoding the Iymphoc yte homing receptor CD44 revca ls atleast 12 a lternatively spliced exon s. Proe Nafl Acad Sci USA 1992 : 89: 12160-4. Jalk an en ST, Barga tze RF, Herron LR , Butcher EC. A Iymphoid cell surface glycoprotein involved in endothelia I cell recognition a nd Iymphoc yte homing in man. Eur J Immunol 1986: 16: 11 95-202. Cooper DL, Dougherty G, Harn HJ et al. The complex CD44 transcriptionalunit ; alternati ve splicing of three internal exons generates the epithelial form of CD44. 8ivehem Biophys Res Cvmmun 1992: 182: 569-78. Fox SB, Fa wcett J, Jackso n DG et al. Normal human ti ss ues , in addition to so me tumors, express multiple djfferent CD44 isoforms. CW1Cef Res 1994: 54: 4539-46. Mackay CR , Terpe HJ, Stauder R , Marston WL, Stark H, Gunthert U Expression and modulation ofCD44 variant isoform s in humans. J Cell Biv/1994 : 124: 71-82. Aruffo A, Stamenkovie l , Melnick M, Underhill CB, Seed B. CD44 is the principal cell surface receptor for hyaluronate . Cell 1990: 6\: 1303- 13. Huet S, Groux H, Caillou B, Valentin H, Prieur AM, Bernard A . CD44 contributes to T cell activation. J Il11muno/1989: 143: 798-801. Miyake K, Medina KL, Hayashi S, Ono S, Hamaok a T , Kincade PW. Monoclonal antibodies to Pgp-ljCD44 block Iympho-hemopoiesis in long-term bone marrow cultures. J Ex p Med 1990: 171: 477-88. Moll J, Khaldoyanidi S, Sleeman JP ct al. Two different function s for CD44 proteins in human myelopoiesis. J C/in InveSl 1998 : 102: 1024··34 . Joo M, Lee HK, Kang YK. Expression of E-cadherin, beta -catenin, CD44s and CD44v6 in gastnc adenocarcinoma : relationship with Iymph node metastasis. An/icaneer Res 2003 : 23 1581-8 . Legras S, Gunthert U, Stauder Ret a l. A strong expression of CD44-6v correla les Wilh shorter survival of palients with acule myeloid leukcmia. B/ood 1998: 9\: 3401-13 . Magyarosy E, Sebes tyen A, Timar J. Express ion of metas ta sis associated protcin s, CD44v6 and NM23-H J, in pediatric acute Iymphoblastic leukemia . An!icancer Res 200 I: 21: 819-23 .
C 2007 The AUlhors Jovrnal compdalion C 2007 Blac, wall M nksgaard
CD44 and CD27 delineate B-precursor development
24. Roederer M. Spectral compensation for now cytomctry: vis ualization artifacts, Iimitations , and caveats Cy/UI1lC'/ry 2001: 45: 194-205 . 25. Zuna J, Muzikova K, Madw J, Krejci O, Trka J, Temperature non-homogeneity in rapid airnow-based cycier significantly aITects real-time PCR . BiUlechniques 2002: 33: 508, 510, 512. 26 . van Zelm MC, van der Burg M, de Ridder O et aL 19 gene rearrangement steps are initialed in early human precursor B cell subse ts and correlate with specific transcription factor expression. J Immunol2005 : 175: 5912-22. 27. Langerak A W , Nadel B, De Torbal A et aL Unraveling the conseculive recombination events in the human IGK locus. J Immunv/2004 : 173: 3878-88 . 28. Pongers-Willemse MJ, Verhagen OJ, Tibbe GJ et aL Real-lime quantitative PCR for the detection ofminimal residual disease in acute Iymphoblastic leukemia usingjunctional region specific TaqMan probes. Letlkemia 1998: 12: 2006-14. 29. Loken MR , Shah VO, Daltilio KL, Civin CL Flow cytometric analysis of human bone marrow. JI. Normal B Iymphocyte development. B/vod 1987: 70: 1316- 24. 30. Ryan OH, Nuccie BL, Ritterman I, Liesveld JL, Abboud CN, Insel RA. Expression of interleukin-7 receptor by lineage-negative human bone maITow progenitors with enhanced Iymphoid proliferative potential and B-Iineage differentiation capacity. Blood 1997: 89: 929-40. 31. Hadd ad R, Guardiola P, lzac B et aL Molecular characlerization ofearly human T j NK and B-Iymphoid progenitor cells in umbilical cord blood. Blvvd 2004: 104: 3918-26. 32. Klein U, Tu Y, Stolovitzky GA et aL Gene expression dynamics during germinal center transit in B cells. Ann N Y Acad Sci 2003: 987: 166-72. 33. Paramithiotis E, Cooper MO. Memory B Iymphocytes migrate to bone marrow in humans. Proe Na!/ Acad Sci USA 1997 94: 208- 12. 34. Agematsll K, Kobata T, Yang FC el aL CD27jCD70 interaction direclly drives B cell IgG and IgM synlhesis. EurJ Immww/1995 : 25: 2825-9 . 35. Kobata T , Jacquol S, Kozlowski S, Agematsu K, Schloss man SF, Morimoto C. CD27-CD70 inleractions regulate B-cell activation by T cells. Prve Nall Acad S ci USA 1995: 92 11249-53. 36. Wiesmann A , Phillips RL, Mojica Mel al. Expression of CD27 on murine hematopoietic stem and progenitor cells. Immtlni!y 2000: 12: 193- 9 . 37. Nolte MA, Arens R , van Os Ret aL Immune aClivation modulales hematopoiesis through interactions betwccn CD27 and CD70 . Na/ Iml11u11012005 : 6: 412-8. 38. Guikema JE, Vellenga E, Abdulahad WH , Hovenga S, Bos NA . CD27-triggering on primary plasma cellleukncmia cells ha s anti-apoplotic effects involving mitogen activated protein kina ses. Br J Haema/U12004: 124: 299-308. 39. Prasad K V, Ao Z , Y oon Y et aL CD27, a member of the tumor necrosis factor receptor family, induces apoptosis and bind s to Siva , a proapoptotic protein. Proe Na!1 A cad Sci USA 1997: 94: 6346-51. 40. Akiba H , Nakano H , Nishinaka Set a L CD27, a member ofthe tumor necrosis factor receptor superfamily, activales
9
E. Mejstříková, strana 136
C044 and C027 delineate B-precursor development
NF-kappa B and stress-aclivaled prolein kina sejc-J un N -lerminal kinase via TRAF2, TRAF5, and NF-kappaB- inducing kina se. J Blol Chem 1998: 273: 13353- 8 41. Deguchi T , Komada Y, Sugiyama K et al. Expression of homing-associated cell adhesion molecule (H-CAMjCD44) on
10
M . Vaskova et aj.
human CD34+ hemalopoielic progeni tor cells. Exp HemalUl 1999: 27: 542- 52. 42 Ghaffari S, Dougherty GJ, Eaves AC, Eaves CJ. Di ve l'se effecls of anti-CD44 antibodies on lhe stromal cell-mediated support of norma I bul not leuk aemic (CM L) haemopoiesis in vitro. Br J Ha emalol 1997: 97: 22-8.
C 2007 The Authors Journal compilation © 2007 BlackweU MUrlksgsard
E. Mejstříková, strana 137
Příloha
9
Childhood secondary ALL after ALL treatment
Zuna J, Cavé H, Eckert C, Szczepanski T, Meyer C, Mejstrikova E, Fronkova E, Muzikova K, Clappier E, Mendelova D, Boutard P, Schrauder A, Sterba J, Marschalek R, van Dongen
.TJ, Hrusak 0 , Stary J, Trka J. Leukemia, 200 7, (fF 6,924)
E.
Mejstříková,
strana 138
Leukemia (2007), 1-5 © 2007 Nalure Publíshing Group AI! rights reserved 0887-6924/07 $30.00
www.nalure.com/leu
ORIG INAL ARTICLE
Childhood secondary AU after ALL treatment J Zuna 1, H C;:Jvé 2 , C Eckert J , T Szczepanski 4 ,s, C Meyer 6 , E Mejstrikova 1, E Frankova), K Muzikova 1 , E Clappier 2 , O Mendelova?, P Bautard B, A Schrauder", J Sterba 7 , R Marschalek 6 , JJM van Dongen 4 , O Hrusak 1, ) Stary 1 and J Trka 1
ICLlP (Childhood Leukaemia Investigation Prague), Department or Paediatric Haematology and Oncology, Charles University, 2nd Medical School, Prague, Czech Republic; 2 Laboratoire de Biochimie Cénétique, H opita l Robert Debré, Paris, France; l Oepartment or Paediatric O ncology/Haematology, Charite Medical Un iversity Berlin, Bertin, Cermany; 4Department or Immunology, Erasmus Me University Medica l Center Rotterdam, Rotterdam, The Netherlands,' SDepartment or Paediatric Haematology and Oncology, Silesian M~dical Academy, Zabrze, Poland; ('Institute or Pharmaceutical Biology/ DCAL, University Frankfurt, Frankfurt/Main, Cermany; / DepJrtment oí Paediatric Oncology, University Hospital, Brno, Czech Republic; ~Paediatric Onco-Haematology Unit, University Hospital, Caen, France and 9Department or Paediatrics, Un iversity Hospital, Kiel, Cermany
ar
Dala on secondary acule Iymphoblastlc leukaemla (sALL) lollowing ALL treatmenl are verv rare, However, the incidence might be underestimated as sALLs without a significanllineage shlft mighl aulomatically be diagnosed as relapses. ExaminaIlon of immunoglobulin and T-cell receptor gene rearrangemenls broughl a new tool thal can help in dlscrimination between relapse and sALL We focused on the recurrences of childhood ALL 10 dlscover the real frequency of Ihe sALL after ALL Irealment. We compared clonal markers in matched presenlation and recu rrence samples of 366 patlents treated according to the Berlln-Frankfurt-Munster (BFM)-based prolocols, We found two ca ses of sALL and another three, where the recurrence is suspicious of being sALL rather than relapse. Our proposal for Ihe 'secondary ALL after ALL' dlagnoslic criteria is as follows : (A) No clonal relatlonshlp between diagnosls and recurrence; (B) significant immunophenotypic shift - slgnlflcanl cytogenetic shift - galn/loss of a fuslon gene, For Ihe sALL (A) plus at leasl one (B) criterion should be fulfilled. With these crileria, the eslimated frequency of the sALL after ALL is according 10 our data 0.5-1,5% of ALL recurrences on BFMbased protocols. Flnally, we propose a Ireatment strategy for Ihe patlenls with secondary disease, Leukemia advance onlíne publication, 26 Apríl 2007; doi:1 0_1 038/sjJeu_2404718 Keywords: secondary acute Iymphoblastíc leukaemia; chíldhood; relapse
Inlroduction Secondary or treatment-related acute leukaelTlia is a well-known complication ol previous ca ncer therapy. The vast majority ol cases in paediatric patients co mprise secondary acute myeloid leukaemia (AMU, whereas secondary acute Iymphoblastic leukaemia (ALU ís cons ídered to be a rare dise~se . Case reports of secondary ALL (sALU in child ren have been described following treatment of various malignant diseases (Wí lm' s tumour, Hodgkin's disease, neuroblastoma, Ewing's sarcoma, osteosa rcom a, medulloblastoma, retinoblastoma, ependymoma and so on)l,l However, although ALL is the most common malignancy in childhood, the data on child hood sALL following ALL treatment are scarce. Correspondence: Dr lan Zuna, (LIP, Department of Paediatric Haematology and Oncology, Charles Univ 'rsity Prague, 2nd Medical School, V Uvalu 84, Prague S - 150 06, Czech Republic. E-mail: jan_zuna wlrmotol.cuni .cz Received 5 March 2007; revised 27 March 2007; accepted 28 March 2007
ln four large studies describing the incidence of secondary neoplasms in children treated for ALL, no secondary ALL w as diagnosed among the total of more than 2S 000 children dev eloping altogether 171 secondary malignancies,l--<" Th e data demonstrate two fa cts (1) ALL treatment bears a relatively low risk in terms of secon dary tumours compared with treatment of other frequent child hood malignancies 7 (2) Diagnosis of secondary ALL after ALL treatment is extremely rare. This rareness might be caused by the fact that ALL recurrence after ALL treatment is usually automatica lly diagnosed as J relapse of the original leukaemia. Comparative analysis of immulloglobulin (lg) and T-cell receptor (T R) gene rearrangements brought a new tool that can help in discrimination between real relapse and sALL clonally unrelated to the origin al disease_ 8 Only very lew case reports have been published adm itting that a supposed ALL relapse might represent a secondary malignancy but no comprehensive study aimed at determining the frequency of secondary ALL after ALL treatment has been presented 50 far. ln our study, we aimed to answer the question w hat is the Jctual frequenc y of this phen omenon and to show that it might be underestimated. By comparison of IgffCR rearrangements and other markers of the malignant clone in matched pre5entation and recurrence samples, we screened a series of 366 patients con secuti vely diagnosed in four cent res and treated according to the Be rlin-Frankfurt-Mun ster (BFM)-ba sed protocols_ We found two cases with secondary ALL and another three in which the second malignancy is suspicious of being sALL rather than relapse. On the basis of our results, we propose guide lines for defining the secondary ALL after ALL treatment.
Malerials and melhods
Patients A total ol 366 childhood pati ents with relapsed ALL were analysed both at diagnosis and recurrence of the disease in four centres (Prague, Paris, Berlin, Rotterdam). AII children were treated according to the BFM or BFM-related protocols (BFM (n = 81), (European Orga nisat ion or Research and Treatment oí Cancer) EORTe (n = 199), (Dut ch Childhood Leukemi a Study Grou p) DCLSG (n = 86)), The patients ar(~ consecutive unselected cases from given protocols in whom marker stability for all recurrences of childhood ALL was assessed at re5pective centres. Informed consent for the therapy and joint research examination was obtained from patients or their guardians, and
•
E.
Mejstříková,
strana 139 ChildhO()d secondary ALL afte, ALL t,eatment JZuna el a/
protocols were reviewed by Institutional Review Boards (IRB) or Research Ethics Commiltees (REC) of respective centres. Patient UPN5 was rcported previously in Ihe study concerning IgfTCR rearrangements "
ImmunorC'ceptor gene rearrangements We examined Ihe IgfTCR gene rearrangement patterns in bone marrow samples of Al.l. patients using PCR primer set s covering the vast majority (> 90%) of Ig-heavy chain (lCH) , 19-I\: (lCK), TCR-ó (TCRO) and TCR-y (TCRC) rearrangements in B-cell precursor (BCP) ALL and the set of rea ction s covering the T-cell acute Iymphobl ast ic leukaemia (T-ALL) rearrangements (TCRO, TCRC and 51L- TALI rearrangements). In cases with no clonal relationship between the samples from diagnosis and recurrence in these loci, we added examination of incomplete ICH rearrangements, and (except for the patient UPN4, where the low amount of available DNA precluded the analysis) both complete and incomplete TCR-{3 (TCRB) rearrangements. AII detected rearrangements were sequenced and compared between the primary and secondary leukaemia to distinguish completely new rearrangcments from related rearrangements with secondary changes . In thc discordant patients, the diagnostic samples were analysed for the presence of all new 'recurrence rearrangements' to exclude their presence even at lower level in the original leukaemic population . The sensitivity of this analysis was 10- 4 _10 - 5 in thc patients UPNl-4 and 10 L in patient UPNS. Sequences of primcrs and PCR conditions were specified elsewhere . I O- U To reliably distinguish clonal PCR prod ucts from polyclonal, we performed heteroduplex analysis of fragments using polyacryl amide gel. 14
Fusion gene determination and cytogenetics Presence of TEL-AML 1, BCR-ABL and MLL-AF4 fus io n genes was exa mined as a part of routine ALL diagnostics al origin al diagnosis and recurrence . MLL fusion sequence in patient UPN1 was establi shcd at the Diagnoslic Centre for Acute Leukaemia (DCAL) in Frankfurt. A long-di sta nce inverse PCR method was ls uscd to determine the genomic fusion break point. Routine karyotyping was performed at diagnosis and recurrence of the disease. In somc cases, fluorescence in silu hybridizalion analysis targeted to MLL gene rearrangements was added at the time of recurrence.
Flow cytometry Flow cytometry immunophenotyp ing of bone marrow aspirates was routinely performed at cl iagnos is and at rel apse using panel of mAbs recommendecl by the European Croup for the 1 Immunologic I Characterization of Leuk emias. ó
Genetic identity confirmation ln three patients (UPN2, UPN3 and UPN4), suspect from the secondary ALL, a palient identity confirmalion of diagnostic and recurrence samples was performed to ru le oul the possibility of sample confusion. Microsatellite testing using AmpFISTR Profilerplus Kit (PE Applied Biosystems, Darmstadt, Cermany)17 or by Prom egaPowerpl ex 16 kit (promega, Madison, WI, USA) was used for th e affi rmalion .
Results Among 366 relapscs analysed in our study, we found five cases without any clonal relationship between diagnosis and ukemia
recurrence of the disease. We aimed to verify lhe discord ance of clonal markers in these patients at different levels - by the anal)'sis of immunophenotype, cytogenetics, fusion genes and IgfTCR rearrangements. Th e IgfTCR rearrangelllents were pre sent as spec ific clonal markers in virtuall)' all ALL pati ents . Thus, we focused on detailed analysis of the rearrangements in the fivc patients and besides routine IgfTCR screening, we compared sequences of the rearrangements at diagnosis and recurrence; and we also attempted to backtrack all new rearrangcmcnts from the recurrence of the disease back to lhe original diagnosis. Neither the comparison of sequences nor thc backtracking showed an)' clonal relationship between th c primary and secondary leukaem ia in these five cases. Thc patienťs characteristics are summarized in Tabl e 1. In addit ion to the completely new pattern of IgfTCR rearrangelllcnts at the time of ALL recurrence, 2/5 patients (UPN 1 .lnd UPN2) showed additional immunophenotypic (Iineage switch from BCP to T-cell leukaemia) and genetic (occurrence of a new MLL fusion gene, loss of TEL-AML 7 fusion gene) features supporting designation as the seco ndary leukaemia . In th e patient UPN1, the recurren ce referred here is the seconcl rcc urrence; the first recurrence of thi s chil d bore all clonal signs of a genuine relapse with immunophenotype, cytogenetics and IgfTCR rearrangements correspo nding to the or iginal diagnosi s. At the second recurrence this patient showed comp lele immunophenolypic switch from BCP to T-ALL, lhe blast cells lost th eir hyperdiploid character (DNA index was 1.17-1.21 and 1.00 at diagnosis, th e first recurrence and the second recurren ce, rcspectivelyl
Discussion To our knowledge , only six cases of ALL recurrcnce that might be considcrcd as being secondary rather than relapsing leukaemia have been described 50 far. Thc rcv iew af the literature is summarized in Table 2. The lisL of cases includes three children wilh 'Iate developing' t(4;11), 18,19 and two ca ses where other translo cations invol ving MLL gene arose (t(11 ;14)20 and t( 11 ;16)·n). In the remaining case (as well as in the t(ll ;14) case), thc assumption that the recurrence is rather a secondary leukacm ia came from the fact that no common marker was founcl Jfter examinatíon of IgfTCR rearrangements at c1iagnosis and recurrence of ALL 9,20 We are aware that definite diagnosis of an indisputable sALL is intricale. In most cases, there is a hypothetical possibility lhat lhc discase has originated in a very early prugenitor with ability to diffcrentiate into very dissímil ar cell populations with seemingly no clona I relationship to cach ether. Unless the leukaemic stem cell is defined, it ís virlually unfeasible to rule out the possibility of a biphenotypic/biclonal clisease at diagnosis w ith a sma ll , undetected subclone outwardly unrelated to the predominant diagnostic clone. Such cel ls cou ld emerge aiter an effective treatment of the major clone and givc rist' to a dominant relapse population . Thus, the on ly virtually indi sputable sALLs are cases where a fusion gene, whi ch is thought to be lhe first hit in leukaemogcncsis (for cxample, TEL -AMU or MLL-AF4l, is lost al the recurrence. To preclude or at least to minimize the risk of a hidden biphcnotypi c/biclonal discJse at diagnosis, absence oi all thc rcc urrcnce-specific rea rran gements should be verified in th c diagnostic sample. However, the probability of a rc currcnce constitulecl by cclls with completely unrelated clonal charilctcristics but stili originated in lhe same leukaemic stem ce ll as the original
E. Mejstříková, strana 140
•
Childhood secondary ALL after ALL treatment J Zuna et a/
Table 1
Characteristics of the 5 patients with possible secondary ALL aher ALL treatment
UPNi
UPN2
UPN3
UPN4
UPN5
5.1 F BCP/BCprr neg/neg/MLL -MAML2 ABCIABC/DEF ALL·BFM95-SRlALLRez BFM96-S2 2.9/2.7
5.8 M BCprr TEL AML 1I neg ABCD/EFGH EORTC-58951-VHR
3.0 M Trr Neg/ neg ALL-BFM 2000-MR
78 F BCP/ BCP Neg/neg AB/CDE EORTC-58881-S
8.2 F BCP/BC P NA/NA ABCDElFGH DCLSG-ALL-8-HR
2.2
1.7
6.5
3.5
No Toxic death in the recurmnce
Ves Death due 10 ex1ramedullary (thymic) pro9resslon No 1350 220 4
Ves Second CR (32) after alloBMT
Ves Second CR (72)
12 No 120 3
No No 240 3
No Death in the recurrence (progressive disease) No 1350 270 7
"a/ien/
1ge at diagnosis (years) 3ender :nmunophenotype (PIR) ;usion genes IP/R) qrrCR (P/ R) Treatment preceding the sALL CR preceding the sALL (years) :R after sALL Outcome (months in CR)
rladiation (Gy)" Etoposide (mg /m2 )' Daunorubícin (mg/ m2)" :Yclophosphamidel !oslamide (g/m2
18 2000 225 102
r'
AB /C
~bbreviations:
alloBMT, allogeneic bone marrow transplanlation ; BCP, B-cell precursor; CR, complete remission; IgrrCR, immunoglobuhn and T·cell receptor; NA, not available; P, presentation; R, recurrence; sALL, secondary acute Iymphoblastic leukaemia. 'g/TCR: in each patient, different letters stand for differenl and unrelated rearrangements, 11 patient UPN 1, we reler to presentation and both first (genuine relapse) and seco nd (sALL) recurrences. 'Cumulative doses of selected drugs preceding the sALL Table 2
Review of the literature - CJ ses of poss ibl p sALL after ALL described
50
far
qeference
Origina/ diagnosis
Secondary ALL
/g/TCR rearrangemen/s
Ou/come
Szczepanski et al. 20
T-ALL T-ALL Mature B-ALL with t(8;14) T-ALL ALL ALL
T-ALL T-ALL wlth t(11 ;14) BCP -ALL with t(4;11) BCP-ALL with t(11; 16) BCP-ALL with t(4;1 1) BCP-ALL with t(4;11)
Unrelated Unrelated Unrelated Unrelated (only TCR-fl tested) Not done Not done
Not reported Not reported CR achieved Not reported CR achieved , alloBMT periormed CR achieved, alloBMT periormed
Millot et a/, 19 hunger et a/, 2' 8rizard et al. 18
Abbreviations: alloBMT. allogeneic bone marrow tfansplantation; ALL, acute Iymphoblastic leukaemia; B-ALL, B-cell acute Iymphoblastic Ieu aemia; BCR B-cell precursor; CR, complete remission; IgrrCR, immunoglobulin and T-cell receptor; sALL, secondary acute Iymphoblastic !eukaemia; T-ALL, T-cell acute Iymphoblastic leukaemia; TCR-fl, T-cell receptor-fl.
