Whole-genome landscape of pancreatic neuroendocrine tumours

TLDR: In this paper, the authors performed whole-genome sequencing of 102 primary pancreatic neuroendocrine tumours and defined the genomic events that characterize their pathogenesis, including a deficiency in G:C,>T:A base excision repair due to inactivation of MUTYH, which encodes a DNA glycosylase.

Abstract: The diagnosis of pancreatic neuroendocrine tumours (PanNETs) is increasing owing to more sensitive detection methods, and this increase is creating challenges for clinical management. We performed whole-genome sequencing of 102 primary PanNETs and defined the genomic events that characterize their pathogenesis. Here we describe the mutational signatures they harbour, including a deficiency in G:C > T:A base excision repair due to inactivation of MUTYH, which encodes a DNA glycosylase. Clinically sporadic PanNETs contain a larger-than-expected proportion of germline mutations, including previously unreported mutations in the DNA repair genes MUTYH, CHEK2 and BRCA2. Together with mutations in MEN1 and VHL, these mutations occur in 17% of patients. Somatic mutations, including point mutations and gene fusions, were commonly found in genes involved in four main pathways: chromatin remodelling, DNA damage repair, activation of mTOR signalling (including previously undescribed EWSR1 gene fusions), and telomere maintenance. In addition, our gene expression analyses identified a subgroup of tumours associated with hypoxia and HIF signalling.

Figures

Figure 1 | Mutational signatures in pancreatic neuroendocrine tumours. a, Five signatures (A–E) were identified in 98 PanNET samples. Signature A is previously undescribed and has been named MUTYH because all tumours dominated by this signature carried a germline inactivating mutation in MUTYH with concurrent loss of the wild-type allele; signatures B to E have been previously described and are reported with their given names. Tumours with a high MUTYH, APOBEC, BRCA or Age signature showed a higher number of mutations per megabase. b, Validation of MUTYH signature in four additional PanNETs and one colon tumour. Three PanNETs and the colon tumour contained a dominant MUTYH signature and harboured a germline-damaging MUTYH mutation with concurrent loss of the wild-type allele, and one PanNET with a benign MUTYH variant did not contain the signature. c, The tumour with the BRCA signature contained a BRCA2 germline variant (R3052W) and genomic instability as seen in pancreatic ductal adenocarcinoma. d, Variants displayed above the MUTYH protein were associated with the MUTYH signature and contained somatic biallelic inactivation of the MUTYH gene (highlighted in bold); those below were not and are either benign or showed no loss of the wild-type allele. Asterisks show variants reported as pathogenic by ref.50 (IDs #5294 & #5293). Variants predicted by

Full text

Scarpa, A. et al. (2017) Whole-genome landscape of pancreatic
neuroendocrine tumours. Nature, 543(7643), pp. 65-71.
(doi:10.1038/nature21063)
This is the author’s final accepted version.
There may be differences between this version and the published version.
You are advised to consult the publisher’s version if you wish to cite from
it.
http://eprints.gla.ac.uk/137698/
Deposited on: 12 December 2018
Enlighten – Research publications by members of the University of Glasgow
http://eprints.gla.ac.uk

