Abstract
Neuroendocrine neoplasms (NENs) are a relatively rare group of heterogeneous tumours originating from neuroendocrine cells found throughout the body. Pancreatic NENs (PanNENs) are the second most common pancreatic malignancy accounting for 1–3% of all neoplasms developing in the pancreas. Despite having a low background mutation rate, driver mutations in MEN1, DAXX/ATRX and mTOR pathway genes (PTEN, TSC1/2) are implicated in disease development and progression. Their increased incidence coupled with advances in sequencing technologies has reignited the interest in PanNEN research and has accelerated the acquisition of molecular data. Studies utilising such technological advances have further enriched our knowledge of PanNENs’ biology through novel findings, including higher-than-expected presence of germline mutations in 17% of sporadic tumours of no familial background, identification of novel mutational signatures and complex chromosomal rearrangements and a dysregulated epigenetic machinery. Integrated genomic studies have progressed the field by identifying the synergistic action between different molecular mechanisms, while holding the promise for deciphering disease heterogeneity. Although our understanding is far from being complete, these novel findings have provided the optimism of shaping the future of PanNEN research, ultimately leading to an era of precision medicine for NETs. Here, we recapitulate the existing knowledge on pancreatic neuroendocrine tumours (PanNETs) and discuss how recent, novel findings have furthered our understanding of these complex tumours.
Introduction
Neuroendocrine neoplasms (NENs) develop from cells of neuroendocrine differentiation distributed throughout the body, most commonly in the small intestine, lung and pancreas (PanNENs) (Dasari et al. 2017). PanNENs were originally called ‘islet cell tumours’ due to the belief that they arise from islet cells. More recently, it has been suggested that they may originate from pluripotent stem cells of the ductal-acinar system (Vortmeyer et al. 2004, Klöppel et al. 2007).
NENs are rare tumours previously described as ‘orphan’ cancers (Kawasaki et al. 2018). However, this rarity may be due to variation in disease coding/nomenclature between (inter)national cancer registries leading to inaccurate incidence and prevalence rates (IR and PR, respectively) (Niederle et al. 2010, Genus et al. 2017). Recent population-based studies indicate higher IRs than previously described (Fraenkel et al. 2015) with a 3- to 5-fold increase during the last 3–4 decades both worldwide and in the United Kingdom (Halfdanarson et al. 2008, Yao et al. 2008, Dasari et al. 2017, Genus et al. 2017). Increased detection rates are attributed to several factors including use of more sensitive imaging/diagnostic modalities, refined markers of neuroendocrine differentiation (for example, chromogranin A, synaptophysin, neuron-specific enolase, pancreatic polypeptide), establishment of accurate records, improvements in training and increased awareness both in the clinical field and the general population.
The IR of PanNENs is currently 0.43/100,000 having been 0.17 in 1973–1977 (2003–2007 SEER) (Lawrence et al. 2011). Due to the indolent nature of some NENs the PR is higher than any other upper gastrointestinal or liver cancer, second only to colorectal cancer when considering gastrointestinal malignancies (Yao et al. 2008, Dasari et al. 2017). PanNENs account for 1–3% of all pancreatic neoplasms and are the 2nd most common pancreatic malignancy (Bosman et al. 2010). Autopsy studies have demonstrated that up to 10% of PanNENs remain clinically undetected, an observation that may impact the reported IR/PR (Kimura et al. 1991, Halfdanarson et al. 2008, Fraenkel et al. 2012). Survival rates vary depending on grade and stage at diagnosis (Yao et al. 2008) with 5-year survival rates post resection for well-differentiated tumours of around 65% (de Wilde et al. 2012a ). In patients with distant unresectable disease (stage IV), most commonly liver metastases, 5-year survival is 16–46% (Mignon et al. 2000, Viúdez et al. 2016, Batukbhai et al. 2019) (Fig. 1A).
Overview of the multi-layered pancreatic neuroendocrine tumours (PanNET) complexity. (A) Discrimination of PanNENs into those associated with a familial syndrome, due to germline mutations in tumour suppressor genes, and sporadic tumours further sub-divided into functional and non-functional. Epidemiological and survival rates (orange box) and tumour classification according to the WHO 2017 criteria (yellow boxes) are also shown. (B) Involvement of major molecular pathways/mechanisms in PanNET pathogenesis (numbered 1–6) and observed genomic and epigenomic alterations in various implicated genes reported in the literature. Disease mutational burden information in relation to pancreatic ductal adenocarcinomas is shown in the green box. Note: Numbers next to the various genes indicate the pathway/mechanism in which each gene belongs to. Pathways not indicated by a number are those previously implicated in PanNETs but their role needs further investigation to be fully elucidated. Panel A data from (1) Batukbhai et al. (2019), (2) Lloyd et al. (2017), (3) Lawrence et al. (2011), (4) de Wilde et al. (2012a), (5) Ries et al. (2007), (6) Fishbein & Nathanson (2015), (7) Zhang et al. (2013), (8) Franko et al. (2010), (9) Marx et al. (2005). Note: in (B), frequency of genetic events described according to studies discussed in the text. Amp, amplification; CA, chromosomal aberrations; CIMP, CpG island methylator phenotype; cnv, copy number variation; GF, gene fusion; GM, germline mutations; HyperM, hypermethylated, HypoM, hypomethylated; LoH, loss of heterozygosity; MI, Mitotic Index; PDAC, pancreatic ductal adenocarcinoma, PD, poorly differentiated; PTC, post-transcriptional control; re, rearrangement; SM, somatic mutations; WD, well differentiated. *More detailed description of miRNA involvement in PanNETs in Supplementary Table 1.
Citation: Endocrine-Related Cancer 26, 9; 10.1530/ERC-19-0175
Functionally, PanNENs are divided into functional and non-functional tumours (F-PanNENs and NF-PanNENs, respectively). F-PanNENs are associated with various clinical syndromes leading to earlier diagnosis due to clinical manifestation of hormone secretion-associated symptoms (Fig. 1A). In contrast, NF-PanNENs pose a significant early-diagnostic challenge since symptoms only become evident later on in the disease course due to the growing tumour mass. At initial diagnosis, approximately 60–80% of cases are detected with locally advanced or metastatic disease (Baur et al. 2016, Viúdez et al. 2016), resulting in a worse prognosis.
Classification of PanNENs, according to World Health Organization (WHO) criteria, is based on the degree of cellular differentiation and fraction of proliferating neoplastic cells as a measure of tumour grade (determined by the mitotic count and the Ki-67 labelling index). PanNENs are classified into well-differentiated (WD), low-to-intermediate-grade (G1 and G2, respectively) pancreatic neuroendocrine tumours (WD-PanNETs, >90%) and poorly differentiated (PD), high-grade, G3 pancreatic neuroendocrine carcinomas (PanNECs). While G1/G2 NETs can be either functional or non-functional, G3 NECs are mainly non-functional due to loss of hormone secretion-associated genes (Sadanandam et al. 2015). This distinction in histological/morphological characteristics and clinical outcomes is also reflected in the molecular/genetic landscape between the two entities (Yachida et al. 2012, Girardi et al. 2017). While TP53 and RB1 recurrent mutations are found in NECs, they are rarely seen in NETs. However, reports indicate that 35–50% of G3 PanNECs are well differentiated with similar clinical course and molecular profiles to G2 PanNETs (Vélayoudom-Céphise et al. 2013, Sorbye et al. 2014, Basturk et al. 2015, Tang et al. 2016). The WHO classification criteria have been recently updated (Lloyd et al. 2017) to address this heterogeneity by further sub-dividing tumours into WD G3 PanNETs and PD G3 PanNECs.
PanNENs are highly heterogeneous and complex tumours, which demonstrate diverse clinical behaviours (rate of development and progression, metastatic potential, treatment response/resistance and survival rates) that ultimately define tumour fate. Although various prognostic indicators and classification systems evolved over time have been clinically useful, they have their own inherent limitations and, most importantly, they do not reflect the biological and clinical heterogeneity of PanNETs. Such heterogeneity is more pronounced in G1 and G2 tumours that, unlike the predictable lethality of G3 NECs, have a variable clinical course ranging from indolent to aggressive (Lloyd et al. 2017).
In this review, we provide an update on the evolving genomic and epigenomic landscape of G1/G2 PanNET development and progression and discuss recent novel findings. As PD G3 NECs represent a separate tumour entity, they are not discussed further.
Genetic landscape of PanNETs
Cancer-predisposition syndromes
Although PanNETs are most commonly sporadic, they also arise in the familial setting due to inherited germline mutations in tumour suppressor genes (TSG) that increase organ susceptibility to cancer development. These cancer-predisposition syndromes (Fig. 1A) account for <10% of PanNETs (Marx et al. 2005) and are discussed below. Recently, germline mutations in DNA damage repair genes, such as CHEK2 and MUTYH, have been identified in 17% of sporadic cases of apparently no familial background (Scarpa et al. 2017).
Multiple endocrine neoplasia type 1 and 4 (MEN1/MEN4)
MEN1 is an autosomal dominant syndrome caused by germline mutations in the MEN1 gene (chr. 11q13). Affected individuals often display loss of the WT MEN1 allele (Larsson et al. 1988), resulting in PanNET development in ~40% of cases. Loss of heterozygosity (LoH) is found not only in MEN1-associated tumours, but also in patients without MEN1 (Friedman et al. 1989). Although most of the resulting tumours are non-functional, ~10–25% of cases develop insulinomas (Anlauf et al. 2007). Other common manifestations include parathyroid tumours (with hyperparathyroidism occurring in 95% of patients), anterior pituitary tumours, duodenal gastrinomas, adrenocortical and carcinoid tumours. There are also cutaneous signs including angiomas and lipomas (Vortmeyer et al. 1998). Germline mutations in MEN1 were also recently identified in sporadic PanNETs of apparently no familial history (Scarpa et al. 2017). It is, therefore, important that any patient presenting with a PanNET and a history of primary hyperparathyroidism or pituitary adenomas is referred for genetic testing.
PanNETs also develop in the context of multiple endocrine neoplasia type 4 (MEN4) due to CDKN1B germline mutations. CDKN1B encodes the cyclin-dependent kinase inhibitor p27 with mutations leading to increased cell cycle progression and proliferation. The clinical spectrum of MEN4 is similar to MEN1, with ~3% of those suspected to have MEN1 actually harbouring a CDKN1B mutation. CDKN1B has been previously found to be mutated in 8% of small intestinal NETs (Francis et al. 2013).
Von Hippel-Lindau (VHL)
VHL arises due to germline mutations in the VHL gene on chromosome 3p (Seizinger et al. 1988) and shows an autosomal dominant mode of inheritance. Inactivation of the WT allele by somatic point mutations, LoH or loss of 3p results in a number of solid tumours such as haemangioblastomas, clear-cell renal carcinomas, phaechromocytomas (Kaelin 2002) and PanNETs in 12–17% of cases (Binkovitz et al. 1990, Libutti et al. 1998). The VHL-encoded protein forms part of a ubiquitination complex that targets and degrades subunits of hypoxia inducible factor (HIF) 1 and 2. Loss results in increased expression of HIF1/2 and other factors, such as vascular endothelial growth factor (VEGF) and promotes formation of vascularised and cystic tumours (Kaelin 2002).
Tuberous sclerosis complex 1/2 (TSC1/2)
Tuberous sclerosis (TS), linked to mutations in TSC1 and TSC2 genes (chr. 9 and 16, respectively), is a high-penetrance autosomal dominant syndrome associated with benign hamartomas, cognitive impairment and epilepsy. Development of PanNETs is less frequent, ranging between 1.8 and 9% (Anlauf et al. 2007, Larson et al. 2012, Koc et al. 2017). These are mainly due to germline TSC2 mutations, consistent with the observation that sporadic PanNETs harbour recurrent somatic mutations mainly in TSC2 (Jiao et al. 2011). LoH or large chromosome 16 deletions subsequently inactivate the remaining, WT allele. The proteins encoded by TSC1 and TSC2, hamartin and tuberin, dimerise to form a regulatory protein upstream of phosphatidylinositol 3-kinase (PI3K) and mammalian target of rapamycin (mTOR) signalling pathway that regulates cell growth. Dysregulation of the mTOR pathway is an intrinsic characteristic of PanNETs and has been successfully targeted by the drug everolimus (Larson et al. 2012, Koc et al. 2017).
Neurofibromatosis type 1 (NF1)
Germline mutations in the NF1 gene (chr. 17) result in neurofibromatosis type 1 (NF1), a highly penetrant autosomal dominant syndrome. NF1 encodes neurofibromin, both a negative regulator of the Ras-MAP kinase pathway involved in cell growth and proliferation (Evans et al. 2004, Brems et al. 2009) and an inhibitor of mTOR (Johannessen et al. 2005). NF1 patients have distinctive cutaneous features of café au lait patches, neurofibromas and axillary freckling. Patients may also develop malignant tumours including optic or brain stem gliomas, phaeochromocytomas, gastroenteropancreatic (GEP) NETs, mainly of the small intestine and rarely pancreatic somatostatinomas or insulinomas (Anlauf et al. 2007).
Sporadic PanNETs
Over the last 10 years, advances in sequencing technologies have enabled unbiased genome-wide analysis of PanNETs by several groups. Such studies have highlighted the molecular heterogeneity of the disease, driven by the crosstalk between various events (e.g. somatic mutations, variations in copy number and DNA methylation changes) operating at distinct levels (genome and epigenome, including post-transcriptional control; Fig. 1B). Additionally, these studies have confirmed the role of previously identified alterations and revealed novel findings (e.g. involvement of germline mutations, gene fusions and clinically relevant molecular subgroups) adding further information to elucidate the molecular characterisation of PanNET development.
MEN1, histone modifications, chromatin remodelling and telomere maintenance
Somatic mutations in MEN1 are frequently detected in both functional and non-functional sporadic PanNETs and are commonly associated with LoH of the remaining WT allele, consistent with its role as a TSG. Moore et al. (2001) reported that 26% of NF-PanNETs had inactivating MEN1 mutations, while allelic loss of chromosome 11q13 (MEN1 locus) was observed in 70% of informative cases. This frequency was confirmed in a subsequent study providing a link between MEN1 mutations, present in 30% of cases and alterations in protein expression due to the fact that the majority of them (mainly truncating mutations) occur within regions of the gene associated with subcellular protein localisation (Corbo et al. 2010). More recently, Jiao et al. (2011) reported higher MEN1 mutational frequencies (44.1%) confirming its significant role in PanNET development. Most importantly, this study provided the first evidence for recurrently inactivating mutations in two genes that had previously not been implicated in disease pathogenesis. Approximately 43% of tumours were found to carry mutations in DAXX (25%) or ATRX (17.6%). Within the cohort, 23.5% of tumours were found to harbour MEN1 mutations in the presence of either DAXX or ATRX, while 20.6% tumours had a single MEN1 mutation. Collectively, 63.2% of tumours had MEN1/DAXX/ATRX mutations. Interestingly, the observed sequence alterations often resulted in truncating DAXX and ATRX mutations with loss of the corresponding protein product through nonsense-mediated decay (Jiao et al. 2011, Fishbein & Nathanson 2015). The fact that mutations in DAXX/ATRX occur in a mutually exclusive fashion in the majority of cases indicates that both belong in the same pathway and is consistent with common functional roles in apoptosis, chromatin remodelling and telomere maintenance (de Wilde et al. 2012b ).
