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.
This is a preview of subscription content
Subscribe to Nature+
Get immediate online access to the entire Nature family of 50+ journals
Subscribe to Journal
Get full journal access for 1 year
only $3.90 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Tax calculation will be finalised during checkout.
Get time limited or full article access on ReadCube.
All prices are NET prices.
Bosman, F. T., Carneiro, F., Hruban, R. H. & Theise, N. D. WHO Classification of Tumours of the Digestive System 4th edn (International Agency for Research on Cancer, 2010)
Corbo, V. et al. MEN1 in pancreatic endocrine tumors: analysis of gene and protein status in 169 sporadic neoplasms reveals alterations in the vast majority of cases. Endocr. Relat. Cancer 17, 771–783 (2010)
Jiao, Y. et al. DAXX/ATRX, MEN1, and mTOR pathway genes are frequently altered in pancreatic neuroendocrine tumors. Science 331, 1199–1203 (2011)
Missiaglia, E. et al. Pancreatic endocrine tumors: expression profiling evidences a role for AKT-mTOR pathway. J. Clin. Oncol. 28, 245–255 (2010)
Neychev, V. et al. Mutation-targeted therapy with sunitinib or everolimus in patients with 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. BMJ Open 5, e008248 (2015)
Elsässer, S. J., Allis, C. D. & Lewis, P. W. Cancer. New epigenetic drivers of cancers. Science 331, 1145–1146 (2011)
Heaphy, C. M. et al. Altered telomeres in tumors with ATRX and DAXX mutations. Science 333, 425 (2011)
Marinoni, I. et al. Loss of DAXX and ATRX are associated with chromosome instability and reduced survival of patients with pancreatic neuroendocrine tumors. Gastroenterology 146, 453–460 (2014)
Song, S. et al. qpure: A tool to estimate tumor cellularity from genome-wide single-nucleotide polymorphism profiles. PLoS One 7, e45835 (2012)
Nones, K. et al. Genomic catastrophes frequently arise in esophageal adenocarcinoma and drive tumorigenesis. Nat. Commun . 5, 5224 (2014)
Waddell, N. et al. Whole genomes redefine the mutational landscape of pancreatic cancer. Nature 518, 495–501 (2015)
Popova, T. et al. Genome Alteration Print (GAP): a tool to visualize and mine complex cancer genomic profiles obtained by SNP arrays. Genome Biol . 10, R128 (2009)
Mermel, C. H. et al. GISTIC2.0 facilitates sensitive and confident localization of the targets of focal somatic copy-number alteration in human cancers. Genome Biol . 12, R41 (2011)
Alexandrov, L. B. et al. Signatures of mutational processes in human cancer. Nature 500, 415–421 (2013)
Nik-Zainal, S. et al. Mutational processes molding the genomes of 21 breast cancers. Cell 149, 979–993 (2012)
Nik-Zainal, S. et al. Landscape of somatic mutations in 560 breast cancer whole-genome sequences. Nature 534, 47–54 (2016)
Patch, A. M. et al. Whole-genome characterization of chemoresistant ovarian cancer. Nature 521, 489–494 (2015)
Al-Tassan, N. et al. Inherited variants of MYH associated with somatic G:C-->T:A mutations in colorectal tumors. Nat. Genet . 30, 227–232 (2002)
Aretz, S. et al. MUTYH-associated polyposis (MAP): evidence for the origin of the common European mutations p.Tyr179Cys and p.Gly396Asp by founder events. Eur. J. Hum. Genet. 22, 923–929 (2014)
Vogt, S. et al. Expanded extracolonic tumor spectrum in MUTYH-associated polyposis. Gastroenterology 137, 1976–1985 (2009)
