Neuroendocrine tumors (NETs) are a heterogeneous group of neoplasms, arising from neuroendocrine cells that are dispersed throughout the body. Around 20% of NETs occur in the context of a genetic syndrome. Today there are at least ten recognized NET syndromes. This includes the classical syndromes: multiple endocrine neoplasias types 1 and 2, and von Hippel–Lindau and neurofibromatosis type 1. Additional susceptibility genes associated with a smaller fraction of NETs have also been identified. Recognizing genetic susceptibility has proved essential both to provide genetic counseling and to give the best preventive care. In this review we will also discuss the knowledge of somatic genetic alterations in NETs. At least 24 genes have been implicated as drivers of neuroendocrine tumorigenesis, and the overall rates of genomic instability are relatively low. Genetic intra-tumoral, as well as inter-tumoral heterogeneity in the same patient, have also been identified. Together these data point towards the common pathways in NET evolution, separating early from late disease drivers. Although knowledge of specific mutations in NETs has limited impact on actual patient management, we predict that in the near future genomic profiling of tumors will be included in the clinical arsenal for diagnostics, prognostics and therapeutic decisions.
Neuroendocrine tumors (NETs) are rare neoplasms originating from neuroendocrine cells most commonly located in the endocrine glands but also in the gastrointestinal and bronchopulmonary systems (Fig. 1) (1). NETs usually display slow proliferation but have a high rate of advanced stage at presentation (2, 3). Neuropeptide hypersecretion is a hallmark of these diseases and often results in distinct hormonal syndromes. These syndromes are associated with both increased morbidity and mortality (2, 4). The median survival of patients with well differentiated NETs is about 10 years, contributing to the relatively high prevalence of these diseases (5, 6).
NET is one of the most heritable groups of neoplasms that feature in at least ten genetic syndromes. Additional susceptibility genes associated with a smaller fraction of NETs have also been identified. In this review we will present relevant genetic information regarding these diagnostic entities. We will also summarize data from tumor sequencing studies that provided comprehensive maps of the genomic landscape of NETs. Finally we comment on the current and future potential of these data as tools for the development of biomarkers as well as novel druggable targets. Below we present syndromes and additional susceptibility genes with increased risk of NET development (Table 1). The present paper reviews genetic profiles, by means of early and late mutations, of relevance for tumorigenesis in NETs.
Genetic syndromes with NET manifestation. Overview of genetic syndromes that involve NETs; penetrance of NETs at the specified location is presented in parenthesis; figures on penetrance are presented when available.
|Name||Gene||NET manifestation||Other manifestation|
|MEN1||MEN1||Parathyroid (>90%), gastroenteropancreatic (50%), anterior pituitary (30%), lung and thymus (10%)||Adrenocortical tumors|
|MEN2A (FMTC)||RET||Medullary thyroid (90–100%), adrenal medulla (20–80%), parathyroid (20%)a|
|MEN2B||RET||Parathyroid, medullary thyroid (100%), adrenal medulla (50%)||Marfanoid habitus|
|MEN4||CDKN1B||Parathyroid, pancreas, pituitary|
|Neurofibromatosis type 1||NF1||Adrenal medulla (1–5%), duodenum||Neurofibroma, cafe-au-lait spots|
|von Hippel–Lindau||VHL||Adrenal medulla and sympathetic ganglia (15%), pancreas (10%)||Hemangioblastoma, renal carcinoma|
|Familial PGL 1-5||SDHA-D, SDHAF2||Sympathetic and parasympathetic paraganglia, Adrenal medulla||GIST, renal carcinoma|
|Familial PCC and PGL syndromes||TMEM127, MAX, FH, MDH2||Adrenal medulla, sympathetic ganglia (TMEM127 30%)|
|Polycytemia paranganglioma syndrome||EPAS1||Sympathetic ganglia, adrenal medulla, duodenum||Polycytemia|
|Tuberous sclerosis complexes||TSC1, TSC2||Pancreas||Hamartoma|
|HPT-JT syndrome||HRPT2||Parathyroid adenoma (80%) and carcinoma (15%)|
Penetrance in MEN2 is specific to the particular mutation.
Multiple endocrine neoplasia type 1
Multiple endocrine neoplasia type 1 (MEN1; OMIM 131100) is one of the first described autosomal dominantly inherited complex endocrine syndromes (7). MEN1 has a high penetrance, over 90% by age 40, and is present in about three per 100 000 individuals. Gene carriers most frequently develop tumors in the parathyroid glands (95%), the anterior pituitary (20–40%), and in the endocrine cells of the pancreas/duodenum (40–80%) (8). Any of these can be the presenting lesion, and typically clinically detectable in the young adult, although rare cases of childhood tumors are reported. By definition MEN1 is present if two of these classical target tissues are affected by tumors, and familial MEN1 includes at least one relative with a corresponding tumor. Apart from the above-mentioned organs, MEN1 carriers frequently develop NETs of the foregut, e.g. bronchial, thymic, and gastric ECLomas. Adrenocortical lesions, mostly non-secreting, hyperplasias or adenomas but also rare cases of adrenocortical carcinomas are also present in MEN1 patients. The most lethal of the MEN1 lesions are the pancreatic and the thymic NETs which frequently develop into metastatic disease. Several non-endocrine tumors are also overrepresented in MEN1 patients, such as facial angiofibromas, collagenomas, lipomas and meningeomas. The typical clinical picture of the disease is highly variable between family members and dependent of the affected organs and the pattern of hypersecreted hormones in each case.