diagnosis is low . Moreover, thi s dilemma is rather academic as from the practi ca l point of view (for cxample, treatment purposes) such ca se should not be considcred as a standard re lapse anyway. Our multicentre study is the first to investigate systematically the clona I relationship between presentation and recurrence in childhoocl ALL. We present a large cohort of patients examined on running trials and analysed subsequently with regard to reveal inci dence of possible sALL after ALL treatment. As repo rts in the literJture are E'xtremely rare, no stJndards dcfining this issue have been postulated 50 far. Our proposal for th, ' diagnostic criteria of '5econdary ALL after ALL treatrllcnť is as lollows:
iA)
18)
No clonal relationship between diagnosis and rec urrence (lglTCR, fusion genes Jt DN A level, cytogenetic marker). • significam immunophenotypic shift (typically lineage switch) • significant cytogenetic shift • gain or loss of a fusion gene
For the diagnosis of secondary leukaemia, (A) plus at least one
lB) criterion should h" fulfilled. ln our study, two patients (UPN1 and UPN2) meet our criteria for sALL. The patients fulfilling only the (A) c riterion (pati ents UPN3, UPN4 and UPN 5 in our study) are ' pos sible' secondary
ALLS, but without additional evidence the diagnosis of secondary leukaemia could be challenged. The number of IglTCR markers changed between di agnosis and recurrence (the (A) criterion) might also have its significance - providing no identical or related marker is maintained between the two time points, lhe more changes detecled the higher is the probability of secondary ALL. Thu s, in our case UPN5 with eight such changes, the sALL is highly possible eve n without any (B) criterion fulfilled , On the other hand, in case UPN3 (oni)' three changes and no (B) criterion), the diagnosis of sALL co u lel be questioned more easi ly, Although the longer remission duration could be considered as a supporting ev idence for secondary rather than relapsed ALL, we did not include the criterion of the time to recurrence into aur proposal. Studies on very late relapses show Ihat even recurrence more than 20 years from diagnosis ar(' clonally related to the original leukaemic cells (1 3/13 very late relJpses 5-24 years from diagnosis 2 Z. 23 ). On the contrar)', ver)' c'lrly recurrences (Iess than 1 yea r from the ori g in al diagnosis) are certainly more lik ely to be genuine relapses. Occurrence of a new fusion gen e is not a guarantcc pN se that the recurrence is a seco ndary leukaemia. For cXJmplc, lhc t(9;22), associated with the BCR-ABL rearrangemcnt, cnuld bc a late appearing, therapy-related secondary event in the evolution of the primary clone,24-2B Leukemia
.eukemia
E. Mejstříková, strana 141 Childhood secondary All after All treatment J Zuna el aJ ALL is the most common childhood malignancy, and thus it can be supposed that sALL after ALL treatment is more common than reported - sALL s without a significant linea ge shift have probabl y been automatically diagnosed as relapses; in cases where an immunophenotypic shift is considerable (such as in Our patients UPN1 and UPN2), the flow cytometry is the first method that draws our attention to a possibl e secondary disease. The main oncogenic factors increasing th e risk of subsequent neoplasms are a genetic susceptibility and a previous therapy, particularly radiotherapy, topoisomerase inhibitors (etoposide, doxorubicinJ, alkylating agents (cyclophosphamide) and some antimetabolites (6- thioguanine). More intensive use of some of th ese treatment options in certain older ALL protocols led to increased frequency of second ary malignancies to almost 5%29 Although tbese components are also used in current BFM-based ALL protocol s, the ov erall extent of th e use is relatively low compared to the treatment of other paediatric malignancies and some are applied only in subgroups of patients (l-cell ALL and high-risk ALL). Also the dosing schedule is adapted to ca use as little late effeets as possible. Only five pati ents in aur study are suspect of suffering treatment-related secondary leukaemia, and all of them were treated according to th e standard protocols as well as the rest of the children in this report. Thus, we cannot draw any rE~aso nahl e conclusion regarcling the relat ionship of primary treatment and th e risk of sALL. Nevertheless, it is of note that 3/5 case in our report (pati ents Ur'N1, UPN2 an d Ul) 15) unde/Went a very intensive therapy before sALL. Patient U PN1 was stratified to standard risk treatmen t but suffered relapse and received another protocol of intensive chemotherapy before her second recurrence identified as a secondary ALL. Patients UPN2 and UPNS responded poorly to initial treatment and were restratified to the very high-risk Mm of the EORTC protocol and high-risk arm of the D LSC; protocol, respeetively. Previous treatment an play a rol e in the risk of sALL. However, whde the link betwecn specific drugs and the risk of secondary AML is very strong, in se condary ALL the effect of previous treatment is less pronoul1ced. This faet suggests that other mechanisms - particularly geneti e susceptibility - might be also involved . Polymorphisms of sev ra l detoxification genes (NQ07 , CYP3A and CS1) have been shown to be related to the increased risk of seeondary leukaem ia, mutation s of ATM gene have been linked specifica lly to T-ALL. 7.JO Notably, 3/5 patients described here suffered ALL recurrenee from T-lineage when 2 of these J T-ALLs represented a lineage shift from BCP-ALL. Detailed polYl1lorphism and mutational ana lysis of these genes might be helpful in unmaskin g the sALL pathogenesis. On the basis of current knowl edge on childhood ALL, there are probabl y three typcs of disease recurrence: 1. genuine relapse from a resistant diagnostic (sub)clone, 2. secondary leukaemi a arising frolll th e ori ginal pre-Ieukaemic c10ne (with some clonal markers milintained and some changed) as demonstrated in late relapses of TEL-AML 7 positiv e cases] I and 3. pure sA LL, clonally unrelated to the original leukaemia . It is very difficult to distinguish between the first and second type of recurren ee using current sta ndard techniques as in both of the se events some c lonal markers are maintained between the original diagnosis and recurrence but other can be altered . The only method descri bed 50 far to distinguish at least some 'secondary leukaemias from pre-Ieukaemic clone ' from 'genuine relapses' is the analy sis of the non -translocated TEL gene deletions in the subgroup of patients with thl' TEL-AML 7 fusion gene. D spite th e number of published cases where the recurren ce is beli eved to be a 'seeonda ry leukaemia from the
original pre -Ieukaemic clone' is less than 10,31 ,.12 the published data suggest that the frequeney of this type of recurrence can be 3 1 as high as 20%, particularl)' in late relapses Th e estimated frequency of the pure sALL after ALL treatment is low but not null - according to our data 0.5-1.5% of ALL recurrences on BFM-ba sed protoc ols. In our slUdy at least two patients belong to the category of pure sALL. ln thl' 'pure sALL' and the 'sALL from the same pre-Ieukaemic clone' cases both the previous and the subsequent treatment strategies should be considered . The frequency of these types of recurrence (togeth er possibly even more than 20% of late events) should be co nsidered in discussions regarding an intensity of preceding frontline treatment strategies - th ese failures might, in fact, occur not because of low intensity of therapy but can be triggered due to overtreatment. As for the adequate subsequent treatment, it should be stressed that we deal in fact with new diseases and not with resistant dones selected by a previous therapy. On the one hand, the 'sA LL from the same pre-Ieukaemic clone' cases might be candidates for standard frontline treatment rather than intensified relapse prolocol; however, as Olentioned above, disclosure of these ca ses and their distinction from genuine relapses in current routine practi ce is intricate. On the other hand, the 'pure sA LLs ' represent second independent ma1ignancies of haematopoietic cell s and thus some (possibly inherited) su sceptibility to the disease must be taken into aeeounl. In such cases, haematopoietic stem cell transplantation (SCT) should be considered to replace the pr ed isposed haematopoietic ce lls. As the 'pure sALLs' (u nlike th c 'sALLs from pre-Ieukaelllic c1one') can be revealed in time to adjust their treatment (immunophenotyping, fusion genes detection and IglTCR anal ysis can all be done within a few days), we suggest an approach applying a frontline therapy followed by SCT for all the indisputable eases . For the 'possible' sALL ca ses (fulfilling only the (A) criterion from the above sALL diagnostic proposa l), we would rather recom mend a standard relapse therapy including stratification according to a protocol. However, in the low-risk groups (where there is only a limited SCT indication in most of the current protocols), an SCT should also be discussed based on a deeper understanding and ev idcnc(' of biological origin of sALL. Therefore, we strongly advocate that all treatment deci sions should be handled with caution and should be guided via study centres to ensure a harmonized clinical approach.
Acknowledgements We acknowledge Professor Willem A Kamps the chairman of DCLSG ALL-8 study. The work was supported by grant 21620813 (MSMT CR) and by the Programme Hospitalier de Recherche Clinique (PHRC-AOM02 003).
References Hunger SP, Sklar J, link MP. Acute Iymp hoblastic léukemia occurring as a second malignant neoplasm in childhood: report of three ca ses and review of the literature. J Clin Onco/1992; 10: 156-163. 2 Geetha N, SreedeviAmma N, Kusumakumary P, lali VS, Nair MK. Acute Iym phobl astic leukemia occu rring as a second malignancy: report of a case and revi ew of literature. Pediatr Hemato / Onco/ 1999; 16: 267-2 70. 3 Neglia JP, Meadows AT, Robison LL, Kim TH, Newton WA, Ruymann FS el a/. Second neoplasms after acute Iymp hoblasti c leukemia in childhood. N Eng/ J Med 1991 ; 325 : 1330-1336 .
Childhood secondary ALL after ALL treatment J l una el al
4 Kimball Dalton VM, Gelber RD, Li F, Donnelly MJ. Tarbell NJ. S~ lI a n SE. Second mali gnancies in patients lreated for child-
5
6
7
8
9
10
11
12
13
14
15
16
17
hood acute Iymphoblastic leukemi a. j Ctin O ncol 1998; 16: 2841>-2853. Loning L, Zi mmermann M , Reiter A, Kaatsch P, Hen ze G, Riehm H et al. Secondary ncoplasms subsequent to Berlin-FrankfurtMunster therapy of acute Iymphobl astic leukemia in childhood: significantl y lower risk without crani'll radi otherapy. Blood 2000; 95 : 2770--2775. Bhatia S, Sather HN, Pabust an OB , Tri gg ME, Gaynon PS , Robison LL. Low incidence of sccond neoplasms among children diagnosed with acute Iymphobl asti c leukemia dltc r 1983 . Blood 2002; 99: 42 'i7-4264. Bhatia S, Sklar C. Sccond cancers in surv ivors o f childhood cancer. Nat Rev Cancer 2002; 2: 124-13 2. Szczcpanski T, Willem se MJ. K,lmps WA, van Wering ER, LangerJ k AW, van Dongen Jl. Mol"cul ,lr di scrimination between rel apsed and secondary ac ute Iymphobl asti c I ukemia: proposal for an easy stratcgy. Med Pediatr Oncol 2001, 36: 352- 358. Szczepanski T, Willemse MJ. Brinkhof B, van W ering ER, van der Burg M , van Dongen jJ . Comparative analysis of 19 and TCR gene rearrangem ents at di agnosis and at relapse of childhood precursorB-ALL provides improved strategies for sele tion of stabl e PCR targets for monitoring of minimal residual disease. Blood 2002; 99: 2315-2 3 23 Pongers-Willemse MJ. Seriu T, Stolz F, ďAniello E, Gameiro P, Pisa P et al. Primers and protoco ls for standardized detection of minimal res idual di sease in ac ute Iy mphoblastic leukemia usin g immunoglobulin and T cell rec ptor gene rearran gements and TAL1 deletions as PCR targets: report of the BIOM ED -I CONCERTE D A CTl O N : invcsti ga tion ol minimal residual disease in acute leukemi a. Leukemia 1999; 13: 11 0 - 118 . Szczepanski T, Willcm se MJ. van Wering ER, van Weerden IF, Kamps WA, van Dongen IJ. Precursor-B-ALL with D(H)-J(H) gene rea rrangements ha ve an immature immunogenotype with a high frequency of oligoclonality and hyperdiploidy o f chromosome 14. Leukemia 2001 ; 15: 141 5 142 3. van der Velden VH, Hochhaus A, Ca zzaniga G, Szezepanski T, Gabert J. van D ongen IJ. Detection o f minimal residual di sea se in hematologic malignancies by real-time quantitative PCR : principles, approaches, and laboratory aspects. Leukemia 2003; 17: 1013- 1034. van Donge n JJ, Langera k AW, Bruggemann M , Evans PA, Hummel M, Lavend er FL et al. D esign and standardization of PC R primers and protocols l or dc tcct ion of clonal immunogl obulin and T-cell receptor gene recombi nations in su spect Iy mphoprol iferations: report of the BIOMED-2 Con ccrted Action BMt-I4-CT98-3936. Leukemia 2003 ; 17: 22 '7- 2317 . Langerak AW, Szczepanski T, van der Burg M , Wol vers-Tcllcro lL, van Dongen JI. Het 'roduplex PCR analysis of rearranged T cell receptor genes for clon ality assessment in suspect T cell proliferations. Leukemia 199 7; 11: 2 192-2199. M eyer C, Schneider B, Reichel tvl, Angermueller S, Strc hl S, Schnittger S el Ol l. Diagnostic tool for the identifícation or MLL rearran gc rn ents including unknown partner genes. Proe Na II Aead Sei USA 2005 ; 102: 449-454 . Bene MC, Casloldi G, KnJpp W , Ludwi g WD, Matutes E, Orfao A et al. Proposals lor the immu no logical classificati on of aeule leukemias. European Group for th e Immunological Characterization ol Leukemias (EGll). Leukemia 1995; 9: 1783- 1786 . MassenkE'iI G, Nagy M, Lawang M , Rosen O , Genv resse I, Gesc rick G el al. Reduced intensity conditioning and prophylactic
18
19
20
21
22
23 24
25
26
27
28
29
30
31
32
DLI can cure patients with high-risk acute leukaemi as if compl ete donor chimeri sm can be achieved. Bone Marrow Transplant 2003 ; 3 1: 339- 345 . Bri za rd A, H urel lL, Benz-Lemoine E, Guilhot F, Giraud C, Tanzer J. Two cases of t(4 ; 11) acute Iymphoblastic leukemia (AL U follo w ing ALL without the t(4; 11) : second or secondary leukemias? Br j HElemalo /1991 ; 79: 130-131. Millot F, Brizard F, Sorel N , Preudhomme C, Cividin M , Guilhot F et al. Therapy-related acute Iy mphoblastic leukemia with MLL rearran gement foll owing treatment of Burkin's leukemia. Leuk Lymph oma 2005 ; 46: 925-927. Szczepan ski T, van der Velden V H, Raff T, lacobs DC, van W erin g ER , Bruggemann M et al. Comparative an alys is of T-cell receptor gene rearran gements al diagno sis and relapse of T-cell a .utc Iy mphoblastic leukemia (T-ALU shows high stability of clonal markers for monitoring of minimal residual disea se and revea ls the occurrence of second T-ALL . Leukemia 2003; 17: 2149-2156. Hunger SP, Tkachuk DC, Amylon MO, Link MP, Carroll AJ, Welborn lL el al. HRX involvement in de novo and secondary leukemia s with divem> chromosome 11 q2 3 abnormalities. Blood 1993; 81 : 3 197-3203. Lo N igro L, Cazzaniga G, Di Cataldo A, Pannun zio A, D ' Aniello E, Masera G el al. Clona I stability in children with acute Iymphoblastic leukemia (ALl) wh o relap sed five or more years aftpr diagno sis. Leukemia 1999; 13: 190--1 95. \fora A, Frost L, Goodeve A, Wilson G, Ireland RM, Lilleym an I et al. Late relapsin g childhood Iymphoblastic leukemia. Blood 1998; 92: 233 4-233 7 Kim M, Lim J, Kim Y, Han K, Kang CS, Kim HJ et al. A case of therapy-related acute myeloid leukemia associated with inv(16 ), with subsequent development of t(9 ; 22). Leukemia 200 6; 20: 746- 748. Tsuchiya H , Migita M, Yamamori S, Kaneko Y, Adachi N, Nakamura T et al. A late-appearing Philadelphia chromosome in acute Iy mphoblastic leukemia confirmed by expression of BCRABL mRNA. Leukemia 1995; 9: 1689-1693 . Tchirkov A, Bons JM, C hassagne J. Schoepfer C, Kanold J. Briancon Gel al. Molecular detcction of a late- appearing BCR -ABL gene in a child w ith T-cell acute Iy mphoblastic leukemi a. Ann HemalOl 1998; 77: 55-59 Miller BA, Reid MM, Nell M, Lipton JM, Sallan SE, Nathan DG el al. T-cell acute Iymphoblastic leukaemia with late develof.'ing Phil ade1phi a chromosome. Br j Haemato/1984 ; 56: 139- 146. Coad JE, Arthur DC, Gajl-Peczal ska KJ. Litz CE. Late-developing Phil adelphia chromosomes in a ca- of T-cell ac ute Iymphohl astic leukemia. Leukemia 1994; 8: 889- 894. Pui CH , Boyen 1M, Rivera GK, Hancock ML, Sandlund JT, Ribciro RC et al. Long-term results of total therap y studies 11, 12 and 13A for c hildho od acute Iy mphobla stic leukemi a at St Jude Children's Rcsearch Hospital. Leukemia 2000; 14: 2286-2294 . LiberlOn E, Avi gad S, Stark B, Zilberstein L Frccdman L, Gorfin e M el al. Germ-line ATM gene alterations are associated with susceptibility to sporadic T-cell acute Iy mphobl astic leukemia in children. Genes Chromosomes Cancer 2004; 39: 161-166. Zuna J, Ford AM, Peham M , Patel N, Sah a v, Eckert C el al. TEL deletion analysis supports a novel vi ew of relap se in childhood acute Iymphoblastic leukemia. Clin Caneer Res 2004; 10: 5355-5360 . Ford AM, Fa sching K, Panzer-Grum ayer ER, Koenig M , H aas OA, Greaves MF. Origins of ' Iate' relapse in childhood K ute Iy mphoblastic leukemia with TEL-AML 1 lusion gene,. Blood 2001; 98: 558-564 .