Menin, encoded by MEN1, is a ubiquitously expressed scaffold nuclear protein that interacts with >40 proteins to control several vital processes including transcriptional regulation, DNA repair, genome stability, cell cycle control and miRNA biogenesis (Agarwal et al. 2005, Balogh et al. 2006, Matkar et al. 2013). It plays a key role in coordinating chromatin-remodelling genes through epigenetic regulation (Kim et al. 2003). MEN1 binds to Akt preventing its plasma membrane translocation and subsequently inhibits the PI3K/Akt/mTOR pathway (Fig. 2A) (Wang et al. 2011). MEN1 alterations can, therefore, trigger a chain of highly orchestrated events resulting in the dysregulation of downstream genes in several inter-connected pathways with catastrophic consequences, the extent of which ultimately shape the fate of a pro-oncogenic cell.
Main pathways and mechanisms implicated in PanNET development and progression. (A) Schematic representation of the interactions between menin/MEN1, PI3K/AKT/mTOR, Ras-Raf-MAPK-ERK and p53 pathways. A phosphorylation cascade initiated by IRS-1 activates PI3K, which in turn results in downstream AKT activation through PDK1-mediated phosphorylation. Subsequently, AKT inhibits TSC2 through phosphorylation resulting in mTOR activation and phosphorylation of the S6K and 4E-BP1 proteins to increase cellular proliferation and survival. PTEN is a negative regulator of PI3K which reduces its activation by dephosphorylating PIP3 to PIP2. TSC2 contains a GAP (GTPase-activating protein) domain that enhances the conversion of Rheb into its GDP-bound inactive state, thus inhibiting mTORC1. In PanNETs, the negative regulators of the mTOR pathway, PTEN, TSC1/2, MEN1 and DEPDC5 are frequently mutated resulting in increased cellular growth. Menin also inhibits the Ras-Raf-MAPK-ERK by preventing the RAS-SOS interaction and indirectly the mTOR pathway by preventing the RAS-mediated activation of PI3K. Also shown, is the p53 pathway. Its function is tightly controlled by the negative regulators MDM2, MDM4 and WIP1. WIP1 also inactivates p53 through dephosphorylation of its upstream activators (ATM, CHK1/2). Upregulation of miR-21 and miR-196 results in PTEN inhibition and AKT regulation, respectively. Approved drugs (sunitinib, everolimus and somatostatin receptor analogues) are also shown. (B) Interaction between the chromatin remodellers ATRX and DAXX in maintenance of telomere integrity and genome stability. Inactive heterochromatin at both telomeric and pericentric regions, typically reflected in the methylation status of lysine residues on histone 3, is recognized by the N-terminal domain of the ATRX protein containing a DNA-binding domain. DAXX is a histone H3.3 chaperone recruited by ATRX to form a heterodimer. The resulting heterodimer guides the H3.3 variant into nucleosomes where it is incorporated at telomeres and pericentric heterochromatin and other repressed areas of the genome. The chromatin-bound deposition complex might also contain the histone chaperone DEK and the histone deacetylase HDACII. Successful H3.3 incorporation into the nucleosomes allows deposition of a histone tail modification that maintains compact chromatin structure to stabilise telomere length. This interaction is inactive in the presence of loss-of-function, mutually exclusive mutations in ATRX and DAXX observed in approximately 43% of PanNETs and leads to the activation of the ALT phenotype. ALT is a telomerase-independent telomere elongation mechanism associated with an altered DNA damage response/repair mechanism involving homologous recombination (HR). ALT-positive cells show uncontrolled expression of telomere motifs (TTAGGG)n responsible for telomeres of heterogeneous length. Also shown is the HIRA complex that mediates deposition of H3.3 at promoters, transcribed regions and facultative heterochromatin in senescent cells. Panel A data from Stevenson et al. (2018), Oberg et al. (2013), Stolz et al. (2011), Hu et al. (2010). Panel B data from Elsaesser & Allis (2010).
Citation: Endocrine-Related Cancer 26, 9; 10.1530/ERC-19-0175
Furthermore, menin regulates the histone methyltransferase activity of a nuclear protein complex including the mixed lineage leukaemia members 1 and 2 (MLL1/2). This promotes trimethylation of histone H3 at lysine residue 4 (H3K4me3) and enhances transcriptional activity (Agarwal & Jothi 2012). Menin-dependent histone methylation maintains in vivo the expression of cyclin-dependent kinase inhibitors (CDKis) p18 Ink4c (CDKN2C) and p27 Kip1 (CDKN1B) through active MLL recruitment to their promoters and coding regions. This increases H3K4m3 and prevents islet cell tumour formation (Milne et al. 2005). Dysregulation of this histone modification complex through menin or MLL loss reduces methylation levels and downregulates CDKis, resulting in uncontrolled cell growth and tumour formation in Men1-mutant mice (Karnik et al. 2005). Collectively, these data highlight a key role for MEN1 in epigenetic control through histone regulation of genes implicated in neuroendocrine cell growth and differentiation. Interestingly, this histone complex is also known to impact the chromatin-remodelling machinery. The involvement of other histone modifications in the aetiology of PanNETs is described in more detail under the epigenetic section of this review.
Mutations in ATRX are present in 17.6% of PanNETs (Jiao et al. 2011). ATRX is a nuclear protein of the SWI/SNF complex of chromatin-remodelling genes and has a vital role in maintaining genomic stability. Its depletion increases local DNA damage and results in loss of structural integrity at telomeres and development of endocrine defects (Watson et al. 2013). Recurrent inactivating mutations and chromosomal rearrangements in other chromatin-remodelling genes (SETD2, ARID2, KMT2C (MLL3) and SMARCA4) were recently documented in PanNETs (Scarpa et al. 2017) with SETD2 and ARID1A mutations present in 18 and 13% of advanced WD-PanNETs, respectively (Raj et al. 2016). SETD2 mutations were exclusively present in intermediate-to-high-grade tumours responding to temozolomide and capecitabine. Mutations in DAXX, a histone H3.3 chaperone recruited by ATRX, occur in 25% of PanNETs (Jiao et al. 2011). Once recruited by ATRX, it forms a heterodimer guiding the H3.3 variant into nucleosomes where it is incorporated at telomeres and centromeres to mediate chromatin remodelling and stabilise telomere length (Lewis et al. 2010) (Fig. 2B).
MEN1 has also been shown to play an important role in the biology of telomeres. Lack of telomerase activity and expression of its protein component, hTERT, has been associated with the gradual shortening of chromosomal ends (telomeres) leading eventually to cell senescence (Lin et al. 2003). It has been previously demonstrated that menin interacts directly with the hTERT promoter to negatively regulate its expression. In contrast, hTERT reactivation due to menin protein loss or MEN1 mutations results in increased telomerase activity and telomere length. In addition, menin depletion results in extended cellular lifespan and immortalisation due to the ability of cells to escape senescence (Lin et al. 2003). A subsequent study, however, has questioned the role of menin in regulating telomerase activity in normal and cancer cells (Hashimoto et al. 2008). In fact, a subset of tumours use alternative means to maintain telomere lengths (Lovejoy et al. 2012). Consistent with an important role for telomere biology in NETs, DAXX and ATRX mutations lead to alternative lengthening of telomeres (ALT; Fig. 2B). ALT is a telomerase-independent telomere elongation mechanism associated with an altered DNA damage response/repair mechanism involving homologous recombination (HR) (Lovejoy et al. 2012). Mutations in genes involved in HR DNA repair have recently been identified in PanNETs (Scarpa et al. 2017) and are described in the following section. It has been previously shown that PanNETs with DAXX or ATRX mutations and/or protein loss show 100% concordance with the ALT phenotype (Heaphy et al. 2011a , Jiao et al. 2011). The ALT phenotype has been described in other human cancers including neuroblastomas, paediatric glioblastomas, oligodendrogliomas and medulloblastomas (Heaphy et al. 2011b , Schwartzentruber et al. 2012).
While neuroendocrine microadenomas, a potential sporadic PanNET precursor lesion, display dysregulated menin expression associated with somatic mutations, they lack DAXX/ATRX mutations and are ALT negative. Therefore, menin/MEN1 alterations are considered as early events in the pathogenesis of PanNETs (de Wilde et al. 2012b , Hackeng et al. 2016). ATRX/DAXX loss/mutations and ALT positivity have been associated with larger tumours (>2–3 cm, G2, T3/T4), chromosomal instability, metastatic disease and reduced relapse-free and tumour-specific survival (Marinoni et al. 2014, Pea et al. 2018) suggesting that these changes are a later event in tumour evolution. This established association may, therefore, have a potential value as a marker of aggressiveness for patient stratification. Multivariate analysis demonstrated that ATRX/DAXX status was an independent prognostic predictor of disease relapse alongside tumour grade and stage (Marinoni et al. 2014). Tumour grade was the only independent predictor of tumour-specific survival. Of note, ALT status was not tested in either multivariate analyses. However, a recent study identified ALT positivity through multivariate analysis as an independent risk factor for liver metastases (Pea et al. 2018).
PI3K/Akt/mTOR pathway
Mutations in the PI3K/Akt/mTOR pathway occur in 14.7% of PanNETs (Jiao et al. 2011). This pathway is a master cancer regulator involved in several vital cellular processes including angiogenesis, cell cycle progression, DNA repair, apoptosis, gene expression and epigenetic regulation (Fig. 2A). These processes are facilitated through crosstalk between upstream regulatory proteins (PTEN, PI3K) and downstream effectors (MDM2, FOXO, GSK-3β) that are, in turn, under the control of various compensatory signalling pathways (Ersahin et al. 2015).
The first evidence suggesting disruption of the mTOR pathway in NF-PanNET pathogenesis was presented in a study describing a metastatic primary case with a PTEN mutation and LoH of the remaining WT allele resulting in loss of protein expression (Perren et al. 2000). Frequent TSC2 and PTEN gene expression downregulation leading to protein level alterations has also been documented in ~80% of NF-PanNETs with more pronounced losses correlating with liver metastases, shorter time to progression and shorter disease-free and overall survival (OS) (Missiaglia et al. 2010). Further evidence implicating the PI3K/Akt/mTOR pathway in disease pathogenesis stems from the first whole-exome sequencing (WES) study of 68 sporadic PanNETs (Jiao et al. 2011) where 14.7% of cases (10/68) had recurrent inactivating (loss-of-function) TSC2 (8.8%) and PTEN (7.3%) mutations. One PTEN-mutant case also harboured a PIK3CA mutation in a region (described as an ‘oncogenic hotspot’) involved in the activation of the kinase domain of the encoded protein (Samuels et al. 2004). Mutations in DEPDC5, another gene in the mTOR pathway, were recently identified (Scarpa et al. 2017). Although PTEN, TSC1, TSC2 and DEPDC5 mutations commonly occur in a mutually exclusive manner, a tumour carrying both PTEN/TSC2 mutations has been reported (Jiao et al. 2011).
Collectively, these findings highlight that dysregulation of the PI3K/Akt/mTOR pathway is an innate characteristic of a significant proportion of PanNETs. This may explain improved survival rates observed in a subset of advanced tumours treated with the mTOR inhibitor everolimus (Yao et al. 2011). However, there is currently insufficient evidence to suggest PI3K pathway mutations as a biomarker of response (Yao et al. 2019). Indeed, current treatment guidelines are based on clinical judgement, while predictive biomarkers of disease recurrence or metastatic progression and treatment response/resistance are still lacking. Recently, a SNP substituting an arginine (R) for glycine (G) in codon 388 of FGFR4 (G388R) was linked to PanNETs of higher stage, risk of liver metastases and reduced response to everolimus (Serra et al. 2012).
DNA damage repair
Single-base excision and HR DNA double-strand break repair systems are both part of the DNA damage repair machinery. Several lines of evidence support the role of the HR DNA repair complex in the development of the ALT phenotype. In telomerase-negative yeasts, genes encoding the HR proteins (such as Rad50, Rad51 and Rad52) are essential for their survival (Cesare et al. 2010). In humans, this complex is also vital for telomere length maintenance and includes several HR proteins such as ATM, CHEK2, BRCA 1 and 2, RAD50 and RAD51. ATM is an important cell cycle checkpoint serine/threonine kinase, which has a genome-wide surveillance role and regulates, upon sensing double-strand DNA breaks, various downstream proteins such as p53 and BRCA1, CHEK2 checkpoint kinase and other RAD checkpoint proteins required for genome stability. ATM has been found mutated in ~5% of PanNETs (Corbo et al. 2012), while germline mutations in BRCA2, CHEK2 and MUTYH, another gene involved in single-base repair, were recently identified in PanNETs (Scarpa et al. 2017).
Germline BRCA2 mutations are associated with familial breast and ovarian cancer, prostate and PDACs. BRCA2 acts with RAD51 to enable double-strand DNA break repair and mutations predispose to cancers harbouring multiple structural genomic rearrangements (Chapman et al. 2012, Stoppa-Lyonnet 2016). CHK2, a serine/threonine protein kinase encoded by CHEK2, is a critical cell cycle checkpoint regulator which is activated in response to double-strand DNA breaks (Stolz et al. 2011). It has several tumour suppressor functions and its dysregulation may result in chromosomal instability. It has been implicated in G1 cell cycle arrest and regulation of apoptosis through p53 stabilisation and phosphorylation, respectively (Ou et al. 2005, Cai et al. 2009, Roeb et al. 2012). CHK2-mediated increase in phosphorylation of NEK6 and CDK–cyclin complexes is also known to result in G2/M cell cycle arrest (Donzelli & Draetta 2003, Lee et al. 2008). In addition, following DNA damage, CHK2 phosphorylates BRCA1 to restore survival, while it also regulates DNA repair through BRCA2 transcriptional stimulation and enhancement of RAD51 interactions with chromatin (Venkitaraman 2001, Jang & Lee 2004). Germline mutations are generally associated with breast cancers, where they confer moderate risk increases, and other carcinomas including sarcomas and brain tumours (Desrichard et al. 2011). MUTYH maintains genome integrity by preventing oxidative damage-induced mutations and coordinates error-free base excision repair (BER), thus preventing tumour development (Mazzei et al. 2013). Germline mutations have been previously linked to MUTYH-associated polyposis (MAP), an autosomal recessive colorectal polyposis syndrome (Sieber et al. 2003) associated with the accumulation of multiple bowel polyps and increased predisposition to colorectal cancer (Al-Tassan et al. 2002). Around 1 in 100 of the general population carry a pathogenic mutation but in both colorectal cancer and PanNETs, evidence suggests that two germline mutations are required for pathogenesis. Given that NETs have a very low mutational background, it is interesting that a significant proportion of patients harbour germline defects in DNA repair pathways.
Copy number alterations
Chromosomal alterations, implicated as potential drivers of PanNET development and progression, have been studied using various approaches such as fluorescence in situ hybridisation (FISH), comparative genome hybridisation (CGH) and genome-wide, high-density SNP arrays. Several themes have emerged from these studies, which are summarised below:
Several studies have confirmed the presence of multiple chromosomal alterations, generally occurring in broad genomic regions which can encompass entire chromosomes (Pea et al. 2018). Specific regions associated with recurrent gains and losses and LoH are summarised in Table 1.