In 1988 the MEN1 gene was linked to chromosome 11q13 and suggested to be a suppressor (9). The gene was finally identified by positional cloning, and heterozygous germline inactivating MEN1 mutations was revealed in 70% of typical index cases (10). The proportion of MEN1 patients with a causative MEN1 mutation was later estimated to 75–95% (11). By now more than 1000 mutations have been recognized. The vast majority causes truncation of the protein. The most likely cases to reveal a germline MEN1 mutation are probands developing lesions at young age and show multiple tumors per organ. There is no convincing genotype–phenotype correlation identified so far. Furthermore if a mutation cannot be found it does not exclude MEN1 and recently alternative genes giving rise to syndromes resembling MEN1 have been suggested, e.g. germline CDKN1B (p27kip) mutations resulting in MEN4 (12). It have also been postulated that common genetic variants within CDKN1B could act as disease modifiers in MEN1, possibly contributing to the observed clinical heterogeneity (13).
Classical MEN1 tumorigenesis depends on a second MEN1 hit, by means of a somatic mutation eliminating also the WT allele. But some NETs in MEN1, such as thymic and duodenal NETs, do not to necessary require a complete inactivation of the gene (14). Somatic homozygous inactivation of MEN1 is also frequently seen in sporadic tumors of the MEN1 target organs.
The MEN1 protein, named menin, is ubiquitously expressed and preferentially located in the nucleus (10). It has been suggested to be a scaffold protein with more than 40 interacting proteins and thus are involved in a large number of biological functions, such as chromatin modification, DNA repair, transcription, cell division, protein degradation, motility and adhesion (15). Several Men1 knock-out mouse models mimicking the human syndrome are available.
Multiple endocrine neoplasias type 2 and familial medullary thyroid cancer
Multiple endocrine neoplasias type 2 (MEN2) and familial medullary thyroid carcinoma (FMTC) are autosomal dominantly inherited disorders characterized by development of multiple endocrine MTC, pheochromocytoma (PCC) and parathyroid adenomas (reviewed in Wells et al. (16)). MEN2 subtype A (OMIM 171400) accounts for about 80% of cases with an almost complete penetrance for MTC, while PCC and parathyroid adenoma are seen in about 50 and 30% of patients (17). MEN2B (Omim 162300) includes more aggressive MTC, dysmorphic marfanoid features but no parathyroid adenoma (18).
MEN2 and FMTC are caused by gain of function mutations in the rearranged during transfection (RET) protooncogene, localized on chromosome 10, which encodes a receptor tyrosine kinase (19, 20, 21). Gain of function mutations in RET result in autonomous activation that transduces activating signals through the RAS/MAPK and PI3K/AKT pathways (22). A majority of MEN2 and FMTC cases reveal constitutional mutations in the cysteine-rich extracellular domains (exons 10–11) of the RET gene (23), while disease causing variants within RET non-cysteine regions (exons 13–16) are less common and the related phenotype is characterized by pronounced heterogeneity (24). Patients with constitutional RET mutations show strong genotype–phenotype correlations, e.g. codon 634 transitions linked to MEN2A and codon 918 mutations linked to MEN2B. It is critical to recognize the specific RET mutation in order to optimize patient and family management, particularly for the timing of prophylactic thyroidectomy and to provide information on the risk of developing PCC and parathyroid adenoma (25, 26).
Multiple endocrine neoplasia type 4
Multiple endocrine neoplasia type 4 (MEN4; OMIM 610755) is a rare syndrome that is thought to predispose development of NETs, mainly parathyroid and pituitary adenomas. The trait is inherited through germline mutations in CDKN1B encoding p27kip (12). MEN4 seems to occur in an autosomal dominant fashion and is linked to loss of function mutations in the cell cycle regulator CDKN1B (12). Future studies have to be performed to confirm the phenotype and penetrance of CDKN1B mutations.