Page 1 of 29
Whole-genome landscape of pancreatic neuroendocrine tumours
Aldo Scarpa
1,2
*§, David K. Chang
3,4, 7,29,36
* , Katia Nones
5,6
*, Vincenzo Corbo
1,2
*, Ann-Marie Patch
5,6
, Peter
Bailey
3,6
, Rita T. Lawlor
1,2
, Amber L. Johns
7
, David K. Miller
6
, Andrea Mafficini
1
, Borislav Rusev
1
, Maria
Scardoni
2
, Davide Antonello
8
, Stefano Barbi
2
, Katarzyna O. Sikora
1
, Sara Cingarlini
9
, Caterina Vicentini
1
, Skye
McKay
7
, Michael C. J. Quinn
5,6
, Timothy J. C. Bruxner
6
, Angelika N. Christ
6
, Ivon Harliwong
6
, Senel
Idrisoglu
6
, Suzanne McLean
6
, Craig Nourse
3, 6
, Ehsan Nourbakhsh
6
, Peter J. Wilson
6
, Matthew J. Anderson
6
, J.
Lynn Fink
6
, Felicity Newell
5,6
, Nick Waddell
6
, Oliver Holmes
5,6
, Stephen H. Kazakoff
5,6
, Conrad Leonard
5,6
,
Scott Wood
5,6
, Qinying Xu
5,6
, Shivashankar Hiriyur Nagaraj
6
, Eliana Amato
1,2
, Irene Dalai
1,2
, Samantha
Bersani
2
, Ivana Cataldo
1,2
, Angelo P. Dei Tos
10
, Paola Capelli
2
, Maria Vittoria Davì
11
, Luca Landoni
8
, Anna
Malpaga
8
, Marco Miotto
8
, Vicki L.J. Whitehall
5,12,13
, Barbara A. Leggett
5,12,14
, Janelle L. Harris
5
, Jonathan
Harris
15
, Marc D. Jones
3
, Jeremy Humphris
7
, Lorraine A. Chantrill
7
, Venessa Chin
7
, Adnan M. Nagrial
7
, Marina
Pajic
7
, Christopher J. Scarlett
7,16
, Andreia Pinho
7
, Ilse Rooman
7
†, Christopher Toon
7
, Jianmin Wu
7,17
, Mark
Pinese
7
, Mark Cowley
7
, Andrew Barbour
18
, Amanda Mawson
7
†, Emily S. Humphrey
7
, Emily K. Colvin
7
,
Angela Chou
7,19
, Jessica A. Lovell
7
, Nigel B. Jamieson
3,4,20
, Fraser Duthie
3,21
, Marie-Claude Gingras
22,23
,
William E. Fisher
23
, Rebecca A. Dagg
24
, Loretta M.S. Lau
24
, Michael Lee
25
, Hilda A. Pickett
25
, Roger R.
Reddel
25
, Jaswinder S. Samra
26,27
, James G. Kench
7,27,28
, Neil D. Merrett
29,30
, Krishna Epari
31
, Q. Nguyen
32
,
Nikolajs Zeps
33,34,35
, Massimo Falconi
8
, Michele Simbolo
1
, Giovanni Butturini
8
, George Van Buren II
23
, Stefano
Partelli
8
, Matteo Fassan
1
, Australian Pancreatic Cancer Genome Initiative‡, Kum Kum Khanna
5
, Anthony J.
Gill
7,27
, David A. Wheeler
22
, Richard A. Gibbs
22
, Elizabeth A. Musgrove
3
, Claudio Bassi
8
, Giampaolo Tortora
9
,
Paolo Pederzoli
8
, John V. Pearson
5, 6
, Nicola Waddell
5,6
§, Andrew V. Biankin Biankin
3,4,7,29,36
§ & Sean M.
Grimmond
37
§
1
ARC-Net Centre for Applied Research on Cancer, University and Hospital Trust of Verona, Verona 37134,
Italy.
2
Department of Pathology and Diagnostics, University and Hospital Trust of Verona, Verona 37134, Italy.
3
Wolfson Wohl Cancer Research Centre, Institute of Cancer Sciences, University of Glasgow, Garscube Estate,
Switchback Road, Bearsden, Glasgow G61 1QH, UK.
4
West of Scotland Pancreatic Unit, Glasgow Royal Infirmary, Glasgow G31 2ER, UK.
5
QIMR Berghofer Medical Research Institute, Herston Road, Brisbane 4006, Australia.
6
Queensland Centre for Medical Genomics, Institute for Molecular Bioscience, The University of Queensland,
St Lucia, Brisbane, Queensland 4072, Australia.
7
The Kinghorn Cancer Centre, Cancer Division, Garvan Institute of Medical Research, University of New South
Wales, 384 Victoria St, Darlinghurst, Sydney, New South Wales 2010, Australia.
8
Department of Surgery, Pancreas Institute, University and Hospital Trust of Verona, Verona 37134, Italy.
9
Medical Oncology, University and Hospital Trust of Verona, Verona, Italy.