Chromosomal aberrations identified in PanNETs.
| Broad genomic regions commonly associated with recurrent chromosomal losses and gains1-2,a | |||
|---|---|---|---|
| Frequency (%) | |||
| Losses | Putative tumour suppressor genes | NF-PanNETs | F-PanNETs |
| 1p | CDKN2C (1p32.3), RUNX3 (1p36.1), RIZ (1p36.21), TP73 (1p36.32) | 28 | 3–27 |
| 1q | RABGAP1L (1q25.1), IL24/MDA7 (1q32.1), PHLDA3 (1q32.1) | 24 | 10–20 |
| 3p | SFMBT1 (3p21.1), RASSF1 (3p21.31), β-catenin (3p22.1), hMLH1 (3p22.2), RARβ (3p24.2), VHL (3p25.3) | 27 | 20 |
| 6q | CCNC (6q16.2), AIM1 (6q21), PTPRK (6q22.33) | 38 | 3–70 |
| 10q | PTEN (10q23.31), MGMT (10q26.3) | 26 | 3–23 |
| 11p | WT1 (11p13) | 34 | 3–23 |
| 11q | MEN1 (11q13.1), ATM (11q22.3), TSG11 (11q23), SDHD (11q23.1) | 39 | 13–23 |
| Gains | Putative oncogenes | ||
| 7p | EGFR (7p11.2) | 28 | 6–37 |
| 7q | HGF (7q21.11), c-MET (7q31) | 36 | 12–47 |
| 9p | CDKN2A (9p21.3), RRAGA (9p22.1), OVC (9p24), JAK2 (9p24.1) | 19 | 6–29 |
| 9q | CDK9 (9q34.11), ABL1 (9q34.12), VAV2 (9q34.2), LMX1B (9q33.3), NOTCH1 (9q34.3) | 27 | 12–43 |
| 17q | ERRB2 (17q12) | 41 | 5–57 |
| 20q | AURKA (20q13.2) | 31 | 6–43 |
| Gene-specific changes due to CNVs or biallelic inactivation through LoH2-3,b | |||
| Chromosomal event | Gene | Frequency (%) | |
| CNV | LoH | ||
| Losses | MUTYH (1p34.1) | 47 | 5 |
| CHEK2 (22q12.1) | 49 | 2 | |
| BRCA2 (13q13.1) | 9 | 1 | |
| SETD2 (3p21.31) | 51 | 4 | |
| MLL3/KMT2C (7q36.1) | 10 | n/a | |
| MEN1 (11q13.1) | 33–70 | 40 | |
| DAXX (6p21.32) | 53 | 20 | |
| ATRX (Xq21.1) | 19 | 3 | |
| PTEN (10q23.31) | 40 | 7 | |
| DEPDC5 (22q12.2-3) | 49 | 2 | |
| TSC1 (9q34) | 17 | n/a | |
| TSC2 (16p13.3) | 43 | 2 | |
| PHLDA3 (1q32.1)c | n/a | 75 | |
| EYA1 (8q13.3) | n/a | n/a | |
| Gains | AKT2 (19q13.2) | 25 | n/a |
| ERBB2 (17q12) | 25 | n/a | |
| ULK1 (12q24.33) | 31 | n/a | |
| PSPN (19p13.3) | 21 | n/a | |
| Prevalent chromosomal aberrations associated with metastatic disease3-5,d | |||
| Frequency (%) | |||
| Losses | 2pq, 3pq, 6pq, 10p, 11pq, 16pq, 22pq | 25–75 | |
| Gains | 4pq, 5pq, 7pq, 9q, 12q, 14q, 17pq, 18q, 19pq, 20q | 25–71 | |
[1] Capurso et al. 2012; [2] Scarpa et al. 2017, [3] Pea et al. 2018, [4] Speel et al. 2001, [5] Zhao et al. 2001.
aFunctional PanNETs (F-PanNETs) in the study by Capurso et al. (2012) included gastrinomas, insulinomas and metastatic insulinomas. Metastatic insulinomas had the highest observed frequencies in the range indicated for each chromosomal event and were similar to those observed in NF-PanNETs. b91 G1/G2 NF-PanNETs included in the study by Scarpa et al. (2017). Genetic analyses in Pea et al. (2018) included 48 primary WD-PanNETs (<3 cm) of which 24 developed liver metastases. cPHLDA3, a novel TSG, is inactivated in up to 75% of PanNETs through LoH and DNAme in the second allele (Ohki et al. 2014). Its loss inhibits AKT-driven suppression of the mTOR pathway and has been implicated in disease progression. dHigher frequency of chromosomal gains in metastasising tumours, while losses were more prevalent in non-metastatic. Chromosomal gains frequently associated with recurrent CN-LOH events.
As previously described, MEN1 is inactivated early in the pathogenesis of PanNETs through somatic mutations in one allele and loss of the remaining WT allele. Therefore, detection of recurrent chromosome 11q losses in smaller tumours <2 cm is not unexpected (Speel et al. 2001). It has been suggested that both DAXX (6q21.2) and ATRX (Xq21.1) are inactivated through mutations in one allele and either loss or epigenetic silencing of the remaining allele in the case of DAXX (6q21.2) or X chromosome inactivation of the second ATRX allele (Xq21.1) indicating a tumour suppressor role for both genes. In fact, copy number changes involving the DAXX and ATRX locus have been found in 53% and 19% of cases, respectively (Scarpa et al. 2017). The notion regarding the involvement of an alternative mechanism, such as epigenetic silencing, was recently supplemented through data showing that DAXX and ATRX-mutant PanNETs are epigenetically more dysregulated compared to WT tumours (Pipinikas et al. 2015). In addition, more pronounced genome-wide epigenetic changes were present in DAXX mutants and in higher grade, G2 tumours. In the case of ATRX, 10/98 tumours analysed by Scarpa et al. (2017) were found to harbour inactivating point mutations. Biallelic ATRX inactivation by LoH was exclusively detected in three out of four female patients. Earlier reports have highlighted a higher frequency of Xq loss in PanNETs of females (45%) compared to males (5%) (Speel et al. 1999) and have provided evidence for the association between sex chromosome losses and characteristics of disease aggressiveness (higher grade, advanced stage, local invasion and metastasis and worse clinical outcome) (Missiaglia et al. 2002). In agreement with the confirmed role of ATRX in maintaining telomere integrity and genome stability, it has been previously suggested that genes on X chromosome may act as negative regulators of cell proliferation through induction of senescence (Missiaglia et al. 2002). Collectively, these findings highlight a potential pathogenic function of X chromosome losses in determining disease development and progression.
LoH at 17p13 (TP53 locus) is limited to metastatic NF-PanNETs lacking TP53 mutations (Beghelli et al. 1998), suggesting the presence of other TSGs in the region. The p53 pathway, regulated by both positive and negative regulators, has a vital role in maintaining genomic stability (Fig. 2A). Among the negative p53 regulators, aberrant activity of MDM2, MDM4 and WIP1 impairs p53 function, which is a significant step in tumourigenesis. MDM2 (an E3 ubiquitin ligase) binds to and degrades p53, while MDM4, required for MDM2 optimal activity, interacts and negatively regulates p53 function (Bond et al. 2005, Marine et al. 2007). WIP1 (a serine/threonine phosphatase) inactivates p53 either directly or indirectly through dephosphorylation of p53 upstream activators (ATM, Chk1, Chk2), while it is also known to induce p53 degradation through MDM2 stabilisation (Lu et al. 2007). Although <3% of PanNETs harbour TP53 mutations (Scarpa et al. 2017), TP53 is a haploinsufficient gene; the smallest deviation from its normal functions can have a huge impact on tumourigenesis (Malkin et al. 1990). MDM2 (22%), MDM4 (30%) and WIP1 (51%) amplification, correlating with mRNA and protein expression, occurs in PanNETs (Hu et al. 2010) in a mutually exclusive fashion to TP53 mutations (Marine et al. 2007). Furthermore, a high frequency functional MDM2 promoter SNP inducing its expression and resulting in p53 attenuation has also been identified in PanNETs (Hu et al. 2010). Collectively, these observations demonstrate a highly impaired p53 PanNET pathway and highlight the potential value of MDM inhibitors in reversing p53 inactivation and sensitising PanNETs to other therapies. Chromosomal alterations included as part of multi-layered data integration studies are described later in this review.
Novel findings
The most comprehensive genomic study of PanNETs to date identified five, whole-genome-derived mutational signatures from 98 G1/G2 PanNETs (Scarpa et al. 2017). While two of the signatures, ‘APOBEC ’ and ‘age ’ were significantly enriched in 4 and 5 cases respectively, 69 cases were found to have the unknown ‘signature 5 ’, which has been previously reported in tumours with FHIT gene deletions (Alexandrov et al. 2013, Volinia et al. 2017). LoH at 3p14.2, the FHIT locus, was previously documented in 40% of NF-PanNETs (Beghelli et al. 1998). A fourth signature, ‘BRCA ’ was enriched in 15 cases of which one found to carry a pathogenic BRCA2 germline mutation coupled with LoH in the remaining allele. However, the most striking finding was the identification of a previously unknown mutational signature (‘novel ’) in five sporadic tumours enriched for inactivating germline mutations in the MUTYH gene. Driven by the detection of germline mutations, Scarpa et al. (2017) screened the entire cohort for germline mutations in DNA damage repair genes (MUTYH, BRCA2, CHEK2) and genes linked to familial PanNETs (MEN1, VHL, CDKN1B). MUTYH germline mutations associated with the same ‘novel ’ signature were also identified through targeted sequencing in a validation cohort of 62 cases. In total, ~18% of tumours were found to harbour germline mutations at a higher-than-expected frequency: 8 MUTYH, 6 MEN1, 4 CHEK2 and 1 in each of BRCA2, VHL and CDKN1B. However, their penetrance among PanNET cases is currently unknown and warrants further investigation possibly within a clinical genetic referral context. This list has recently been expanded through the identification of novel non-synonymous germline mutations in PIF1 (n = 2), involved in transcriptional regulation and chromatin remodelling, MAPKBP1 (n = 1), a member of the MAPK/ERK signalling pathway (Vandamme et al. 2019) known to be co-activated with the PI3K/Akt/mTOR pathway in neuroendocrine tumours through PI3K (Carracedo et al. 2008), and identification of a pathogenic germline variant leading to somatic loss of ATM (Lawrence et al. 2018). Although no somatic alterations in MAPK/ERK have been previously reported in PanNETs, recent identification of novel, high-abundance MAP4K2 and MAPKBP1 mutations in ~16% of tumours potentially implicates the pathway in disease pathogenesis and highlights a rationale for a combined MAPK-mTOR inhibition (Vandamme et al. 2019). The use of MEK inhibitor PD0325901 was shown to enhance the antiproliferative effect of the PI3K-mTOR inhibitor BEZ235 in BON-derived mouse xenografts (Valentino et al. 2014). Similarly, the antitumoural effect of mTORC1 inhibition by rapamycin both in vitro and in a xenograft mouse model was enhanced through MAPK inhibition (Carracedo et al. 2008). Interestingly, Vandamme et al. (2019) identified mutational hotspots in DAXX and CYFIP2, a novel proapoptotic gene, and multiple low-abundance recurrent mutations associated with another hotspot in the PTCH2 gene, involved in Hedgehog signalling. Mutations in SMAD4, previously described in small intestinal NETs, were also detected in three tumours (Vandamme et al. 2019).
Another novel finding, previously observed in 2–3% of all cancers and in up to 25% of bone tumours (Stephens et al. 2011), is the first ever documentation of chromothripsis in PanNETs (Scarpa et al. 2017). Copy number changes and structural variations, compatible with chromothripsis, were identified in nine PanNETs. Four tumours were associated with complex structural rearrangements on 11q13 resulting in complete MEN1 inactivation in two of them (Scarpa et al. 2017). In addition, 17 tumours had less complex rearrangements resulting in inactivation of TSGs: ARID2 (5), MTAP (4), SMARCA4 (3), KMT2C/MLL3 (3), SETD2 (1) and CDKN2A (1). Of note, in three cases, rearrangements resulted in oncogenic gene fusions of the EWSR1 with BEND2 and FLI1. EWSR1 fusions have been proposed as being part of a novel mTOR activation mechanism. Both complex genomic rearrangements and EWSR1 gene fusions were associated with shorter telomeres supporting the idea that chromothripsis may arise as result of telomere exhaustion.
A key limitation in validating novel findings in PanNET research is the lack of reliable cell lines and in vivo models, with BON1 and QGP1 cell lines still in use despite increasing evidence that they are not representative of the disease phenotype (Boora et al. 2015, Vandamme et al. 2015). Rip-Tag mouse models for MEN1 and insulinoma have been successfully produced (Bertolino et al. 2003, Crabtree et al. 2003, Olson et al. 2011); however, creating DAXX/ATRX-knockout mice has proved more challenging. There are currently ongoing efforts to produce neuroendocrine tumour organoids, which, when validated would provide a much awaited resource.
Integrated multi-omic data approaches
Recently, extensive chromosomal alterations involving whole chromosome arm losses or gains were used to classify tumours into four clusters with distinct mutational profiles and ALT status (Scarpa et al. 2017). Cluster 1 characterised by a recurrent pattern of whole chromosomal losses (RPCL) in regions previously reported to be affected by loss (1, 3, 6, 8, 10, 11) as well as losses in additional chromosomal regions (2, 15, 16 and 22). This cluster included mainly G2 tumours and was enriched for the ALT phenotype with 56% of the ALT+ cases harbouring DAXX or ATRX mutations. Additionally, this cluster had the highest frequency of mutations in the mTOR pathway genes and included tumours with extra-pancreatic spread. Cluster 2 , including G1 and G2 tumours, was characterised by limited copy number alterations confined mainly to chromosome 11 and mutations limited to MEN1. Approximately 83% of the cases were ALT negative with 4/6 ALT+ cases carrying a DAXX mutation. Cluster 3 included polyploid tumours with gains in all chromosomes and the highest somatic mutational rates. Tumours in this cluster were mainly ALT negatives and generally mutationally silent for MEN1, DAXX/ATRX and mTOR genes. Cluster 4 included aneuploid tumours with predominantly whole chromosome gains affecting several chromosomes. Interestingly, a 100% concordance between ALT+ and DAXX/ATRX mutations was observed. Despite the novelty of the described findings, deducting useful information regarding their potential prognostic value is not straightforward. However, in this study series, G2 tumours (coinciding with the RPCL cluster) harbouring mutations in DAXX/ATRX and correlating with mutations in mTOR genes were associated with poor prognosis.
Another recent targeted/whole-genome sequencing study reported three CN-based clusters (Lawrence et al. 2018) similar to those identified in the Scarpa study. Correlation with clinical data demonstrated that tumours in cluster 1 had the poorest prognosis, cluster 2 tumours had more favourable clinical/pathological outcomes, while those in cluster 3 had variable clinical outcomes. Similar clusters have also been reported by Pea et al. (2018). Supplementary Figure 1A, B and C (see section on supplementary data given at the end of this article) summarises the findings from these three studies. Recently, integration of DNA methylation and gene expression profiles demonstrated the presence of two separate clusters of WD-PanNETs, with one consisting exclusively of MEN1/DAXX/ATRX-mutant cases (Chan et al. 2018) enriched for a gene expression signature characteristic of the α-cells of pancreatic islets. In contrast, an earlier study identified mature islet β-cells as the possible cell of origin of MEN1/DAXX/ATRX WT PanNETs (Sadanandam et al. 2015; Supplementary Fig. 1D).
Epigenetic landscape of PanNETs
PanNETs are generally mutationally silent compared to their exocrine counterpart, pancreatic ductal adenocarcinoma (PDAC) (Jiao et al. 2011) and other tumour types (Lawrence et al. 2013) (Fig. 1B). The frequency of MEN1/DAXX/ATRX mutations in PanNETs is between 56.1 and 63.2% (Jiao et al. 2011, Scarpa et al. 2017) and is similar to the frequency of mutations in MEN1 and other chromatin-remodelling genes (ARID1A and PSIP1) reported for lung carcinoids (Fernandez-Cuesta et al. 2014). Since these mutations affect the epigenetic machinery, the epigenome is likely to be a causative tumorigenic mechanism, acting synergistically with the genome to provide somatic cells a survival and growth advantage. The epigenetic machinery (from the Greek prefix ‘επι’ meaning ‘superimposed upon’ genetics) represents an additional layer of stable information inherited through mitotic division that affects gene function without changing the DNA sequence and has a fundamental role in cancer development (Jones & Baylin 2002).