Neurofibromatosis type 1
Neurofibromatosis type 1 (NF1; OMIM 162200) is an autosomal dominant syndrome that is characterized by multiple endocrinopathies and nervous system manifestations (28). The most common features are fibromatous skin tumors, lichen eye nodules, optic gliomas and café-au-lait spots (29). Endocrinopathies are less common and include PCCs and duodenal NETs (30). Neurofibromatosis type 1 is caused by loss of function mutations in NF1 that has been linked to deregulation of both rat sarcoma viral oncogene homolog (RAS) proteins and the ERK/MAPK signaling pathway (31, 32).
von Hippel–Lindau syndrome
The von Hippel–Lindau syndrome (VHL; OMIM 193300) syndrome has an incidence of ∼1/36 000 individuals (33) and is an autosomal dominant disease. VHL is characterized by increased risk of tumors and cysts; retinal and CNS hemangioblastomas, PCC, paragangliomas (PGLs), renal clear cell carcinomas, renal cysts, pancreatic NETs, pancreatic cysts and endolymphatic sac tumors.
The syndrome is caused by inactivating mutations in the VHL tumor suppressor gene, located in 3p25 that is involved in the oxygen-sensing pathway through regulation of hypoxia-inducible factors (34). Truncation of VHL results in decreased ubiquitination of HIF transcription factors resulting in increased expression of target genes such as VEGFA etc. VHL is subclassified into type 1 (truncating mutations) and type 2 (missense mutations) that differ in clinical presentation (35). Subtype 1 presents without PCC whereas type 2 have a high penetrance of PCC. Subtype 2 may be further subclassified into type 2A without renal cell carcinoma or pancreatic cysts that may be present in subtype 2B.
PCC and PGL syndromes
There are at least 12 genes that have proved to confer susceptibility to PCCs and/or PGLs: SDHA (36), SDHB (37), SDHC (38), SDHD (39), SDHAF2 (40), FH (41), VHL (42), EPAS1 (43), NF1 (32), RET (20), TMEM127 (44) and MAX (45). The classic genetic syndromes MEN2, NF1 and VHL have been presented above. In the following section we will give an overview to more recently described diagnoses that show pronounced heterogeneity both in term of the affected organs, disease penetrance and mode of inheritance.
Familial PGL syndromes types 1–5 are transmitted in an autosomal dominant manner. These syndromes are caused by the loss of function mutations in SDHD (PGL 1), SDHAF2 (PGL 2), SDHC (PGL 3), SDHB (PGL 4) and SDHA (PGL 5). SDHx genes (SDHA, SDHB, SDHC and SDHD) encode succinate dehydrogenase subunits that are catalyzing reactions in tricarboxylic acid cycle and in the respiratory electron transfer chain. Due to the disruption of the tricarboxylic acid cycle the associated PGL harbor unique metabolic profiles with accumulation of oncometabolites (46). Although familial PGL syndromes share pathogenic mechanisms, with disruption of the succinate dehydrogenase complex, the phenotype varies between the different subtypes (47). The molecular rationale for this heterogeneity remains to be identified. Paternal transmission has been shown to occur in PGL 1 and 2.
Familial PGL type 1 (OMIM 168000) have an almost complete penetrance for parasympathetic tumors in the head–neck region (48, 49). Furthermore, unilateral PCC and/or sympathetic PGLs are seen in about 25% of patients respectively (50, 51).
Familial PGL type 4 (OMIM 115310) is associated with significant morbidity and increased mortality due to the substantial risk of development of malignant sympathetic PGLs (51, 58, 59). PCC and parasympathetic PGL also occur. In addition, patients with PGL 4 have an increased risk of developing gastrointestinal stromal tumors (GIST) as well as renal cell carcinoma (50).
Familial PCC and PGL syndromes
Patients with familial PCC and PGL of the TMEM127 subtype (OMIM 613403) have an estimated penetrance of PCC of 30% whereas abdominal PGL are less frequent (44, 62, 63, 64). No other phenotypes have been described. This disorder presents in patients with loss of function mutations in the Transmembrane protein 127 gene that is linked to dysinhibition of the mammalian target of rapamycin (mTOR) pathway.
Myc-associated factor X (MAX) associated familial PCC and PGL (OMIM 154950) is an autosomal dominant disorder with a suggested paternal mode of transmission (45, 65). These patients show susceptibility to PCCs and PGLs with unknown penetrance. No other phenotypes have been described. This syndrome is caused by loss of function mutations in MAX that results in deregulation the MYC–MAX–MXD1 pathway (45).
Fumarate hydratase (FH) associated PCC and PGL (OMIM 136850) is an autosomal dominant syndrome previously known to result in susceptibility to leimyomatosis and renal call cancer (OMIM 150800 (66)). PCC and PGL was recently recognized as a feature of this syndrome and the penetrance is currently unknown (41, 67, 68). The syndrome is caused by function mutations in FH that results in reduced enzymatic activity with fumarate accumulation.
A recent study described germline mutations in MDH2 as a cause of PGL (69). Although this finding needs to be confirmed, it is indeed remarkable that MDH2 is the third Krebs cycle gene suggested to be involved in PGL tumorigenesis (69).