Stephens, P. J. et al. Massive genomic rearrangement acquired in a single catastrophic event during cancer development. Cell 144, 27–40 (2011)
Rausch, T. et al. Genome sequencing of pediatric medulloblastoma links catastrophic DNA rearrangements with TP53 mutations. Cell 148, 59–71 (2012)
Georgitsi, M. et al. Germline CDKN1B/p27Kip1 mutation in multiple endocrine neoplasia. J. Clin. Endocrinol. Metab. 92, 3321–3325 (2007)
Francis, J. M. et al. Somatic mutation of CDKN1B in small intestine neuroendocrine tumors. Nat. Genet . 45, 1483–1486 (2013)
Lubensky, I. A. et al. Multiple neuroendocrine tumors of the pancreas in von Hippel-Lindau disease patients: histopathological and molecular genetic analysis. Am. J. Pathol . 153, 223–231 (1998)
Dong, X. et al. Mutations in CHEK2 associated with prostate cancer risk. Am. J. Hum. Genet . 72, 270–280 (2003)
Gonzalez-Perez, A. et al. IntOGen-mutations identifies cancer drivers across tumor types. Nat. Methods 10, 1081–1082 (2013)
Gerlinger, M. et al. Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. N. Engl. J. Med . 366, 883–892 (2012)
Li, B. E. et al. Distinct pathways regulated by menin and by MLL1 in hematopoietic stem cells and developing B cells. Blood 122, 2039–2046 (2013)
Tang, M. et al. The malignant brain tumor (MBT) domain protein SFMBT1 is an integral histone reader subunit of the LSD1 demethylase complex for chromatin association and epithelial-to-mesenchymal transition. J. Biol. Chem. 288, 27680–27691 (2013)
Asai, A. et al. High-resolution 400K oligonucleotide array comparative genomic hybridization analysis of neurofibromatosis type 1-associated cutaneous neurofibromas. Gene 558, 220–226 (2015)
Baba, T. et al. Persephin: A potential key component in human oral cancer progression through the RET receptor tyrosine kinase-mitogen-activated protein kinase signaling pathway. Mol. Carcinog . 54, 608–617 (2015)
Lindahl, M. et al. Human glial cell line-derived neurotrophic factor receptor alpha 4 is the receptor for persephin and is predominantly expressed in normal and malignant thyroid medullary cells. J. Biol. Chem. 276, 9344–9351 (2001)
Alers, S., Löffler, A. S., Wesselborg, S. & Stork, B. Role of AMPK-mTOR-Ulk1/2 in the regulation of autophagy: cross talk, shortcuts, and feedbacks. Mol. Cell. Biol . 32, 2–11 (2012)
Sturm, D. et al. New brain tumor entities emerge from molecular classification of CNS-PNETs. Cell 164, 1060–1072 (2016)
Delattre, O. et al. Gene fusion with an ETS DNA-binding domain caused by chromosome translocation in human tumours. Nature 359, 162–165 (1992)
May, W. A. et al. Ewing sarcoma 11;22 translocation produces a chimeric transcription factor that requires the DNA-binding domain encoded by FLI1 for transformation. Proc. Natl Acad. Sci. USA 90, 5752–5756 (1993)
Sankar, S. & Lessnick, S. L. Promiscuous partnerships in Ewing’s sarcoma. Cancer Genet . 204, 351–365 (2011)
Stockman, D. L. et al. Malignant gastrointestinal neuroectodermal tumor: clinicopathologic, immunohistochemical, ultrastructural, and molecular analysis of 16 cases with a reappraisal of clear cell sarcoma-like tumors of the gastrointestinal tract. Am. J. Surg. Pathol. 36, 857–868 (2012)
Brohl, A. S. et al. The genomic landscape of the Ewing Sarcoma family of tumors reveals recurrent STAG2 mutation. PLoS Genet . 10, e1004475 (2014)