DNA methylation
The most studied and best understood epigenetic mark is DNA methylation (DNAme), the covalent addition of a methyl group to cytosines preceding guanines in a CpG dinucleotide. CpG islands (CGIs) are CpG-rich regions present at high frequency in gene promoter-regulatory regions. Although CGIs have low methylation levels in normal cells, cancers are characterised by aberrant CGI-promoter hypermethylation resulting in gene silencing. On the other hand, extensive methylation within the gene body has an opposite gene expression effect (Yang et al. 2014). Overall, cancers are globally hypomethylated, a phenomenon that has been linked to chromosomal instability, another hallmark of cancer (Eden et al. 2003, Fraga et al. 2004). DNA hypermethylation is an early event in cancer initiation and dictates the rate of disease development and progression.
The first studies of DNAme in NETs were candidate gene driven. However, recent development of genome-wide DNAme approaches has provided a more detailed insight into the importance of the mechanism in disease pathogenesis (Pipinikas et al. 2015, Karpathakis et al. 2016, 2017).
Hypermethylation of the RASSF1A gene promoter, which is regulated by p53 and DAXX (Zhang et al. 2013), is the most commonly epigenetically dysregulated region observed in 75% of NF-PanNETs (House et al. 2003) and is most frequent in metastatic compared to non-metastatic tumours (Malpeli et al. 2011) and less frequent in PDACs (Dammann et al. 2003). Functional RASSF1A is known to suppress miR-21 (discussed in the next section), which is associated with liver metastases (Roldo et al. 2006, Ram et al. 2014). It also leads to p53 stabilisation and cell cycle arrest upon DNA damage through disruption of the DAXX–MDM2 axis (Song et al. 2008). CDKN2A, TIMP3 and MGMT promoter hypermethylation has been reported in 40–44% of PanNETs and is associated with more aggressive tumours and worse prognosis (House et al. 2003, Wild et al. 2003). MGMT promoter hypermethylation leading to loss of protein expression has been suggested as a predictive biomarker of temozolomide response (Kulke et al. 2009). Increased DNAme across multiple CpG loci, known as CIMP (CpG island methylator phenotype) and LINE1 and Alu hypomethylation, correlating with early recurrence and disease aggressiveness, have also been identified in PanNETs (Arnold et al. 2007, Choi et al. 2007, Stricker et al. 2012). Finally, PanNET subgroups with distinct global hypomethylation and gene-specific hypermethylation profiles associated with different clinical outcomes have been described (Stefanoli et al. 2014). Genes aberrantly methylated in PanNETs are summarised in Table 2.
Genes with aberrant DNA methylation in sporadic, non-functional pancreatic neuroendocrine tumours (NF-PanNETs).
| DNAme pattern | Function/pathway | Gene | Chromosome | DNAme frequency | Disease association | References |
|---|---|---|---|---|---|---|
| HyperM | Cell cycle regulation/apoptosis | RASSF1A | 3p21.31 | 75–80% | Larger tumours and presence of metastatic disease | [1], [2], [3] |
| Cell cycle regulation/apoptosis | CDKN2A (p16INKA4a) | 9p21.2 | 0%, 17%, 40%, 52% | Early tumour recurrence, metastasis and reduced OS | [1], [2], [4], [5], [6], | |
| Cell cycle regulation/apoptosis | P73 | 1p36.32 | 17% | Higher DNAme frequency in larger tumours associated with metastasis | [2] | |
| Cell cycle regulation/apoptosis | APC | 5q22.2 | 21–48% | Increased proliferation and CIN | [2], [4] | |
| Cell cycle regulation/apoptosis | HIC-1 | 17p13.3 | 93% | Worse prognosis (as part of CIMP) | [4] | |
| Inhibition of tumour invasion | E-cadherin | 16q22.1 | 23% | Higher DNAme frequency in larger tumours associated with metastasis | [2] | |
| Inhibition of tumour invasion | TIMP3 | 22q12.3 | 0–57% | Increased risk of metastatic disease (LNs and liver) | [2], [4], [7], [8], [9] | |
| Gene expression regulation | RAR-β | 3p24.2 | 25% | Higher DNAme frequency in larger tumours associated with metastasis | [2] | |
| Genome integrity/DNA repair | hMLH1 | 3p22.2 | 0–23% | Higher DNAme frequency in larger tumours associated with metastasis | [2], [4] | |
| Genome integrity/DNA repair | O-6-MGMT | 10q26.3 | 17%, 40% | Hypermethylation possibly predicts response to temozolamide | [2], [4] | |
| Angiogenesis - Hypoxia pathway (HIF) activation | VHL | 3p25.3 | 6% | Reduced DFS | [10] | |
| Multiple functions and pathway interactions | MEN1 | 11q13 | 19% | Worse prognosis (as part of CIMP) | [4] | |
| Regulation of apoptosis, angiogenesis, EMT, cell migration/invasion | RUNX3 | 1p36.11 | 7% | Worse prognosis (as part of CIMP) | [4] | |
| CpG Island Methylator Phenotype | CIMP | – | 29–83% | Correlation with higher Ki67 and poor prognosis | [4] | |
| Regulation of AKT activity, cell proliferation and apoptosis | PHLDA3 | 1q32.1 | 100% | Disease progression and poor prognosis | [11] | |
| Cellular differentiation | HOPX | 4q12 | G1: 7%, G2: 12%, G3: 67% | Higher grade and worse prognosis | [12] | |
| Cytoskeletal protein with potential role in tumour initiation-progression | α-internexin | 10q24.33 | N/A | Advanced stage, metastasis, recurrence and reduced OS | [13] | |
| HypoM | Transposable element | LINE1 | – | Variable methylation levelsa | Higher tumour grade, LN metastasis and worse prognosis | [9], [14], [15] |
| Transposable element | Alu | – | Variable methylation levels | Higher tumour stage and poor prognosis | [14] |
LINE1 methylation levels <58% observed in 21% of tumours were associated with worse prognosis (Stefanoli et al. 2014). In Arnold et al. (2007), overall DNAme frequencies were given across different PanNET subtypes (benign and malignant functional and non-functional PanNETs). [1] Dammann et al. 2003; [2] House et al. 2003; [3] Malpeli et al. 2011; [4] Arnold et al. 2007; [5] Muscarella et al. 1998; [6] Serrano et al. 2000; [7] Wild et al. 2003; [8] Arnold et al. 2008; [9] Stefanoli et al. 2014; [10] Schmitt et al. 2009; [11] Ohki et al. 2014; [12] Ushiku et al. 2016; [13] Liu et al. 2014; [14] Choi et al. 2007; [15] Stricker et al. 2012.
aLINE1 median methylation levels were 66% in G1 (range 61–72%) and 63% in G2 (range 56–64%) PanNETs compared to a median methylation level of 73% in normal pancreas (Stricker et al. 2012).
Alu, short interspersed element (Arthrobacter LUteus Homologues); CIN, chromosomal instability; DFS, disease-free survival; EMT, epithelial-to-mesenchymal transition; LINE1, long interspersed element; LN, lymph node; NF-PanNETs, non-functional PanNETs; OS, overall survival.
Histone modification (acetylation and methylation)
Several histone modifications, including methylation, acetylation and demethylation, have been described. Histone methylation of genes encoding other CDKis, such as p16 Ink4a (CDKN2A) and p14 ARF , has been identified as an early driver in NET pathogenesis. Enhanced H3K9me3 activity through menin–Daxx interactions leads to epigenetic inhibition of a pro-proliferative gene, Mme, and results in NET suppression in mice (Patel et al. 2018), while Men1−/− mouse embryonic fibroblasts have reduced global methylation of histone H3K9 and H3K4. In addition, menin interacts with the SUV39H1 (suppressor of variegation 3-9 homolog protein family) and mediates H3K9 methylation. This menin-dependent recruitment of SUV39H1 is essential for chromatin remodelling and is reduced through introduction of patient-derived MEN1 mutations into the SUV39H1 interaction domain (Yang et al. 2013). Furthermore, downregulation of MEG3 and HOX genes has been associated with development of non-functional pituitary adenomas and parathyroid tumours, respectively (Shen et al. 2008, Cheunsuchon et al. 2011). A study in mouse embryonic stem cells demonstrated that both the Meg3 and Hox loci are regulated by menin in a H3K4me3-dependent manner. Loss of menin results in H3K4me3 loss and gene downregulation and gives rise to MEN1-like sporadic pancreatic tumours (Agarwal & Jothi 2012).
Post-transcriptional control (microRNAs and lncRNAs)
Non-coding RNAs (ncRNAs) are post-transcriptional gene regulators implicated in cancer development and progression (Gibb et al. 2011). They are classified into small/short and long non-coding RNAs (sncRNAs and lncRNAs, respectively). miRNAs are sncRNAs (~22 nucleotides long) with a post-transcriptional RNA interference role leading to gene silencing through their imperfect 5’-base pairing with target mRNA sequences (Ambros 2004). Under specific conditions, however, they also induce gene expression (Vasudevan et al. 2007, Ørom et al. 2008) indicating a dual role in switching on/off gene expression and have, therefore, both oncogenic and tumour suppressing properties. Up to 1000 miRNAs present in the human genome regulate ~30% of the transcriptome with a single miRNA targeting or being the target of hundreds of mRNAs (Berezikov et al. 2005, Holley & Topkara 2011). Their differential expression not only between different cancer types but also between subtypes of the same tumour and correlation with various clinical characteristics coupled with their stability in circulation, makes them promising candidates for the development of non-invasive diagnostic/prognostic tools. Long non-coding RNAs (>200 nucleotides long) regulate gene expression, both transcriptionally and post-transcriptionally (splicing and translation) (Hauptman & Glavač 2013), while the involvement of individual lncRNAs in chromatin remodelling suggests an active role in moulding the epigenome of cancer cells (Gupta et al. 2010).
MicroRNAs
One of the earliest studies of miRNAs in PanNET development was a study comparing their regulation in normal pancreas, F-/NF-PanNETs and pancreatic acinar cell carcinomas (PACCs) (Roldo et al. 2006). Identified miRNAs were able to discriminate tumours from normal pancreas, with those shared in common between PanNETs and PACCs having a possible involvement in pancreatic tumourigenesis (Supplementary Table 1). Similarly, miRNAs upregulated in PanNETs in comparison to PACCs were suggested as having a possible role in either normal neuroendocrine differentiation or NET pathogenesis. Compared to normal pancreas, miR-103 and miR-107 were upregulated and miR-155 downregulated in PanNETs, while miR-204, miR-211 and miR-203 were upregulated in insulinomas compared to NF-PanNETs. With regards to clinical presentation, miR-21 (targets and down-regulates PTEN) was strongly associated with both higher Ki67 and metastatic disease. PDCD4 mRNA expression, a tumour suppressor gene and putative target of miR-21, was downregulated in metastatic tumours. Grolmusz et al. (2018) subsequently used this dataset to identify prognostic miRNAs correlating with altered proliferation, tumour stage and survival rates. Hierarchical miRNA clustering identified two clusters: C1 including 82% of G1 tumours and C2 including mainly G2 (11/18) and further confirmed miR-21 upregulation in metastatic disease. In a follow-up study, positive correlation with Ki-67 and metastatic disease was demonstrated for miR-642 and miR-210, respectively, but not miR-21 (Thorns et al. 2014). As miR-210 is induced by HIF-1a, it has been suggested to contribute to tumour adaptation to hypoxia-related stress and to determine disease initiation and metastatic potential of PanNETs (Huang et al. 2009). Overexpression of miR-196 (regulates AKT signalling and HOX gene expression) and miR-27b has also been linked to more aggressive tumours, disease recurrence and decreased OS. miR-196 was also associated with shorter DFS suggestive of a potential value in patient stratification (Lee et al. 2015). Recently, a 30-miRNA signature demonstrated the presence of three clusters matching with mRNA-based disease subtypes with distinct clinical characteristics (Sadanandam et al. 2015). miR-Cluster-1 tumours were enriched in the MEN1-like/intermediate mRNA subtype, composed of NF-PanNETs associated with a moderate metastatic potential. miR-Cluster-2 was enriched in the metastasis-like primaries (MPL) mRNA subtype and included tumours with high metastatic potential. miR-Cluster-3 included mainly insulinomas, of which none was associated with metastatic disease and was enriched in the ‘islet/insulinoma’ (IT) mRNA cluster. Distinct gene mutational rates in mTOR, MEN1/DAXX/ATRX and DNA damage pathways were associated with each cluster (Supplementary Fig. 1D). Additional studies have investigated miRNA regulation in PanNETs for a variety of purposes (Li et al. 2013, Zhou et al. 2016, Zimmermann et al. 2018, Panarelli et al. 2019). Supplementary Table 1 provides a summary of miRNAs implicated in PanNETs.
Long non-coding RNAs
Studies of lncRNAs in PanNETs are scarce with most having evaluated MEG3, which is activated by menin in an H3K4me3-dependent manner and CpG hypomethylation at the promoter region (Agarwal & Jothi 2012). In BON1 cells, MEG3 upregulation demonstrated an anti-tumour effect that was eliminated in the presence of miR-183 overexpression targeting BRI3 to promote tumourigenesis through regulation of the p38/ERK/Akt and Wnt/β-catenin pathways. The MEG3/miR183/BRI3 axis may, therefore, play a significant role in PanNET development (Zhang & Feng 2017). Furthermore, MEG3 directly targets the c-MET proto-oncogene with human MEN1-associated PanNETs demonstrating MEG3 downregulation and c-MET upregulation. In human sporadic insulinomas, this discordant MEG3–c-MET expression is associated with MEG3 promoter hypermethylation, while demethylating agents reactivate MEG3 and suppress cell proliferation in mouse insulinoma cells. Therefore, epigenetic MEG3 reactivation and c-MET suppression may induce a MEN1-dependent anti-tumour response and represents an attractive therapeutic approach worth evaluating (Modali et al. 2015).
Two lncRNAs, the HOX antisense intergenic RNA (HOTAIR; 12q12.13) and the metastasis-associated lung adenocarcinoma transcript 1 (MALAT1; 11q13), were recently implicated in PanNETs, with both tumour-promoting (Hajjari & Salavaty 2015, Zhao et al. 2018) and tumour-suppressing (Cao et al. 2016, Kwok et al. 2018) overexpression observed in various human cancers. Their tumour-promoting overexpression is due to activation of downstream pathways including Wnt/β-catenin (Hirata et al. 2015) and ERK/MAPK (Wu et al. 2014) and induction of epithelial-mesenchymal transition (EMT) through epigenetic regulation of genes such as E-cadherin downregulation and induction of gene members of the MMP family (Hajjari & Salavaty 2015, Hirata et al. 2015). In PDACs, overexpression of both lncRNAs is associated with more aggressive disease and poorer disease-free survival (Kim et al. 2013, Pang et al. 2015). HOTAIR is a chromatin-modifying lncRNA involved in re-programming the global chromatin state of cancer cells (Valastyan et al. 2012). It has been previously implicated in prostate cancer neuroendocrine differentiation (Chang et al. 2018), while its overexpression results in alterations in H3K27me and increases the metastatic potential of breast cancer cells (Gupta et al. 2010). Targeting HOTAIR has been shown to reduce cell invasiveness through induction of apoptosis and cell cycle arrest (Kim et al. 2013). However, a recent study on 83 primary GEP-NENs (72% G1, 20% G2, 8% G3) of which 52% developing in the pancreas, demonstrated that upregulation of both lncRNAs is associated with less aggressive disease and better prognosis (Chu et al. 2019). Another lncRNA, telomeric repeat-containing RNA (TERRA), which regulates telomerase activity and heterochromatin formation, is suppressed in the presence of active ATRX (Cusanelli & Chartrand 2015).