Polycytemia and PGL syndrome
Disruption of oxygen sensing is also a recurrent disease mechanism in NETs. The recently recognized polycytemia and PGL syndrome was shown to occur as a result gain of function mutations in EPAS1 that resulted in a pseudohypoxic state through reduction in HIF2α degradation mediated by disturbed VHL binding (43, 70, 71, 72). The mode of inheritance for the polycytemia and PGL syndrome (OMIM 603349) is currently unknown due to near exclusive presentation in mosaicism. These patients are prone to develop PGL, PCC, polycytemia and somatostatinoma (72). A recent study also suggested ocular manifestations in patients with polycytemia and PGL syndrome (73). A majority of patients with mosaic mutations have had female gender and the underlying mechanism for this phenomenon is unknown (72).
Tuberous sclerosis complex
Tuberous sclerosis complex subtypes 1 and 2 (OMIM 191100 and 613254) are autosomal dominant syndromes that occur in 1:6000–10 000 individuals (74, 75, 76). These patients show multiorgan manifestations; mainly hamartomas but also angiomyolipomas, renal cell carcinoma and pulmonary lymphangioleiomyomatosis (reviewed by Curatolo et al. (77)). Pancreatic NET is a less frequent manifestation (78). These syndromes are due to the loss of function mutations in TSC1 and TSC2 that encode proteins forming the tuberin–hamartin complex that is essential for mTOR signaling (79). Somatic mutations in TSC2 occur frequently in pancreatic NETs (80).
Hyperparathyroidism-jaw tumor syndrome
Hyperparathyroidism-jaw tumor syndrome (OMIM 145001) is an autosomal dominant syndrome caused by truncating mutations in the HRPT2 located on 1q31.2 (81). It is characterized by adenomatous or malignant lesions of the parathyroids, jaw tumors, as well as renal and uterine tumors (82, 83).
Genetic landscape of NETs
Compared with other tumor types the genetic landscape of NETs is characterized by relatively few mutations and chromosomal aberrations per tumor (Fig. 2). So far, around 24 genes have been associated with neuroendocrine tumorigenesis (Table 2). These genes seem to interconnect to important pathways involved in cell metabolism, chromatin modification and growth control (Fig. 3).
List of genes involved in neuroendocrine tumorigenesis. Overview of genes involved in NET tumorigenesis. ALT, alternative lengthening of telomeres.
|Gene||Gene function||Chronological classification||Organ specificity|
|ATRX||ALT||Late||Pancreas, adrenal medulla, paraganglia|
|CDKN1B||Cell cycle||Unknown||Pancreas, small intestine, parathyroid, anterior pituitary|
|EPAS1||Cell signaling||Early||Paraganglia, adrenal medulla, duodenum|
|H-, K-RAS||Cell signaling||Early||Thyroid C-cell, adrenal medulla|
|FH||Metabolism||Early||Adrenal medulla, paraganglia|
|KTMD2||Chromatin modification||Unknown||Adrenal medulla|
|MAX||Cell signaling||Early||Paraganglia, adrenal medulla|
|MEN1||Unknown||Early||Parathyroid, anterior pituitary, endocrine cells in pancreas, duodenum, gastrium, lung, thymus|
|NF1||Cell signaling||Early||Adrenal medulla, duodenum|
|RET||Cell signaling||Early||thyroid C-cell, adrenal medulla, parathyroid|
|SDHx||Metabolism||Early||Paraganglia, adrenal medulla|
|TERT promoter||Telomere maintenance||Late||Paraganglia|
|TMEM127||Cell signaling||Early||Paraganglia, adrenal medulla|
|TP53||Chromatin integrity, Cell signaling||Unknown||Endocrine pancreas, adrenal medulla|
Parathyroid adenoma and carcinoma
Constitutional mutations in CDC73 (81), MEN1 (84), RET (20) and possibly CDKN1B (12) and CASR (83) cause susceptibility to parathyroid adenoma. Two exome sequencing studies investigated the genomic landscape of parathyroid adenomas and discovered somatic MEN1 mutations in 35% of cases (85, 86). A third study investigated the exomes of parathyroid carcinomas identified recurrent mutations in the PRUNE2 gene (87). The median number of somatic amino acid substituting mutations per adenoma was eight (range 2–100) (86). Parathyroid carcinomas harbored more mutations (synonymous average 51, range 3–176) that showed an apolipoprotein B mRNA editing enzyme, catalytic polypeptide-like (APOBEC) like mutational signature (87). APOBEC associated genomic instability is a result of cytosine-to-uracil deamination that is catalyzed by APOBEC3A-H enzymes and have been shown to occur in multiple different types of carcinomas (88).