Crompton, B. D. et al. The genomic landscape of pediatric Ewing sarcoma. Cancer Discov . 4, 1326–1341 (2014)
Tirode, F. et al. Genomic landscape of Ewing sarcoma defines an aggressive subtype with co-association of STAG2 and TP53 mutations. Cancer Discov . 4, 1342–1353 (2014)
Lovejoy, C. A. et al. Loss of ATRX, genome instability, and an altered DNA damage response are hallmarks of the alternative lengthening of telomeres pathway. PLoS Genet . 8, e1002772 (2012)
Sadanandam, A. et al. A cross-species analysis in pancreatic neuroendocrine tumors reveals molecular subtypes with distinctive clinical, metastatic, developmental, and metabolic characteristics. Cancer Discov . 5, 1296–1313 (2015)
Lin, S. Y. & Elledge, S. J. Multiple tumor suppressor pathways negatively regulate telomerase. Cell 113, 881–889 (2003)
Bar-Peled, L. et al. A Tumor suppressor complex with GAP activity for the Rag GTPases that signal amino acid sufficiency to mTORC1. Science 340, 1100–1106 (2013)
Fang, M. et al. MEN1 is a melanoma tumor suppressor that preserves genomic integrity by stimulating transcription of genes that promote homologous recombination-directed DNA repair. Mol. Cell. Biol . 33, 2635–2647 (2013)
Wang, Y. et al. The tumor suppressor protein menin inhibits AKT activation by regulating its cellular localization. Cancer Res . 71, 371–382 (2011)
Matkar, S., Thiel, A. & Hua, X. Menin: a scaffold protein that controls gene expression and cell signaling. Trends Biochem. Sci . 38, 394–402 (2013)
Landrum M. J. et al. ClinVar: public archive of interpretations of clinically relevant variants. Nucleic Acids Res . 44, D862–D868 (2016)
Biankin, A. V. et al. Pancreatic cancer genomes reveal aberrations in axon guidance pathway genes. Nature 491, 399–405 (2012)
Lawrence, M. S. et al. Discovery and saturation analysis of cancer genes across 21 tumour types. Nature 505, 495–501 (2014)
Kambara, T. et al. Role of inherited defects of MYH in the development of sporadic colorectal cancer. Genes Chromosom. Cancer 40, 1–9 (2004)
O’Callaghan, N. J. & Fenech, M. A quantitative PCR method for measuring absolute telomere length. Biol. Proced. Online 13, 3 (2011)
Bailey, P. et al. Genomic analyses identify molecular subtypes of pancreatic cancer. Nature 531, 47–52 (2016)
Wilkerson, M. D. & Hayes, D. N. ConsensusClusterPlus: a class discovery tool with confidence assessments and item tracking. Bioinformatics 26, 1572–1573 (2010)
Law, C. W., Chen, Y., Shi, W. & Smyth, G. K. voom: Precision weights unlock linear model analysis tools for RNA-seq read counts. Genome Biol . 15, R29 (2014)
Fang, H. & Gough, J. The ‘dnet’ approach promotes emerging research on cancer patient survival. Genome Med . 6, 64 (2014)
Lewis, T. B., Coffin, C. M. & Bernard, P. S. Differentiating Ewing’s sarcoma from other round blue cell tumors using a RT-PCR translocation panel on formalin-fixed paraffin-embedded tissues. Mod. Pathol . 20, 397–404 (2007)
Rossi, S. et al. EWSR1-CREB1 and EWSR1-ATF1 fusion genes in angiomatoid fibrous histiocytoma. Clin. Cancer Res . 13, 7322–7328 (2007)
Cai, Z., Chehab, N. H. & Pavletich, N. P. Structure and activation mechanism of the CHK2 DNA damage checkpoint kinase. Mol. Cell 35, 818–829 (2009)
Krieger, E., Koraimann, G. & Vriend, G. Increasing the precision of comparative models with YASARA NOVA--a self-parameterizing force field. Proteins 47, 393–402 (2002)