•
Leukem;a
E. Mejstříková, strana 142
P ř íloha
10
Allogeneic stem cell transplantation in children with leukemia using human leukocyte antigen-mismatched unrelated donors
Sedlaeek P, Fonnan.kova R , Mejstrikova E, Keslova P, Hubaeek P, Dobrovolna M, Vrana M, Kupkova L, Pittrova H, Stary J. Pediatrie Transplantation, 2007, (lF 1,505)
E. Mejstříková, strana 143
Pedi(ll r Troll:splrml(lf; ulI 2fJ08: I .'. !4- 3/
Copy riglu © 2007 Block we/l MlIllk.\gaard
Pediatrie Transplantation 001: 11I.IIII U.1J9Y·J046.2007.1J0762 ..\'
Allogel1eic stem cell transplantation in children with leukemia using humal1 leukocyte antigen-misll1atched Ul1related donors Sedlacek P, Formankova R, Mejstrikova E, Keslova P , Hubacek P , Oobrovolna M , Vrana M, Kupkova L, Pittrova H , Stary J. Allogeneic stem cel! transpl a ntation in children with leukemia using human leukocyte antigen-mismatched unrela ted donors Pediatr Transplantat ion 2008: 12: 24-31. © 2008 Blackwell Munk sgaa rd Abstract: Al10geneic HSCT is a curative treatment , when chemotherapy fai1s, for certain malignant diseases . ln Europe, on ly 15 % of the indicated children have an HLA-matched sibling available; in 65- 70% of others, HLA al!ele-matched (9- 10/ 10) UOs can be identified. For the re st, it is necessary to identify other alternative donors (HLA-mismatched family or unrelated cord blood). We present our d a ta of HSCT using HLA partially alle1e-mismatched (7-8/l0) UO s in 24 children with leukemia. Uniform GvHO proph ylaxis was used (rATG, CsA and MTX) . A cute GvHO grade II was diagnosed in 70.8 % of the patients and grade III· IV in 12.5 % . Overal! incidence of chronic GvHO was 38.7% (extensive in 30%). The probability of EFS was 60 .3% (95% CI 35.5- 78.1) and OS was 74.9 (95% C I 49 . 1-88.9) . No difference in survival between PBSC and BM recipients was observed. TRM at day + 100 was 4%, and overal1 was 12.5%. We conclude that used combination of drugs for GvHO prophylaxis is efficient even for patients transplanted with grafts from a HLA-mismatched UOs. lt enables stable engra ftment , good contr01 of G vHO, full reconstitution of immunity, and is not connected with unacceptable transplanHelated mortality.
Petr Sedlacek', Renata Formankova', Ester Mejstrikova', Petra Keslova\ Petr Hubacek', Marie Dobrovolna 2, Milena Vrana 2, Libuse Kupkova 3, Helena Pittrova 4 and Jan Stary' 1Department ol Pediatric Hematology and Oncology, University Hospital Motol. 2nd Medical School, Charles Universi ty, Prague, Czech Republic, ' Institut e of Hematology and Blood Transfusion. Prague, Czech Republ ic, JCzech Stem Cell Registry, Prague, Czech Republic, 'Czech National Marrow Donors Registry, Pilsen, Czech Republic
Key words : allogeneic hematopoie ti c stem cell lrans plantation - human leukocyte antigen· mismatched - Ieukemla - unrelated donor - chlldren Petr Sedl acek. MO. PhD. HemalO poletlc Stem Cell Transplant Unit, Department ol Pediatric Hematology and On cology, UniverS ity Hospital Motol and 2n d rviedica l Sch ool, Charles University, V Uvalu 84. 150 06, Prague, Czech Republic Tel. . + 420 22443 6552 Fax: + 420 22443 6519 E-mail: [email protected] Accepted lor publication 11 May 2007
Allogeneic HSCT is a potentially curati ve treatment for cerlain malignant diseases. It is often indicated ln patients wilh leukemia where response to chemotherapy is inadequate . Unfor-
tunately, in about 15-20% of the patients, we are not able to find a HLA-matched RD or UD. Infusion of a graft from a HLA partially mismatched donor, unrelated cord blood , or
Abbreviations: *, with 12 men and 12 women; ~ pair, one pair out of 24 pairs (donor-recipient); ABiL, acute biphenotypic leukemia ; ABLC, amphotericin B lipid complex injections; AdV, adenovirus; AML, acute myelogenous ieukemia ; ANC, abso lute neutrophil co unt; ATG, antithymocyte globulin; I3KV , BK virus; BM , bone ma rrow; cGvHD, chronic graft vs. host disease; CM L, chronic myelogenous leukcmia ; CMV, cy tomegalo virus; CsA, cyclosporine A; DLI, donor Iymphocyte infusio n; EBV, Epstein-Barr virus; EFS, event-free su rvival; ESG-MRDALL, European study group on detection of MRD in acute Iymphoblas tic leukemia; FPIA , fluore sce nce polariza tion immunoassay; G vHD , graf! vs . host discase; HHV6, human herpesvirus 6; HHV7, hum an herpesvirus 7; HLA , human
leukocyte a ntigen; HLA-A ,B,DR, hum a n leukocyte antigens-A,B,DR; HSCT, hematopoietic stem cell transplantati on; HSV, herpes si mplex virus; IFI, invasive fungal infection; IgHjTCR, immunoglobulin heavy chain rearrangementJT cell receptor; 1ST, immunosupprcssive therapy; MRD, minimal residua I disease; MTX, met hot rexate; OS , overaJl survival ; PBSe, peripheral blood stem cells; PCR, polymerase chain reaction ; PCR-SSP, polymerasc chain reaction-sequence specific primer; QoL, quality of liťe; STR , short tandem repeats; rATG, rabbit antithymocyte globulin; RD, related donor; TRM , transplant-related mortality; UD, unrelated dono r; VNTR, variable number tandem repeats; VZV, Varicella zoster virus.
24
E.
Mejstříková,
strana 144
HseT in leukemia using HLA-mismatched donors haploidentical family donor with graft manipulalion is then considered with preference mostly based on local experience and/or availability (I). However, all these alternatives have an increased risk of post-transplant morbidity/mortality because of an increased risk of GvHD, graft failure , and slow immune reconstitution (2, 3). Furthermore, in patients wilh malignancies , manipulation of the graft may adversely affect a fa vorable alloreactive effec t direcled against residual disease (4). Since the beginning of 1999, HLA typing using PCR methods was routinely available at our institution enabling high-reso lution (allele) typing o f class I (A *, B* , Cw*) and dass Ir HLA antigens (DRBI *, DQB 1*) in all patient-donor pairs. Based on level of HLA allele match between donor and recipient, we were able to identify an acceptable HLA partially mismatched donor (7-8/10) for a majority of patients lacking a fuHy matched (910 out of 10 a]Jeles) donor . Here , we report our dinical experience with uniform GvHD prophylaxis using a combination of CsA, MTX , and rATG prior to unmanipulated graft administration in children with leukemia (5, 6). Patients and methods Between January 1999 and December 2006, in ou r unit , 24 consecutive child ren with leukemia (detailed di agnoses are shown in Table I) underwent 26 allogeneic HS CT usíng an HLA-mi smatched UD because we failed to identify a suitab le HLA-ma tched donor in the time frame availab le. The interval between search and day of HSCT ranged from 98 to 492 days, med ian 147 day s. These 24 patients (a ge range 2.2-18. 1 yr; medían 11.3) were tra nspla nted with unmani pulated gra fts fro m UD mismatched in tw o (13 patients) or three (II patients) HLA a lle les. The locations of the various a ll e le mismatches prospec tively ide ntined by PCR-ssr were as follows (A * -2x, B* -1 3x, Cw· -27x, DR BI- -4x, DQB I * -13 x). Six out of 24 donor- recipient pairs were matched in six out of six alleles in loci A-, B*, or DRBI*' Another 17/24 (71%) pairs were mismatched in nve out of six alleles, 1/24 pair was mismatched in B*, DRB I· (and Cw·) allele s (four out of six al leles); details in Table I . The median age of UD was 34 yr (age range 22-49) with 12 males and 12 females. Conditioning regime ns used for nrst HSCT were fully myeloab lative in 23/ 24 patients (Table 2). Uniform GvHD prophylaxis consisted of combination of CsA, MTX, and rATG. CsA starting at day I before infusion of the gra ft was given in two to th ree daily infusio ns (over two h) to maintain the trough serum leve ls of 200250 ~g/ L (met hod of detection - FPIA) . MTX was given on 2 day -f I (15 mg/m or on ly 10 mg/ m 2 in more advanced disease) , day + 3 (10 mg/m"), and day + 6 (10 mg/ m 2) with ieucovorine rescue (15 mg/m 2) in a single dose 24 h following every dose of MTX. Thymoglobuline was used in four patients in a daily dose 3.75-4 mg/ kg for days -4 through -I (total 15- 16 mg/kg). Later on, within further 20 patients, ATG Frese niu s S was used in a daily do se 10 mg/ kg fo r days -4 through -I (total 40 mg/ kg).
Primary grafts were PBSC (n = 13) and BM (n = ll) . The nna l decision about type of graft was made by the donor and/or loca l harvest center based on trans pi a nt center preference. C harac teristics of grafts are shown in T a ble 3. Two patients later received a second graf! (PBSC) from the same donor, both for leukemia relapse. ACLlte and chronic GvHD were díagnosed a nd graded usíng established cri teria (7, 8) and were prim ar ily treated with prednisone, CsA, tacro limus, sirolimus, or MMF in standard doses (9-11 ). Da y of neutrophil e ngraftment was defined as the nrst out of three consec utive days when the ANC reached 0.5 x 109 / L or more. Platelet engraftment was denned as plate let count 20(50) x 109 /L or more for seve n consecutive days without transfusion. Chimerism was assessed by using PCRbased a nalyses of polymorphic VNTR/STR on recipient fro m un separa ted peripheral blood frequently starting on day + 14 and then once weekly until day + 100, later less frequently in patients with stable full donor chimerism , to connrm efficie nt engraftment and to rul e out risk of la te graft failure/ rejection or relapse ( 12). PCR assay of specinc fusion genes and IgH /TCR receptors acco rding to the type of leukemia was used for MRD monitoring pre and a fter HSCT. The testing was proceeded and evaluated according to the crite ria of the ESG-MRD-ALL (13, 14). A marrow sample was ta ken routinel y in patients with any denned leukemia target two to three wk before HSCT, a t days + 28, + 60, + 100, 6 and 12 months after transplantation, or in ca se of decreasing overall chimerism or positive M RD in previous samp les. A complete hematologica l re mi ssion was defined as less than 5% blasts in the marrow aspirate and func tional hema topo iesis. Surveillance of viral infections as a common cause of transplant-related morbidity and mortality was based on quantitati ve real-time PCR technique on DNA from whole blood. EBV and CMV were scree ned routinely as part of weekly testing. 0ther viruses, HSV , VZV , H HV6 , HHV7, ade noviru ses group A-C, and BK virus (15-1 7), were tested only in ease of clinical suspicion. Results for leukotrophi c viruses (CMV, HHV6, HHV7, and EBV) were normalized to 100000 huma n genomic equivalents assessed by quantincation of the albumin gene. AII these studies were approved by loeal ethica l comm ittee and all pa rents signed informed consents.
Results
Hematopoietic engraftment and chimerism
Full trilineage-stable primary engraftment was ac hieved in aH 24 children (100%). Engraftment characteristics are listed in Table 3. Com plete donor chimerism was observed in all 24 patients, and could be documented after a median of 21 days (range 14-98) with no difference between PBSC and BM. Reappeara nce of mixed chimerism was detected only in patients with emerging \eukemia relapse. GvHO incidence and severity
Acute GvHD grade ]J was diagnosed in 17 (70 .8%) patients; grade III-IV in three (12.5 %) . Overall incidence of chronic GvHD in 23 evalu-
25
tT1
...
(fl
N
CTI
Q.
~
~ n
~.
~
'"'1<
...
~ ~
Vl
~
~
O
<
1" Vl
Iq
Ta ble 1.
UPN
Dg at SCT
114 11 6 122 142 143 148 153 159 173 183 200 200 211 216 223 233 244 245 245 251 253 258 266 268 282 285
CML MOS ALL CML CML CML MOS NHL ALL CML AML
~
Stage
AP RAEBI/AML CR2 CPI CPI CPI RC CR3 CR2 CPI PRl CR2 Ph+ALL CRI AML CR3 CR l Ph+ALL RAEBI MOS Ph+ALL CR l ASiL CRl CR2 AML CR2 ALL CR3 Ph +ALL CR2 sMOS RAEB ALL switched to AML CRl sAML CRI MOS RC/man
Not eval.. not eva luable
Oate ol SCT 1st MM 2nd MM X-99 XI-A9 111-00 111-01 IV-Ol VI-Ol IX-Ol XI I-Ol X-02 111-03 IX-03 1-05 111-04 IV-04 VII-04 XI-04 11-05 VI-05 VII-06 VI-05 VI-05 IX-05 1-06 1-06 VII-06 IX-06
Cw Cw DOBl B B Cw Cw B A B ORBl
Cw Cw DOBl Cw Cw DOBl Cw Cw Cw Cw Cw
B ORBl B B A S
Cw Cw Cw Cw DOBl Cw
S ORSl B S ORBl Cw S
Cw S Cw Cw DOBl DOBl Cw
3rd MM rATG brand
Graft
(DL I) C03 + (kg) bw
DOBl DOBl
BM BM BM BM PBSC BM BM BM BM PBSC PBSC PBSC PBSC PBSC PBSC PBSC BM PSSC PSSC PBSC PBSC PBSC PBSC PB SC BM BM
(1) 10/6
DOBl DOBl DOBl Cw DOBl Cw
DOBl
Cw Cw
Thymo 16 mg/kg Thymo 16 mg/kg Thymo 16 O1g/l:g Fresenius Freseni us Frese nius Freseni us Fresen ius Fresen ius Fre senius Frese nius Fresenius 50 mg/kg Fresenius Fresenius Fre sen:us Fresenius Fresenius Fresenius Fresenius Frescnius Thymo 16 mg/ kg Frese ni us Fresenius Fre se niu s Frese nius Frese nius
AGVHO Grade
cGVHO grade
III O III II II II O
Extensive None Not eval Extenslve Extensive Limited None None None Li mited None None Extensive None None Extensive None None
(1) 10/6
(4 ) 10/ 5: 5 x 10/6:1017: 5 x 10/6
(2) 10/6; 5 x 10/6
IV II II III-I V II
1ST in Event
Outcome
Alive/Well Alive/well EBV PTLO Oied Al ive/well Alive/we ll Al ive/Well Alive/Well A1ive/Well CM V pn eu010ni a Died Alive/we ll Relap se Alive/well Alive/well MRO positive Alive in relapse Al ive/well Ext. cGVHO Oied Relapse Oied Relapse MOF Oied Alive/well None Extensive Relapse Al ive in relapse None Ahve/well None Aliv8/Well Exten sive Alive/well Alive/well None none Alive/well Cytog.relapse
SU~J l vo r s
None None None None None None None None None None None None
None None None None CsA laper None MMF
::l
l~
E. Mejstříková, strana 146
HSCT in Icukcmia using HLA-mismatched donors Table 2. Characteristics of cond itioning regi mens used prior first HSCT (n
= 24 patien ts)
Fract ionated total body irradiat lon (rTB I) rTBI 14.4 Gy rTBI 12 Gy Oral busulphan (Bu) Bu 16 mg/kg UPN 153
Fludarab lne (Flu) 2 Flu 160 mg/ m
Cyclophosphamide 120 mg/kg Etoposide 60 mg/kg Cyclophosphamide 120 mg/ kg Melphalan 140 mg/ m2
9
Thiotepa 15 mg/kg
Characteristics ol condllioning regimen s used prior first HSCT (n = 2 patients ) UPN 200
150 mg/m2
Fludarabtne Melphalan Cyclosporine A Methotrexate rATG Fresen ius rTBI Etoposide Cyclosporine A Methotrexate rATG Fre se nius
UPN 245
140
3 mg/ kg LV. 10 mg/m 2 50 mg/kg 12 Gy 60 mg/ kg 3 mg/ kg iv 10 mg/m 2 40 mg/kg
Ta ble 3. Characteri stlcs ol primary grafts. engraf1ment and graf1 versus host disea se
NC/kg bw x 108 median/range C034+ cells/ kg bw x 106 median/ range C03+ cells/kg bw x lOB median/range engraftment of ANC >0.5 x 109/1 med lan/range engraf1ment ol platelets >20 x 109/1 median/range engraftment of platelet ~ >50 x 109/1median/ra nge aGVHD none or I aGVHO ll/ IV aGVHO none or Ilmiled aGVHO extensive alive/well allve wilhoul event
PBSC (n = 131
BM (n = 11)
12 (57-291 8 (5-1 61
4. 3 (2 6-6 61 5 (24- 11 8)
37 (06-158) OJ (02-071
Tota1 (n = 241 6J 6 1.15
16 (12-221
22 (19-271
195
23 (16- 941
28 (19--411
24
23 (16-1661
28 (19- 100)
24
2 (11)
2 (91
9 (5)
8 (2)
II
9
mg/mz
4 20/24 (83%) 17 7/ 23 (30%) 19 16
able patients was 38.7%; 8.7% experienced 1imited and 30% extensive, respectively (Tables 2 and 3) . Relapse rate. prevention, and treatment
Leukemia relapse occurred in five (hematological in 4, cytogenetic in I) out of 24 (20.8 % ) patients 309- 544 days after HSCT (median 410 days) . Until now, two patients died as a consequence of leukemia relapse at a median 134 days after relapse confirmation. Eight donor Iymphocyte infusions directed according to the level of M RD post-transplant or mixed and increasing chimerism were givcn to four patients (CML, AML,
(0-70-3) (0-2) (start O-I) (0+ 1. +3. +6) (0-20+21 (0-40-21 (0-11 (start 0-11 (0+ 1. +3. +61 (0-40 - 11
AML, ABiL) 120-505 days after HSCT (median 246 days) in doses of I x 10 5 to 5 x 107;kg CD3 + cells (median 2.5 x 10 6 ) . In two patients (AML, ABiL), it failed to prevent hematological relapse (18-20). These two patients, subsequently, underwent high-dose chemotherapy, achieved CR, and were retransplanted with PBSC [rom the same donors as before (III and 155 days following 1st HSCT) (Table 2). One patient remains in remission; the other died of regimen and early GvHD-related toxicity. Two other (AML and CML) continue in compJete remission 26 and 72 months following the last dose of DLI. Infecti ous complications
In 16 out of 24 patíents (67%) , reactivation of CMY, EBY, BKY, or AdY was detected . The common pathogens include CMY in II paticnts (46%) with one patient who died due to CMY pneumonia; EBY in eight patients (33%) with one patient who died as a consequence of EBY Iymphoproliferative disease; hemorrbagic cystitis in rour patients where BKY was detected in urine in all four, and three patients, where AdY (serotype 31) was detected in blood, but none of them deve\oped c1inical symptoms dcspite no therapy was given. Fungal infections were not frequent in lhis cohort with very high risk of developing IFI. Mycotic pneumonia was proved (Asperf{illus species) in one and pro babic in second patient, in botb only pretransplant. Both werc trcalcd during early post-transplant period with ABLC,
27
E.
Mejstříková,
strana 147
Sedlacek et al. 100 80
~
60
(fl
\
'::i~
n=24
o
40
I I ' •
I
II
40
20
20
o
o o
I
(fl
lL.
W
III
~ 60
I I III
I II
n=24
2
3
4
5
7
6
o
8
2
3
Yr
4
5
6
7
8
Yr
Fig. I. Probability of ove rall survival and event-free surviva l in entire cohorl of patients.
100 80
~ 60 (fl
lL.
w
PBSC (n = 13) (n =1 1)
~
100
---- PBSC (n = 13) -+- BM (n = 11)
-+- BM
80
...
I I I
~
~
+
60
(fl
o
40 20
40 20
P = n.s
O O
2
3
4
5
6
7
p = n.s
O
8
Yr
O
2
3
4
5
6
7
8
Yr
Fig. 2. Event-free survival and overal! surviva l in patients transplanted using bo ne marrow and peripheral blood stem cells.
which was later switched to oral voriconazole. None of them suffercd from reactivatíon of I 1. Al! other pat ients receíved prophylactic oral suspension of ítraconazole. In patients with extcnsive chronic GvHD , we usually used prophylaxis with oral voriconazole. No patient ín this cohort díed as a consequence of fungal infection and that was also confirmed on autopsies. QoL followlng HSCT
Patients on no continuous 1ST are usu a lly fully active, have no Iimitations, and continue with normallífe. Out of seven patients who did suffer from chronic extensive GvHD, one died with activc extensive cGvHD, one is on CsA tapering with no signs of cGvHD and five are off any 1ST (Table I). So far, it appears that there is no difference in QoL when comparing this group witb the other patients transplanted at our institution using 9- 10/10 HLA-matched UDs . Overall outcome
EFS is calculated from the date of transplantation to tbe last follow-up or first event (death or
28
relapse of the primary disease whatever occurred first). Probabilities of EFS and OS are estimated using the KapJan~Meier method. Median followup till first event or last follow-up of to tal cohort is 1.57 yr (range 0.2~7.2 yr). According to the type of graft, median follow-up is 1.65 for PBMC (range 0.85~5.8 yr) and 1.24 for BM (range 0.27.2 yr). The probability of EFS was 60.3% (95% CI 35.5~78.1) and OS was 74.9 (95% Cl 49.1~ 88.9) (Fig. I). EFS of BM group was 56.8 (95% Cl 21.3~81.3) and OS was 68.2% (95% CI 29.7~ 88.6). EFS of PBMC group was 60 .6 (95% Cl 24.7~83.5) and OS was 77% (95% Cl 31.2~94.3). Difference between PBMC and BM groups in EFS and OS is not statístically significant (Cox~ Mantel test) (Fig. 2). TRM at day + 100 was 4% with overal! TRM 12.5%. Altogether, five patients died at a median of 1.54 yr post-transplant (range 0.2~1.65), two (8.4%) died as a consequence of leukemia relapse, one of CMY pneumonia (0.68 yrs), one of EBY lymphoproliferative disease (0.2 yrs) , and one of gastrointcstinal bleeding because of extensive GvHD (1.65 yr).