Clinical implications
Molecular targeted therapies
The most successful targeted therapies in PanNETs are those against the somatostatin receptor (SSR), known to be overexpressed in approximately 70% of cases (Grozinsky-Glasberg et al. 2008). The CLARINET study demonstrated a significantly improved PFS with the somatostatin receptor analogue (SSA) lanreotide over placebo (Caplin et al. 2014) in PanNETs, while efficacy of octreotide was extrapolated from the PROMID trial in midgut NETs (Rinke et al. 2009). SSAs result mostly in stable disease and have become standard first-line treatment for slow growing, G1 or G2 PanNETs and those with a functional syndrome (Pavel et al. 2016). Peptide receptor radiotherapy (PRRT) targets the SSR using an SSA attached to a radioactive isotope such as Lu 177 or Y 90 and is effective in those tumours overexpressing the SSR on radionuclide imaging. The NETTER-1 study (Strosberg et al. 2017) demonstrated the efficacy of Lu-177-dotatate in midgut NETs and results of this and other retrospective studies in PanNETs (Ramage et al. 2018) have led to its FDA approval in advanced GEP-NETs. However, to date, it has not been possible to drug all of the pathways identified as PanNET drivers. The two drugs that have been the most successful for therapy target the mTOR and VEGF pathways. Everolimus, an oral mTOR serine-threonine kinase inhibitor, is now used routinely to treat pancreatic and intestinal NETs along with typical/atypical bronchial NETs (Yao et al. 2011). Sunitinib, an oral tyrosine kinase and potent inhibitor of VEGF and platelet derived growth factor (PDGF) receptors (Faivre et al. 2006), has proved effective in targeting activated angiogenic PanNET pathways (Raymond et al. 2011, Faivre et al. 2017). Results are awaited from an ongoing study, which aims to select between everolimus and sunitinib depending on whether tumours harbour predominantly mutations in the mTOR or angiogenesis pathways (Neychev et al. 2015). However, preliminary results suggest that only 7.7% of tumours had actionable mutations in either pathway. Combination therapies targeting both mTOR and VEGF pathways or using targeted agents in combination with somatostatin analogues (SSAs) have not shown clinically meaningful benefits and have been associated with several adverse events (Kulke et al. 2015, 2017).
Inhibition of mTOR results in increased PI3K-Akt signalling and, therefore, dual targeting of these pathways has been explored, with some success, in cell lines and pre-clinical models (Antonuzzo et al. 2017). However, the phase II trial of the PI3K mTOR inhibitor BEZ235 combined with downstream mTOR inhibition resulted in non-significant improvement in survival and high adverse side effects rates (Salazar et al. 2018), while the pan-Akt inhibitor MK-2206 (phase II trial) also showed modest benefits (Reidy-Lagunes et al. 2012). Resistance can also develop following VEGF inhibition through HIF-α and c-MET activation (Sennino et al. 2012). Cabozantanib, a multi-tyrosine kinase inhibitor targeting VEGF and c-MET, has shown responses in early-phase trials, though final results are currently awaited (Chan et al. 2017). The CDK4/6 inhibitors, palbociclib and ribociclib, commonly used in breast cancer, are currently being trialled in PanNETs due to the observed overexpression of CDKs (Tang et al. 2012, Pavel & Sers 2016).
Epigenome-based targeted therapies
Given the high CIMP prevalence in PanNETs (Stefanoli et al. 2014) and DNMT activation following MEN1 loss (Yuan et al. 2016), studies have examined the efficacy of DNA methyltransferase inhibitors (DNMTis). Despite success in cell lines, these drugs have not yet been trialled in humans (Alexander et al. 2010). Interest has also been shown in histone deacetylase inhibitors (HDACis) due to the involvement of MEN1 in chromatin remodelling. Although response rates for the HDACi panobinostat were low, stable disease was demonstrated in a proportion of patients (Jin et al. 2016). Inhibitors of the bromo and extraterminal domain (BET) protein family, which interact with acetylated histone residues, have also shown success in pre-clinical studies. One compound, JQ1, reduced growth and induced apoptosis in both cell lines and mouse PanNET models (Lines et al. 2017). RRx-001, inhibiting both DNAme and HDACs, is being trialled to induce global epigenetic changes and resensitise patients to chemotherapy (Carter et al. 2015, Padda et al. 2018).
Immunotherapy
A deeper understanding of the genomic PanNET landscape will greatly improve patient stratification for immunotherapies. Phase Ib (KEYNOTE-028) and phase II (KEYNOTE-158) trials of PD1 inhibitor pembrolizumab in PanNET patients have shown modest activity limited mainly to patients with PD-L1 >1%, with the best response being stable disease (Mehnert et al. 2017, Strosberg et al. 2019). Disease control in WD-PanNETs was shown in an ongoing trial for the PD1 inhibitor spartalizumab (PDR001) (Yao et al. 2018), while PD-L1 positivity and tumour mutational burden (TMB) are currently evaluated as biomarkers of efficacy as part of a trial for the PD1 inhibitor sintilimab (IBI308) (Jia et al. 2018). TMB is a recognised biomarker for immunotherapy because of its relationship to neoantigen burden, which in turn stimulates T-cell proliferation and increases immune reactivity (McGranahan et al. 2016). Although PanNETs generally have a low TMB (Yao et al. 2019), it is possible that immunotherapy may still prove effective for rare cases with advanced disease and high TMB or other biomarker, when conventional therapies have failed.
Due to modest responses to single-agent checkpoint inhibition, dual blockade is being considered. The DUNE phase II trial recruits a range of NET patients including G1-G3 PanNETs for treatment with anti-CTLA4 and anti-PD-L1 combination therapy (Hernando-Cubero et al. 2018). Another option to improved checkpoint inhibition response may be addition of metronomic chemotherapy (Chen et al. 2017) to promote inflammation and induce mutations to increase neoantigen presentation. There have also been promising results from early studies in oncolytic virus therapy and CAR-T cell therapy (Vedvyas et al. 2016, Yamamoto et al. 2017).
Conclusions
Over 100 years have passed since Siegfried Oberndorfer first used the term ‘Karzinoid’ to describe this rare, unusual tumour entity. Figure 3 illustrates some of the most important findings in PanNET research. Although our understanding surrounding NET biology is far from being complete, recent PanNET studies have benefited from sequencing advances and have started to provide a deeper understanding of their complex molecular landscape. However, tumour heterogeneity remains poorly defined and has delayed the development of better diagnostic/prognostic tools and novel, more efficacious molecular-based therapies. Recent higher detection frequencies have reignited the interest in NET research leading to concentrated efforts to decipher their molecular landscape with recent findings holding the promise of shaping the future of NET research. Recurrent schemes have emerged from various studies all pointing towards key driving roles of the MEN1/DAXX/ATRX/ALT axis and genes in the mTOR and DNA damage pathways in disease pathogenesis alongside a plethora of chromosomal and telomere alterations and a highly dysregulated epigenetic machinery. Several novel findings have also highlighted a higher-than-expected frequency of germline mutations in sporadic PanNETs and presence of complex chromosomal rearrangements.
Timeline of key molecular, therapeutic and disease model developments in neuroendocrine tumour research with a special focus on PanNETs, and future perspectives. [1] Oberndorfer (1907)*, [2] Oberndorfer (1929)*, [3] Maher et al. (2011), Davison et al. (1936), [4] Wermer (1954), [5] Hanahan (1985), [6] Larsson et al. (1988), [7] Pless (2005), [8] Latif et al. (1993), [9] Chadrasekharappa et al. (1997), [10] Crabtree et al. (2003), [11] Pellegata et al. (2006), [12] Missiaglia et al. (2010), [13] Jiao et al. (2011), [14] Heaphy et al. (2011a), [15] Blumenthal et al. (2012), [16] Stefanoli et al. (2014), [17] Mohamed et al. (2014), [18] Sadanandam et al. (2015), [19] Pipinikas et al. (2015), [20] Scarpa et al. (2017), [21] Lloyd et al. (2017), [22] Guo et al. (2017), [23] https://www.fda.gov/drugs/resources-information-approved-drugs/fda-approves-lutetium-lu-177-dotatate-treatment-gep-nets; accessed 05/05/2019, [24] Kunz et al. (2018), [25] Chamberlain et al. (2018). (a) Original references (1) and (2) from Modlin et al. (2004); (b) 2018 – E2211 trial shows temozolomide and capecitabine achieved the longest PFS in advanced low or intermediate-grade PanNETs. ALT, alternative lengthening of telomeres; CIMP, CpG island methylator phenotype; cfDNA, cell-free DNA; CTCs, circulating tumour cells; FDA, Food and Drug Administration; iPSC, induced pluripotent stem cells; PDX, patient-derived xenografts.
Citation: Endocrine-Related Cancer 26, 9; 10.1530/ERC-19-0175
Further studies integrating various levels of ‘omic’ data are required to fully address the multi-layered disease heterogeneity and, by doing so, to characterise in depth the molecular features of the disease, predict clinical behaviours, improve disease management and optimise treatment outcomes. Such integrated approaches are expected to have a direct clinical impact leading eventually this challenging, yet, intriguing tumour entity into the era of true precision medicine. It is strongly believed that future classification system updates would implement molecular profiles for better prognostic stratification and choice of personalised treatments.
Supplementary data
This is linked to the online version of the paper at https://doi.org/10.1530/ERC-19-0175.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this review.
Funding
This work did not receive any specific grant from any funding agency in the public, commercial of not-for-profit sector.
References
Agarwal SK & Jothi R 2012 Genome-wide characterization of menin-dependent H3K4me3 reveals a specific role for menin in the regulation of genes implicated in MEN1-like tumors. PLoS ONE 7 e37952. (https://doi.org/10.1371/journal.pone.0037952)
Agarwal SK, Kennedy PA, Scacheri PC, Novotny EA, Hickman AB, Cerrato A, Rice TS, Moore JB, Rao S, Ji Y, et al. 2005 Menin molecular interactions: insights into normal functions and tumorigenesis. Hormone and Metabolic Research 37 . (https://doi.org/10.1055/s-2005-870139)
Alexander VM, Roy M, Steffens KA, Kunnimalaiyaan M & Chen H 2010 Azacytidine induces cell cycle arrest and suppression of neuroendocrine markers in carcinoids. International Journal of Clinical and Experimental Medicine 3 .
Alexandrov LB, Nik-Zainal S, Wedge DC, Aparicio SA, Behjati S, Biankin AV, Bignell GR, Bolli N, Borg A, Børresen-Dale AL, et al. 2013 Signatures of mutational processes in human cancer. Nature 500 . (https://doi.org/10.1038/nature12477)
Al-Tassan N, Chmiel NH, Maynard J, Fleming N, Livingston AL, Williams GT, Hodges AK, Davies DR, David SS, Sampson JR, et al. 2002 Inherited variants of MYH associated with somatic G:C-->T:A mutations in colorectal tumors. Nature Genetics 30 . (https://doi.org/10.1038/ng828)
Ambros V 2004 The functions of animal microRNAs. Nature 431 . (https://doi.org/10.1038/nature02871)
Anlauf M, Garbrecht N, Bauersfeld J, Schmitt A, Henopp T, Komminoth P, Heitz PU, Perren A & Klöppel G 2007 Hereditary neuroendocrine tumors of the gastroenteropancreatic system. Virchows Archiv 451 S29–S38. (https://doi.org/10.1007/s00428-007-0450-3)
Antonuzzo L, Del Re M, Barucca V, Spada F, Meoni G, Restante G, Danesi R, Di Costanzo F & Fazio N 2017 Critical focus on mechanisms of resistance and toxicity of m-TOR inhibitors in pancreatic neuroendocrine tumours. Cancer Treatment Reviews 57 . (https://doi.org/10.1016/j.ctrv.2017.05.001)
Arnold CN, Nagasaka T, Goel A, Schraf I, Grabowski P, Sosnowski A, Schmitt-Gräff A, Boland CR, Arnold R & Blum HE 2008 Molecular characteristics and predictors of survival in patients with malignant neuroendocrine tumors. International Journal of Cancer 123 . (https://doi.org/10.1002/ijc.23690)
Arnold CN, Sosnowski A, Schmitt-Gräff A, Arnold R & Blum HE 2007 Analysis of molecular pathways in sporadic neuroendocrine tumors of the gastro-entero-pancreatic system. International Journal of Cancer 120 . (https://doi.org/10.1002/ijc.22569)
Balogh K, Rácz K, Patócs A & Hunyady L 2006 Menin and its interacting proteins: elucidation of menin function. Trends in Endocrinology and Metabolism 17 . (https://doi.org/10.1016/j.tem.2006.09.004)
Basturk O, Yang Z, Tang LH, Hruban RH, Adsay V, McCall CM, Krasinskas AM, Jang KT, Frankel WL, Balci S, et al. 2015 The high-grade (WHO G3) pancreatic neuroendocrine tumor category is morphologically and biologically heterogenous and includes both well differentiated and poorly differentiated neoplasms. American Journal of Surgical Pathology 39 . (https://doi.org/10.1097/PAS.0000000000000408)
Batukbhai BDO & De Jesus-Acosta A 2019 The molecular and clinical landscape of pancreatic neuroendocrine tumors. Pancreas 48 (https://doi.org/10.1097/MPA.0000000000001189)
Baur AD, Pavel M, Prasad V & Denecke T 2016 Diagnostic imaging of pancreatic neuroendocrine neoplasms (pNEN): tumor detection, staging, prognosis, and response to treatment. Acta Radiologica 57 . (https://doi.org/10.1177/0284185115579932)
Beghelli S, Pelosi G, Zamboni G, Falconi M, Iacono C, Bordi C & Scarpa A 1998 Pancreatic endocrine tumours: evidence for a tumour suppressor pathogenesis and for a tumour suppressor gene on chromosome 17p. Journal of Pathology 186 . (https://doi.org/10.1002/(SICI)1096-9896(199809)186:1<41::AID-PATH172>3.0.CO;2-L)
Berezikov E, Guryev V, van de Belt J, Wienholds E, Plasterk RH & Cuppen E 2005 Phylogenetic shadowing and computational identification of human microRNA genes. Cell 120 . (https://doi.org/10.1016/j.cell.2004.12.031)
Bertolino P, Tong WM, Galendo D, Wang ZQ & Zhang CX 2003 Heterozygous Men1 mutant mice develop a range of endocrine tumors mimicking multiple endocrine neoplasia type 1. Molecular Endocrinology 17 . (https://doi.org/10.1210/me.2003-0154)
Binkovitz LA, Johnson CD & Stephens DH 1990 Islet cell tumors in von Hippel-Lindau disease: increased prevalence and relationship to the multiple endocrine neoplasias. American Journal of Roentgenology 155 . (https://doi.org/10.2214/ajr.155.3.1974734)
Blumenthal GM, Cortazar P, Zhang JJ, Tang S, Sridhara R, Murgo A, Justice R & Pazdur R 2012 FDA approval summary: sunitinib for the treatment of progressive well-differentiated locally advanced or metastatic pancreatic neuroendocrine tumors. Oncologist 17 . (https://doi.org/10.1634/theoncologist.2012-0044)
Bond GL, Hu W & Levine AJ 2005 MDM2 is a central node in the p53 pathway: 12 years and counting. Current Cancer Drug Targets 5 . (https://doi.org/10.2174/1568009053332627)
Boora GK, Kanwar R, Kulkarni AA, Pleticha J, Ames M, Schroth G, Beutler AS & Banck MS 2015 Exome-level comparison of primary well-differentiated neuroendocrine tumors and their cell lines. Cancer Genetics 208 . (https://doi.org/10.1016/j.cancergen.2015.04.002)
Bosman FT, Carneiro F, Hruban RH & Theise ND 2010 WHO Classification of Tumours of the Digestive System. Lyon, France: IARC Press.