Gastric, duodenal and pancreatic NETs
Constitutional mutations in MEN1 (84), VHL (42), NF1 (32), EPAS1 (somatostatinoma) (43), as well as TSC1 and 2 (74, 75) cause susceptibility to pancreatic NETs. CDKN1B have also been suggested as a susceptibility locus, but limited numbers of cases have been presented (12). In sporadic pancreatic NETs ATRX, DAXX, MEN1, TP53, ATM and mTOR pathway related genes are commonly mutated in a somatic fashion (80, 89, 90, 91). Pancreatic NETs harboring somatic mutations in ATRX and DAXX have been associated with chromosomal instability and reduced survival (92). It has been suggested that ATRX and DAXX mutations are not involved in the initial phase of tumorigenesis. Instead these mutations are thought to be late events that result in malignant transformation of PNETs (92, 93). The proposed pathogenic mechanism of ATRX and DAXX deficiency is the activation of the alternative lengthening of telomeres phenotype (94, 95). This phenotype has been associated with genome rearrangements, defects in the G2/M checkpoint and altered double-strand break repair (95). It have also been suggested that ATRX/DAXX mutations lead to differential epigenetic deregulation (96). The mean number of amino acid substituting somatic mutations in non-functioning pancreatic NETs were 16 (80). The degree of chromosomal aberrations is more variable and ranges from no somatic copy number alterations to genome-wide aneuploidy (92, 97).
Insulin producing pancreatic NETs with manifest hypoglycemic symptoms, i.e. insulinomas, have been identified with recurrent somatic mutations in YY1. This gene encodes a nuclear transcription factor and the described recurrent mutation Thr372Arg has been shown to result in neomorphic effects and altered transcription (98, 99, 100). Insulinomas exhibit a low mutational burden and the average amino acid substituting mutations were four per tumor (98, 99, 100). High chromosomal instability has been recognized in a subset of insulinomas especially in those with malignant features (101).
Medullary thyroid carcinoma
Germline mutations in RET cause MEN 2 and FMTC; both syndromes showing high penetrance of MTC (20). MTC from MEN2 patients rarely show any other somatic driver mutations (103). Investigating the genomic landscape of sporadic MTCs revealed mutually exclusive oncogenic mutations in RET and RAS subtypes K and H in 75 and 15% respectively (104, 105, 106). Furthermore there were an average of 18 amino acid substituting somatic mutations per MTC (104). ALK gene fusions were recently described in two out of 98 investigated cases suggesting that ALK inhibition might further be evaluated for treatment of these rare tumors (107, 108).
PCC and PGLs
Mutations in up to 20 genes have been suggested with PCC and PGL tumorigenesis. A total of 12 genes is believed cause of genetic susceptibility to PCC and PGL and a further six loci have been suggested but remains to be validated. These include KIF1Bβ (109, 110), IDH1 (111), MDH2 (69), BAP1 (112) as well as EGLN subtypes 1 and 2 (113, 114). Somatic mutations in HRAS have been shown to cause these diseases (115, 116). Several genes have also been suggested to work as the disease modifiers of PCC and PGL. ATRX have been identified in a subset of malignant PCC and PGL having the alternative lengthening of telomeres phenotype (117). TERT promoter mutations have been identified to occur exclusively in succinate dehydrogenase deficient PGL (118, 119). A recent study also described the recurrent somatic mutations in KMT2D. Whether this gene acts as a disease driver or modifier remains to be identified (120). The genomic landscape of PCC and PGL is associated with relatively low degree of genomic instability having an average of 19 non-synonymous mutations per tumor (117, 121, 122). Clinical criteria have classified these tumors accordingly to location into PCCs (adrenal medulla) and PGLs (extra adrenal). Accumulating evidence show that it is possible to sub-classify PCC and PGL into at least three distinct molecular clusters (based on methylome, transcriptome and siRNAome) having unique mutational profiles as well as different clinical characteristics (41, 121, 123). Cluster 1a is enriched with noradrenaline producing PGLs that show mutations in genes involved in cell metabolism through the Kreb cycle (SDHX, FH and MDH2) (121). Cluster 1b has noradrenaline producing PCC and PGL with a pseudohypoxic phenotype (VHL and EPAS1) (43, 123). Cluster 2 has PCC and PGLs with a mixed adrenaline/noradrenaline phenotype having aberrantly activated signaling in the MAPK and PI3K/AKT signaling pathways (RET, NF1, TMEM127, MAX and HRAS) (45, 115, 124, 125) (Fig. 4)
NETs of the lung
This section describes the genomic findings of well differentiated NETs derived from the broncopulmonary system. These have classically been denoted typical and atypical pulmonary carcinoids based on morphological criteria (126). Pulmonary carcinoids occur in about 5% of MEN1 patients (127). One study examining genome-wide mutation status in 44 sporadic pulmonary carcinoids identified three genes with recurrent mutations: MEN1, PSIP1 and ARID1A (128). Network analysis of genes with somatic mutations extended the number of cases with relevant mutations, 40% had covalent histone modifiers and 22% subunits of the SWI/SNF complex (22%) (128). The occurrence of somatic MEN1 mutations in sporadic pulmonary carcinoids have previously been highlighted (129). There were the averages of 13 protein-altering mutations per sample (128). Differential expression of miRNAs in typical vs atypical lung carcinoids has also been reported (130).