E. Mejstříková, strana 148
HSCT in leukemia using HLA-mismatched donors Discussion
Improvements in HLA typing at the al1ele level, wider spectrum of efficient drugs for GvHD prophylaxis and therapy, and prospective PCR quantitative monitoring of viral load have led to decrease in TRM in patients transplanted from HLA-malched UDs. Which alternative donor is better in the case of no available HLA-matchcd donor remains to be resolved. Different centers have differenl preferences mostly based on local experience. Many centers successfully use unrelated eord blood where the na"ive immune system permits reduced stringency of HLA and, therefore, wilhin the acceptable level of mismatch, it is possible to find a suitablc donor for the majority of children . Its wide use although is limiled by the efficient cell dose availahle for older children and adults (21, 22). This can be overcome by using double cord transplants (23). Other disadvantages include naivete of immunity against viruses and unavailability of cord blood for potential adoptive immllnotherapy or re-transplantation. Use of haploidentical family donors is possible, but large and frequent experience of cooperaling laboratories for preparation of T-celldepleted graft is essen tial to limit the risk s of non-engraftment and GvHD. High incidence oť viral infections in the early post-transplant period increases the risk of TRM. Therefore, this method is more restricted to several eenters with large experienee . New techniques of depletion may improve the immune reeonstitution and graft vs. leukemia effect without the enormous risk of serious GvHD (24). We present our experience with another alternative. HLA-mismatched UDs were prospectively selected ba sed on level of allele mateh. We and others speculate that allele Ol' antigen mismatch is equal1y ad verse to survival. In our clinical experienee, UD with upto three allele mismatches (no more than one in loci A*, B*, or DRB I *) could be used for a patient with malignant disease wilh acceptable risk for toxicity iť adequate serotherapy is given together with a myeloablative conditioning regimen. On the contrary, we speculate that the practice still used in many eenters to select donors based on HLA match in A *, B*, or DRB I * loci, with no respect to numbers of potential mismatehes in Cw* and/or DQB1* loci, is not efficient. Such attitude may explain inferior outcome results compared with those achieved in our cohort. It is not rare to have many Cw* and / or DQB I * mismatches even in donor-recipient pairs, other vice allele matehed in 5-6/6 in standard A*, B*, and DRBI* loci.
ln our series we did not observe engraftment problcms. All patients experienced primary and stable engraftment with full donor chimerism. Reappearance of mixed chimerism was detected only as a consequence oť emerging relapse of leukemia . Highly incident acute GvHD mostly of grade II was manageable by standard corticosteroids. Rate of leukcmia relapse as well as the incidence of falal viral infections was low. We speeulate that lhc dose of rATG given in our cohon of patienls is safe in preserving the graft vs. leukemia effect (high incidence of acute GvHD , low incidence oť relapses), and is efficient in protecting the patient against moderate to severe GvHD without incrcasing the risk of post-transplant fatal infections (very low TRM). There is consenSllS that matehing of UDs and patienls for HLA c1ass II allclcs improves the outcome of HSCT. However, the significancc of HLA class I allelic mismatches for transplant outcome is under ongoing discussion, and reports on long-term effecls like chronic GvHD are rare. Some studies, especially published earlier, are biased by the fact lhat HLA typing was nol performed by PCR methods al high-resolution level (four digits) in all typed alleles. Other studies are biased by different proportion of minorities relevant to the different incidence of certain HLA alleles among patienls and/ or recipients (25). Serologically undisclosed HLA dispari ties account for the increased rate of post-lransplant complications. Whereas, a HLA-ABDR-serologically identicaJ donor can be identified in the International Registry for> 90% of the patients. only upto half of them can benefit of a highly compatible donor if donor selection is based on allele level matehing for HLA-A/B/Cw/DRB I / B3 / B5/ DQBI loci among the Caucasian population. Most of the ineompatibilities are cJustered in a limited num ber of serotypes that can be targeted first d uring the searches. Because of linkagc disequilibrium (e.g., B-Cw or DRB l-DQB I), incompatibilities at a given locus are often associated with disparities at adjacent loci (26). Schaffer et al. ha ve pu blished an analysis of outcome in 104 donor-recipient pairs, transplanted in between 1988 and 1999, retrospectivcly typed for HLA class I and elass II by PCR-SSP. They eoncJuded that genomic HLA class I and class II typ ing may improve the outcome after unrelated HSCT and also that the awareness of HLA class I and II mismatches, not detccted by older methods, in a recipient-donor pair makes it possible to give appropriate pre- and posl-transplantation treatment (27). In addition, olhers investigated the association oť HLA class I allele
29
E. Mejstříková, strana 149
Sedlacek et aJ.
mismatches and outcome. In cohort of 144 patients givcn a HSCT from an UD who were matched for HLA-DRBI, DRB3 /4/ 5, and DQBI alleles the risk of chronic GvHD was significantly increascd in patients with c1ass I-mismatched donors (mismatch either detectcd by low- or high-resolution typing). A single HLA c1ass I allele mismatch significantly increased the risk of chronic GvHD in multivariate analysis. OS was significantly reduced in patient-donor pairs with more than one allele c1ass I mismatch (28). On the contrary, Duggan et a!. reported 57 patients receiving UD HSCT and matched for the disease and stage with other 57 recipients of genotypically matched RD HSCT. AI! UD recipients were matched serologically for A and Band by high resolution for DR and DQ antigens AII patients received CsA and short-term MTX . UD HSCT recipients also received rATG (Thymoglobulin) over three days pretransplant. They concluded that UD HSCT recipients matched as above, and given pretransplant ATG have similar outcomes to recipients of matched RD HSCT using conventional drug prophylaxis (5). Based on our results and together with information published so [ar (29), we also conclude that high resolution of HLA al!eles, both c1ass I and c1ass II plays an important role in the selection of a suitable U D. However, when a fully matched donor is not available, we show that GvHD prophylaxis with use of rATG enables the use of an unmanipulated, partially mismatched donor without excessive risk of poor outcome because of severe acute GvHD. Several differen t brands of A TG are a vaila ble, and therefore, when using ATG in conditioning regimen, one needs to consider the ATG brand , the adeq ua te dose a nd the proper timi ng. Exact correlation between different brands is not clear yet as they have different activity against different populations of cells. Also the dose may vary based on the type of donor and conditioning regimen . Lower doses (Thymoglobulin 6-10 mg/ kg total; rATG Fresenius 20-25 mg/kg total) are currently used in reduced intensity conditioning regimen when fully matched donor is used (30) or in patients who underwent SCT using T cell highly depleted graft from haploidentical donor (31). Much higher doses (Thymoglobulin 15 mg/ kg total; rATG Fresenius 60 mg/kg total and more) are used in transplants using mismatched donors and unmanipulated graft (32 , 33). Conclusions
Our study shows that combination of CsA, short-term MTX and rATG in GvHD prophy-
30
laxis prior to HSCT using unmanipulated grafts of HLA-mismatched UDs is efficient to prevent occurrence of very severe acute GvHD grade 111IV. Serotherapy (r A TG Freseni us; Freseni usBiotech) was well tolerated and in doses given (40 m g/kg total) did not increase post-transplant mortality by long-Iasting depression of immunity or increase risk of leukemia relapse . Incidence of chronic GvHD was not increased compared with series where fully matched UDs were used, more over in majority of patients it disappeared within the time without serious consequences (Table I). Overall outcome is satisfactory, and therefore, it is possible to use such alternative donors in patients with advanced leukemia lacking a HLAmatched donor as a reasonable alternative to unrelated cord blood or haploidentical family donor. We strongly recommend to extend HLA typing for clinically relevant Cw* and DQB 1* loci in centers stili using only A*, B* and DRBI * for selection of suitable donor. Acknowledgments We wish to thank our eollaborators from CPH (Czeeh Pediatrie Hematology Working Group) for referring patients, nation a l registries of donors in Pilsen and Prague, HLA laboratories namely in Institute of Hematology and Blood Transfusion a nd CLlP (Childhood Leukemia Investigation Prague) in Prag ue. We thank David Jaeo bsohn for editorial help. This work was partly supported by grants CEZ 23736000 I and VZ 00064203.
References J MIANO M, LMOPIN M, HARTMAN N O. el al. Haematopoietic stem cell tran splantation lrends in children over the last three decades: A Sllrvey by the paediatric diseases working party of the ElIropean group for blood and marrow transplantation. Bone Marrow Transplant 2007: 39: 89-99. 2. Kl.l. G~B I EL T. HANDGRHINGER R, LANG P, BADER P, NIL'TIIAMM~ R D. HaploidenticaJ transpl an tation for acute lymphoblastic lellkemia in childhood Blood Rev 2004: 18: 181192. 3. LA NG p, GRElL J, BAD~R P, et al. Long-t erm outcome after haploidentical stem cell transpl an tation in children. Blood Cell s Mol Dis 2004: 33: 281-287. 4. DEY BR, SPITZLR TR. Current status oť haploidentical stem cell transplantation. Br J Haema tol 2006: 13 5: 423--437. 5. D UGG AN P, BOOTH K, ClIAUD[IRY A , et al. Unrelat ed donor BMT recipients given pretransplant low-dose antithymocyte globulin have outcomes equivalent to matched sibling BMT: A matched pair analy sis . Bone Marrow Transpla nt 2002: 30: 681686. 6 FIN KE J, SCIIMOOR C, LANG H, POTTIIOI·. K, BERT[ H. Malched and mi smatched allogenejc stem-cell transplant:ltion from unrelated donors using combined graft-versus-host discasc proph ylaxis including rabbit anti-T lymphocyte globulin. J Clin Oncol 2003: 21' 506- 5 13. 7. PRZEPI ORKA D, W~IS l)ORF D, MARTIN P. et al. 1994 Consensus conference on acute GVHD grading. Bone Marrow Transplant 1995: 15: 825-828.
E.
Mejstříková,
strana 150
HSCT in leukemia using HLA-mismatched donors 8. SII ULMAN HM , SULLlVAN KM , WEIOLN PL, et al. Chroni c gra ft- versus-host syndrome in man. A long-term c\inicopatho logic study of 20 Seattle patients. Am J Med 1980: 69 : 204217. 9. CARPlNTl'I\ PA , S,\N OI,RS JE . Steroid -re fractor y graft,vs.-host disease: Pas t, préscnt and future. Pedi a tr Tran splant 2003: 7(Suppl. 3): 19-3 1. 10. SIMPSO N D. New development s in the prophylaxis and treatmen t of graft ve rsus host disease. Expert Opin Pharmacother 200l: 2: 1109-111 7. II. VOGlLSANG G B, ARAI S. Mycoph enol a te mofetil fo r the prevention and treatmenl of graft-versus-host diseasc foliowing stem cell transp lantati on: preliminary findín gs. Bone Marrow Transplant 2001. 27: 1255-1262. 12. B,\OLR p, BL 'K J, FR LY A, et a l. Seri,,1 and quantitative anaIysi s of mi xed hematopoietíc chimerism by pe R in patient s with ac ute leukemias aliows thc prediction of rel apse after allogeneic BMT Bo ne Marrow Transpl ant 1998: 2 l: 487-495. 13. KRU CI O, VAN DER VLW I'N VH , B,\ IlLR P, et al. Level of minimal residual discasc prior to haemato poietic stem cell transplantation predicts prognosis in paedi atric patients with acute Iymphobl astic leukaemia: A report of the Pre-BMT M RD study group . Bone Marrow Transplant 2003 : 32: 84985 1. 14. ECKLRT C, SCRIDLU A, TA UUL T. et a l. Compari so n between TaqMan a nd LightCycl er technologies for quantificat ion of minimal resid ual disease by using immun oglobulin and T-cell receptor gene s conse nsus probes. Leukemia 200 3: 17: 25172524. 15. KI MURA H, MORITA M, YABlITA Y, et al. Quantitalive a nalysis of Epstcin- Barr virus load by usin g a real-time PC R assay. J Clin Microbiol 1999: 37 132- 136. 16 . TANAKA N, Kl MUR A H, IIOA K, ct al. Quantitative anal ys i, of cytomegalovirus load using a rea l-time PCR Hssay. J Med Virol 2000: 60: 455-462. 17. PHIi'TAKIS P, BOGDA NOVIC G . KOKIIAU P. M ~LL~-rWT H , DALl ANIS T. BK virus (BKV) quan ti fi ca tion in urine sam ples of bone ma rrow transplanted patients is helpful for d iagn os is of hemorrha15ic cystitis, allhou15h wide indi vidua l variation s exisl. J C1in Virol 2003: 26 7 1-77 . 18. BCCK JF, KLlNG~O ILL T, K RI!Y~ NBERG H. et al. [Relapse of childh ood ALL, AML and MOS after alloge neic stem cell transpla ntation can be prevented by donor Iymph ocy te infusion in a critical stage of increasing mixed chimerism]. Klin Padiatr 2002 214: 201 - 205. 19. SANCiIU. .I , SmH ,\NO J, GOMI.'Z P, et al. C1inical va lue of immunological monitoring of minimal residu al disease in acute Iymp hoblastic Ieukaemia after allogeneic transplantation. Br J Haemalol 2002: 11 6: 686-694. 20. YOSIIIMI A, BA DER P, MATrIIES-M AHTI N S. et ul. Donor leukocyte infusion ufter helJ1atopoietic Slem celltranspl an tation in patients with ju venile mye lomonocytic leukemia. Le ukemia 2005: 19: 971-977. 21 WALL DA, CARTIR SL, K~HNA N NA, et al. Busulfanjmelphalan jant ithymocytc globulin followed by lInrelated donor cord
22.
23.
24.
25.
26.
27.
28 .
29.
30.
31.
32.
33.
blood transpl antatio n for treat ment of infan t leukemia and leukemia in young children: The co rd blood tra nsplantation stud y (COBLT) experience. Biol Blood Marrow Transplant 2005 II . 637- 646. PARKMAN R, COHEN G, CARTER SL, el al. Successful immune reeonstitution decreases leukemic relapse and impro ves su rvival in recipíenl s of unrelated cord blood tran sp la nta ti on. Bi o l Blood Marrow Transp lant 2006: 12: 919-927. BARK U, JN. Who should get cord blood transp lants? Biol Blood Marrow Tra nspl an t 2007: 13(Suppl. I) 78-82 . LANG P, SCIIUMM M, GREIL J, et al. A eomparison between three graft manipulation methods fo r hap loidentica l stel11 cell transplanla ti on in pe
31
Delection of residual B precursor Iymphoblaslic leukemia by uniform gating flow
Abstract
cytomelry
Residua! di seaSé (R O) is 3n impo rt ant prognoslic faclor in acutc Iymp ho blast lc leukemia (AL L). Flow cy tomctry ( FC )· ba.sed RD d erectlo n IS easy to perform. bUl interpret3tion req uire s ex'pert 3nalysis d ut:' to indi\IiduaJ
Running lilie: Leukemia early response by flow cylomelry ES ler Mej střikoval ,l. EV3 Fror1kovó"!. ~omáš Kal in a "", Marek Omelk a'}·IO, Drafů Bafin ,e. Kl ara Dubravč ie , Klára Po sp iš dová ' J, Man ill3 Vaškova l., Drnrit Luna", A lice Suk I-Iang Chen g', M argaret Ng~, Yonna Leu ng \, Janos KappelmZlyer6, F lora KIss!", S hai luae l, 1, Balia Stfirk\ M art in Sc hrappe' , Jan Trk a l 1, Jan St3ry 2, Ondřej
differe nces 81ll ong pati ents. We focused al the desI g n of stand ard rzed and rep ro duci ble RO m oni tori ng in ALL
RD was investi gMcd by a uniform gnt lng strategy. whic h was dcsl gn ed internationa lly a nd tesled in one cen ter by IglTCR rearrnngeme nts. For each &a ft:, positi \" ity cut·off value was a s:signtd usin g qu antifu::ali o n of non·
Hrusák l ~
}CLlP - Ch i/dhood Leukemia In ~'esljgalioll Pmgue. ~Departm enr oj Pediatrie Hematolog)' and Onc% gy 1,,·/ Medieul School. Uniw>niry Ho.fp i/al Moto/. Pragut.' Czech Repuh/h:, JClinieal/n slitutl' o/Labom/ory Diag nosis. Ull i\'ersi/y Clinlea/ Hospiwf Cen tre. l.ogreb. C,·OOl io. ~ Cenlre o( Pediatrie He mal% gyIOncology. Schm!idcr Chi/drcn's Medical Cemre oJ/sroe/. Petah Tk/l'a (Oepurlm elll oj Anarom h:a/ olld Cel/ular POIhology. Prin ce oj Wales HUjpiral. Nmzg Kong. ~Df.:'par/mellf ofClinica l BiochemiJ"II) ' al/d Mo /ecu /ur Potlwlogy. Medh.:a / ulld f/ ea /rlt Sdenees Cenlrc. U"ú'ersiIY o{Ot.'hrt'ét!n. HUl/gal)'. ~Ped;atric H emOl%gy t.ll1d Ollc% gy . Scheba Mt~{I/(:aJ Cen1C~r. Te / Haslro mer. Israel. ~Oťpa rrll1 e,,/ o!Pr:diatr;c.\' . Uni\,(~".i"ily f/ospi/o l Sclde.su·ig ~J-lolslein. Klel. Cerman)'. 9 JaroslOl1 Háj ek Cen/cr f or TheorClica/ and App/ied SlUlis/ ies . F'atultJ' ofMathemarics and Physics. Charles U"i"''er.'ri"IY. Progllf!. C=eC'h Republic Correspon d ing BUlhor '
Ond rej Hrusak tel/fa." +4202244364771641 3 (llhlr~J. J1[I.ls:1k ~lClfmQtQI cunl q. Oepar1mcm ofPcdlatric He mato lo gy s nd Oncolo gy nd 2 Medical School and Uni ve rsity Hosp ital Motol V Uvalu 84 Prague C.zec h Republi c
Acknowledgements Wc thank J. Ridoskov a, 1.. Go ndo re mova.., O. Thť.Jrner. P. Scmerak. K. ,\f u7.il.;ova. L Reznickol/u.. T S koeQyska for tce hn ical as$lstan c e in pro vldmg RD stud,es . 1\ 1. KOV Ole and V. Pel ko\'<J. helped us w lIh c arc ťul readmg o f" lhe man usenpt A Lub helped us wI lh des ig n ing. th e prcsentati on 01' sup plcrncnl al data Co ll abo ration ofCzec h Ped Iatrie He mato logy (C PH) ecnt ers (da ta man ag C;.~rs A . Vrza lo\'a" K. KrRma í.lOva, leJders : B. Blazek (Ostrava ),
Z. Cema (Pl zen). Y. Jabali. (C<ske Bu d_Jo vlce). V.
~'lJh al
(Olomouc ), D. Proc hazko". (USll nad Labem), J.
SIC' rba (8rno ). J. Hak and K. To uso vska (1 lrndec K.ral ove) IS bigh.ly apprcci ated . T hc ,,",o rle was su ppo r1ed by ~rants
VZMSMT MS MOO2 16208 13. IGA MZ NR8269'3i200S, N R/9SJ I-31200 7. GAU K 7543/2007 and by Ihc Israch Cancer ASSOClallo n. T K. ,,'as suppo rted by GACR 30 1106tP1 62
N L(j
,....... ~
~\..o
....
'Jl
,~~
;;-
o ;r:: .... ..!G
'Jl
'v ~ ~
leuk emie background. Co mparing to IgrrCR al 0.1 % level , 83 of 103 spccirnens were correcll y diagno scd by FC Th e predi ctivc valu c of Fe RD at day I S was then analyzed ln 8 l1n cage :\LL. da)" 15 FC sig niticantly corrclatcd wl th IglTC R res ults at day 33 and/ar weck 12 (p
PCR~dele c l Ab l e
M RD in 8 precursor ALL.
frequencies in speci me ns w hose M RD negativi ty was evident either becausc Ihe pat lt~ n ts suffered from ALL of thc o pposit e IIncage (T versus B) or becaus c ;\,t RD was p rovc n
n egatl v ť"
by PCR
Introduction Thc s peed of leuk emia c learance dunn g Iherapy is a major prognostlc faCIOT. R ťs l duaJ dlsease (RD) dunng the carl y rhase of th crapy can be assessed by mi croscopy (I) or by more sensitive tec hniqucs, such as n o ....'
Patients and methods
cytometry (Fr) a r po l y m e r a~ť chain rcacti on (peR) ofthe rearrangements ofim munoglobulin (lg) or T ce ll
Patients receptor (TCR) gen e segm ents (2-4). The strong prcdictivc value of RD i5 a lso ll1aintalned a1 thc level of minlma! RD (MRD) durin g com plete rcmi ssion (CR) (5-7) . Severa l strateg ies are uscd w hen utill zi ng
Patien ts were recflJited from the A.LL IC BF~t:! OO2 protoco! (staned in Xl/2002). Th c prolOcol Vvas dC: 5ig.ncd fo r
in formation rrorn previ olls r~tro spccti\'č studi es: slratili cation is based o n MRD Icvel rrom eilher a single lime-
co untric s \\Iith c1inic
point (8.9) ar consec uti ve time-poinl s (7, 10). Real -lim e quantitati\."c (RQ) PCR·based detection ať specific 19
tcc hniques . MRD
and TCR ge ne rearrangeme nls represenlS the CUlTen! gold stan dard for MRl). Standardizcd RQ PCR deleclS
PCR) treatmenl protocol AIEOP- BF M ALL 2000. Laboratori e~ rrom the Czech Re pu b hc, rsrael , Croalia.