Page 2 of 29
10
Department of Pathology, General Hospital of Treviso, Department of Medicine, University of Padua, Italy.
11
Department of Medicine, Section of Endocrinology, University and Hospital Trust of Verona, Verona, Italy.
12
The University of Queensland, School of Medicine, Brisbane 4006, Australia.
13
Pathology Queensland, Brisbane 4006, Australia.
14
Royal Brisbane and Women’s Hospital, Department of Gastroenterology and Hepatology, Brisbane 4006,
Australia.
15
Institute of Health Biomedical Innovation, Queensland University of Technology, Brisbane, Australia.
16
School of Environmental & Life Sciences, University of Newcastle, Ourimbah, New South Wales 2258,
Australia.
17
Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education/Beijing), Centre for
Cancer Bioinformatics, Peking University Cancer Hospital & Institute, Beijing 100142, China.
18
Department of Surgery, Princess Alexandra Hospital, Ipswich Rd, Woollongabba, Queensland 4102, Australia.
19
Department of Anatomical Pathology, St Vincent’s Hospital, Sydney, New South Wales 2010, Australia.
20
Academic Unit of Surgery, School of Medicine, College of Medical, Veterinary and Life Sciences, University
of Glasgow, Glasgow Royal Infirmary, Glasgow G4 OSF, UK.
21
Department of Pathology, Queen Elizabeth University Hospital, Greater Glasgow & Clyde NHS, Glasgow
G51 4TF, UK.
22
Department of Molecular and Human Genetics, Human Genome Sequencing Center, Baylor College of
Medicine, One Baylor Plaza, MS226, Houston, Texas 77030-3411, USA.
23
Michael E. DeBakey Department of Surgery and The Elkins Pancreas Center, Baylor College of Medicine,
One Baylor Plaza, Houston, Texas 77030-3411, USA.
24
Children’s Hospital at Westmead, Westmead, New South Wales 2145, Australia.
25
Children’s Medical Research Institute, The University of Sydney, Westmead, New South Wales 2145,
Australia.
26
Department of Surgery, Royal North Shore Hospital, St Leonards, Sydney, New South Wales 2065, Australia.
27
University of Sydney, Sydney, New South Wales 2006, Australia.
28
Tissue Pathology and Diagnostic Oncology, Royal Prince Alfred Hospital, Camperdown, New South Wales
2050, Australia.
29
Department of Surgery, Bankstown Hospital, Eldridge Road, Bankstown, Sydney, New South Wales 2200,
Australia.
30
School of Medicine, Western Sydney University, Penrith, New South Wales 2175, Australia.
31
Department of Surgery, Fremantle Hospital, Alma Street, Fremantle, Western Australia 6160, Australia.

Page 3 of 29
32
Department of Gastroenterology, Royal Adelaide Hospital, North Terrace, Adelaide, South Australia 5000,
Australia.
33
School of Surgery M507, University of Western Australia, 35 Stirling Highway, Nedlands, Western Australia
6009, Australia.
34
St John of God Pathology, 12 Salvado Rd, Subiaco, Western Australia 6008, Australia.
35
Bendat Family Comprehensive Cancer Centre, St John of God Subiaco Hospital, Subiaco, Western Australia
6008, Australia.
36
South Western Sydney Clinical School, Faculty of Medicine, University of New South Wales, Liverpool, New
South Wales 2170, Australia.
37
University of Melbourne Centre for Cancer Research, University of Melbourne, Melbourne, 3010, Victoria,
Australia.
†Present addresses: Oncology Research Centre, Vrije Universiteit Brussel, Brussels, Belgium (I.R.); Pancreatic
Cancer Translational Research Group, Adult Cancer Program, Prince of Wales Clinical School, Lowy Cancer
Research Centre, UNSW, Sydney, Australia (A.M.).
‡A list of participants and their affiliations is provided in the Supplementary Information.
*These authors contributed equally to this work.
§These authors jointly supervised this work.
The diagnosis of pancreatic neuroendocrine tumours (PanNETs) is increasing owing to
more sensitive detection methods, and this increase is creating challenges for clinical
management. We performed whole-genome sequencing of 102 primary PanNETs and
defined the genomic events that characterize their pathogenesis. Here we describe the
mutational signatures they harbour, including a deficiency in G:C>T:A base excision
repair due to inactivation of MUTYH, which encodes a DNA glycosylase. Clinically
sporadic PanNETs contain a larger-than-expected proportion of germline mutations,
including previously unreported mutations in the DNA repair genes MUTYH, CHEK2
and BRCA2. Together with mutations in MEN1 and VHL, these mutations occur in 17%
of patients. Somatic mutations, including point mutations and gene fusions, were
commonly found in genes involved in four main pathways: chromatin remodelling,
DNA damage repair, activation of mTOR signalling (including previously undescribed