Brems H, Beert E, de Ravel T & Legius E 2009 Mechanisms in the pathogenesis of malignant tumours in neurofibromatosis type 1. Lancet: Oncology 10 . (https://doi.org/10.1016/S1470-2045(09)70033-6)
Cai Z, Chehab NH & Pavletich NP 2009 Structure and activation mechanism of the CHK2 DNA damage checkpoint kinase. Molecular Cell 35 . (https://doi.org/10.1016/j.molcel.2009.09.007)
Cao S, Wang Y, Li J, Lv M, Niu H & Tian Y 2016 Tumor-suppressive function of long noncoding RNA MALAT1 in glioma cells by suppressing miR-155 expression and activating FBXW7 function. American Journal of Cancer Research 6 .
Caplin ME, Pavel M, Ćwikła JB, Phan AT, Raderer M, Sedláčková E, Cadiot G, Wolin EM, Capdevila J, Wall L, et al. 2014 Lanreotide in metastatic enteropancreatic neuroendocrine tumors. New England Journal of Medicine 371 . (https://doi.org/10.1056/NEJMoa1316158)
Capurso G, Festa S, Valente R, Piciucchi M, Panzuto F, Jensen RT & Delle Fave G 2012 Molecular pathology and genetics of pancreatic endocrine tumours. Journal of Molecular Endocrinology 49 R37–R50. (https://doi.org/10.1530/JME-12-0069)
Carracedo A, Ma L, Teruya-Feldstein J, Rojo F, Salmena L, Alimonti A, Egia A, Sasaki AT, Thomas G, Kozma SC, et al. 2008 Inhibition of mTORC1 leads to MAPK pathway activation through a PI3K-dependent feedback loop in human cancer. Journal of Clinical Investigation 118 . (https://doi.org/10.1172/JCI34739)
Carter CA, Degesys A, Oronsky B, Scicinski J, Caroen SZ, Oronsky AL, Reid T, Cabrales P & Roswarski J 2015 Flushing out carcinoid syndrome: beneficial effect of the anticancer epigenetic agent RRx-001 in a patient with a treatment-refractory neuroendocrine tumor. Case Reports in Oncology 8 . (https://doi.org/10.1159/000441775)
Cesare AJ & Reddel RR 2010 Alternative lengthening of telomeres: models, mechanisms and implications. Nature Reviews: Genetics 11 . (https://doi.org/10.1038/nrg2763)
Chadrasekharappa SC, Guru SC, Manickam P, Olufemi SE, Collins FS, Emmert-Buck MR, Debelenko LV, Zhuang Z, Lubensky IA, Liotta LA, et al. 1997 Positional cloning of the gene for multiple endocrine neoplasia-type 1. Science 276 . (https://doi.org/10.1126/science.276.5311.404)
Chamberlain CE, German MS, Yang K, Wang J, VanBrocklin H, Regan M, Shokat KM, Ducker GS, Kim GE, Hann B, et al. 2018 A patient-derived xenograft model of pancreatic neuroendocrine tumors identifies sapanisertib as a possible new treatment for everolimus-resistant tumors. Molecular Cancer Therapeutics 17 . (https://doi.org/10.1158/1535-7163.MCT-17-1204).
Chan CS, Laddha SV, Lewis PW, Koletsky MS, Robzyk K, Da Silva E, Torres PJ, Untch BR, Li J, Bose P, et al. 2018 ATRX, DAXX or MEN1 mutant pancreatic neuroendocrine tumors are a distinct alpha-cell signature group. Nature Communications 9 4158. (https://doi.org/10.1038/s41467-018-06498-2)
Chan JA, Faris JE, Murphy JE, Blaszkowsky LS, Kwak EL, Jackson McCleary NJ, Fuchs CS, Meyerhardt JA, Ng K, Zhu AX, et al. 2017 Phase II trial of cabozantinib in patients with carcinoid and pancreatic neuroendocrine tumors (pNET). Journal of Clinical Oncology 35 (Supplement 4) . (https://doi.org/10.1200/JCO.2017.35.4_suppl.228)
Chang YT, Lin TP, Tang JT, Campbell M, Luo YL, Lu SY, Yang CP, Cheng TY, Chang CH, Liu TT, et al. 2018 HOTAIR is a REST-regulated lncRNA that promotes neuroendocrine differentiation in castration resistant prostate cancer. Cancer Letters 433 . (https://doi.org/10.1016/j.canlet.2018.06.029)
Chapman JR, Taylor MR & Boulton SJ 2012 Playing the end game: DNA double-strand break repair pathway choice. Molecular Cell 47 . (https://doi.org/10.1016/j.molcel.2012.07.029)
Chen YL, Chang MC & Cheng WF 2017 Metronomic chemotherapy and immunotherapy in cancer treatment. Cancer Letters 400 . (https://doi.org/10.1016/j.canlet.2017.01.040)
Cheunsuchon P, Zhou Y, Zhang X, Lee H, Chen W, Nakayama Y, Rice KA, Tessa Hedley-Whyte E, Swearingen B & Klibanski A 2011 Silencing of the imprinted DLK1-MEG3 locus in human clinically nonfunctioning pituitary adenomas. American Journal of Pathology 179 (https://doi.org/10.1016/j.ajpath.2011.07.002)
Choi IS, Estecio MR, Nagano Y, Kim DH, White JA, Yao JC, Issa JP & Rashid A 2007 Hypomethylation of LINE-1 and Alu in well-differentiated neuroendocrine tumors (pancreatic endocrine tumors and carcinoid tumors). Modern Pathology 20 . (https://doi.org/10.1038/modpathol.3800825)
Chu YH, Hardin H, Eickhoff J & Lloyd RV 2019 In situ hybridization analysis of long non-coding RNAs MALAT1 and HOTAIR in gastroenteropancreatic neoplasms. Endocrine Pathology 30 (https://doi.org/10.1007/s12022-018-9564-1)
Corbo V, Beghelli S, Bersani S, Antonello D, Talamini G, Brunelli M, Capelli P, Falconi M & Scarpa A 2012 Pancreatic endocrine tumours: mutational and immunohistochemical survey of protein kinases reveals alterations in targetable kinases in cancer cell lines and rare primaries. Annals of Oncology 23 . (https://doi.org/10.1093/annonc/mdr048)
Corbo V, Dalai I, Scardoni M, Barbi S, Beghelli S, Bersani S, Albarello L, Doglioni C, Schott C, Capelli P, et al. 2010 MEN1 in pancreatic endocrine tumors: analysis of gene and protein status in 169 sporadic neoplasms reveals alterations in the vast majority of cases. Endocrine-Related Cancer 17 . (https://doi.org/10.1677/ERC-10-0028)
Crabtree JS, Scacheri PC, Ward JM, McNally SR, Swain GP, Montagna C, Hager JH, Hanahan D, Edlund H, Magnuson MA, et al. 2003 Of mice and MEN1: insulinomas in a conditional mouse knockout. Molecular and Cellular Biology 23 . (https://doi.org/10.1128/mcb.23.17.6075-6085.2003)
Cusanelli E & Chartrand P 2015 Telomeric repeat-containing RNA Terra: a noncoding RNA connecting telomere biology to genome integrity. Frontiers in Genetics 6 143. (https://doi.org/10.3389/fgene.2015.00143)
Dammann R, Schagdarsurengin U, Liu L, Otto N, Gimm O, Dralle H, Boehm BO, Pfeifer GP & Hoang-Vu C 2003 Frequent RASSF1A promoter hypermethylation and K-ras mutations in pancreatic carcinoma. Oncogene 22 . (https://doi.org/10.1038/sj.onc.1206582)
Dasari A, Shen C, Halperin D, Zhao B, Zhou S, Xu Y, Shih T & Yao JC 2017 Trends in the incidence, prevalence, and survival outcomes in patients with neuroendocrine tumors in the United States. JAMA Oncology 3 . (https://doi.org/10.1001/jamaoncol.2017.0589)
Davison C, Brock S & Dyke CG 1936 Retinal and central nervous system hemangioblastomatosis with visceral changes (von Hippel-Lindau’s disease). Bull Neurological Institute New York 5 .
de Wilde RF, Edil BH, Hruban RH & Maitra A 2012a Well-differentiated pancreatic neuroendocrine tumors: from genetics to therapy. Nature Reviews: Gastroenterology and Hepatology 9 (https://doi.org/10.1038/nrgastro.2012.9)
de Wilde RF, Heaphy CM, Maitra A, Meeker AK, Edil BH, Wolfgang CL, Ellison TA, Schulick RD, Molenaar IQ, Valk GD, et al. 2012b Loss of ATRX or DAXX expression and concomitant acquisition of the alternative lengthening of telomeres phenotype are late events in a small subset of MEN-1 syndrome pancreatic neuroendocrine tumors. Modern Pathology 25 . (https://doi.org/10.1038/modpathol.2012.53)
Desrichard A, Bidet Y, Uhrhammer N & Bignon YJ 2011 CHEK2 contribution to hereditary breast cancer in non-BRCA families. Breast Cancer Research 13 R119. (https://doi.org/10.1186/bcr3062)
Donzelli M & Draetta GF 2003 Regulating mammalian checkpoints though Cdc25 inactivation. EMBO Reports 4 . (https://doi.org/10.1038/sj.embor.embor887)
Eden A, Gaudet F, Waghmare A & Jaenisch R 2003 Chromosomal instability and tumors promoted by DNA hypomethylation. Science 300 455. (https://doi.org/10.1126/science.1083557)
Elsaesser SJ & Allis CD 2010 HIRA and Daxx constitute two independent histone H3.3-containing predeposition complexes. Cold Spring Harbor Symposia on Quantitative Biology 75 . (https://doi.org/10.1101/sqb.2010.75.008)
Ersahin T, Tuncbag N & Cetin-Atalay R 2015 The PI3K/AKT/mTOR interactive pathway. Molecular Biosystems 11 (https://doi.org/10.1039/c5mb00101c)
Evans DGR, Komminoth P, Scheithauer BW & Peltonen J 2004 Neurofibromatosis type 1. In Pathology and Genetics: Tumours of Endocrine Organs. WHO Classification of Tumors, pp . Eds DeLellis RA, Lloyd RV, Heitz PU & Eng C. Lyon, France: IARC Press.
Faivre S, Delbaldo C, Vera K, Robert C, Lozahic S, Lassau N, Bello C, Deprimo S, Brega N, Massimini G, et al. 2006 Safety, pharmacokinetic, and antitumor activity of SU11248, a novel oral multitarget tyrosine kinase inhibitor, in patients with cancer. Journal of Clinical Oncology 24 . (https://doi.org/10.1200/JCO.2005.02.2194)
Faivre S, Niccoli P, Castellano D, Valle JW, Hammel P, Raoul JL, Vinik A, Van Cutsem E, Bang YJ, Lee SH, et al. 2017 Sunitinib in pancreatic neuroendocrine tumours: updated progression-free survival and final overall survival from a phase III randomized study. Annals of Oncology 28 . (https://doi.org/10.1093/annonc/mdw561)
Fernandez-Cuesta L, Peifer M, Lu X, Sun R, Ozretic L, Seidel D, Zander T, Leenders F, George J, Muller C, et al. 2014 Frequent mutations in chromatin-remodelling genes in pulmonary carcinoids. Nature Communications 5 3518. (https://doi.org/10.1038/ncomms4518)
Fishbein L & Nathanson KL 2015 Inherited and somatic genetics of pancreatic neuroendocrine tumors. In Management of Pancreatic Neuroendocrine Tumors, 1st ed., Chapter 2, pp . Ed Pisegna JR. New York, NY, USA: Springer-Verlag. (https://doi.org/10.1007/978-1-4939-1798-3)
Fraenkel M, Faggiano A & Valk GD 2015 Epidemiology of neuroendocrine tumors. Frontiers of Hormone Research 44 (https://doi.org/10.1159/000381970)
Fraenkel M, Kim MK, Faggiano A & Valk GD 2012 Epidemiology of gastroenteropancreatic neuroendocrine tumours. Best Practice and Research: Clinical Gastroenterology 26 (https://doi.org/10.1016/j.bpg.2013.01.006)
Fraga MF, Herranz M, Espada J, Ballestar E, Paz MF, Ropero S, Erkek E, Bozdogan O, Peinado H, Niveleau A, et al. 2004 A mouse skin multistage carcinogenesis model reflects the aberrant DNA methylation patterns of human tumors. Cancer Research 64 . (https://doi.org/10.1158/0008-5472.CAN-03-4061)
Francis JM, Kiezun A, Ramos AH, Serra S, Pedamallu CS, Qian ZR, Banck MS, Kanwar R, Kulkarni AA, Karpathakis A, et al. 2013 Somatic mutation of CDKN1B in small intestine neuroendocrine tumors. Nature Genetics 45 . (https://doi.org/10.1038/ng.2821)
Franko J, Feng W, Yip L, Genovese E & Moser AJ 2010 Non-functional neuroendocrine carcinoma of the pancreas: incidence, tumor biology, and outcomes in 2,158 patients. Journal of Gastrointestinal Surgery 14 (https://doi.org/10.1007/s11605-009-1115-0)
Friedman E, Sakaguchi K, Bale AE, Falchetti A, Streeten E, Zimering MB, Weinstein LS, McBride WO, Nakamura Y & Brandi ML 1989 Clonality of parathyroid tumors in familial multiple endocrine neoplasia type 1. New England Journal of Medicine 321 . (https://doi.org/10.1056/NEJM198907273210402)
Genus T, Bouvier C, Wong K, Srirajaskanthan R, Rous B, Talbot D, Valle J, Khan M, Pearce N, Elshafie M, et al. 2017 Incidence and prevalence of neuroendocrine tumours in England. Endocrine Abstracts 52 OC3. (https://doi.org/10.1530/endoabs.52.OC3)
Gibb EA, Brown CJ & Lam WL 2011 The functional role of long non-coding RNA in human carcinomas. Molecular Cancer 10 38. (https://doi.org/10.1186/1476-4598-10-38)
Girardi DM, Silva ACB, Rêgo JFM, Coudry RA & Riechelmann RP 2017 Unraveling molecular pathways of poorly differentiated neuroendocrine carcinomas of the gastroenteropancreatic system: a systematic review. Cancer Treatment Reviews 56 . (https://doi.org/10.1016/j.ctrv.2017.04.002)
Grolmusz VK, Kövesdi A, Borka K, Igaz P & Patócs A 2018 Prognostic relevance of proliferation-related miRNAs in pancreatic neuroendocrine neoplasms. European Journal of Endocrinology 179 . (https://doi.org/10.1530/EJE-18-0305)
Grozinsky-Glasberg S, Shimon I, Korbonits M & Grossman AB 2008 Somatostatin analogues in the control of neuroendocrine tumours: efficacy and mechanisms. Endocrine-Related Cancer 15 . (https://doi.org/10.1677/ERC-07-0288)
Guo D, Wu F, Liu H, Gao G, Kou S, Yang F, Abbas N, Zhou T, Cai X, Zhang H, et al. 2017 Generation of non-integrated induced pluripotent stem cells from a 23-year-old male with multiple endocrine neoplasia type 1 syndrome. Stem Cell Research 18 . (https://doi.org/10.1016/j.scr.2016.12.002)
Gupta RA, Shah N, Wang KC, Kim J, Horlings HM, Wong DJ, Tsai MC, Hung T, Argani P, Rinn JL, et al. 2010 Long non-coding RNA HOTAIR reprograms chromatin state to promote cancer metastasis. Nature 464 . (https://doi.org/10.1038/nature08975)
Hackeng WM, Brosens LAA, Poruk KE, Noe M, Hosoda W, Poling JS, Rizzo A, Campbell-Thompson M, Atkinson MA, Konukiewitz B, et al. 2016 Aberrant menin expression is an early event in pancreatic neuroendocrine tumorigenesis. Human Pathology 56 . (https://doi.org/10.1016/j.humpath.2016.06.006)
Hajjari M & Salavaty A 2015 HOTAIR: an oncogenic long non-coding RNA in different cancers. Cancer Biology and Medicine 12 . (https://doi.org/10.7497/j.issn.2095-3941.2015.0006)
Halfdanarson TR, Rabe KG, Rubin J & Petersen GM 2008 Pancreatic neuroendocrine tumors (PNETs): incidence, prognosis and recent trend toward improved survival. Annals of Oncology 19 . (https://doi.org/10.1093/annonc/mdn351)
Hanahan D 1985 Heritable formation of pancreatic beta-cell tumours in transgenic mice expressing recombinant insulin/simian virus 40 oncogenes. Nature 315 . (https://doi.org/10.1038/315115a0)
Hashimoto M, Kyo S, Hua X, Tahara H, Nakajima M, Takakura M, Sakaguchi J, Maida Y, Nakamura M, Ikoma T, et al. 2008 Role of menin in the regulation of the telomerase activity in normal and cancer cells. International Journal of Oncology 33 .