MRD bOlh specificall y and reproducibly, wilh a delcction limit of 10·-1 to JO.s Ncv enhclcss. s uch PCR me l ho d s
Hun gary and Hong Kong parti c ipated in tl1e
are laborious and coslly. Flo w cytorlll..: lry is a met hod wit h ac ~c ptablc sensiliv ity and w id ť availabi lity in many
guardians signed InfoTrned consent, and thc ::;wdy was approvc d by th e insUtulÍonal ct hlc s committee . In tOlal ,
ce nterslco umries fo r hcmatological and immunologi cal invcstigul ions. \VhiJ e
~
Fe invcsligales anligcns o n inl acl
t::
...... Vl
'<"l~
:>
O
~
>1-.
......
Vl 'či)
~ ~
Fe j\tRD study . Patients cntered the study atler Iheir parent~ or
11 0 patien ts wi,h newly di"gnosed ALL entered ,he st udy (90 Bep ALL and 20 T ALL). The eohort reported
cells. in large lrials co vc ring $Cveral conl incnl s, il lS im po ssi ble to concentr3te MRD d iagnoslics inIo one lab Thc interprctation ofFC data \5 compllcated anu rcquires hi gh ly experienced and skillcd cxpens. Therefo re. II is
of the patienls ' ch:lracteristics is s hown in table I. The diagnosis of ALL was establ is hcd acco rding co nvc nli o na l
imponant to set clear definilion s and slandards, espec ially for inter-Iah studie s. The unfortunate reality is lhat
FAB and immunological crileria. AII samples that mel th ese criteria and were successfull y a na lyzed bolh by FC
CUlTenl publishcd papers us uall y lack detailcd mformation on ~atlllg str aJ:e~te s due to dtfTcrcnces in
and PCR were included in the slud y ln 101aL 6 12 ~ npl es wece centrall y cva luated (dl agnosis : 110, day 8 bone
interpreta lion arnong patienl s On ly a ne published study by Coustan·Smllh et aJ. cva luated two sirn ple
marr ow(B~I):
prcde tined subpopu lation s. C D 1 ~CDI 9r- and CDI <j>C~C D341"" . .u1 il single tj me-point
parienl s wť' rc excluded as ha \'ing an un s uiwblc imrn u nophcn otypc for Fe RD monitori ng
Oll
day 19 (a ti me -poi nt
10J , day 8 peripheral blood(Pll): 8J dal' 15
coJcu l ,u i~ ns.
bc fo re thc achicvcmcnt o f complctc rcm issio n ). Th u.s, lakln g into accoun( d,Ocrcf1ccs among patle.nts, it is
For bac.kgro und
diff1cuh or nearly im possible to rl.:p roduce FC '\'1RD d la~n oSllC!!l . ThlS< IS e.speclíl ll y true for ccn lcr s!coun tfl e$ lhal
vice versa (c ro ss
are j ust staru ng aut wlth
Fe j\·tRD eva lualion. Th us, \"'c d esl~ned a s tudy ln both l3 precu rsor 3Jld T .1\ L L paticn15
li nc a~ e
fI~ l :
108 , dal' J 3 BM: 108, "'eek 12 BM: 100) No
49 s pecimens of B lineogc pati ents wcrc meas urcd w ith T Ilneage co mbi nn ll on s o r
conl(015)
Risk grouping, treatment and time-points
thm e\"'l.Iluated ldenl lcal. p rc ··dcti ncd su b po pu lallo n sl~tes. regard lcss o f lhe in iti al jmmunop~en ot ~pe. w~
<"l <"l I-.
not used for paticnt stratitication Th ls ran in pa rallel to the MRD·based ( lgrfCR RQ
he re containcd patients lreated in 8 ('ze'ch center s, anal yzed and acquircd in a singlů Fe laboratory. A summary
cr) ,..-<
W3S
collected cell freq ucr1cies
ln
eat h ofthcsc ~at es nt indi v idual l"i mc-po int s. The gates werC' mad c up of ant igen
cornbi nalions co mmo nly used in Fe RD studies ln tot a!.
w(:
dcfi ncd 29 g.ateslsubpo puhnions in B ccll precursor
(BCP ) acul C Iymphobl astic h:ukemitl ( A l L.) and ti ve subpop ul at ions in T .~\ LL. In th is stud) . v.c Included the sam ples wllh a\." ada bl~ RD usmg bOl h
Fe an d
PCR~ whlch enabk-d an I!xacl asses.sment
('I f ho lh mct hods for
MRD cval u81 10n at indi vidualllm e- po lOls . T hc object lve o fthe study was to dC\·cl op a roh u~ t M d rep roduc ible FC RD approach for ch ild hood AL L. cVťn for centcrsf.co untri es wllh linuted experiencc in and 10 prcc ise\y
quan tiťy
Fe
M RD analy sis.
the backgr ou nd rH·escnl dunng Iherap y Thís backg roun d was deri \"cd fro m cell
PauC'nts wtre 8SSIgncd 10 th e st an dard ri sk group (SR) when th ey ťul f1 l!ed Ihe folloV<.l ng cnlen a agc < 6 years, im lt alleukocytosis <2 0000j ~I L . good prcdnl:llone Ics:pon sc « 1000 bltlSl slJ.1 L)
ll!
day 8
ln
periphera! blood .
ab se nce of BCR/ A8L and MLU/\ F4 fuslon genes. and aChH!\ mg a CR al dlly )3 . Paticn ts who fu lfdled a l1 of Ihese SR criteri3 and had mo re Ihan 25° 0 blaslS in BM by microscop) a l dny I S were strati fi cd into thc Inlermcdi att risk group (lRG). Other pa1iems Wil h mo re than 25'% b lasts ln B ~ I by nllcrosc.opy J r d a ~' 15
""t;T e
strnt illed int o th e hlgh Ti sk grou p ( HRG). Olher HR O cntc.ri a weíf BCRlA BL or MLlJ '-\F-I- fusiol1 genes, poor
responsc to pred ni so ne at day 8 (>1000 blas t s!~L in per iphera l blood), a r >; %) blasts al dlly 33 in BM . AlI
rcmalni ng patienls were .a.sslgned to the IRG
CO 19 PE (elone SJ25C I), COlO FľfC (clone l2 í) , CD45 PerCP and APC (clone 2DI) and C f), 9 FlT C (cio ne
Alltime-points show n here retl ec[ rt!rn iss ion induction therapy (day 8. day 15. day 33 and wee1c 12). Remission induClion therapy \Na,:; sch~d u l ed over 9 \""ťcks a nd induded a 7-day steroid prc phasc \vit h dai ly o ral prcdnisone (60 mg/ml ofbody surfa~ť area dai ly , c umul ati ve dase of prcd ni!ionc al cl ay 7 had 10 be g rc31e r Ihan 2 \ O m g/ml) and a s ingle dose of intratheca\ mct hot rCx31C (agc aJJusled) on da)' 1, followed by predni 50 ne (60 mg./ m J dail y) from day 8
10
Monoclonal anlibodies
day 28, t3pcrcd thereafter over 9 days . It also included 8 do ses of L-as pa ragm 3se (5000 U/ m 2/day
on days 12, 15 , 18 , 21 . 24, 27 , 30 and 33), daunorubicin (30 mglm} on days 8 and 15 ,
( WO
o lher doses recei ved
TO 12) were purchJS cJ from BO Biosciences (San Jase. CA. OSA ). CDIO FIT C (cione A l.Il 2), CD7 PF (c hme 8H8 . 1), C03 PCSIP C7 (cio ne UCl-rf l) , CD5 PCS/ PC? (clone BLlaJ. COl9 PCS / PC? (c1011e .l4 119), C034 APC (clo no L1 Q), CD58 FLTC (clOIl(" MC DC D58) .nd C066c PE (clone KOR-SA3544) were pllrchasc-J from Im m unótéch (Marseille, France) . COlO PE (clone SS2/36) and TdT flTC (clone HT-6J we re purchast:d from DAKO (Gl oSlrllp, Oenm ark) . SYTO-16 (grccn tluorescent nucleic aci d sta in) was pur c h n ~~ J from In vilrogen Molecular Probes (Carl sb.d. CA, USA)
JR and HR patients at day 22 and 29), vincristin e (1.5 m w m1 on dJys 8. 15,22, a nd 29) and two doses of
inlra theca l melhotrexa te (ag('-adJ usted) o n days 12 .and 33 (2 additional doses o n Jays 18 and 27 were
Dala analysis
ad mini slered in cascs "" il h leukemlc CNS invol vc ment or with traumatic lum bar puncture). From da)' 36 to 64, Ih,,' rcgi men included : two doses of intra thccal mt;t hotrcxate ( ágť -
Oll
days 36 and 64). c)' lar3bme in 4 blocks (75 mg/m ' dail y o n days 38-4 1,45- 48 .
5~-55. _nd 59 6 2) and 6-me rcap lopurine (60 m g/m ' dat ly fro m day 36 10 day 62). Palien" who had aplastic BM
at day 33 underv.rcnt another BM punctu ré onc week Jater (a samp le wjlh com plet e remi ss ion includin o rcge ncralion in BM
\\:
analY L.ed
3S
day 33). Palle nts ",,'ha rcce ivcd a non-ALL
le therapy
bet\vecn day 33 and
Gati ns defi nitions and strategles a re shown in flgures I and 2. Unifo rm templales were designed in thc s oft v.:a rť'
appl11.:ations, Cellquest (BD, San Jase, CA, USA) an d PlowJo (Tree Star, Oregon, USA). G ale po sit ions were defined acc.:ording
[O
rcgcncratin g BM samples in pati ent s \o,.'ith ncgalive
~1RD
by 19- TCR RQ -PC R and/a r using
QSC bcaJ , (Quantum Simple Ce llul " r. Bangs Labs, Fi shers, IN, USA). In all B line age and T lin cage mAb co mbinat io ns. the mAb reacting w iTh a lineage-dctinlng anti gen (CDI9 and C07, respectively)
W 3S
present. In
F1J
wcck J:! (4/6 BC RJ:-\BL positi\'c paticnls rcccivcd imalinib I1lcsy Itne 3ccQrding to thc EsPh i\LL prolOcol : one pallC n! rece ivcd thcrapy acco rd ing to th c [nterfant 2003 prolocol after da)' 33) were ex cludcd fro m lhe E1 na lyses
all RO subsels, the target populalion was ca lculated as a fr action ofn ucleated cell s (SYTO I6 <).
Reported subpopulalions
thal compnsed wcek 12 . Fu rt her Iherapy ,\a5 not e volualed in thl 5 S:lu d ~ Th e definilion o f subpopulation s is show n in fígures I and 2. A II subpopu lmions were repoTl ed in
Sample processing
wl1h A LL oft he respecll\'e lin eagc, rcga rd less oflhe dctai lcd prcscn t in g immunophenotypc The truC' fraclio n of
Sam ples we re proce ssed \Vllh ln 2-1 ho un after Ihey were co lJccted from patlents Sa mp lc preparatton co ns isl ed o f
"
a 15-mi nule Inc li b.Hion w lth monoc lon lll Wl tibodies ( mAb) - sam plc-to- mAb volume rat lOS \Vere use d as
lf"l ,.-<
recornmcn ded by the manufactun!rs . Red blood cell s ","erc then I) zed in a 15-minutc incuba ci o n (ammonium
<:\j
c:<:\j
..... ..... V')
chlo ride), followed by 5 m inul es ol' ce ntrifugation (500g t d iscard ing of the supe rn Gtanl . adding of PBS an d
cvenlS ""ithln a Jť'flned regio n per to ta l of Iymph o id li nci1gc cells (CD I
c:o m bmed them \\'Ith thc cros~ lincage cOf"!l ro ls. For cach ,cpcrted \ (!.Iue ot aJ l lime -po inl s. \~C ca1t: ulated Ihť" 98 1'1 1
im medi ate dalaacqui!ut io n Fo r intracd lu lar stai ning. the h x & Perm kit (An Dťr Gruh Biorescarc h. J\USlna)
pt'n::énulc , ln o rder to cnco mp3SS the ran dorn POI sson distnbuliol') o frea l data aro und Ihe 98 perce mi\e oťlhc th
'<:\j-
:>
was uscd t\1I e\'enIS w ere acq ulred and slored in hstmode fl tes. and no
l J \ C'
gat( strat egy
w :1S
useJ . Al d13g.nosis.
o
a mini mu m of 20.000 ev('n ls per l ube we re J.cq Ul re-d . Ourmg RD foll ow-up_ th e t~ct minimum co unlS \Vť: r c
.....
50.000 ev cnl S (SYTO 16 contaJnlO~ mA b com bma l1ons). 300,000 e \ i: nt s ( the mQsI infomHlII\'c mAb
~ > .....
represem91ive su b-cohort, we considered thc three-fo ld multipli catlon ofoS* percentile to be thc l>ackgro und CUI-otT value. T li neage 1,,'alues werc c.QlcuJaled usmg B hncage cro ss Ill1eage co ntrols onl'1 _ A ll
Fr daw repot1cd
V')
'v ~ f..t.i
co m bmatio n) o r 100.000
(! ·..
ents (all o the r lub es)
Q.$
pn._-de:fi ned subpopulat io ns werc co mpared \\Iith the 5ub population an d timc poi nt-spcC llic backg rou nd c ut-oIT
va lues; onl y when thcy V.o-eTe hlgher did we considcred them to be above c ut offwas consid e red an
posltiv ~
The highest
PCP>l~
subpopulation
Fe RD (cut ofTvalu es ilre shown in table 2.)
respectl ve ly The cutoff"alucs rcprcsenting the background of cach repor1ed subpopulatlon are Il st ed in Table 2
Although il is frequentl y assu med that th e non-mahgnant B precursor background is very low ar absent at en rly ti me- painls, eur rcsuhs do shO\\' non-zero background levels evcn 31 da)' 15 , \vhich must bc taken inlO accounl in
RQ PCR ol IgfTCR rearrangements
Ihe data interprclation (sirn ilar situ atio l1 appe,ared ut day 33, JOla nOt sho\\n)
Follow-up BM or PB samples were proccssed by er)' thrcu;:yte l)' :>is and sloTed at -80°C. Genomic DN A \.\'as
Graphs sho wing cell frequencies af cach in divi dua l FC subpopulation a l day 15 as well as a l days 8. 33 an d week
iso lmcd by the QIAarnpljš' DNA Blood Mini Kit (QI AG EN GmbH , Hilden, Gcnnany ), and Ihe DNA
12
" ':1S
stored
ar ~
3\"nil ilblc (s('(: supplernental Information),
at -20°C unlil furthcr proc cssi ng, Rearrangemc nls of irnmunoglobul in heovy chain (lGH) , immunoglobulin hghr
RD evaluation and comparison with RQ·PCR Ig-TCR ch a in bppa (lCK), T cell receptor delta (TCRD), T cell recep tor gamma (TCRC) and T cell receptor beta (TCRB) Vo'crc identified using sin g le o r muhiplex PCR (II, 12). C lonalir)' assessmcnt. sequenc ing, palient-
specific prim cr desi g n nnd RQ-PCR wilh family- spccltic reverse primcrs and probes for JGH , JGK. TCRD,
TCRB and TCRC were performed as described previously (13 - 1S) The albumin gen e was used to normalize DNA concentrntlon an d quahty. The ESG-M RD -ALL (European Study Group on Minimal Residual Discasc in ALL ) crit oria for RQ-PC R scnsiti vity and qu antitati ve rangc \I,i'cre used ( 16). For stati stic.s of cont inuous ...·ariablcs. numerical MRD valucs (reprcsenli ng th c mean va lu e of posili \'c \' alu es within the triplic,nc) were
wen~ p O"S ItIVt.'
also by FC (71 of 81 and 19 of 20
ln
B prec ursor and T ALL, respectively)
Elt
day 15
Swrting al day 33, most of Ihe FC data were below Ihe cutoff values, and the p c r centag~ of PCRlWs specimcns idcntifíed ils FCptI~ drc ppcd profoundly (dat a not shown) Thcrcforc , th~ cUloff- based, four-co lor FC fOCll scd at the cellular subpopulat lOns deseribed here mig hl serve as a basis for Ihe deSign of RD
inv es t l ~lion
bUl
jl
cannol
Prediction ol molecular remission at day 33 and week 12 using FC stratification at
Statistics FC scnsiri" lty ...vas. de fmed as Ihe percen lagc of sa:m plcs positlve by both meth ods amo ng all PCRfk:s samples. FC specificiry was deti ned as thé pcrccn tage of smnples
L/") L/")
When compcil in~ these FC RD data tO Ihe leukemie ce ll fn.:qucncies ass..::-st:d by PCR, majority oť PCR~.)' spccimens
be used as an RD r:va]uation al day J3 and week 12 of ALL tnerapy,
always rcported
Spearman Rank
The FC RD \aluc d;JtJ. represenled the frcquency of the mosl prominent su bpopulation above ilscutotTvalue
corrť'lm ion
n c.!g(H l \ é
by bo th method s ,.mong peR""1 ss mpks
coefticient Vlas used for correlations ofRD betwcen IIldl \'1 dual lime-points and was
day 15 Next, we asked whether da)' I S FC RD could prcd ict RD at da v J J an d \,,'cc k 12 as cvalualed by PCR. Usi ng Spearman ran k co rrelati ons, \lil: cornparcd FC RD va lucs Ine
calcul.ted " ' ''' g STATfSTICA ,otl",are (StatSon. Inc (2006). STATfSTIC A (data an aly sis software system ),
33 and at wee].,; 12 . ln pilllelltS with B precursor ALL, thi s an alys ls Hldi cated a slg ni ficanl corrclation bct'\\o'cen
vcrsion 7 I " 1). .... "j r·wmO sg m , USA). A li !ll:ar-by-lin ea r associatio n mode l (1 7) was used for analyses of
day 33 or weck 12 (PCR) and day 15 (FC) (R O 5q and O j l, p
corrclations betwee n Fe RD stJbgroups at days 8 nnd 15 and PCR RD subgroups al dny 33 and week 12 ln order
p:a ti enlS, whose d.ata dr d nOI
10 fít lin eM by lrTlca r assoc r311On , we used thc R computmg e.nv; ronment (w'\vw r-proJl:cl.ors)
n!specti\ cly ), Thcrefon.:, wC ana.lyzcd the levels uf day 33 anJ week 12 RD (peR) in categorics dc tincd b) Ihe
rea~ h
:!)lgnific ancc
111
any
oť,h ese
IS
in contraSI 10 T A LL
companson5 (R 0 04 and 0.22, p > O 3,
,.......
~
t::
Fe RD
~
~ .....,
Vl
Results
,ti'
FI!;U rC 4 ) A Imear-by-lint.'.ar associat lo n mod el ( i e , a model fOJ" cont mgmcy ts bles where hol h row'S and
> O
Background identification lor each subpopulation
.....,
Specirne ns wn.h
~ >~
colum ns a"8 d 8b l ~
PCR and FC RD an aly scs Sl dOl) 15 were included in the study (n= 104, B precu rsor
Vl
. Ci)
~
fI-i
level at da> 15, e xc l l,lsi v ~ly in palients wi1h B precursor ALL As shown in Fi gure 4, day J 5 Fe RD
C-..1tegories (defincd by ;nler\"als o f FC RD ;n ascendmg o rde- r) correlsted wlIh dny 33 RD b)' PCR (Prll1htr < 10''',
ALL : ""XX, T AI L.. '(.Xx ) ln total. x..:"t.x and xx..-.:. speclmens (xxx cross-lineage control specimens, xxx PC Rpro ven neg3 livc) served 85 .J basi5 for
non- 111 3Ii!:~Jl(lnt
background identdi cation in Band T 11IlC'.ngC' "L L,
ar~
ordered an d the Séores of the rOW!i and t:" olumns are 8" en by Iheirs NlO ks ) ~ ~owed slg.nlficant
a550C1I1110n bclween day 15 and dtl) 33 (p< 10.7• odds ral;o is 2.321 ( 9 5~o Cl 1.603 - 3 360) The odds ratio dcscri bcs tht nUlo of proportlons of P31 icIlIS in co lumn s an d rO'\\o5 of any given 2.-,, 2 region w;lh bOlh day 33 calegories (peR) as we ll as day 15 c31cgones (FC') ordered asccndl/l gly Sirnilar bli t wť'aker correl at ians \!,,'ere
obscn:cd in B precursor ALL p ati~nt s bet\\'!cc n Fe RD in B ~,t ar periphe ral blood at day 8 and RD (peR)
al
later
time-poin ts (da ta not shown) .
th
pa nel values (25 percentile just
CDSI"'"'CD3 nt "
CD99
b
'N IL') ,
O). Simi lar reduc cion appears to be po ss ib le in T -ALL
' ....CD5"'"'
day 15. lf w c rcpo l1
and Td-r- su bpo pul at io ns a f the CD 71""'"' cell S, th e RD ynJue rcmai ns
iden t ical in 18/19 pos iti \'c specimt ns al1d is reduced by a fact or of O.92 in Qne case.
Contribution of individual valu es on day 15 \Ve an3 1 ~' zed w hlch or lhe reco rd ed subpopula lio ns co ntribu!ed to th c
Fe RD ... alue.