Bell, D. W. et al. Genetic and functional analysis of CHEK2 (CHK2) variants in multiethnic cohorts. Int. J. Cancer 121, 2661–2667 (2007)
We thank E. Missiaglia, S. Beghelli, N. Sperandio, G. Bonizzato, S. Grimaldi, F. Pisani, C. Cantù, G. Zamboni and P. Merlini for assistance at the ARC-Net Research Centre and Verona University; C. Axford, M.-A. Brancato, S. Rowe, M. Thomas, S. Simpson and G. Hammond for central coordination of the Australian Pancreatic Cancer Genome Initiative, data management and quality control; M. Martyn-Smith, L. Braatvedt, H. Tang, V. Papangelis and M. Beilin for biospecimen acquisition; D. Gwynne and D. Stetner for support at the Queensland Centre for Medical Genomics; and The Kinghorn Centre for Clinical Genomics for genome sequencing of validation samples. Funding support was from: Italian Ministry of Research (Cancer Genome Project FIRB RBAP10AHJB); Associazione Italiana Ricerca Cancro (AIRC n. 12182); Fondazione Italiana Malattie Pancreas – Ministero Salute (CUP_J33G13000210001); National Health and Medical Research Council of Australia (NHMRC; 631701, 535903, CDF 1112113, PRF 1025427, SRF 455857, 535903); The Queensland State Government Smart State National and International Research Alliances Program (NIRAP); Institute for Molecular Bioscience/University of Queensland; The Royal Australasian College of Physicians, Sidney Catalyst, NHMRC, Pancare Australia; Australian Government: Department of Innovation, Industry, Science and Research (DIISR); Australian Cancer Research Foundation (ACRF); Cancer Council NSW (SRP06-01, SRP11-01. ICGC); Cancer Institute NSW (10/ECF/2-26; 06/ECF/1-24; 09/CDF/2-40; 07/CDF/1-03; 10/CRF/1-01, 08/RSA/1-15, 07/CDF/1-28, 10/CDF/2-26,10/FRL/2-03, 06/RSA/1-05, 09/RIG/1-02, 10/TPG/1-04, 11/REG/1-10, 11/CDF/3-26); Garvan Institute of Medical Research; Avner Nahmani Pancreatic Cancer Research Foundation; R.T. Hall Trust; Petre Foundation; Philip Hemstritch Foundation; Gastroenterological Society of Australia (GESA Senior Research Fellowship); Royal Australasian College of Surgeons (RACS); Royal Australasian College of Physicians (RACP); Royal College of Pathologists of Australasia (RCPA); QIMR Berghofer Medical Research; The Keith Boden Fellowship (K.N.); NHGRI U54 HG003273; CPRIT grant RP101353-P7; Wellcome Trust Senior Investigator Award (103721/Z/14/Z); CRUK Programme (C29717/A17263 and C29717/A18484); CRUK Glasgow Centre (C596/A18076); CRUK Clinical Training Award (C596/A20921); Pancreatic Cancer UK Future Research Leaders Fund; The Howat Foundation; and the University of Glasgow.
The authors declare no competing financial interests.
Reviewer Information Nature thanks S. Chanock and the other anonymous reviewer(s) for their contribution to the peer review of this work.
A list of participants and their affiliations is provided in the Supplementary Information.
Extended data figures and tables
The chart shows the workflow of analyses conducted on the discovery set of 98 PanNETs and on the validation set of an additional 62 PanNETs and 1 colorectal cancer. CNA, copy-number analysis.
a, Stability plot indicates there are five mutation signatures (>0.9). b, The profile of the five mutational signatures (A–E) and what function has been assigned to these signatures (MUTYH, APOBEC, BRCA, Age and ‘Signature 5’).