NET of the thymus is a part of MEN1 syndrome with a penetrance of <10% (8). CGH analysis found recurrent copy number alterations with the most commonly disturbance identified being gain of 8q24 (MYC locus) (131). Mutational screening of tumor tissue has been performed on a per case-basis with the finding of a few somatic MEN1 mutations (132).
Small intestinal NETs
Small intestinal NETs (SI-NETs) have also been described to occur with familial aggregation in a small fraction of patients (133, 134). Despite substantial efforts no definitive genetic basis for this phenomenon has been described. A recent study used linkage analysis and exome sequencing that revealed IPMK germline mutations in a single family (135). Although experimental investigations seem to indicate a pathogenic effect of the IPMK mutation a validation effort that included 32 additional families did not reveal the presence of any additional mutations (135). Thus IPMK inactivation is not likely to be a significant cause of familial SI-NET. The genomic landscape of SI-NETs has been investigated with exome coverage in close to 100 tumor samples (136, 137). Recurrent mutations were only identified in the CDKN1B gene with a mutation prevalence of 9% of SI-NET patients (136, 137, 138). Instead, the most frequent genomic alteration is hemizygous deletions affecting chromosome 18q (133, 136, 137, 139). Recent data have shown that SI-NETs can be sub-classified based on tumor methylome into three distinct clusters (140, 141). Sub classification of SI-NETs based on gene expression may also provide relevant information (142).
Genetic heterogeneity and tumor evolution
As clinicians seek to select treatments based on genetic biomarkers, knowledge of spatial and temporal heterogeneity could be important and should be researched further. Indeed genetic heterogeneity within and between paired tumors has been described in most NET types. Consistent with the theoretical assumption that MEN1 patients experience parallel development of independent NET clones, studies using loss of heterozygosity (LOH) LOH markers were able to show that paired tumors from MEN1 patients have unique genetic compositions (143, 144). In contrast, X-chromosome inactivation studies of sporadic parathyroid adenomas, gastrinomas, gastric-NETs and MTCs indicate that these tumors develop in a monoclonal fashion (145, 146, 147, 148). Multiple intestinal tumor lesions have been observed in about 20% of SI-NET patients (149). Whether these are the consequences of independent tumor formation or metastatic spread remain to be settled (144, 145). Recent studies highlighted that a subset of SI-NETs show pronounced genetic heterogeneity within and between tumor lesions (136, 138). Korpershoek et al. (150) determined, by using LOH markers, that there is genetic heterogeneity within different areas of selected PCCs and PGLs. Furthermore they recognized that adrenal medullary hyperplasia in MEN2 patients is in fact a precursor lesion for PPGL (151). Widespread genetic heterogeneity within these tumors as well as and between paired lesions has been recently shown (152). Of particular notice was the observed differences between paired primary and metastatic tumor lesions as well as indication of parallel evolution of different metastatic clones (152).
Authors perspectives and future implications
Precision medicine, where the unique properties of patients and tumors are used to tailor diagnostic and therapeutic procedures is now the standard of care of selected cancer types. For instance in carcinomas of the lung and gastrointestinal system knowledge of tumor mutation status is used to predict the response of treatment and detect the emergence of resistance during treatment with kinase signaling inhibitors (153, 154). The concept of precision medicine has also been useful in the management of patients with NETs, mainly for identification of those with genetic syndromes, enabling genetic counseling followed by appropriate diagnostic measures. However, knowledge of tumor mutation status is not yet used on a routine basis in NETs. Ongoing research is currently investigating whether specific RET mutations could predict the success of treatment with receptor kinase inhibitors (NCT01945762) in MTC. Another hypothesis that needs to be tested is that mutations in genes involved in mTOR signaling could be used as biomarkers of rapalogue response (155, 156).
Recent times also saw the publication of experimental findings in NETs that could be important for the development of future therapies: both combination therapy with FAK and mTOR inhibitors (157) as well as inhibition of B-catenin in MEN1 deficient pancreatic NETs (158). Immunotherapy is also an emerging branch in cancer treatment where the knowledge of tumor genetics and biology has proved to be important for prediction. Through determination of the degree of genomic instability and the neo-antigen immunoreactivity the success of PD-1 and CTLA-4 blockade may be estimated (159, 160). On the experimental level mapping the unique somatic mutations present in a particular tumor can also be used to design personalized tumor vaccination protocols (161). Furthermore, genetic instability could be targeted in ATRX or DAXX mutated tumors as they harbor the unique alternative lengthening of telomeres phenotype. Recent experimental data suggest that such tumors are sensitive to ATR inhibition (162).