Amon g a ll spec lmcns on
day \5 , onl y one af 34 recorded subpopul ation s (C D4 5"CI:C Dl 01NJ-C D66cP""C O 19J».» re mained bel ow thc cutoff
Discussion
value in all cases. In add .1Hl n, sorn e val ues rcuched po si ti\,iIY bUl thc ir le\'c ls \'Jere lower Ih an the maxim um
Study design positivc value in
m isslO g. howe ..'cr, and no co ns\!nsus has becn rcachcd as to \ .... ha t constitutes .ln adeq uate negali 'l:c con trol. LAIPs
pa nd RD valu e Upo n exd udlng th c CD58/C DI 0/C D 19/C 0 3.t tube, thc RO mfo rm at1o n was acce ptabl e in
usua lly desc ribc a subpopu lmion of ce ll s of a gi.... c n lineage al a pa.n icu lar dlfTere nti ation stage w ith (a) abcrra nt th
pcrc t:: nt ile, 1.0) . Further rcd uct io n tO a set of just t\\lO rnA b combin ati ons (CD I 0/C066c1CO 19/CD45 and C D20/C D I0IC DI 9/C D3-4 ) sti H ge ne rated accep tahlc resuh s in 99% specim ens, wh ereas selecti ng, a sct of a ny
O1 he r Iwo mAb co mbin ations reduced th e numbe r o ť spee imen s to fewer than 90%. Exd udin g thc
SYTO I6ICDI 9ICD4 5. CD IOIC D66clCDI 91C D4 5 or CD20ICDI OIC D I9IC D34 mAb co mb;na l;o n led to a rcdu ctio n to l O ~'Ó. 82% or 83% ot: ce ptabl c res u!ts. re spec tl\ d y. Exclud ing any ofrh e 1w O T· lin cage rn Ab co rnbm:u io ns redu ced the frequency o f acc eptable resu!ts to fewer than 90% spe..::i mens. Simd ar ly. rcdu c mg. thc B-li ll L'age pane l j ust to a single rnAb com bl na tiQT1 re duced the frequency of accep tab lc rcsults to fewer lhun 900'0 spcc imc ns in a ll casts (data
11 0 1
shown)
-..o
CD IOIC D66c1C D 19/C D4 5) . II \va s poss ib le to fu nher sim ph fy t he proc edure b) reducing th c number ol'
......
recQrded subpo pulat ions . [n a red uced pa ncI . W t' propOSf! to reco rd three C o l ~ su bpopu lal io ns w ilh rc1 at i...dy
If) ~ ~
V> , ~~
;>
o ;;:: ..., ~
hig.h cutoff .... al ut.:s (CO]-I"'''', CD I O"" and C Ol o-I201K,
).
follo wcd by daughter subscts of these C D I ~
subropulations ( CDI O""" J 4 ..... C DIO-·20"'·J 4~". CD IO.. •.... C D I0"""66c"'" an d C D IO......66.0""" - tho i.n or Iwo
~
~
are ph}' siolog icaH}" expressed
a1
the given dl ťferc llIi ali o n stage Such molec ular d lffe rences of leu ke mi e eells
may be hlghly spec ifíc wh en co mpart,;d to BM of hea lthy su bj ecIs. Re m lss lon 0 1\,1 specl mens, howe ver. contain non-mah gnant ee ll s in VnriOll$ phases of regenc mtion or olhcrwisc affcc:t ť J by eh emolh erapy (1 8). Mos t stri kingly , the B lin eagc regcnermes, a nd B prcc ll rsors are abu nJant at late phases o ť therapy, leadin g to pOlenl ial prob lcm s wich RD detecli on A lt hough th is
p h (: n o l11 ~ n on
is wi dd y accep ted. th e possi bili ry th aJ subpopulatl ons
of th c rcgenerati ng cell s an a ln LAJP is often ignored . In addlt lO n, stan dardizcd " pproac hcs a re necessary for RO detcclion in Iargc o.nd inte rnat ion a l protoco ls So far. thc intCr prelat ion of FC has bee n large ly depen dent on the
T he presemed ap proach largcly dJsregards subjec ti\ c C\'slUlu ion of
Fe dala. T he di sti neti on bctween leukem ie
eclls and no n-malignBnt bad :H,rou nd cell s is q UélJll1 tau\'e T o Dur know ledge. this js Ihe fírsl stu dy th al sct s thc basel lllc Ols negative B;\·1 specimens fro m Ihe same chcmo thc-rupy lime·polnt. T he presented appro ach may miss so me cl earl y ntyp lcal ce ll s . hov. c\"er, whc n lh .::)' do not fi l imo tlle pre -ddi ned rcgl otls . Th ls appears 10 bc Ihc reason w h)' 111c: pres.énted ap proar.;.h d id nOl lead 10 a useful correl ofl on al day 33 an d week 12 O fn o!\! , C'..-tn
J
also separate ly in CD·15'M'. C 0 4 S - (Ind ( D4 5""'" fracuons; in 10taJ, 14 rt'corded subsets) Obvlously. :lIl daLlghte r SUbSel:i are less Lhan or ~q lJal
tO
ttl f' lr respecu \ e parem CO 19"'" subpopul sllOns. w hich enables sk lppm~
somt:: oflhc analysc s Such a redu ced panel led let l d ~n t lca l resu lt s in 7 1 o r 83 pos itl \e speci mens. T hrce ofl hc
V>
·v
mo lecu lar ex presslon patt crn s. (b) asynch rony. and/o r (c) profound ov er- or undcrcxpressio n o ťmolecul es lhal
expert experien ce ol' a cytorn clri sl
Ane r reduc ing the 8 1ineage panc i to LWO m Ab corn hi natlons (Co:!O/CO 10lC O 19/C DJ 4 ond
c: ...,;....
measuremen l T h!! ge ne r
in form ali on was cons ldered accepla bl e if the reduccd panel g...:nerated Oln Ro va lue higher than O I o f thc full
100% specimcns (quanutBU"'c1y , the reduced pa ne l rcd uced th c va lucs to 0. 64 to 1.0 ofth e full pancl ..'alues; 5
T he speed ofF C, as wc:11 as its si ng lc·ce ll. qu anl il alive nature, portrays FC as a n idea l roo l fo r RO and M RO
83 spec imen s wou ld be co nsi dered nťgati ..'e. althougl1 very 10 \\1 leve l [1 0 .... to 1O-~ ] MR D was detec tcd by the Cu lI panel: onl y onc o ť th cse p31ie nts had (In ~'IRD abo\'c 10"" al day 15, and thc Slil pe rcentile wa!' 0 52 o ft he full
pa l ~ en l - 1 allo red ~ IRo
Invcsugall on by 4-c olor I~ C ma}' fal l fO g eneratc c lini call y reklble dn:ta. T he abso lute
number o ť rdapsed pa!ierll s illno ng Lhe Fe i\ t RD -ne:gali\'c subsel (day 29) was the highc'St nmo ng all su ata
ln
a
large pub lished trio.J lJl\'olvin g ~ , 1 1 3 pnlient s by Ch il dren'!; Oncology G rou p Study (2 ). Thus. co rn pored to a pre\ lous st ud\' (1 0 1 wlth rcccn tly updated resu lts by Flo hr t t :ll. ( 19) (w hi ch \"I:i '5 th c ba~ls for Ih e AI EOP· BF M
AU . 2000 protocol ),
Fe MRD ar theend ofinduclio n Ireatment mi sscd a much grealer propor1ion ofpati ents
than 19.'TCR MRD Recenl data from Borowitz et a l. (2) werc less en couraging Iha n several ofrhe
preVl0U S
Fe
o fe D I OP~~ and/ar CDJ4P"1' B Iy mphobl asts, howe\'er, ma)' be frec of RO, at te-asI al day 15 of BFM-l>ased protocol s
MRO reports . Here we show usefulness af day 15 Fe RD; the facI lha l aur approach failcd mlater time-points
The relevance ol non-malignant background SUpPOr1S th c reccn! caUrlon (2), despltc using a veT)' diITercnt approach ;jnd a differcnl ch emoth erapy protocol. Cl earl y, the Usé of Fe ar dar 29 ar Jater fo r MRD monitoring is still highl y imponan l AI day 29 or laler, Il rcli ab ly dClects .a high M RD bu rden. whercss its use al the level of 0.0 I % should rcly on a rnulticol or a pp ro ach and)o r other ll1c lhodological impro vcmc nt s
The Sl nltegy ofus ing a wide pand ofantibodies and repon mg predefined subpopulations as dcscri bcd here Wá.s rec en tlv Pllbli shed in a childhood AML MRD stu dy. where heterogencny 3J1 d phenot)' pe insrablh lY in leuk~mic cell s represenled a problem in repol1ing MRD (3 0. 31). Ahhoug h some data comparing FC MRD speclfie LAIP to PCR MRD appear prornising (32 , 33) . no speciťíc
rč p roducib le
usin~
pati e nt
gati ng strategy eon sidcrin g a time-point
baek ground has bcen published.
Alrhough 4-color a pproac hes are frequ en ll y used for RD detc..-cÍ lon , current equipmenl allo ws Ihe use of an
MAb combinations - reasoning The S YTO 16/CO 19 PE/C045 PerCP and SYTO 16/CD7 PE/CD45 PerCP combinations we re used to quantify
JOcrcasin g number of moleeules simuhaneously . Thl5 may further improve: the accuraey
oť
distingui shing
thc propo rtio n of Band T li neage cell s, respecti vcl y, among nuc1eatcd cells. In addiLio n, immature CD4S d ,m,o
leukemie from non-mali gnanl ce lls. even ifno ne w anligens are introduced into the diagnosti cs. Knowledge of
'''' C O I 9~" ~lI s \Vere rec ord,d. The CD2Q FllCi CDIO PE/CD 19 PC7IC D34 APC e>alua!cd B ce ll
the non-malign anl background, howcve r. is \"ital for the correcl interpreta tion of such RD data The presented
d.ffe renuati on (20. 2 1). Changes ln CDI O exp ľess ion leve ls in the conlexl of (' D 20 nnd C034 are frequent in BC P AU
lho C058 FITCICDIO PE/COl 9 PC 7/C[J 34 APC combinat ion was Introduced to detect CD58
rcsults show lhal thc background le vel of immature cells of bOlh Ihe Band T lineage may excced O 1% even dur ing th c il1duc.ti on th c-rapy
ovorexpreSSlOn, commonl ~ fou nd in BCP ALL (22- 24) The CO I O FITCICD66c PElCD t 9 PC 7/CD45 APC cvaJuated C 0 66(.;. which is th e mosl cOlllmon aberrant myc loid ~n,i ~c n in eh ildhood Al.t. a nd u$ually docs no t
chaose bct,,,'cen diagnosis and rela.pse (25). Possiblc tc mporary changes in ils c.xpreSSlon , howe\cr, would decro" sc ilS valuc in M RD. The CD99 FlT C/C D7 PE/C D 5 PO/C D3 APC evaluated CO J ncgativi ry , nd hypeco xpr. s"on of CD99, both of which are common in T A LL (2 6, 27) The intra-TdT FITC ICD7 PElC D3 PC 7li nt ra CDJ A PC cvaluatcd the 'rdT exprcssJOn in T cells, whtch 1$ predom inantl) presenl in the thymus (1 8) Dala inlcrprctallon W1d dcfínition of indl \'id uaI subpop ulations was do ne US dcscr ibed in ti g ure 1 "lr1 d:!
r---..
cr,
..-< ~
c: ~
~ .....,
al day 19 on the St. Jude pro10cal (29 ). Th e defi nitions ofpopul ations shown in Ih lS stud) are
,
tnc poss! bl e presence of non -m alignant CD l r
Vl
'V
~ ~
~
cr MRD calegories, e spec ially al day 33. Day 15 cy tometric RD deleclion is, thus, a notable
alt ernative for PCR-bascd ~'1RD al day 33 , espccia lly in setting.s where PCR-based \ltRD is una" ailable A lth oll g.h morphology patients .....ilh
a
Hl
hypoplastic BM. which i:s \)' plca!ly prcse nt at day 8 and 15 . can co rrectl y idcnlify
high le uke mIe burdcn (BM infdtrat ion over 25%), Fe ean speciflcall y
c3tegO rl ZC
paticnts uSlOg
lin ~a.!!c
A LL mighl h3\'1.: been intluenced by the lower num bt.:r ofpalients in thi s ALL su bSCl A largcr study
for l ' lIneage AL L should test whelher RD in Ihese patlcntS req Ul res a q uontitalivdy di ffe rent slratifícallon . c1 0SCSI
to CD I o""'"'C D 19"" " no CD 19"' C 0 J4~~ Here, we show 3 I1lgh baekground ol' th e,e subpo pulations 3t da)' 15
of Ihe BFM-bascd ALL- IC BFI\ f 2001 protocol Thertfo re. the publlshe-d findUlgs should nOt
'>~ ..... .....,
pre:dicti ve
T
Rt."'Ccrnly Cousl i.ln-Smith et al showeJ the prognost ic rc! c\":tncc ol' CD IO"'" 'CD 19'* M d CD 19;-CDJ4 P"
Vl
:>
As shown herc, thc evaluation of RD on day 15 uSln g;' C}' lOmelric strategy with predefined gatin g \\!as
lowe.r and mean i ngťu l thresholds (Flgure 4). Thc fac I Ihat car ly FC RD d id nOI eorrelate wlth late r II mc -points in
Simplilied RD testing fr eqll ~ncy
The predictive value ol early RD
g.enerahzed. and
or CD3 4r--"'1 8 l.·el15 should be considered in BM. e\'en st enrl~'
phJSCS of ~ hc m o lh ér3py ln ag/ceme nt ",ilh lne ma ln cond usion ofCous lan-Smith el nI • negati\ c: findmgs w lln thesc s, mplitied tec hni que s are likcly to spt:t:.ifica ll y ldť nl lfy palienls \\'lt h low RD Patients with hi gher numbers
80 th the IO stru mcrJ\ a.n d the ga tl ng should be slandardized for reliabl e RD mon itoring. es peciall y in a muhiC"enterselfing. The pre-senled stud y sho \I,'s backgroun d thresholds in precisely defln ed gnles nnd Ihe relevan ce of
a 'Ime-po lnt speciftc RD . Th c highcSl pOlcnt ial for the use o f FC appe.ars to be day 15 of a I3FM-baseJ prolOcol .
These RD
vaJ \lť s
ma)' ldenu fy those patients a( both 11Igh and 10"" risk ol' ~ IR D ar the cnd ol' lnd ucllon and al
\\c(!k 12 ln BCP A.LL.
Reduced panel for day 15 RD A ral hcr extenSlve pand w·as design cd, and quite a h igh number af su bpopul utions were reponcd in this study For practical reaso ns, lirni tmg thc nu mber ofa nalyze d subpop ulation s is \ Ital , especi a ll y in B lineag,c A LL [n
Titles and legends to figures and tables Table 1. Pa tients , characteristics
a ur coho rt, a reduction in th e numbe r of subpop u lauons by o ne -h alf( 14 subpopulati o ns ), omitting one mAb
Hyperdi pl oid patien ts were nee,ative for all fusíon genes listed in thi s lablc S ta ndard, intcrmcdi ate and high risk
co rnbinalion entirel y aod reducing the SYTO 16/C O 19/C D45 10 SYTO 16/ CD 19 p rovl dcd com parable rC5u lt s 10
g,roup (SRO, IRG, HRO) criteria are d escribed in mat e rials and melhods
th e rull pane l af 28 subpopulation s w ilh 4 mAb comb inations ln T A LL, \Ve not iced a decrc3s~d Intensity af TdT exprcssion in resld ua l T I)' mphob lasts (data not sh own) Not surpns ingly. the two subsets o fTdTJY.'~CD7po> s ubpopu laltons were a lready too infrequent and did not con tribute to the ovcrall RD values . Neverthdess, our
Table 2. Cut-off values Va lues represent thc thresho ld fo r RO positivity (A B lineagc, 8 T lin e.ogc) l'
dala showed no predictive \'alu e ol' FC at day 15 in T AL L. Thercfore, an alternali \'c 1:!pproach shou ld be used for these paticnts. Pc rl1aps day 15 could bc rcp lac ed by ano lh er Ji me -po int in T ALL patienls
Table 3. FC sensitivity and specificity at different time-points The number o f FCPO' '~y nnd PCRf"D' ""· ( levcl 10-
1 )
pati enls al respec li ve lirne-poinLs and Ih e correspondlng
senSUt VHy nnd speci ticity of FC RD
References: Figure 1. Gating strategy in B lineage ALL AlI reporlcd ~a l es fo r cac h m Ab cornb malion are show n in pscudocolor plots. Pl ols s how residuallcukcmk: cc ll s
(PCR RD 2: 10.
1 )
in regcncrating bone marrO\.... (week 12). In the CO 10/CD66c1C DI9/C045 cornbtnation.
s ubpopulat lo ns acco rd lng
10
che levels of ('045 expression \\,:ere a lso reported (C0 45 - -, C04 5<.1'I11 , CD4S n , ..
detined by expression on C04 5"1I· g ranulocytes)
Figure 2. Gating strategy in T ALL A. 1I repOI1M g aics for each rnAb combinal ion are 5hown in psc udoco lor pIOIS. Plols s h ow res idu alleukcrnic cells (PCR-RO > 10.
1 )
in a bone mafrow (day 15) logether \Vit h non-mal ignant T Iy mph ocytes
Figure 3. Correlation of RQ-PCR RD and FC RD at day 15 FC RD (calc ulared as lhe h ighesl frequency of che subpopulation which excecded Ihe lime point-speciflc
00
backg round "'alue)
15
co mp ared to RQ-PCR RD ca!c ularcd ac cording to ESO principles. On ly sarn ples not used
Lr'l ,-<
for bac k @. ro u n d q u átl tlflcatio n a re s h own . Th enu mber oť paue n ts in
cach q uadran t i::. shown
~
t:: ~
I-. .....
Vl
,~~
;>
O
...!<:
;c ..... Vl
'Q)
~
w
Figure 4. Correlation between resldual disease by PCR at day 33 and residual disease by FC at day 15 in BCP ALL Pat1enlS wcr c c31eogorized besed on PCR-dclccled RO a t da" 33 inl o negati,:c (\\ hite) , bc-ID\\.' 10-1 (ob liqul!' halching). below 10 " (.~W .J Ilnd < 10'.1. \'en ical halching) and ut or above 10·" ( black) The freq uencies are
shown in su hscts o ť palien ts dc1i ned by FC RD at day 15
References Steinherz PG. Gaynon PS, Breneman
Je. Chcrlow JM, Grossman NJ, Kersey JH. et al
Cytoredllction
and prognosls ln a Culť Iy mphoblastlc leukerma --the Importance of early marrQ w respollsc. report from the Childrens Conce r Group. ] Clm Onco l 1996: 14(2).389 ·98 . 1. Borowi t.7. MJ. Devldas ~ f , Hu nger SP, Bowman WP , Carroll AJ, Carroll WL , el 0.1 Cliníca\ sigm fícance o f minimal n:sid ual di sease ln ..-:hildhood Ocule Iymphoblast ic Icukcmia and ilS relati o nship to other progn ostic factors A Chi ldre n's Oncology G ro up study Blood 2008 . 3. Borowitz MJ, Pullen OJ, Sh uster Jl , Vl swa na lh n D. Montgomery K. Willman CL. el al . Minllnal res idual dlseasc delccllon in ch lldhood pn:cursor- B-cell ac ut ~ lymphohlastl c leukemla: relation 10 other ris k faclors. A Chlldren's Oncology Group study L. u'em,a 2003 ; 17(8): 15664 Pan7...er-Grumaye r fR, Schneider M, Pan zer S. Fasching K, Gadner H. Rapid molecular response during earl y induction che mothernpy predicts a goud outcome in childhood acule Iymphoblasli c leukemi a B100d 2000;95 (3): 790-4 5. Dw'O rm "I N. Frosc hl G , PrinlZ D. Mann G, Potschger U. Muhl eggcr N, ct aL Prognostic signiti cance .:Ind modahu es of n ow cytometric minimal resid ual dl scasc delcctlon in childhood acute lymphoblastic leukemia Bl ood 2002,99(6 ) 195 2-8 6 Bjorklund E, Maz ur J. SoderhaJl $ . Porwit-MacDonald A Flow cylomctr;c follow-up ofminimal
n
residual discasc in bo ne
C'-
If)
....... ~
t:: ~
I-.
