prof. dr. P. Devilee, p/a Marian Kantebeen secretariaat Nijbakker-Morra Stichting Postbus 9600, L1-Q 2300 RC LEIDEN tel: 071-526 6624 fax: 071-5248 158
Dr. Lodewijk A.A. Brosens Afdeling Pathologie, UMC Utrecht (H04-312) Heidelberglaan 100 3584 CX Utrecht Tel: 088-7574762 Email:
[email protected]
Utrecht, 1 juni 2015
Betreft: eindverslag Nijbakker-Morra Stipendium
Geachte heer Devilee, Van 1 mei 2014 tot 1 mei 2015 ben ik als postdoctoral research fellow werkzaam geweest in de onderzoeksgroep van Prof. Dr. Hruban en Dr. Laura Wood in the Sol Goldman Pancreatic Cancer Research Center van het Johns Hopkins Ziekenhuis. Per 1 mei ben ik weer werkzaam in Nederland als gastrointestinaal patholoog in het UMC Utrecht. Bijgevoegd een verslag van het werk dat ik Baltimore heb verricht. Door het uitstel van mijn stage is het onderwerp van mijn research anders dan ik in mijn initiële aanvraag had aangeven. Mijn research heeft zich gericht op de genetische karakterisering van het moment van invasie in intraductaal papillair mucineus neoplasme (IPMN). Door middel van whole exome sequencing van de niet invasieve component van een IPMN en het hieruit ontstane pancreas carcinoom hopen we genetische of epigenetische veranderingen te vinden die cruciaal zijn voor het ontstaan van carcinoom uit een IPMN. In een volgend stadium van dit onderzoek zal ook naar genetische heterogeniteit gekeken worden, maar daar ben ik in het afgelopen jaar niet aan toe gekomen. Het onderzoek en de samenwerking met Dr. Wood zal ik de komende jaren voorzetten met behulp van een Career Development Grant van de Maag Lever Darm Stichting die ik tijdens mijn verblijf in de VS in 2014 heb ontvangen. Naast het wetenschappelijk werk heb ik in het Johns Hopkins Ziekenhuis ook volop kunnen meekijken met gerenommeerde pathologen op het gebied van gastrointestinale pathologie waaronder Dr. Montgomery en Dr Hruban en mij zodoende verder kunnen specialiseren. Het jaar in Johns Hopkins is niet alleen wetenschappelijk zeer succesvol geweest maar heeft ook sterk bijgedragen aan mijn verdere specialisatie als patholoog in de gastrointestinale pathologie. Ik ben de Nijbakker-Morra Stichting dan ook zeer dankbaar voor het in mij gestelde vertrouwen en het mede mogelijk maken van dit belangrijke jaar in mijn ontwikkeling. Met vriendelijke groet,
Lodewijk Brosens
Fellowship Project Report Dr. L.A.A. Brosens PROJECT TITLE: Characterization of the Moment of Invasion in Pancreatic Cancer OBJECTIVES: 1. Define the genetic events that underlie the moment of invasion by comparing the somatic genetic alterations in the non-invasive components of IPMNs to those in associated minimally invasive carcinomas arising from these lesions. 2. Determine the extent of heterogeneity in these alterations at the level of single cells. WORK CARRIED OUT: Selection of cases for genetic analysis The Pathology Archive of the Johns Hopkins Hospital was searched for pancreatic resection (Whipple) specimens with intraductal papillary mucinous neoplasm (IPMN) and an associated carcinoma (Figure 1). In addition, cases from multiple institutions in the USA as well as Europe, and Asia were collected through our international network of pancreatic cancer researchers. In total 157 Whipple resections with IPMN associated pancreatic cancer have been meticulously reviewed by four dedicated pancreatic cancer pathologists (Dr Hruban, Dr. Offerhaus, Dr. Wood and Dr. Brosens). Many cases appeared not just right for WES due to several reasons. An important issue appeared to be the size of the invasive carcinoma associated with IPMN. The invasive cancer associated with an IPMN cannot be too small since it is then impossible to obtain sufficient DNA for whole exome sequencing (WES). However, if the cancer is too large it troubles the relationship between the IPMN and cancer and distinction between a dilated pancreatic duct with cancerization by an adjacent cancer or IPMN is often impossible. We therefore decided to only include cases with and invasive component <2 cm clearly arising from an IPMN. Also, carcinomas arising from IPMNs are often colloid (mucinous) carcinomas and therefore poorly cellular and not suitable for WES. Although many cases therefore appeared to be not suitable for WES, we have now collected and processed 20 cases for WES. In addition, we have collected 30 cases where the invasive carcinoma is too small for DNA isolation for WES, but which can be used for targeted sequencing approaches once genes involved in malignant transformation of IPMN have been identified.