Page 4 of 29
EWSR1 gene fusions), and telomere maintenance. In addition, our gene expression
analyses identified a subgroup of tumours associated with hypoxia and HIF signalling.
Pancreatic neuroendocrine tumours (PanNETs) are the second most common epithelial
neoplasm of the pancreas and have a mortality rate of 60%
1
. The World Health Organization
(WHO) classification, which assesses the proliferative fraction of neoplastic cells, divides
PanNETs into three groups: low grade (G1), intermediate grade (G2), and high grade (G3).
While G3 tumours are invariably lethal, 90% of PanNETs are grade G1 or G2. These have an
unpredictable clinical course that varies from indolent to highly malignant. Our current
understanding of the molecular pathology of G1 and G2 PanNETs is insufficient for their
clinical management, where the challenge is to predict the aggressiveness of individual
tumours in order to identify patients who will benefit from early aggressive therapy and to
minimize harm from the inadvertent overtreatment of patients with indolent disease.
PanNETs are usually sporadic but also occur as part of three hereditary syndromes:
multiple endocrine neoplasia type 1 (MEN-1), von Hippel-Lindau syndrome (VHL), and
occasionally tuberous sclerosis complex (TSC)
1
. Somatic mutations of MEN1 occur in 35%
of PanNETs
1–3
. Recent expression profiling and exome sequencing have highlighted the
importance of activated mTOR signalling as a druggable mechanism in 14% of patients
3,4
and, although the mTOR inhibitor Everolimus is approved by the FDA for the treatment of
advanced PanNET
5
, it is not yet possible to use molecular analysis to select patients who will
benefit. In addition, the apoptotic regulator DAXX or the chromatin modifier ATRX are
mutated in up to 40% of PanNETs
3,6
, where they promote alternative lengthening of
telomeres (ALT) and chromosomal instability
7,8
.
Our comprehensive molecular analysis of 102 clinically sporadic PanNETs defines
their molecular pathology and identifies several novel candidate mechanisms that activate
mTOR signalling, including novel gene fusion events. We have uncovered an important role
for germline MUTYH variants through a novel G:C>T:A mutational signature. In addition,
we have identified a larger-than-anticipated germline contribution to clinically sporadic
PanNETs, delineating future challenges in the clinical assessment of susceptibility.
Genomic landscape of PanNETs
The study workflow is illustrated in Extended Data 1. Patients were recruited and consent for
genomic sequencing obtained as part of the International Cancer Genome Consortium
(http://www.icgc.org). All cases were classified according to WHO criteria
1
. The cohort

Frequently asked questions

Q1. What is the role of mTOR in solid tumours?

Telomere maintenance: upregulation of TERT and telomere lengthening is a well-established pro-survival mechanism in solid tumours. 

Q2. What is the role of mTOR in the development of panNETs?

2) The mutational status of DAXX, ATRX and mTOR pathway genes could be used to stratify the prognosis of intermediate grade (G2) PanNETs, the subgroup with the least predictable clinical behaviour. 

Q3. What is the role of mTOR in the development of a new therapeutic target?

3) The identification of previously undescribed mechanisms that activate mTOR signalling may lead to the development of biomarkers that could be used to predict therapeutic responsiveness to mTOR inhibitors such as everolimus, which are currently poorly defined. 