'-'
Vl
,ci > o ~ ;i: '-' Vl
. Ci)
~ ~
m á rrůw
gi ..'es prognos tic information in childrcn wilh acu le lym phobL1st ic Icukemia
Lcukemia 2003 , 17( I) : 138·48 7. Willcmse ~'1J . Serill T, Hetti nger K. ď A ni ello E. Hop \VC . Panzer-Grum ayer ER. el a l. Detec lion of min imal resldual dlsease ldentifics di ffc:rcnccs ln Ircatmcn l responsť' between T-Al..L and precursor B-ALL. Blood 2oo2:99( 12) 43 86-9 3 8. Pui C ll. Sa nd lund JT , Pe l D, Campana D. Ri vera GK, R ibeiro RC. ct al. Impro ved oulcOrnc for child ren wilh acule Iy m p hobl astlc leukcm ia. resuhs ofTolal ·fh-.!rapy Study XIIIB nt St Jude Chddren ts R~search Ilo'I'"a l Blood 2004. 104(9) .2690-6 9 CouS IM- Srmth E, Beh m FG. Sanchcz J. Boyen JM. Hancock ML. Raimo ndi se, et a l. Immuno logical de tectio n o f mm im al resldual di sea.se ln chi ldren ",ilh acute lymp hob laslIc leu kae mia. Lancet 1998.35 1(9 102) 550-4. 10. van Do ng cn JJ , Seriu T. Panzer-Grurnayer ER , Bi ondi A. Ponijc rs-\Villcrnsé M J, Co rral L, et al Prognostic ....a luc of mi ni nwl resl dual dl SCasC in acutc !y mphoblastic \cukaernia in chi1dhood. Lance l 1998,352(9 I 42) 173 1-8 II Pongers-\Vl1 lemse M J, Seri u T, S to l7. F, d'Am cll o E, G ame iro P. Pi s.::l p. et al. Prim e rs a nd protocol s fo r sla ndard lzcd dc.tec lion o fm lnlm al resi dua! dlse ase in acute Iy mphobl astic leukemi 3 using immuno g lobulin a nd T cell receplo r s ene r<:arran gcments and TA L I dc lelions os PC R la rgel'· report of Ihe BIOMED·I CONCERT ED ACTIO N. in\'e.stlg ation ofminimal resldll nl dlsease in a.;ute leu kemia. Leukem lB 1999: 13(1) 110-8. 12 van Dongcn JJ. Lan ge rnk A\V. Bruggc mann M, EV8nS PA. H umm el M. La \ endc r FL. CI al. DeSIg n í.Ind standardizatJ on ofPC R pflme rs and prolocol s for dctection ofcJonal immunogl obuli n and T-cell rc-ceptor ~('n e recomblO3uons 111 sus pe ct Iy mpho proli ferauons: repon ofthe B10M ED-2 Concert cd AClion BMH-I- C'T9 8-3 9 36. Le ukcm,a 2003. I 7(1 2) .1 25 7· J I i 13. \'8n de r Vddcn Vil, Hochhnu s t\, Cazz.oni ga G. Szcz.cpanski T, Gaben J. VO Il Dong~ JJ . DClccti o n of minimal resid ual disease in hema lolog ie mali g nanc lcs by real-timc quantltOl lVC PCR : princi p lcs, ap proac hcs. aHd I.boralory aspec". I. eukemi. 2003 . 17(6) 1013 -34. 14 . Bruggema nn M. "' JI de r V. ld.n VH, Raľf T. Droese J, Rilgen M, PO!! C. CI al. Rea m mged T-cell rece ptor bela g cnes rcprescnt po\\erfu l Illrgets for quanufic8l lon o f mIn ima! residual dlseasc in chlldhood and aduh T-c cll acule Iym phob lasu c I. ukem,a Le uk emia 2004 ; I 8(4) 709-1 9 S ra mkov Q. L. ~lu 2l kova K, Fra nkova E. KrcJ c i O. Sed lacek P, Form ankova R. el ol DCl ectable min lln al 15. resldua l d iseasc beton.: alloge nclc hemalop<,i eu e stem cell trun s pl::uu3I1o n predic ts c xt rcme ly po o r prognos is in chil dre n \\j th ac ut e Iymphobl astl c lcuk cmm PedlO tr B!ood Cal1ce r 2006 . 16 von der \ ·elden V H. Cazzamga G. Sch rauder A, Hancoc k J. Bader P. Panzer-Grumayc r ER , CI al An a l)' sis Of mlnHllal resid ua l disease by IgffC R ge ne rearrangements gu idchnes for intcrpre.uuion o freal-lim e quantHativt peR data Lcuk~m i a 200 7. 17 Agresti A CaJego rical Data Analvs is: John \ Vlley & Sons. 2003 18. \ 3n Lo che m EG. \ Vlegers Y ~1 . van de n Becmd R. Hahlen K , \'an Dongen JJ . Hoolj kaas H Regcnc rruion panem o f prec ursor-B-cells in bone mafTOW of 3cute Iy mphoblasllc leukemi a pati en ls depend s on Ihc 'Ype or prcc.
20. Lucio p. G aipa G , "an Lochem EG. van Weri ng ER, Porwil-Moc Dona ld A. Furt . T, 01.1 BIOM E D·I co ncened ac tion report: no\\' C)'lOm elTlC Imm unophc noly pmg ofprecurso r B-ALL w lth standardized triplcs tai nings. BIO ~,tED-1 Conccrt eJ Ac ti o n InveSll go I1 o n of Mlnlmal Resid ual Disease in Acute l eu kemi a · Internatlo nal Standardiza1lon and Chm ca l Eva luatlon Lcukerma 200 1, 15(8) 11 85-92 21 Lucic P, Parrcira A. van den Bec md ~-tW . van Lochem EG. van W<:ring ER. Baar3 E, et al. Flow cyto metric anal ys is of norm a l B cell d ifTerenti ation : a ľrame of reference fo r the dctection o f min imal residual di sease in precurso r-B ·ALL Leukemia 1999. 13(3):41 9· 27 22. N eale GA. CousUln -Smilh E, Pan Q. C hen X, Grohn 8 , Stov. P, et al. Tande m application oft1ow cy tom etry and pol ymerase c hain reactlon for comprehen.sl\"e dť..1 é(..'1 10 n of m inima l residual dlsease in ch lldhood acule IymphoblaSlic I
E.
Mejstříková,
strana 160
Figure 1,
SYTO 16/ CO 19/C045
Co20/Co ID/CO 19/C034
Co58/Co 1O/CO 19/C034
CO IO/Co 66c/Co 19/C045
"
<5' c
;o Cll
-<:
d O;
Č)
CJ ';-J
()
CJ
"'"
Úl
()
CJ
19
Č)
CJ ';-J
()
CJ (11 Č)
CJ
w :.:;
~
~
___
- ~CJ ___ , _ _ -bl g
'"
'-----------'
;,t ~
-;-t ()
3
CJ
"O
Ol"
';-J
o 3 o ~ o
()
~ n'
IQ
co
CJ
w
:5'
;,t
()
CJ
W
E. Mej stříková, strana 161
O l _I _
_ _ _ _ _ _ _
_
~
o
<3 ' e:
čD
Ci> I
o ~ o
oo 00
q
6)
q,~ o
<>0
Q)o
oQ(f>cP
'6 0
o~
I
9
o
00
o
~ tePo COcJl,0 :;::
./>ol I
Ol
o
I
--- - ------I
"O>
'<
::J CI)
Ol
co CI) dI
o 00
o o
«>'0 0 o
00
80
Frequency of day 33 PCRpos patients according FC RD categories at day 15
..... o
o~
A
P "'Tl
;J?
() p
:o o
&.
;,\!
Ci: 'f' Dl '<
o
-'" ;J?
~
o
N (J1 ;J?
~
'"
;J?
Ul
o
:$!. o
8:$!. o
Table I.
Table 2. Characteristics at diagnosis (n) 67 43 68 42
Male Female <6 years ~6
years
BCP total BCP SRG BCP IRG BCP HRG prae-B cALL AHL (cALUMy+)
90 47 33 10 17 70 l'
pro-B
Prednisone good response
85
Prednisone poor response
5
VVBC at diagnosis 220000luL VVBC at diagnosis <200001uL CR1 achieved at day 33 CR1 achieved late( (till week 12) Death p(ior day 33 Genotype subgroups (BCP ALL) TEUAML1 BCRfABL hype(diploidy (DNA index 21 .16<1.6) 28 MLLfAF4 others
TALL SRG IRG HRG mature T ALL mature T ALL TCR alphal betaPO' mature T ALL TCR gammaldeltaPO' AH L (mature T ALL TC R gammaldelta" ' IMy+)" inte(mediate T ALL prae-T Prednisone good response Prednisone poor response
A B lineage 20 O 12 8 5 2
CD101><"C034\>'O'C019~ O '
0.11
COl On'IICD20 n-eC034~o,C 01g0 U
0.0072
CD34Co.CD19;MO
0.29
C010 bdot"' CD19~'"
0024 3.8 0.58 0.014
CD 1Ob"' '"C058 WlVIlf CD34 ~~' CD19!J<>'
3 12
8 14
65 86
WBC at diagnosis <200001uL CR1 achieve d at day 33
19
7
0.0056
CD10~"'CD19-
VVBC at diagnosis 220000luL
29
3.2
CD10 1>l!t I'l':CD34po~CD19r ' .
CD1 On··CD20MgCD19~·
25
CR1 achieved later (lili week 12) Death prior day 33
day 15 cut olf value
sytopoo·C045<1''''CD19~~·
6
O
C010/l104C058 t.n,·CD34 "' · C019~'"
0.13
CD10n"'CD58q Il1CD34po'CD19pn
0.049
CD10~f'PCCDS8b",!oO CD19P'"
0.03
CD10""·CD58 1"'.MCD19 P'"
0.52 0.95
C010no' CD5SbdtftlCD19l><>' C010~· ·CD66c''''~
CD10 ~"·CD66c~·" CD19~o,
0.2 0.017
CD10PG 'CD66c",tI"" CD19 f10 •
0.028
C04Sflttt "'C0101>O'CD66c Q'" CD191>01 CD4Sdlm CD10 PO 'C066c;>
0.052 0.12
CD4S ..... CD101l'O.. CD66c"""· CD1911'01
0.083
CD19P<1 '
CD45MilMCD101>:loIll C06ecl'D- CD19""
O 0.01
C04S·,m CD10 Io.tltl lt~ CD66c"" C019 ~'
C045"'i1CD10 "",IICCD66(:"~ CD1g lHl "
O 26
0.012
CD45~ri,hlC0101>O IC066(:""'" CD19"- '
0.0069
CD4Sd'"'CD10~~C066(:WIfAt C019-
0.021 0.0043
C045"~CD1 O""'"CD66c 01llM C019-
CD45
CU
0.12
CD19 g < ,.
0.26
B T lineage
N '-D
,....-< ~
c:
~
'-<
'-' Vl
,~
:>
O
~
' ...... >'-<
'-' Vl
·v ~ ~
day 15 cut olf value
CD5~uCD99 ~'~MC07P.'
0.14
COSpuC03.... COr.. •
0.18 0.36
ITdT'o., CD7 P"" ITdT ;~ 'CD3P~'iC03 ~"
Cor""
iT d-r:°sC DJ""'IIiC 03"" CD
r
s
0.22 0.014
E. Mejstříková , strana 163
6.4.
Přiložené
P říloh a
publikace bez impakt faktoru
12
Imunofenotypizace
d ěts kých
leul<émií
Tomáš Kalina, Ester Mej s tříkov á, Martina Vášková,
Ondřej
Transfúze a hematologie dnes, 2006, (bez impakt faktoru)
Hrušák
E. Mejstříková, ~ trana 164
Imunofenotypizace K alina
Tl, 2, 3, Mejstříková E.I , 2,
dětských
leukemií
3, Vášková MJ, 2, 3, Hrušák
0 ,1,2,3
lÚstav imunologie, 2Klinika dětské hematologie a onkologie UK 2. LF a FN Motol, 3CL IP Childhood Leukernia Investigation Prague
Souhrn lmunofenotypizace je již dnes standardní metodou v diagnostice hemato-onkologických onemocnění. V současné d o b ě zažíváme rozvoj mnohobarevné průtokové cytometrie, j ak ve výzkumu, tak i v klinické diagnostice. Práce shrnuje diagnostik u a sledování akutní lymfoblastické leukemie pomocí průtokové cytometrie. Klí č o vá sl ov a: imunofenotypizace, mnohobarevná pr ůtoková cytometrie, akutní leukemie, expresní profilování, m inimální reziduální nemoc S ummar y Kali na T.,
Mejstříková
E., Váš ková M., Hr ušák O.: lmmunophenotyping of childhood leukaemias
lmmunophenotyping is a standa rd method used io diagnostics of hematological malignancies. Progress in number of multicolour cytometry as well as new targets identified by genornics bring new possibilities in research and diagnostics. These aspects are summarized in the article. K ey w ords : immunophenotyping, multicolour flow cytometry, acute leukaemia, expressioll profiling, minimal residua] disease Trans. Hemal. dnes, 12, 2006, No. J, p . 20-25.
Úvod
Popis metody
lmunofenotypizace neboli stanovení povrchových b un ěk pomocí průtokové cytometrie zažívá v posledlÚ dekádě ohromný rozmach, jak po stránce dostupnosti, tak po stránce teclmické vyspělosti instrumentů a rozvoje technik a postupů uplatňujících se ve výzkumu i v diagnostice. To vede na jednu stranu k možnosti získálÚ množství klinicky relevantních informací, ale zárove ň je potřeba tyto informace zpracovat a klinicky správně interpretovat. Záměrem prezentovaného čl ártku je poskytnout přehlednou informaci klinickým i laboratorním hematoonkologů m, která pomůže v uvažovárú bě hem diagnostického a léčebn é ho procesu. Výhodou prutokové cytometrie je její komplexlÚ přístup a její schopnost dát relevantní odpověď i v situaci, kdy ošeul.Ijící lékař má více otázek než jasných odpovědí. V hemato-onkologii je cytometrie centrální metodou, která řeš í otázky diferenciálně diagnostické, typizuje leukemické buňky a má vliv na zařaze rú do léčebných protokolů. Zatím výzkuJllIlě se cytometrie používá ke sledovárú průběh u l éčby (minimální reziduální nemoc) a k hledárú nových znaků k predikci prognózy. Leukemické buňky se vyvíjejí v kostní dřerú, v období p lně rozvinuté choroby je nacházíme v perifem hvi nemocných, kudy recirkulují, podobně jako nezralé krevlÚ buňky, ze kterých vycházejí. Schopnost zkoumat imunofenotyp jednotlivých bun ěk ve směsi buněk jakou je kostlÚ dřeň i pe rifem! krev před s tavuj e hlavní výhodu prutokové cytometrie. V posledrú dob ě se do diagnostické plůtokové cytometrie dostává možnost mnohobarevné analýzy. Současně plicházejí nové informace z genomiky a proteomiky, které mohou ovlivnit diagnostické možnosti imunofenotypizace.
Průtoková cytometrie používá jako zkoumaný materiál buněčnou suspenzi (např. kostní dřeň, perifemí krev či likvor). Metoda vyžaduje čerstvý materiál (živé a neagregované buňky, obvykle do 24 hodin po odběru). K detekci jednotlivých povrchových nebo cy toplazmatických molekul používá monoklonální protilátky konju c gované s fluorescenčlÚmi značkami. Suspenze buněk je po navázání fluorescenčně značených protilátek nasáta do průtokového cytometru, kde jsou buňky seřazeny do úzkého proudu kapaliny a vysokou rychlosti proudí komorou a jsou ozářeny laserovým paprskem. Laserové světlo excituje na buňk ác h navázané konjugáty protilátka-fluorescenční zna čka, které emitují svě tlo o vlnové délce určené typem fluorescenční značky. Souča s né prutokové cytometry umožňují zapojením jednoho až tří laseru excitovat 3 až 9 různých fluorescenční ch značek zároveň. Pro každou buňku je zapsána kvantita každého fluorescenčního znaku a dva nefluorescenč nÍ parametry popisující velikost a granularitu buňky . V jednom vyšetření se obvykle použije simultánní kombinace znaků (v diagnostice leukemií cca 2S znaků, které jsou vyšetře ny v cca lS oddělených alikvotech zkoumané suspenze), během sekund až minut se změří 20-200 tisíc buněk v každém alikvotu. Ke grafické analýze těchto S až II parametrových dat se používá speciální software, kde se provádí výběr ("gating") podskupin buněk na základě přítomnosti jednotliv ých znaků (obr. 1). Komplexita z m ě řených dat narůstá se zvyšováním počtu parametru a vhodně zvolená strategie analýzy je velmi důležitá pro ínterpretaci dat. Interpretaci dat mu sí provád ě t vždy zkušený odborník-
znaků
20
E.
Mejstříková,
strana 165
-cytometris ta. Při počtech nových diagnóz v dětském věku kolem 80 ro č ně je vhodná koncentrace vyšetření na jedno pracoviště, což je v Č R díky spolupráci všcch dět ských hematologických center zajištěno.
Cytometrie a akutní lymfoblastická leukemie
vždy byla dostatečně dlouhá doba sledování (I) nebo se hodnotili pacienti s BCP-ALL a T- ALL společně. Vzhledem k léčebnému ú s pěc hu léčiva založeného na monoklonální protilátce anti- D3 3 a cytostatika calicheamicinu u myeloidní leukemie (3) je zajímavý nález korelace CD33 pozitivity u BCP-ALL se špatnou prognózou (4).
Predikce rizika
l munofenotypizace při diagnóze
Přítomnost
Cytometrie je díky rychlosti provedení vyšetře ní základním laboratorním vyšetřením vedoucím k diagnóze leukemie. Základní otúzky, které diagnostické vyšetření odpovídá jsou: • Jedná se o leukemii, o nemaligní proliferaci prekurzorů nebo o změnu procentuálního zastoupení subpopulací bun ě k např. při útlumu v kostní dřeni? • Vychází leukemie l myeloidní nebo z Iymfoidní řady? • Jedná se o ALL z B-prekurzorů nebo z T-řady? Odpovědi na tyto diagnostické otázky mají (gatovántj význam pro potvrzení nebo vyloučení maligního onemocnění a pro zařazení pacienta do léčebného protokolu.
jednotlivých molekul na blastech ALL je in tenzivně zkoumána z pohledu potenciálních predjktorů úspěšnosti terapie. Velká pozornost byla v ě nována aberantní expresi myeloidních znaků na buňkách ALL. Jednotlivé studie nalezly nepříznivý nebo žádný prognostický význam exprese těchto molekul při diagnóze. Tato diskrepance byla způsobena urč itými nedostatky v designu studií. Rů z né myeloidní znaky byly často hodnoceny společně (např. CD13 a/nebo CD3 3) (1, 2) . Ne výběru buněk
Struktura
c J CDI 0-2D r ( !1 . .2 .." ::
....
h) '
e-:.
~ C· 'i---o-~.a.;.;;,,...,
-I eo
.~
,.-,. j1
' Po
."t
J'
Ol l)
·.I~
'o.
~
: .~.,"
5
se .
CDu'· '
~
u
\ -" ',; ...~:
10;.'
~ 1~----~----------~ ~ ,,.,:
o u ' o·
I
I
(,
~
~L. c
'o'
,.
.-..-, ..,,-o' H. I
Ilr...l
I t;
16
~ ó
U'
29 J
10-
!C'
Iq'.T
T.
10!·\
O llr. 1. Piíklad pou žití mnohobarevné kombinace CCD34 FlTC/CD J O PElintra CD I 79a PCS/CD20 A40S/lg M Dy647) znázorřlUjící vývoj B řady v nemaligní KD . Vlevo nahoře jsou zobrazené CD 19+ huňky ze širokého lym fomonocy tarního gatu pod le exprese CD 10 a CD20 antigenu. Během vývoje B (-ady buiíky postupně získávají antigen C020 a ztrácí an tige n D J O. Zral é B buňky, které jsou CD20pos a CD I Oncg nebo jen s l abě + pak o pou štějí KD. Na dalších obrázcíc h je pak rozvedena podle jed no tlivých subpopul ac í exprese intrace llllámího V preB receptoru Cintra CD J 79a II povrchového IgM). Buňky postupně v průběhu svého vývoje přechodně ex primují Vf'rc B recepto r a postupně více buněk exprimuje na povrchu IgM . Osy zobrazují intenzilu Iluo res 'c nce v grafu jmenovaných znaků.
B-prekurzorová akutní lymfoblastická leukemie (BCP-ALL) Kostru dřeň je plimámírn I1Ústem vývoje prekurzorů krevních buněk. Sebeobnova a vývoj hematopoetických kmenových buněk , vývoj prekurzorů B Iymfocytů a myeloidních buněk probíhá pouze v kostní dřeni, zatímco prekurzory T lymfocytů dozrávají v thymu. Imunofenotyp Bprekurzorových leukemií připomíná imunofenotyp nemaligních prekurzorů B řady. Klíčovým úkolem je tedy nalézt spolehlivé znaky, které odliší nonnální prekurzory od maligních buněk . Kombinace změn jako např. aberantní exprese, asynchronní exprese, hyperexprese, útlak normálních vývojových stádií či přítom nost prekurzorů v periJcmÍ krvi umožní stanovit správný diferenciálně diagnostický závěr. Imunofenotypová klasifikace BCP-ALL podle návr-
--_ ._- - - - 21
E. Mejstříková, strana 166
----------------------------------------------------------------------------
hu skupiny GIL (5) (tab. I), kterou používáme v rámci BFM léčebných protokolů i v České republice rozlišuje tři - kategorie: • proB ALL • commonALL • preB ALL Tato klasifikace vychází z vývojového schématu B Iymfopoézy. Ačkoliv imunofenotyp proB ALL je spojen s přestavbou MLL genu a horší prognózou, rozdíl v prognóze či biologických vlastnostech mezi common ALL a preB ALL nebyl prokázán . Zároveň se z pohledu ALL řeší i otázka hybridních leukemií, kdy jednotlivým zn akům, je dáno hybridní skóre a pokud je skóre vyšší než 2, je leukemie klasifikována jako akutní hybridní leukemie (AHL - AL UMy+ nebo AML/Ly+). V následujícím přehl edu diskutujeme znaky používané v diagnostice ALL (tab. 2), více informací o jednotlivých D znací h, pokud není uvedena reference, lze dále získat na webové tránce HLDA (Hu man Cell Differcntiation Molecules Antigens Wokshop). Tab. I.