However hypotheses based on extrapolation of data from basic NET research needs to be thoroughly tested in a clinical setting. There is an on-going debate whether currently available NET models are representative for human NET disease, supported by the discrepancies of the genomic landscape between NETs and commonly used cell lines (163, 164). Interpretation of basic science data must therefore be made with caution.
In order to screen for genetic disturbances in the established NET genes in a cost effective manner, reliable protocols for novel deep sequencing techniques should be implemented (165, 166). These techniques could also facilitate improvement of diagnostic yield of patients with suspicion of heritable disease. It has also been proposed that deep sequencing can improve the sensitivity in the detection of somatic mutations, especially in scenarios with a low tumor cell fraction (167). Current strategies to analyze tumor genomes are limited by the need of sufficient amount of good quality tumor tissue. In the future peripheral blood samples (liquid biopsies) containing either circulating tumor cells or circulating cell free DNA might instead be used for recognition of the genomic landscape of the individual tumors (168, 169).
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the review.
Our research is supported by Lions Cancerforskningsfond Uppsala and Swedish Cancer Foundation. Joakim Crona holds a research position funded by Akademiska Sjukhuset.
We thank Professor Peter Stålberg for his valuable suggestions and comments.
YaoJCHassanMPhanADagohoyCLearyCMaresJEAbdallaEKFlemingJBVautheyJNRashidA. One hundred years after "carcinoid": epidemiology of and prognostic factors for neuroendocrine tumors in 35,825 cases in the United States. Journal of Clinical Oncology2008263063–3072. (doi:10.1200/JCO.2007.15.4377).
LonguiniVCLourencoDMJrSekiyaTMeirellesOGoncalvesTDCoutinhoFLFranciscoGOsakiLHChammasRAlvesVA. Association between the p27 rs2066827 variant and tumor multiplicity in patients harboring MEN1 germline mutations. European Journal of Endocrinology/European Federation of Endocrine Societies2014171335–342. (doi:10.1530/EJE-14-0130).
ChenHSippelRSO'DorisioMSVinikAILloydRVPacakK. The North American Neuroendocrine Tumor Society consensus guideline for the diagnosis and management of neuroendocrine tumors: pheochromocytoma, paraganglioma, and medullary thyroid cancer. Pancreas201039775–783. (doi:10.1097/MPA.0b013e3181ebb4f0).
Crowe F Schull W Neel J. A Clinical Pathologicala and Genetic Study of Multiple Neurofibromatosis. Springfield IL: Charles C Thomas 1956
AstutiDLatifFDallolADahiaPLDouglasFGeorgeESkoldbergFHusebyeESEngCMaherER. Gene mutations in the succinate dehydrogenase subunit SDHB cause susceptibility to familial pheochromocytoma and to familial paraganglioma. American Journal of Human Genetics20016949–54. (doi:10.1086/321282).
RichterSPeitzschMRapizziELendersJWQinNde CubasAASchiaviFRaoJUBeuschleinFQuinklerM. Krebs cycle metabolite profiling for identification and stratification of pheochromocytomas/paragangliomas due to succinate dehydrogenase deficiency. Journal of Clinical Endocrinology and Metabolism2014993903–3911. (doi:10.1210/jc.2014-2151).
BurnichonNRohmerVAmarLHermanPLeboulleuxSDarrouzetVNiccoliPGaillardDChabrierGChabolleF. The succinate dehydrogenase genetic testing in a large prospective series of patients with paragangliomas. Journal of Clinical Endocrinology and Metabolism2009942817–2827. (doi:10.1210/jc.2008-2504).
NeumannHPPawluCPeczkowskaMBauschBMcWhinneySRMuresanMBuchtaMFrankeGKlischJBleyTA. Distinct clinical features of paraganglioma syndromes associated with SDHB and SDHD gene mutations. Journal of the American Medical Association2004292943–951. (doi:10.1001/jama.292.8.943).
KorpershoekEFavierJGaalJBurnichonNvan GesselBOudijkLBadoualCGadessaudNVenisseABayleyJP. SDHA immunohistochemistry detects germline SDHA gene mutations in apparently sporadic paragangliomas and pheochromocytomas. Journal of Clinical Endocrinology and Metabolism201196E1472–E1476. (doi:10.1210/jc.2011-1043).
PantaleoMAAstolfiAIndioVMooreRThiessenNHeinrichMCGnocchiCSantiniDCatenaFFormicaS. SDHA loss-of-function mutations in KIT-PDGFRA wild-type gastrointestinal stromal tumors identified by massively parallel sequencing. Journal of the National Cancer Institute2011103983–987. (doi:10.1093/jnci/djr130).