Figure 1. An IPMN with high-grade dysplasia (left lower corner) and an associated adenocarcinoma. Tissue processing and DNA analysis In order to perform both WES and targeted analyses of intratumoral heterogeneity from the same cases, a specific tissue processing system was developed (Figure 2). For DNA isolation for WES, DNA was isolated from the invasive and non-invasive components of the IPMNs using core-ing instead of laser microdissection. Importantly, before the cores for WES are taken from the paraffin blocks, 20 membrane slides and 7 IHC slides were prepared from each
case for future laser capture microdissection for targeted sequencing of specific already known mutations to address genetic heterogeneity and immunohistochemistry and stored at -20°C. For core-ing, areas with invasive cancer and areas with non-invasive IPMN were identified on the HE slides and matched to the paraffin block. Then 0.6 mm thick cores from invasive and non-invasive areas were taken separately from the paraffin block. After core-ing, a new HE slide was made to see if the areas of interest were cored successfully. If necessary, additional cores were taken. DNA was isolated from the tissue cores using Proteinase K digestion and a Qiagen FFPE DNA isolation kit. Currently 20 cases have been submitted for Whole Exome Sequencing using the Illumina HiSeq sequencer. Sequencing results are currently ready for 12 of the 20 cases. The last 8 cases are currently being sequenced.
Methods 20 membrane sections for LCM for targeted sequencing + 10 IHC sections
DNA isolation, library preparation, Whole Exome Sequencing
Figure 2: Tissue processing protocol RESULTS OBTAINED: Currently 12 cases have been successfully sequenced using the Illumina HiSeq sequencer and preliminary data analysis is done. The last 8 cases are currently being sequenced. Quality parameters of WES are summarized in Table 1. Table 1. Summary of quality parameters of WES. NORMAL Alias Distinct Reads per Base MTP1 128 MTP2 112 MTP3 92 MTP4 115 MTP5 81 MTP7 68 MTP8 105 MTP6 71 MTP9 81 MTP11 67
HIGH GRADE DYSPLASIA CARCINOMA % Bases >10 Distinct Reads % Bases >10 Mut-based Distinct Reads Reads per Base Reads Cellularity per Base (%) 93 191 94 68 196 92 182 93 68 194 90 175 93 61 194 87 175 93 45 191 91 82 93 67 59 90 124 92 39 191 93 166 94 81 187 87 178 93 33 189 90 127 92 71 96 90 77 92 50 101
% Bases >10 Mut-based Reads Cellularity (%) 94 94 93 93 93 92 94 93 93 92
55 59 31 45 45 23 27 57 38 26
MTP13 60 MTP14 82
87 90
147 143
92 93
60 52
123 151
93 93
28 66
Mean 88.5
90
147.3
92.8
57.9
156
93.1
41.7
Preliminary sequencing results are summarized in Table 2. In brief, between 24 and 67 somatic sequence alterations were found in non-invasive IPMN and 20 and 88 in invasive cancer, respectively. Between 4 and 38 genes were mutated in the invasive cancer but not in the noninvasive IPMN, and there thus represent potential candidate genes with a role in initiation of invasive growth. Between 16 and 56 genes were mutated in both non-invasive IPMN and invasive cancer and these included genes well known to be important in pancreatic cancer (KRAS, GNAS, P53, CDKN2A, ATM and RNF43). Genes only mutated in the non-invasive IPMN but not in the invasive cancer likely represent passenger mutations without an important role in carcinogenesis. Table 2. Preliminary sequencing results
MTP1 MTP2 MTP3 MTP4 MTP5 MTP7 MTP8 MTP6 MTP9 MTP11 MTP13 MTP14
Total Mutations in mutations HGD 45 33 101 67 57 52 51 47 72 56 64 60 91 59 56 49 29 24 60 57 48 44 96 64