Q4. How many tumours contained telomeres that were 1.5 longer than matched normal?

Twenty-four tumours contained telomeres that were 1.5× longer than matched normal DNA and 36 contained telomeres that were 1.5× shorter than matched normal DNA. 

Q5. How many mutations were found in ATRX or DAXX mutant tumours?

In total, 22 of 26 ATRX or DAXX mutant tumours displayed ALT, and in DAXX mutations were more frequent (19/22) than ATRX mutations (3/22), in contrast to in vitro studies in which ATRX alterations are more prevalent43. 

Q6. What is the significance of the fusion events in ALT tumours?

ALT tumours have been reported to undergo recurrent regions of gene copy gain and loss in a panel of human cancer cell lines in vitro43; in contrast, whole-chromosome loss of specific chromosomes predominates in PanNETs. 

Q7. what is the ad hoc protocol for a phase II clinical trial?

Mutation-targeted therapy with sunitinib or everolimus in patientswith advanced low-grade or intermediate-grade neuroendocrine tumours of the gastrointestinal tract and pancreas with or without cytoreductive surgery: protocol for a phase II clinical trial. 

Q8. What is the MUTYH-associated colorectal polyposis syndrome?

Germline biallelic inactivation ofPage 6 of 29MUTYH causes the autosomal recessive MUTYH-associated colorectal polyposis syndrome and is associated with somatic G:C>T:A transversions in the APC gene, the driver of colorectal polyps18. 

Q9. What was the significance of the ALT test?

Telomere integrity and PanNET molecular subtypes Telomere repeat content was quantified using whole-genome sequencing data, and ALT was assessed using C-tailing qPCR in 86 cases (Extended Data Fig. 8a). 

Q10. What is the common variant of MUTYH in the Italian cohort?

the two missense MUTYH mutations identified in the Italian cohort (c.536A>G, p.Y179C; c.1187G>A, p.G396D) are the most common MAP-linked variants in populations of European origin and have been shown to be founder mutations in a recent haplotype analysis of 80 families with MAP from Italy and Germany19. 

Q11. What is the morphological and immunophenotypical features of PanNETs?

The three tumours that conatined EWSR1 fusions had morphological and immunophenotypical features typical of PanNETs, had absent or weak staining for CD99, and lacked any clinicopathological features of Ewing sarcoma. 

Q12. What other genes encoded negative regulators of mTOR signalling?

PTEN mutations were mutually exclusive with mutations in TSC1 (n = 2) and TSC2 (n = 2), which encode other negative regulators of mTOR signalling (Supplementary Table 5). 

Q13. What is the average rate of recurrent chromosome loss in panNETs?

Structural rearrangements are less common in PanNETs (mean, 29 events per tumour; range 3–216) (Supplementary Table 6) than in pancreatic ductal adenocarcinoma (119 per tumour; range 15–558)11. 

Q14. How did the authors identify the three tumours?

To verify their findings, the authors used amplicon sequencing of MUTYH on 62 additionalPanNETs and identified three tumours bearing pathogenic germline mutations coupled with LOH. 

Q15. How many significantly and recurrently mutated genes were defined?

Sixteen significantly and recurrently mutated genes were defined using IntOGen27 analysis (Q < 0.1) (Extended Data Fig. 6a, Supplementary Table 12). 

Q16. what is the role of mTOR in the regulation of autophagy?

S., Löffler, A. S., Wesselborg, S. & Stork, B. Role of AMPK-mTOR-Ulk1/2in the regulation of autophagy: cross talk, shortcuts, and feedbacks. 

Q17. What mutations were found in the recurrent catastrophic rearrangements?

Four of these nine tumours had recurrent catastrophic rearrangements on chromosome 11q (Extended Data Fig. 4b), all involving 11q13, and two of these rearrangements led to loss of MEN1. 

Q18. What was the average mutation rate in the RPCL subtype?

Thepolyploid group had the highest somatic mutation rate (P 0.002, Mann–Whitney test) with an average of 1.98 mutations per Mb (Extended Data Fig. 7c).