Přehled
specifické znaky pro B-Iymfocytámí
zařazení
přítomnost
řadu
alespoň
dvou z následujících tří znaků: CD19 Znak definující příslu š nost k B řadě je především znak CD 19, který se objevuje záhy po Iirůovém rozhodnuti společného lymfoidního prckurzoru směrem k B řadě, ve kterém hraje hlavní úlohu transkripční faktor Pax (Nutt, 1998). Tento znak zůstává na povrchu B bu ně k ve všech stádiích, ztrácí se až na zralé plazmatické buňce. Hraje nepostradatelnou roti jako koreceptor B-bun ěč ného receptoru ve fyziologickém vývoji, aktivaci a diferenciaci B buněk. CD79a CD79a (lg-alfa) je signální molekula, je rovn ě ž B-liniově specifickým znakem, jeho cytoplazma tická exprese dokonce předchází znak CD19 a je první specifickou molekulou oddělující B-prekurzor od společného Iymfoidního prekurzoľU (7). CD22
Podtřída
Krité ria • 2 nebo 3 z nás ledujících : CDI9pol (intra)CD79apoL and CD22pol • C0 3 n
B prekurLOr ová
zralá B
LL- -
--
• 2 nebo 3 l. nás ledujících : CD 19poZ, (intra)CD79aPO' and C02 2PO' o C D3neg o inLraC03neg o Kpol nebo tJx>L --c--• (intra)C03PO' o C07po7 -
Tab. 2. Panely protilátek. které používáme
proB ALL
Kritéria CD loneg CD20 ne g
cALL
CO IOPO>' intral gMneg
__ preB ALL
• KnC.S a Ancg
I---
při
~.
proT ALL preT ALL intermediámí T- ALL zralá T-ALL TCRa~ POl T-ALL TCRyOPOz T-ALL
diagnostice akutní leukemie v naší
labo ratoři .
T-A L L
BCP -ALL
AML
i-MPO/i - D79acy/i-3 i-TdT/mCD7 /rn3/i-3 ;-C D22 045114 C 066c/l9 CD4/8/3 C 020/l 0/19/34 C099/7/5/3 CD I5/ 117/33/34 stanovení DNA cyklu
CD34/3817
CD 10/33119 C065/2IHLADR C0101I3/19 CD34/38/l 9 CD 79a CD22/24 COI9/NG2 CD64(pokud je AHL skóre 2) i-lgM/ mCD19 K/CD 19 IJCOl9 ml gM
Syto-16/33/45/34 CD 19/56/33/34 C02l7/33/34 HLA ORJ38/3 3 i-CDI3 C0331l3 CD41 C042b C06 1 glyko forin A (CD235a) CD65/33 C03 3/NG2
i-C022 COla (PE) TcRybrrcRa~
C07/NG2 CD64/J 3 (pokud je AHL skóre2) Je-li TcRyOPOz resp. TcRa~Poz , nastaví se Vo res p _ V~ pane l
m
AHL sk(,re Vo runel
povrchové značení intl'Ucclulúmi značení viz text protilátky prol i jednotlivým typLIJll
v~ punel
protilátky proli jednotlivým Vf)
Sylo- ltí
nuo re sccnč ní barvi č ka , značící buňky
V ň řctČ l.CU řetě7.c cl
obsahující DNA, pumáhá k odlišení drti
---
intralgMpoL
-
--_.-
Bez další subk/asifikace
zákJadn í panel
Vysvět1i v ky:
vyžadujeme
klasifikace podle návrhu skupiny EGIL.
Kategorie
U
Liniově
Pro
C02 neg C05 neg C08 ncg CD2pOl. a/nebo CD5PO' a/nebo CD8pol CDlapoz C03PoZ CD I aneg TCRa~POl
T CRyoPO'
CD22 je specifický znak pro B Iymfocytárni řadu, v časných fázích vývoje je exprimován jen cytoplazmaticky, k povrchové expresi do chází v normálním vývoji v pozdních stadiích spolu s IgD a CD21 (8 , 9). U leukemií z B řady se vyskytuje ve 3/4 případů (obvykle cALL a preB ALL). Molekula je příbuzná molekule CD33. Testuje se terapeutický účinek protilátky proti CD22 konjugované s cytostatikem calicheamicinem u různých CD22pos malignit (l0).
vývojová stadia
B -prekurzorů
Další znaky popisují v normální kostní dřeni vývoj (CD 10, CD34, CD20, cytoplazmatické TdT, povrchové i cytoplazmatické IgM). COlO Zn ak COlO, tzv. CALLA i.lntigen, byl popsán jako typický znak pro "common
E. Mej stříkov~,.-.-:s::t~ra:.:n:.:a=---=1..::6.:...7____________ _ _ ______________
ALL" (1 1- 14) . B lymfocyty v periferní krvi tento znak nemají, na rozdíl od většiny B-prekurzorů v kostní dřeni. Leukemické blasty jej nesou ve více než 95 %, velmi často ve vysokých množstvích (tzv. hyperexprese CDlO). V periferní krvijsoll B buňky většinou CDlO negativní,jen malá subpopulace je D 10 slabě pozitivní (tzv. transitional B buňky). CDIO hyperexprese v perifcmí krvi je tedy specifická pro leukemii (obr. 2). Nelú liniově specifický, vyskytuje se také u granulocytú (15) a ve stadiu kortikáLního thymocytu i v rámci vývoje T řady. Část T-ALL je rovněž poz.itivllí, intenzita je ale zpravidla nižší než u BCPALL (1 J).
CD3..1 časných progenitorů a kmenových buněk. Není specifický. Lm unoglobulin M QgM), těž ký řetězec V nom1ální B lymfopoéze je po úspě š né VDJ rekombinaci nejprve exprimován tě ž ký ře t ězec IgM cytoplazmaticky, poté je spolu s VpreB molekulou vystaven na
Z nak
l i niově
10'
-r----
-
některé
leukemie a lze proto využít pro sledování MRN
( 17).
Aberantní znaky u B-lymfocytární
řady
Znaky om zené v normálním vývoji pro jiné řady, ale vyskytující se často na leukemických buňkách (18) . CD66e CKOR-SA3544 antigen) Aktivačn í znak granulocytů, u nemaligních B prekurzorů se nevysky tuje, proto jej můžeme na B prekurzorech považovat za znak malignity (6). CD33. C D l3 Znak monocytů a granulocytů, aberantně se vyskytuje na BCP-ALL. CDl5, CD65 Znak granulocytů, aberantně se vyskytuje na BCP-ALL. NG2 Molekula chondroitin sulfátu, ve fyziologické hematopoéze tato molekula není exprimována. NG2 molekula se aberantně vyskytuje u leukemií (ALL i AML) s přestav bou genu MLL (obr. 3).
- - - ----,
3
T-akutní lymfoblastická leukemie CT-ALU Primárním místem 10' 101 T lymfopoézy je thymus, 10' o o o leukemické buňky však o Ci Ci <.l <.l <.l 10' 10' cirkulují v periferní krvi 'O' a kostní dřeni. Přítomnost nezralých T lymfocytů 10\ 1o~ 10' 10 ' 10' IO v periferní krvi či kostní C020 dřeni tedy ukazuje na diaObr. 2. Obnizek vlevo ukazuje normálni vývoj B řady podle exprese antigenu C020 a CO 10, uprostřed na gnózu T-ALL. Typicpozadí normální KD (kontury) je zobraze na COlO negati vní pro S ALL a vpravo cALL. kde u většiny blastů je hyperexprcse CD 10. Všechny grafy j sou 7. gam CD 19pos bun ě k S odpovídajícími op tickými vlastnostmi. kou známkou malignity u T- ALL je koexprcse CD7 a CD5 na CD3 negativních buňkách nebo kocxprese povrch, což spouští signalizaci umožňující proliferaci CD4 a CD8 v periferní krvi nebo kostní dřeni. prekurzoru s ú s pě š ně přes tavěným antigenním receptorem. Klas ifikace leukemií používá detekci cytoplazmatického IgM jako marker pro preB ALL, povrchové IgM Liniově specifické znaky pro T-lymfocytární řadu definuje zralou B-Ieukemii, která j nejméně obvyklá CD7 a liší se prognózou i léčbou . CD7 je glykoprotein exprimovaný na zralých T lymCD20 focytech, NK buňkách a thymocytech, ale i na některých se objevuje zhruba paralelně s dokončením přestavby nediferencovaných progenitorech. Je to senzitivní martěžkého řetězce a s jeho cytoplazmatickou expresí. U leuI D' 10' kemií je často přítomen, někdy asynchronně s hyperexpresí CD 10. Z pravidla je pozi tivní jen na subpopulaci 10' 10' blastů, což. limituje eventuální vyuŽ.ítí specifické antiW W C D20 léčby u BCP-ALL. Zralá B-ALL je typicky vyso'. o.. 10' 0... 102 N N ce pozitivní. (') (') 10'
10 '
J
z
Z 10'
10'
Tenninální deoxy-nukleotidyl transferáza CTdT) Cytoplazmatická TdT je typicky přítomna u leukemických bUllěk, v normální kostní dřeni se vyskytuje jen u ma lé subpopulace pre kurzorů (především CD 19+34+ 10+ až ++), lze využít pro odlišení maligní a ne maligní B lyrnfoproliferace. CD58 Je znak nespeeifiek)! pro B řadu, ale jeho vysoká exprese spolu s vysokou expresí CD 10 je charaktelistická pro
10'
10'
",'
C019
,,~
10'
C019
Obr. 3. Aberantní exprese NG2 na BC? ALL blastech u pacientky kojeneckou ALL s proká zaný m fú zním genem MLUAF4 (vlevo). Vpravo obrázek u téže pac ientky 2 roky od diagnózy onemoc něn í v kompletní remisi a negati vním různím genem MLUAF4 . Je zře te lné nenulové pozadí v regenerující KO, které limituje vyui ití při detekci MRN .
?l
E.
Mejstříková,
strana 168
ker T-ALL, ale není specifický, jelikož jej exprimují také n ěkteré typy myeloidních leukemií. Malá část nezralých progenitorů ve fyziologické KD je CD7+33+. CD3 CD 3 je součástí komplexu T-bun ěč ného receptoru (TCR). CD3 j e exprimován v cytoplazmě časnými thymocyty. Cytoplazmatická exprese CD3 je základním znakem, který zařazuje buňky k T řadě. CDS.í! C02 Zn aky CD2 a COS jsou exprimovány na thymocy tec h a zralých T Iymfocytech. COS také definuje podskupi nu B I ym fo cytů (tzv. T independentní BI lymfocyly). CD2 nalézáme také na NK buňk ác h. Oba antigeny jsou exprimovány na více než 90 % T-A LL a používáme je jako podpůrn é z naky. C0 2 je aberantně exprimována u čás ti AML, ty picky u AML M 3v (6).
Znaky definující vývoj ová stad ia a piny
fu nkční
podsku-
,04 Znak definující "T-helper" podskupinu T lymfoc y tů v periferní krvi, zároveň je exprimován také monocy ty. V thymu se objevuje nejprve u dvojitě pozitivních thymocytů (spolu s CO S). Je koreceptorem komplexu CD3 -TCR. Bývá pozitivní u AML s monocytárním vyzráváním . C08 Definuje cytotoxické T lymfocyty (antigen specifické z abij eč e). V th ym u se objevuje nejprve u dvojitě pozitivních thymocytů (s polu s CD4). Je koreceptorem komplex u C0 3-TCR. TCRcd3 TCRyO T-buně č ný receptor, je základním nástroj e m T bun ěk při specifickém rozeznávání antigenů. D 1a Je znak ex primovaný nezralými T Iymfocyty v thymu. Definuje intermediární T-ALL (19). Q222 CD 99 j e exprimováno na řadě hematopoetic kýc h bun ě k i progenitorů, v případě thymocytů a T-ALL je exprese kvantitativně zvýšená (20).
'"o
'0'
10'
1O'
1O'
,,'
u
1O'
'"o
u
'0·
Sledovállí průběhu
léčby
Současný přístup
k sestavování l éčebných protokolů ALL používá strategii terapie šité na míru rizika selhání léčby. Jedním z parametru je odpověď na léčbu. Již přes dvě dekády se hodnotí rychlost redukce počtu blastů při prednisonové předfázi. Nyní se zavádí v některých protokolech detekce minimální reziduální nemoci (MRN) v průběhu terapie jako kritérium pro přeřazení mezi rizikovými skupinami. K detekci MRN se používá me toda kvantitativní P R a zatím pouze výzkumně také průtoková cytometrie. Ne docela vyře šeným probl émem aplikace cytometrie ve sledování MRN je značná podobnost leukemických buněk a regenerujících B-Iymfocytů, které v některýc h časových bodech terapie tvoří významné pozadí. Pro klinicky použitelné měření MRN je třeba detekovat leukemickou buňku na pozadí 1 000 až 10 000 buněk. Zlepšení specifity očekáváme od zavedení mnohobarevných protokolů (více než 4 fluorescenční znaky).
Nové prediktory Studie expresního profilování (EP, expression profi!ing, gene chi ps, microarrays) přinášejí data o expresi deseti ti s íců genů. Jejich vyhodnocením je možné hledat sk upiny (například ALL pacientů) , které mají podobné expresní vzorce ("expression patterns"), tzn. je pro n ě charakteristická exprese některých ge nů , nebo hledat geny, které jsou typické pro jednotlivé skupiny (a vztahovat je k biologii, le ukemogenezi a prognóze) (21). Odhady počtu genů, které j so u kličové pro idcntifikaci určité skupiny pacientů ( např. genotypové a rizikové), se liší. Někteří autoři předpokládaji, že ge nů určujících např. genotypové poskupiny nemusí být více než 20 (např. Downing (22). Takový počet genu, či odpovídajících proteinů je potom možné studovat metodami molekulární biologie (RT-PCR) či prutokové cytometrie. Průtoková cytometrie má výhodu simultánního stanovení více molekul na jedné buňce a relativní nenároč no s ti na kvalitu, množství vzorku a na provedení. V naší s tudii využíváme dat z EP k nalezení těc hto klíčových molekul. Poda ři lo se nám nalézt dvě molekuly (CD44 a CD27) předpovída jící přítomnost fúzního genu TEL/AMLl (23). Chen et al. (17) navrhli na základě dat z EP detekci molekuly CDS8 při vyšetřování minimální residuální nemoci. Stále se také intenzivně pracuje na standardizaci EP pro diagnostické účely, ale jako racion á lnější se jeví využití této metody pro výzkumné účely a pro nalezení nových diag nostických a prognostických znaků.
'0·
Závěr O br. 4. Srovnání imunofenotypu nemaligních T lymfocytll (v kontunic h) a intermcdiámí T-ALL (zvýraz.něná populace obdélníkem). Na levém obrázku (kombinace CD99/7/S/3, ga te CD7p<Jl bez hyperexprese CDSI"'z pro kontury a CD7 ""S,I'Ibé p Ol CD99++) typicky vysoká exprese CD99 a nízká až' negativní povrchová CD3. Na pravé m obrázku koexprese CD4 a CD8 z kombinace CD4/8/3.
Vhodné a metodicky dobře provedené použití průtokové cytometrie má významnou roli pIi diagnostice a diferenciální diagnostice leukemie. lmuno[cnotypizací odlišíme nemaligní prekurzory od leukemických buněk v aspirátu kostní dřeně či petifemí krve. Na základě liniově spec ific-
E.
Mejstříková,
strana 169
kých znaků stanovíme typ leukemických buněk. Další znaky definují srupdí zralosti a můžou před povědět i riziko selhání terapie. Mnohobarevná průtoková cytometrie je výzkumn ě používána k vyhledávání a stanovení prediktivní hodnoty dalších molekul a ke sledováni MRN.
Seznam použitých zkratek
KD Bep MRN
TCR ALL AML
MLL
kostní dřeň B cell precursor minimální reziduální nemoc T cell receptor akutní lymfoblastická leukemie akutní myeloidní leukemie mixed Iineage leukemia
Práce byla podpořena grantem lGA MZ CR 7430-3. Poděkováni
Výsledky jsou umožněny spoluprací všech center v rámci Pracovní skupiny pro dětsko u hematologii (B. Blažek, Z. Cerná, Y Jabali, V Mihál, D. Procházková, 1. Sta rý, 1. Stěrba, 1. Hak, K. Toušovská). LaborantkLim a laborantům Ústavu imunologie a Kliniky dětské hematologie a onkologie za pomoc při získávání a analýze dat (1. Ridošková, K. Posp íšilová, L. Gondorčínová, D. Thurner, P Semerák, K. Mužíková, L. Řezníčková, K. Krejčíková, A. Brabencová).
Literatura
--------------------------
I. Uekun FM, Sathcr flN, G aynon PS, Arlhur DC, lligg ME, Tubergen DG, cl a l. Clinical Features and Trentrnent Outcome of Chi ld ren With Myeloid Antigen Positivc Acute Lymphoblastic Leukemia: A Report From the Children's Cancer Group. Blood 1997; 90( I): 28-35. 2. Putti MC RondeW R Cocito MG , Arico M , Sainati L, Conler V, et al. Exp'ression of MyeloiďMarkers Lacks Prognostic Impact in Children Treated for Acute Lymphoblastic Leukemia: Italian Experience in AIEOP-ALL 88-91 Stud ies . Blood 1998; 92(3): 79580 1. 3. iles F, Estey E, O' Bnen S. Ge mtu Lumab ozogamicin in the treatment of acutc myeloid leukemia. Cancer 2003; 98( IO): 2095-2 104. 4. Mejstriko\'a E, Kalina T, Trka J, Stary J, Hrusak O. Correl ation of CD33 with poorer prognosis in chi ldhood ALL implicates a potential of anti-CD33 frontl ine th erapy. Leukemia 2005; 19(6):1092-1094. 5. 8cne MC, Castoldi G , Knapp W, L udwig WD, M atutes E, Orfao A, et al. Proposn ls for thc immunologica l c1assífication of acute leukemias. Europcan Group for the Immunological Charac te rization of Leukemias (EGll..). Leukemia 1995; 9(10): 1783- 1786. 6. Hrl 'ak 0, Porwit-MacDonald A. Antige n expression pattems rcflectin genotype of acute Iellkemias. Leukemia 2002; 16(7): 12331258. 7. Dworwk MN, l"ritsch G, F roschl G, Printz D, Gadner H. Fourcolor flow cytometric investigalion of termi nal deoxy nucleotidyl transferase-posi tive Iymphoid precurso rs in pediatric bone marrow: CD79a expression precedcs CD 19 in early B-cell ontoge ny. Blood 1998; 92(9): 3203-3209. 8. Loken MR, Shah VO, Dattilio KL, C i\'in Cl. Flow cytometric analysis of human bone nwnow. ll. Normal B Iymphocyte developmen t. Blood 1987 ; 70(5) : 13 16-1324.
9. Colernan M, Goldenberg DM, Siegel AB, Ketas JC, Ashe M, Fiore JM, et a l. Epraruzumab : Targeting B-Ce ll Malignancies through CD22. Clin Cancer Res 2003; 9(10): 399 1-3994. 10. DiJoseph J l", Armellino D , 8 0ghaert ER, KJlandke K, Dougher MM. Sridharan L, ct al. Antibody- targeted chel11olherapy with CMC-544: a CD22-targeted immunoconjugale ol' calicheamicin fo r the treatment of B-Iymphoid malignanc ies. Bl ood 2004; 103(5) 1807-1814. ll. Sbipp MA , Slefano GB, Switzer SN, G riffin ro, R einher.t EL. CD tO (CALLA)/neutral endopeptidase 24 .11 modlllates innarrunatory peptide-induced changes in neutrophil morphology, migrdlion, and adhesion proteins and is it~elť regulated by neutrophil activation. Blood 1991; 78(7): 1834-1 841. 12. Shipp MA, Slefano GB, D ' damío L, Switzer SN, Howard FD, Sinisterra J, et al. Dowru'egulation of enkephalin-mediated innammatory responses by CD 10/neutral endopeptidase 24. 11. Narure 1990; 347(629 1) 394-396. 13. Letarte M, Vera S, Tran R, Addis J8, O nizu ka RJ , Quackenbush EJ, et al. Common acute Iymphocytic leukemia anti gen is identical to neutral endopeptidase. J Exp Med 1988; 168(4): 1247-1253 . 14. Brown G, Greaves MF, Lister TA, Rapson N, Papamichael M. Expression of human T and B Iymphocyte cell -surface markers on leukaemi c cells. Lancet 1974; 2(7883): 753-755. 15. Bene Me. lrrununophenotyping of acute leukaemias. lmmunol Len 2005; 98(1): 9-2 1. 16. Barcena A, Muench MO, Galy AH, Cupp J, Roncarolo MG, Phillips JH, et al. Phenotypic and functional analys is o f T-cell precursors in the human fetal liver and thymus: CD7 express ion in the early stages of T- and myeloid-cell develop ment. Blood 1993; 82(11) : 3401-3414. 17. C hen JS, C oustan-Smíth E, Suzuki T, Neale GA, Mihara K, Pui CH, et al. ldentification of novel markers for monitoring minimal residual disease in acu te Iymphoblastic leukemia. Blood 200 1; 97(7): 2 11 5-2120. 18. Kalina T, Vasko\'a M, Mejstriko\'a 1:. Madzo J, Trka J, Sl
Klinika
MUD" Tomáš Kali"a . Ph.D. hemarologie a onkologie LF2UK a FN MOlol CLlP-Childhood leukemia In ves/i g(uion pJ"{//ur V Úva lu 84, Praha 5. 15006
dělské
Korespondenční
ou/o r: MUD/: Esrer Mejslríková Klin ika dětsk,l hemalologie a onkologie LF2 UK a FN MOIOI V Úv,,11I84 /5006 Pr,,110 5 e-m.oil: eSle/:1I1ejs/rikol'a@!{molol.culli.cz Dodcíno: 5. I. 2006
Nijato: 23. J. 2006