Mutations in Cancer 41 88 32 49 61 57 69 46 25 59 20 58
Mutations in HGD ONLY 4 13 25 2 11 4 22 10 4 1 28 38
Mutations in Cancer ONLY 12 34 5 4 16 1 32 7 5 3 4 32
Shared Mutations in HGD and cancer 29 54 27 45 45 56 37 39 20 56 16 26
MEAN Min Max
64.2 29 101
50.4 20 88
13.5 1 38
12.9 1 34
37.5 16 56
51 24 67
Further analysis of this data will include visual inspection of raw sequencing data and confirmation of mutations by a second technique (e.g. Sanger sequencing, Ion Torrent). Data will be prioritized by using common sense about the type of mutations and types of genes where these mutations are found (e.g. is the mutation predicted to result in protein change or predicted to be silent?, is a mutation present in multiple tumors? is a gene predicted to be a tumor suppressor or oncogene?). In addition, direct comparative sequencing of non-invasive IPMN and associated invasive carcinoma will allow us to isolate only those mutations that underlie clinically relevant transitions (IPMN to cancer) and to distinguish genetic drivers from passengers. Finally, targeted sequencing assays using Ion Torrent will be developed to confirm mutations found by Illumina sequencing and to address genetic heterogeneity. During my postdoc year at Johns Hopkins I was awarded a Career Development Grant of the Dutch Digestive Foundation which allows me to continue and finish these studies during the flowing years.
Articles published/submitted during Fellowship: 1 2 3
4 5 6
Brosens LA, Hackeng WM, Offerhaus GJ, Hruban RH, Wood LD. Pancreatic adenocarcinoma pathology: changing “landscape”. J Gastrointest Oncol. In press. Brosens LA, Offerhaus GJ, Canto MI, Montgomery EA, Giardiello FM. Simultaneous Juvenile Polyposis Syndrome and Neurofibromatosis type 1. In press Basturk O, Hong S-M, Wood LD, Adsay NV, Albores-Saavedra J, Biankin AV, Brosens LA, Fukushima N, Goggins M, Hruban R, Kato Y, Kern S, Klimstra DS, Klöppel G, Krasinskas A, Longnecker DS, Maitra A, Matthaei H, Offerhaus GJA, Shimizu M, Takaori K, Terris B, Yachida S, Esposito I, Furukawa T. Revised Recommendations for Assessing Precursor Lesions in the Pancreas. Am J Surg Path Submitted Tieu AH, Edelstein D, Axilbund J, Romans KE, Brosens LA, Wiley E, Hylind L, Giardiello FM. Clinical Characteristics and Natural History of Multiple Colorectal Adenoma Patients without Germline APC or MUTYH Mutations. Journal of Clinical Gastroenterology Submitted Brosens LA, Offerhaus GJ, Giardiello FM. Hereditary Colorectal Cancer: Genetics and Screening. Surgical Clinics of North America. Accepted. Brosens LA, Giardiello FM, Offerhaus GJ, Montgomery EA. Syndromic gastric polyps: at the crossroads of genetic and environmental cancer predisposition In: Stem Cells, Pre-neoplasia, and Early Cancer of the Upper Gastrointestinal Tract. Editors: Sir Nicholas Wright and Marnix Jansen Springer, in press
Other scientific activities during fellowship: -
Intraductal Neoplasia Group PanIN Classification Meeting The Johns Hopkins Medical Institutions (June 16-18, 2014) 4th Annual Practical Genomics: From Biology to Biostatistics Workshop, Center for Computational Genomics, Johns Hopkins, Baltimore, USA (August 18-20, 2014) 14th Annual Current Topics in Gastrointestinal and Liver Pathology, Baltimore, USA (October 11-12, 2014) (15 CME credits) Johns Hopkins Pathology Grand Rounds (May 2014-May 2015; 20 CME credits) Guest editor Surgical Pathology clinics issue on pancreatic pathology (co-guest editor with Dr. L Wood)