a, Four PanNet samples, three of which harboured a pathogenic MUTYH germline variant, and a colon tumour with a pathogenic MUTYH mutation underwent WGS to validate the association of MUTYH biallelic inactivation with the MUTYH mutation signature. b, Family pedigree of the patient with colon cancer. The 64-year-old male patient with colon cancer was identified as a candidate for MUTYH mutation analysis owing to the presence of two synchronous cancers in the proximal colon, each arising in a contiguous tubulovillous adenoma, as well as approximately 50 adenomatous polyps predominantly in the caecum and ascending colon. The index patient’s brother presented with colorectal cancer at 45 years of age and his sister presented with colorectal cancer at 64 years of age and with breast cancer at 59 years of age. The index patient’s son had polyps removed at 36 years of age. Mutation signature analysis was performed using the 98 discovery PanNET samples and the colon and 4 PanNET validation samples. c, Stability plot showing the solution for the five mutational signatures (>0.75). d, The profile of the five mutational signatures (A–E) and what function has been assigned to these signatures (MUTYH, APOBEC, BRCA, Age and ‘Signature 5’). e, The contribution of each signature (mutations per Mb) and proportion of the signatures in each tumour are shown.
a, Top, the number and type of somatic structural rearrangements in each tumour. Bottom, tumours with more events tended to have longer telomeres. b, Two methods were used to determine clusters of somatic structural rearrangement breakpoints. Orange squares, chromosomes with a significant cluster of events as determined by a goodness-of-fit test against the expected distribution (P < 0.0001, Kolmogorov–Smirnov test). Blue squares, chromosomes deemed to harbour a high number of breakpoints because they had a chromosomal breakpoint per Mb rate that exceeded the 75th percentile of the chromosomal breakpoint per Mb rate for the cohort by five times the interquartile range. Red squares, chromosomes for which both of these criteria were met. Clusters of events were reviewed and nine tumours were found to harbour regions of chromothripsis. c, Recurrent chromothripsis for chromosome 11 was detected in four tumours. The chromothripsis event caused loss of the MEN1 gene locus in two of these samples.
a, CHEK2 structure indicating the positions of the germline variants. Mutations are highlighted by rendering as magenta sticks with protein domains coloured as indicated in the adjacent keys. The model includes a superimposed phosphopeptide (red). b, A summary of the CHEK2 variants and their predicted impact on protein structure. To functionally test the CHEK2 variants, a panel of FLAG–CHEK2 constructs encoding P85L, ∆77–82, D177H and E282K was generated. c, d, FLAG western blot of transfected HEK293T whole cell lysates (c) or anti-FLAG immunoprecipitates (d) showed that, compared to the wild type, there was normal expression of P85L but reduced expression of ∆77–82, D177H and E282K. e, Assessment of kinase activity of CHEK2 variants. Immunoprecipitated proteins were incubated either with GST alone (−) or with GST–CDC25C amino acids 200–256 (+) in the presence of γ-P32 ATP. Input and kinase activity were assessed by film radiography (top) and coomassie staining (bottom). Immunoprecipitates of ∆77–82, D177H and E282K had significantly reduced kinase activity in terms of both autophosphorylation and phosphorylation of CDC25C whereas the activity of P85L was normal. f, Quantification of expression levels by western blotting expressed as a fraction of wild type. Data points represent independent experiments. Error bars are mean ± s.e.m. g, h, Quantification of kinase activity. P32 counts for CDC25C (g) and CHEK2 (h) bands were scintillation counted. Corresponding bands from untransfected controls were used for background subtraction. Background-corrected P32 counts per minute were then standardized to wild type for each experiment. Data points represent independent experiments. Error bars are mean ± s.e.m. i, j, Quantification of kinase activity relative to protein expression. Kinase activity (from i and j) was standardized to protein expression level (from f). D177H was not analysed in this manner owing to its very low expression level. Error bars are mean ± s.e.m. Once the low expression level of ∆77–82 is taken into account, it is evident that the expressed protein retains normal kinase activity. On the other hand, E282K is kinase defective even after adjusting for its reduced expression. D177H expression is so low that it is not possible to reliably correct kinase activity for relative expression level, so it is unclear whether D177H is kinase dead as well as unstable. Data are summarized in Supplementary Table 16.
a, The number of SNVs and indels within the genome of each patient (n = 98) is shown in the histogram. The driver plot displays the somatic mutations in key genes or those identified as significantly mutated (Intogen Q < 0.1). SETD2 is also reported, although its Q value was 0.15, as it was recurrently inactivated in six samples and multiple independent deleterious SETD2 mutations were observed in one tumour (a nonsense present at 3%, a missense at 14%, and a frameshift at 11%; only the nonsense is shown but the case is highlighted with a black arrow), suggesting strong selection for SETD2 inactivation in that tumour. b, Somatic mutations in MEN1 are predominantly nonsense mutations or insertions–deletions causing frame shifts and premature protein termination, and occur throughout the protein.