Hauptman N & Glavač D 2013 Long non-coding RNA in cancer. International Journal of Molecular Sciences 14 . (https://doi.org/10.3390/ijms14034655)
Heaphy CM, de Wilde RF, Jiao Y, Klein AP, Edil BH, Shi C, Bettegowda C, Rodriguez FJ, Eberhart CG, Hebbar S, et al. 2011a Altered telomeres in tumors with ATRX and DAXX mutations. Science 333 425. (https://doi.org/10.1126/science.1207313)
Heaphy CM, Subhawong AP, Hong SM, Goggins MG, Montgomery EA, Gabrielson E, Netto GJ, Epstein JI, Lotan TL, Westra WH, et al. 2011b Prevalence of the alternative lengthening of telomeres telomere maintenance mechanism in human cancer subtypes. American Journal of Pathology 179 . (https://doi.org/10.1016/j.ajpath.2011.06.018)
Hernando-Cubero J, Manzano JL, Benavent M, Lopez C, Teulé R, Garcia-Carbonero R, Carmona-Bayonas A, Crespo G, Cubillo A, Jimenez-Fonseca P, et al. 2018 Durvalumab plus tremelimumab for the treatment of patients (pts) with advanced neuroendocrine neoplasms (NENs) of lung or gastroenteropancreatic (GEP) origin. A phase II multicohort trial (DUNE Trial/GETNE 1601). Neuroendocrinology 106 (Suppl 1) 228 abstract J10. (https://doi.org/10.1159/000487699)
Hirata H, Hinoda Y, Shahryari V, Deng G, Nakajima K, Tabatabai ZL, Ishii N & Dahiya R 2015 Long noncoding RNA MALAT1 promotes aggressive renal cell carcinoma through Ezh2 and interacts with miR-205. Cancer Research 75 . (https://doi.org/10.1158/0008-5472.CAN-14-2931)
Holley CL & Topkara VK 2011 An introduction to small non-coding RNAs: miRNA and snoRNA. Cardiovascular Drugs and Therapy 25 . (https://doi.org/10.1007/s10557-011-6290-z)
House MG, Herman JG, Guo MZ, Hooker CM, Schulick RD, Lillemoe KD, Cameron JL, Hruban RH, Maitra A & Yeo CJ 2003 Aberrant hypermethylation of tumor suppressor genes in pancreatic endocrine neoplasms. Annals of Surgery 238 ; discussion 431. (https://doi.org/10.1097/01.sla.0000086659.49569.9e)
Hu W, Feng Z, Modica I, Klimstra DS, Song L, Allen PJ, Brennan MF, Levine AJ & Tang LH 2010 Gene amplifications in well-differentiated pancreatic neuroendocrine tumors inactivate the p53 pathway. Genes and Cancer 1 . (https://doi.org/10.1177/1947601910371979)
Huang X, Ding L, Bennewith KL, Tong RT, Welford SM, Ang KK, Story M, Le QT & Giaccia AJ 2009 Hypoxia-inducible mir-210 regulates normoxic gene expression involved in tumor initiation. Molecular Cell 35 . (https://doi.org/10.1016/j.molcel.2009.09.006)
Jang ER & Lee JS 2004 DNA damage response mediated through BRCA1. Cancer Research and Treatment 36 . (https://doi.org/10.4143/crt.2004.36.4.214)
Jia R, Li YL, Jin YS, Chen Y, Zhang Y, Wan Y, Liu RR, Zhou H, Jiang YB, Xu Q, et al. 2018 Efficacy and safety of anti-PD-1 antibody (IBI308) in treating advanced neuroendocrine neoplasms. Neuroendocrinology 106 (Suppl 1) 230 abstract J12. (https://doi.org/10.1159/000487699)
Jiao Y, Shi C, Edil BH, de Wilde RF, Klimstra DS, Maitra A, Schulick RD, Tang LH, Wolfgang CL, Choti MA, et al. 2011 DAXX/ATRX, MEN1, and mTOR pathway genes are frequently altered in pancreatic neuroendocrine tumors. Science 331 . (https://doi.org/10.1126/science.1200609)
Jin N, Lubner SJ, Mulkerin DL, Rajguru S, Carmichael L, Chen H, Holen KD & LoConte NK 2016 A phase II trial of a histone deacetylase inhibitor panobinostat in patients with low-grade neuroendocrine tumors. Oncologist 21 . (https://doi.org/10.1634/theoncologist.2016-0060)
Johannessen CM, Reczek EE, James MF, Brems H, Legius E & Cichowski K 2005 The NF1 tumor suppressor critically regulates TSC2 and mTOR. PNAS 102 . (https://doi.org/10.1073/pnas.0503224102)
Jones PA & Baylin SB 2002 The fundamental role of epigenetic events in cancer. Nature Reviews: Genetics 3 . (https://doi.org/10.1038/nrg816)
Jonkers YM, Ramaekers FC & Speel EJ 2007 Molecular alterations during insulinomas tumorigenesis. Biochimica et Biophysica Acta 1775 . (https://doi.org/10.1016/j.bbcan.2007.05.004)
Kaelin WG Jr 2002 Molecular basis of the VHL hereditary cancer syndrome. Nature Reviews: Cancer 2 . (https://doi.org/10.1038/nrc885)
Karnik SK, Hughes CM, Gu X, Rozenblatt-Rosen O, McLean GW, Xiong Y, Meyerson M & Kim SK 2005 Menin regulates pancreatic islet growth by promoting histone methylation and expression of genes encoding p27Kip1 and p18INK4c. PNAS 102 (https://doi.org/10.1073/pnas.0503484102)
Karpathakis A, Dibra H, Pipinikas C, Feber A, Morris T, Francis J, Oukrif D, Mandair D, Pericleous M, Mohmaduvesh M, et al. 2016 Prognostic impact of novel molecular sub-types of small intestinal neuroendocrine tumor. Clinical Cancer Research 22 . (https://doi.org/10.1158/1078-0432.CCR-15-0373)
Karpathakis A, Dibra H, Pipinikas C, Feber A, Morris T, Francis J, Oukrif D, Mandair D, Pericleous M, Mohmaduvesh M, et al. 2017 Progressive epigenetic dysregulation in neuroendocrine tumour liver metastases. Endocrine-Related Cancer 24 L21–L25. (https://doi.org/10.1530/ERC-16-0419)
Kawasaki K, Fujii M & Sato T 2018 Gastroenteropancreatic neoplasms: genes, therapies and models. Disease Models and Mechanisms 11 dmm029595. (https://doi.org/10.1242/dmm.029595)
Kim H, Lee JE, Cho EJ, Liu JO & Youn HD 2003 Menin, a tumor suppressor, represses JunD-mediated transcriptional activity by association with an mSin3A-histone deacetylase complex. Cancer Research 63 .
Kim K, Jutooru I, Chadalapaka G, Johnson G, Frank J, Burghardt R, Kim S & Safe S 2013 HOTAIR is a negative prognostic factor and exhibits pro-oncogenic activity in pancreatic cancer. Oncogene 32 . (https://doi.org/10.1038/onc.2012.193)
Kimura W, Kuroda A & Morioka Y 1991 Clinical pathology of endocrine tumors of the pancreas. Analysis of autopsy cases. Digestive Diseases and Sciences 36 . (https://doi.org/10.1007/BF01297144)
Klöppel G, Anlauf M & Perren A 2007 Endocrine precursor lesions of gastroenteropancreatic neuroendocrine tumors. Endocrine Pathology 18 . (https://doi.org/10.1007/s12022-007-0025-5)
Koc G, Sugimoto S, Kuperman R, Kammen BF & Karakas SP 2017 Pancreatic tumors in children and young adults with tuberous sclerosis complex. Pediatric Radiology 47 . (https://doi.org/10.1007/s00247-016-3701-0)
Kulke MH, Hornick JL, Frauenhoffer C, Hooshmand S, Ryan DP, Enzinger PC, Meyerhardt JA, Clark JW, Stuart K, Fuchs CS, et al. 2009 O6-Methylguanine DNA methyltransferase deficiency and response to temozolomide-based therapy in patients with neuroendocrine tumors. Clinical Cancer Research 15 . (https://doi.org/10.1158/1078-0432.CCR-08-1476)
Kulke MH, Niedzwiecki D, Foster NR, Fruth B, Kunz PL, Kennecke HF, Wolin EM & Venook AP 2015 Randomized phase II study of everolimus (E) versus everolimus plus bevacizumab (E+B) in patients (Pts) with locally advanced or metastatic pancreatic neuroendocrine tumors (pNET), CALGB 80701 (Alliance). Journal of Clinical Oncology 33 (Supplement 15) (https://doi.org/10.1200/jco.2015.33.15_suppl.4005)
Kulke MH, Ruszniewski P, Van Cutsem E, Lombard-Bohas C, Valle JW, De Herder WW, Pavel M, Degtyarev E, Brase JC, Bubuteishvili-Pacaud L, et al. 2017 A randomized, open-label, phase 2 study of everolimus in combination with pasireottide LAR or everolimus alone in advanced, well-differentiated, progressive pancreatic neuroendocrine tumors: COOPERATE-2 trial. Annals of Oncology 28 . (https://doi.org/10.1093/annonc/mdx078)
Kunz PL, Catalano PJ, Nimeiri H, Fisher GA, Longacre TA, Suarez CJ, Yao JC, Kulke MH, Hendifar AE, Shanks JC, et al. 2018 A randomized study of temozolomide or temozolomide and capecitabine in patients with advanced pancreatic neuroendocrine tumors: a trial of the ECOG-ACRIN Cancer Research Group (E2211). Journal of Clinical Oncology 36 (15_Supplement) 4004. (https://doi.org/10.1200/JCO.2018.36.15_suppl.4004)
Kwok ZH, Roche V, Chew XH, Fadieieva A & Tay Y 2018 A non-canonical tumor suppressive role for the long non-coding RNA MALAT1 in colon and breast cancers. International Journal of Cancer 143 . (https://doi.org/10.1002/ijc.31386)
Larson AM, Hedgire SS, Deshpande V, Stemmer-Rachamimov AO, Harisinghani MG, Ferrone CR, Shah U & Thiele EA 2012 Pancreatic neuroendocrine tumors in patients with tuberous sclerosis complex. Clinical Genetics 82 . (https://doi.org/10.1111/j.1399-0004.2011.01805.x)
Larsson C, Skogseid B, Oberg K, Nakamura Y & Nordenskjöld M 1988 Multiple endocrine neoplasia type 1 gene maps to chromosome 11 and is lost in insulinoma. Nature 332 . (https://doi.org/10.1038/332085a0)
Latif F, Tory K, Gnarra J, Yao M, Duh FM, Orcutt ML, Stackhouse T, Kuzmin I, Modi W & Geil L 1993 Identification of the von Hippel-Lindau disease tumor-suppressor gene. Science 260 . (https://doi.org/10.1126/science.8493574)
Lawrence B, Blenkiron C, Parker K, Tsai P, Fitzgerald S, Shields P, Robb T, Yeong ML, Kramer N, James S, et al. 2018 Recurrent loss of heterozygosity correlates with clinical outcome in pancreatic neuroendocrine cancer. NPJ Genomic Medicine 3 18 (https://doi.org/10.1038/s41525-018-0058-3)
Lawrence B, Gustafsson BI, Chan A, Svejda B, Kidd M & Modlin IM 2011 The epidemiology of gastroenteropancreatic neuroendocrine tumors. Endocrinology and Metabolism Clinics of North America 40 , vii. (https://doi.org/10.1016/j.ecl.2010.12.005)
Lawrence MS, Stojanov P, Polak P, Kryukov GV, Cibulskis K, Sivachenko A, Carter SL, Stewart C, Mermel CH, Roberts SA, et al. 2013 Mutational heterogeneity in cancer and the search for new cancer-associated genes. Nature 499 . (https://doi.org/10.1038/nature12213)
Lee MY, Kim HJ, Kim MA, Jee HJ, Kim AJ, Bae YS, Park JI, Chung JH & Yun J 2008 Nek6 is involved in G2/M phase cell cycle arrest through DNA damage-induced phosphorylation. Cell Cycle 7 . (https://doi.org/10.4161/cc.7.17.6551)
Lee YS, Kim H, Kim HW, Lee JC, Paik KH, Kang J, Kim J, Yoon YS, Han HS, Sohn I, et al. 2015 High expression of microRNA-196a indicates poor prognosis in resected pancreatic neuroendocrine tumor. Medicine 94 e2224. (https://doi.org/10.1097/MD.0000000000002224)
Lewis PW, Elsaesser SJ, Noh KM, Stadler SC & Allis CD 2010 Daxx is an H3.3-specific histone chaperone and cooperates with ATRX in replication-independent chromatin assembly at telomeres. PNAS 107 . (https://doi.org/10.1073/pnas.1008850107)
Li A, Yu J, Kim H, Wolfgang CL, Canto MI, Hruban RH & Goggins M 2013 MicroRNA array analysis finds elevated serum miR-1290 accurately distinguishes patients with low-stage pancreatic cancer from healthy and disease controls. Clinical Cancer Research 19 . (https://doi.org/10.1158/1078-0432.CCR-12-3092)
Libutti SK, Choyke PL, Bartlett DL, Vargas H, Walther M, Lubensky I, Glenn G, Linehan WM & Alexander HR 1998 Pancreatic neuroendocrine tumors associated with von Hippel Lindau disease: diagnostic and management recommendations. Surgery 124 . (https://doi.org/10.1067/msy.1998.91823)
Lin SY & Elledge SJ 2003 Multiple tumor suppressor pathways negatively regulate telomerase. Cell 113 . (https://doi.org/10.1016/S0092-8674(03)00430-6)
Lines KE, Stevenson M, Filippakopoulos P, Müller S, Lockstone HE, Wright B, Grozinsky-Glasberg S, Grossman AB, Knapp S, Buck D, et al. 2017 Epigenetic pathway inhibitors represent potential drugs for treating pancreatic and bronchial neuroendocrine tumors. Oncogenesis 6 e332. (https://doi.org/10.1038/oncsis.2017.30)
Liu B, Tang LH, Liu Z, Mei M, Yu R, Dhall D, Qiao XW, Zhang TP, Zhao YP, Liu TH, et al. 2014 α-Internexin: a novel biomarker for pancreatic neuroendocrine tumor aggressiveness. Journal of Clinical Endocrinology and Metabolism 99 E786–E795. (https://doi.org/10.1210/jc.2013-2874)
Lloyd RV, Osamura RY, Kloppel G & Rosai J 2017 WHO Classification of Tumours of Endocrine Organs, 4th ed. Lyon, France: IARC Press.