Copy number was determined using Illumina SNP arrays in a cohort of 98 PanNETs. a, Copy number events were mainly comprised of whole chromosome arm loss or gain. Cluster analysis of the chromosome arm level copy number state stratified the tumours into four subtypes. Group 1: recurrent pattern of whole chromosomal loss, affecting specific chromosomes (1, 2, 3, 6, 8, 10, 11, 15, 16 and 22); group 2: samples with a limited number of events, many with loss affecting chromosome 11; group 3: polyploid tumours, with gain of all chromosomes; and group 4: aneuploid tumours, containing predominantly whole chromosome gains affecting multiple chromosomes). b, The proportion of bases within the genome affected by copy number change. c, The mutations per Mb (SNPs and small insertion deletions). d, GISTIC analysis showing recurrent gains (red) and losses (blue) of the entire cohort.
Whole genome sequence data were used to estimate telomere length in PanNETs relative to the matched normal sample. a, Telomere length estimated by whole-genome sequencing correlated with the telomere length calculated from qPCR (R2 = 0.8091). Values are plotted on a log10 scale. b–e, Boxplots were used to show the association of relative telomere length and DAXX or ATRX and MEN1 mutation status. Mann–Whitney tests were used to determine significant associations (P < 0.05). b, c, Tumours harbouring DAXX or ATRX mutations contain longer telomeres. d, Tumours harbouring MEN1 mutations contain longer telomeres. e, Telomere length is shown in relation to DAXX or ATRX and MEN1 somatic mutations.
Unsupervised clustering, network and gene enrichment analysis for available RNA-seq data identify PanNET subgroups associated with hypoxia and metabolic reprogramming. a, Unsupervised clustering identified three distinct PanNET subgroups (1–3). b, A gene signature defining three expression groups previously described in PanNETs showed enrichment of expression of the intermediate-group genes43 in Group 1 and the metastasis-like PanNET (MLP) genes43 in Group 3. c, Network analysis identified a significant sub-network of genes differentially expressed between Group 3 and other groups (Group 1 and Group 2). Red nodes represent genes upregulated in Group 3 and green nodes represent genes upregulated in other groups. Shaded areas represent network communities. d, Gene enrichment analysis for genes belonging to the sub-network shown in b. e, Heatmap showing the differential expression of genes belong to the identified sub-network. Somatic mutations in some of the recurrently mutated genes are shown (MEN1, DAXX, ATRX and members of the mTOR pathway: DEPDC5, MTOR, PTEN, TSC1 and TSC2).
Kaplan–Meier survival curves. a, b, Tumours harbouring DAXX or ATRX mutations had a poor prognosis in the whole cohort (a) and in the G2 cohort (b). c, Tumours with telomere lengths that were neither short or long had a better prognosis. d, Tumours harbouring mutations in genes that activate the mTOR pathway had a poor prognosis in the G2 cohort (log rank test was used in all instances).
About this article
Cite this article
Scarpa, A., Chang, D., Nones, K. et al. Whole-genome landscape of pancreatic neuroendocrine tumours. Nature 543, 65–71 (2017). https://doi.org/10.1038/nature21063
DNA methylation reveals distinct cells of origin for pancreatic neuroendocrine carcinomas and pancreatic neuroendocrine tumors
Genome Medicine (2022)
APOBEC3 mutational signatures are associated with extensive and diverse genomic instability across multiple tumour types
BMC Biology (2022)
Immunophenotypic and molecular characterization of pancreatic neuroendocrine tumors producing serotonin
Modern Pathology (2022)