Lovejoy CA, Li WD, Reisenweber S, Thongthip S, Bruno J, de Lange T, De S, Petrini JH, Sung PA, Jasin M, et al. 2012 Loss of ATRX, genome instability, and an altered DNA damage response are hallmarks of the alternative lengthening of telomeres pathway. PLoS Genetics 8 e1002772. (https://doi.org/10.1371/journal.pgen.1002772)
Lu X, Ma O, Nguyen TA, Jones SN, Oren M & Donehower LA 2007 The Wip1 phosphatase acts as a gatekeeper in the p53-Mdm2 autoregulatory loop. Cancer Cell 12 . (https://doi.org/10.1016/j.ccr.2007.08.033)
Maher ER, Neumann HP & Richard S 2011 Von Hippel-Lindau disease: a clinical and scientific review. European Journal of Human Genetics 19 . (https://doi.org/10.1038/ejhg.2010.175)
Malkin D, Li FP, Strong LC, Fraumeni JF Jr, Nelson CE, Kim DH, Kassel J, Gryka MA, Bischoff FZ & Tainsky MA 1990 Germ line p53 mutations in a familial syndrome of breast cancer, sarcomas, and other neoplasms. Science 250 . (https://doi.org/10.1126/science.1978757)
Malpeli G, Amato E, Dandrea M, Fumagalli C, Debattisti V, Boninsegna L, Pelosi G, Falconi M & Scarpa A 2011 Methylation-associated down-regulation of RASSF1A and up-regulation of RASSF1C in pancreatic endocrine tumors. BMC Cancer 11 351. (https://doi.org/10.1186/1471-2407-11-351)
Marine JC, Dyer MA & Jochemsen AG 2007 MDMX: from bench to bedside. Journal of Cell Science 120 . (https://doi.org/10.1242/jcs.03362)
Marinoni I, Kurrer AS, Vassella E, Dettmer M, Rudolph T, Banz V, Hunger F, Pasquinelli S, Speel EJ & Perren A 2014 Loss of DAXX and ATRX are associated with chromosome instability and reduced survival of patients with pancreatic neuroendocrine tumors. Gastroenterology 146 453–60.e5. (https://doi.org/10.1053/j.gastro.2013.10.020)
Marx SJ & Simonds WF 2005 Hereditary hormone excess: genes, molecular pathways, and syndromes. Endocrine Reviews 26 . (https://doi.org/10.1210/er.2003-0037)
Matkar S, Thiel A & Hua X 2013 Menin: a scaffold protein that controls gene expression and cell signalling. Trends in Biochemical Sciences 38 . (https://doi.org/10.1016/j.tibs.2013.05.005)
Mazzei F, Viel A & Bignami M 2013 Role of MUTYH in human cancer. Mutation Research 743–744 . (https://doi.org/10.1016/j.mrfmmm.2013.03.003)
McGranahan N, Furness AJ, Rosenthal R, Ramskov S, Lyngaa R, Saini SK, Jamal-Hanjani M, Wilson GA, Birkbak NJ, Hiley CT, et al. 2016 Clonal neoantigen elicit T cell immunoreactivity and sensitivity to immune checkpoint blockade. Science 351 . (https://doi.org/10.1126/science.aaf1490)
Mehnert JM, Rugo HS, O'Neil BH, Santoro A, Schellens JHM, Cohen RB, Doi T, Ott PA, Pishvaian MJ, Puzanov I, et al. 2017 Pembrolizumab for patients with PD-L1-positive advanced carcinoid or pancreatic neuroendocrine tumors: results from the KEYNOTE-028 study. Annals of Oncology 28 (Supplement 5) mdx368. (https://doi.org/10.1093/annonc/mdx368)
Mignon M 2000 Natural history of neuroendocrine enteropancreatic tumors. Digestion 62 . (https://doi.org/10.1159/000051856)
Milne TA, Hughes CM, Lloyd R, Yang Z, Rozenblatt-Rosen O, Dou Y, Schnepp RW, Krankel C, Livolsi VA, Gibbs D, et al. 2005 Menin and MLL cooperatively regulate expression of cyclin-dependent kinase inhibitors. PNAS 102 . (https://doi.org/10.1073/pnas.0408836102)
Missiaglia E, Dalai I, Barbi S, Beghelli S, Falconi M, della Peruta M, Piemonti L, Capurso G, Di Florio A, delle Fave G, et al. 2010 Pancreatic endocrine tumors: expression profiling evidences a role for AKT-mTOR pathway. Journal of Clinical Oncology 28 . (https://doi.org/10.1200/JCO.2008.21.5988)
Missiaglia E, Moore PS, Williamson J, Lemoine NR, Falconi M, Zamboni G & Scarpa A 2002 Sex chromosome anomalies in pancreatic endocrine tumours. International Journal of Cancer 98 . (https://doi.org/10.1002/ijc.10223)
Modali SD, Parekh VI, Kebebew E & Agarwal SK 2015 Epigenetic regulation of the lncRNA MEG3 and its target c-MET in pancreatic neuroendocrine tumors. Molecular Endocrinology 29 . (https://doi.org/10.1210/me.2014-1304)
Modlin IM, Shapiro MD & Kidd M 2004 Siegfried Oberndorfer: origins and perspectives of carcinoid tumors. Human Pathology 35 . (https://doi.org/10.1016/j.humpath.2004.09.018)
Mohamed A, Blanchard MP, Albertelli M, Barbieri F, Brue T, Niccoli P, Delpero JR, Monges G, Garcia S, Ferone D, et al. 2014 Pasireotide and octreotide antiproliferative effects and sst2 trafficking in human pancreatic neuroendocrine tumor cultures. Endocrine-Related Cancer 21 . (https://doi.org/10.1530/ERC-14-0086)
Moore PS, Missiaglia E, Antonello D, Zamò A, Zamboni G, Corleto V, Falconi M & Scarpa A 2001 Role of disease-causing genes in sporadic pancreatic endocrine tumors: MEN1 and VHL. Genes, Chromosomes and Cancer 32 . (https://doi.org/10.1002/gcc.1180)
Muscarella P, Melvin WS, Fisher WE, Foor J, Ellison EC, Herman JG, Schirmer WJ, Hitchcock CL, DeYoung BR & Weghorst CM 1998 Genetic alterations in gastrinomas and nonfunctioning pancreatic neuroendocrine tumors: an analysis of p16/MTS1 tumor suppressor gene inactivation. Cancer Research 58 .
Neychev V, Steinberg SM, Cottle-Delisle C, Merkel R, Nilubol N, Yao J, Meltzer P, Pacak K, Marx S & Kebebew E 2015 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. (https://doi.org/10.1136/bmjopen-2015-008248)
Niederle MB, Hackl M, Kaserer K & Niederle B 2010 Gastroenteropancreatic neuroendocrine tumours: the current incidence and staging based on the WHO and European Neuroendocrine Tumour Society classification: an analysis based on prospectively collected parameters. Endocrine-Related Cancer 17 . (https://doi.org/10.1677/ERC-10-0152)
Oberg K, Casanovas O, Castano JP, Chung D, Delle Fave G, Denefle P, Harris P, Khan MS, Kulke MH, Scarpa A, et al.2013 Molecular aspects of neuroendocrine tumors: implications for current and future therapeutic approaches. Clinical Cancer Research 19 . (https://doi.org/10.1158/1078-0432.CCR-12-3458)
Oberndorfer S 1907 Karzinoide tumoren des dünndarms. Frankfurter Zeitschrift für Pathologie 1 .
Oberndorfer S 1929 Die Geschwülste des Darms. In Handbuch der speziellen pathologischen Anatomie und Histologie, vol. 4, pp . Eds Henke F & Lubarsch O. Berlin, Germany: Springer.
Ohki R, Saito K, Chen Y, Kawase T, Hiraoka N, Saigawa R, Minegishi M, Aita Y, Yanai G, Shimizu H, et al. 2014 PHLDA3 is a novel tumor suppressor of pancreatic neuroendocrine tumors. PNAS 111 E2404–E2413. (https://doi.org/10.1073/pnas.1319962111)
Olson P, Chu GC, Perry SR, Nolan-Stevaux O & Hanahan D 2011 Imaging guided trials of the angiogenesis inhibitor sunitinib in mouse models predict efficacy in pancreatic neuroendocrine but not ductal carcinoma. PNAS 108 E1275–E1284. (https://doi.org/10.1073/pnas.1111079108)
Ørom UA, Nielsen FC & Lund AH 2008 MicroRNA-10a binds the 5′UTR of ribosomal protein mRNAs and enhances their translation. Molecular Cell 30 . (https://doi.org/10.1016/j.molcel.2008.05.001)
Ou YH, Chung PH, Sun TP & Shieh SY 2005 p53 C-terminal phosphorylation by CHK1 and CHK2 participates in the regulation of DNA-damage-induced C-terminal acetylation. Molecular Biology of the Cell 16 . (https://doi.org/10.1091/mbc.e04-08-0689)
Padda SK, Brzezniak CE, Waqar SN, Wakelee HA, Ma PC, Rose MG, Reid T, Spira AI, Wong EYM, Zeman KG, et al. 2018 QUADRUPLE THREAT: A phase II trial of RRx-001 followed by etoposide-platinum re-challenge in previously treated small cell lung cancer. Journal of Clinical Oncology 36 (Supplement 15) e20575. (https://doi.org/10.1200/JCO.2018.36.15_suppl.e20575)
Panarelli N, Tyryshkin K, Wong JJM, Majewski A, Yang X, Scognamiglio T, Kim MK, Bogardus K, Tuschl T, Chen YT, et al. 2019 Evaluating gastroenteropancreatic neuroendocrine tumors through microRNA sequencing. Endocrine-Related Cancer 26 . (https://doi.org/10.1530/ERC-18-0244)
Pang EJ, Yang R, Fu XB & Liu YF 2015 Overexpression of long non-coding RNA MALAT1 is correlated with clinical progression and unfavorable prognosis in pancreatic cancer. Tumour Biology 36 . (https://doi.org/10.1007/s13277-014-2850-8)
Patel P & Galoian K 2018 Molecular challenges of neuroendocrine tumors. Oncology Letters 15 . (https://doi.org/10.3892/ol.2017.7680)
Pavel M, O'Toole D, Costa F, Capdevila J, Gross D, Kianmanesh R, Krenning E, Knigge U, Salazar R, Pape UF, et al. 2016 Enets consensus guidelines update for the management of distant metastatic disease of intestinal, pancreatic, bronchial neuroendocrine neoplasms (NEN) and NEN of unknown primary site. Neuroendocrinology 103 . (https://doi.org/10.1159/000443167)
Pavel ME & Sers C 2016 WOMEN IN CANCER THEMATIC REVIEW: Systemic therapies in neuroendocrine tumours and novel approaches toward personalized medicine. Endocrine-Related Cancer 23 T135–T154. (https://doi.org/10.1530/ERC-16-0370)
Pea A, Yu J, Marchionni L, Noe M, Luchini C, Pulvirenti A, de Wilde RF, Brosens LA, Rezaee N, Javed A, et al. 2018 Genetic analysis of small well-differentiated pancreatic neuroendocrine tumors identifies subgroups with differing risks of liver metastases. Annals of Surgery [epub]. (https://doi.org/10.1097/SLA.0000000000003022)
Pellegata NS, Quintanilla-Martinez L, Siggelkow H, Samson E, Bink K, Hofler H, Fend F, Graw J & Atkinson MJ 2006 Germ-line mutations in p27Kip1 cause a multiple endocrine neoplasia syndrome in rats and humans. PNAS 103 . (https://doi.org/10.1073/pnas.0603877103)
Perren A, Komminoth P, Saremaslani P, Matter C, Feurer S, Lees JA, Heitz PU & Eng C 2000 Mutation and expression analyses reveal differential subcellular compartmentalization of PTEN in endocrine pancreatic tumors compared to normal islet cells. American Journal of Pathology 157 . (https://doi.org/10.1016/S0002-9440(10)64624-X)
Pipinikas CP, Dibra H, Karpathakis A, Feber A, Novelli M, Oukrif D, Fusai G, Valente R, Caplin M, Meyer T, et al. 2015 Epigenetic dysregulation and poorer prognosis in DAXX-deficient pancreatic neuroendocrine tumours. Endocrine-Related Cancer 22 L13–L18 (https://doi.org/10.1530/ERC-15-0108)
Pless J 2005 The history of somatostatin analogs. Journal of Endocrinological Investigation 28 .
Raj NP, Soumerai T, Valentino E, Hechtman JF, Berger MF & Reidy DL 2016 Next-generation sequencing (NGS) in advanced well differentiated pancreatic neuroendocrine tumors (WD pNETs): a study using MSK-IMPACT. Journal of Clinical Oncology 34 246. (https://doi.org/10.1200/jco.2016.34.4_suppl.246)
Ram RR, Mendiratta S, Bodemann BO, Torres MJ, Eskiocak U & White MA 2014 RASSF1A inactivation unleashes a tumour suppressor/oncogene cascade with context-dependent consequences on cell cycle progression. Molecular and Cellular Biology 34 . (https://doi.org/10.1128/MCB.01506-13)
Ramage J, Naraev BG & Halfdanarson TR 2018 Peptide receptor radionuclide therapy for patients with advanced pancreatic neuroendocrine tumors. Seminars in Oncology 45 . (https://doi.org/10.1053/j.seminoncol.2018.08.004)
Raymond E, Dahan L, Raoul JL, Bang YJ, Borbath I, Lombard-Bohas C, Valle J, Metrakos P, Smith D, Vinik A, et al. 2011 Sunitinib malate for the treatment of pancreatic neuroendocrine tumors. New England Journal of Medicine 364 . (https://doi.org/10.1056/NEJMoa1003825)
Reidy-Lagunes DL, Vakiani E, Segal MF, Hollywood EM, Tang LH, Solit DB, Pietanza MC, Capanu M & Saltz LB 2012 A phase 2 study of the insulin-like growth factor-1 receptor inhibitor MK-0646 in patients with metastatic, well-differentiated neuroendocrine tumors. Cancer 118 . (https://doi.org/10.1002/cncr.27459)
Ries LAG, Young JL, Keel GE, Eisner MP, Lin YD & Horner M-J (eds) 2007 Cancer of the pancreas. In SEER Survival Monograph: Cancer Survival Among Adults: U.S. SEER Program, 1988-2001, Patient and Tumor Characteristics, Chapter 7, pp . SEER Program, NIH Pub. No. 07-6215. Bethesda, MD, USA: National Cancer Institute.