YaoLSchiaviFCasconAQinYInglada-PerezLKingEEToledoRAErcolinoTRapizziERickettsCJ. Spectrum and prevalence of FP/TMEM127 gene mutations in pheochromocytomas and paragangliomas. Journal of the American Medical Association20103042611–2619. (doi:10.1001/jama.2010.1830).
AbermilNGuillaud-BatailleMBurnichonNVenisseAManivetPGuignatLDruiDChupinMJosseaumeCAffresH. TMEM127 screening in a large cohort of patients with pheochromocytoma and/or paraganglioma. Journal of Clinical Endocrinology and Metabolism201297E805–E809. (doi:10.1210/jc.2011-3360).
ToledoSPLourencoDMJrSekiyaTLuconAMBaenaMECastroCCBortolottoLAZerbiniMCSiqueiraSAToledoRA. Penetrance and clinical features of pheochromocytoma in a six-generation family carrying a germline TMEM127 mutation. Journal of Clinical Endocrinology and Metabolism2015100E308–E318. (doi:10.1210/jc.2014-2473).
SmitDLMensenkampARBadeloeSBreuningMHSimonMEvan SpaendonckKYAalfsCMPostJGShanleySKrapelsIP. Hereditary leiomyomatosis and renal cell cancer in families referred for fumarate hydratase germline mutation analysis. Clinical Genetics20117949–59. (doi:10.1111/j.1399-0004.2010.01486.x).
CasconAComino-MendezICurras-FreixesMde CubasAAContrerasLRichterSPeitzschMMancikovaVInglada-PerezLPerez-BarriosA. Whole-exome sequencing identifies MDH2 as a new familial paraganglioma gene. Journal of the National Cancer Institute2015107Article ID: djv053doi:10.1093/jnci/djv053).
Comino-MendezIde CubasAABernalCAlvarez-EscolaCSanchez-MaloCRamirez-TortosaCLPedrinaciSRapizziEErcolinoTBerniniG. Tumoral EPAS1 (HIF2A) mutations explain sporadic pheochromocytoma and paraganglioma in the absence of erythrocytosis. Human Molecular Genetics2013222169–2176. (doi:10.1093/hmg/ddt069).
PollakMRBrownEMChouYHHebertSCMarxSJSteinmannBLeviTSeidmanCESeidmanJG. Mutations in the human Ca(2+)-sensing receptor gene cause familial hypocalciuric hypercalcemia and neonatal severe hyperparathyroidism. Cell1993751297–1303. (doi:10.1016/0092-8674(93)90617-Y).
YuWMcPhersonJRStevensonMvanEijkRHengHLNeweyPGanARuanoDHuangDPoonSL. Whole-exome sequencing studies of parathyroid carcinomas reveal novel PRUNE2 mutations, distinctive mutational spectra related to APOBEC-catalyzed DNA mutagenesis and mutational enrichment in kinases associated with cell migration and invasion. Journal of Clinical Endocrinology and Metabolism2015100E360–E364. (doi:10.1210/jc.2014-3238).
PerrenAKomminothPSaremaslaniPMatterCFeurerSLeesJAHeitzPUEngC. Mutation and expression analyses reveal differential subcellular compartmentalization of PTEN in endocrine pancreatic tumors compared to normal islet cells. American Journal of Pathology20001571097–1103. (doi:10.1016/S0002-9440(10)64624-X).
SadanandamAWullschlegerSLyssiotisCAGrotzingerCBarbiSBersaniSKornerJWafyIMafficiniALawlorRT. A cross-species analysis in pancreatic neuroendocrine tumors reveals molecular subtypes with distinctive clinical, metastatic, developmental, and metabolic characteristics. Cancer Discovery201551296–1313. (doi:10.1158/2159-8290.CD-15-0068).
MarinoniIKurrerASVassellaEDettmerMRudolphTBanzVHungerFPasquinelliSSpeelEJPerrenA. Loss of DAXX and ATRX are associated with chromosome instability and reduced survival of patients with pancreatic neuroendocrine tumors. Gastroenterology2014146453–460.e455. (doi:10.1053/j.gastro.2013.10.020).
de WildeRFHeaphyCMMaitraAMeekerAKEdilBHWolfgangCLEllisonTASchulickRDMolenaarIQValkGD. 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 Pathology2012251033–1039. (doi:10.1038/modpathol.2012.53).
LovejoyCALiWReisenweberSThongthipSBrunoJde LangeTDeSPetriniJHSungPAJasinM. Loss of ATRX, genome instability, and an altered DNA damage response are hallmarks of the alternative lengthening of telomeres pathway. PLoS Genetics20128e1002772. (doi:10.1371/journal.pgen.1002772).
LichtenauerUDDi DalmaziGSlaterEPWielandTKuebartASchmittfullASchwarzmayrTDienerSWieseDThaslerWE. Frequency and clinical correlates of somatic Ying Yang 1 mutations in sporadic insulinomas. Journal of Clinical Endocrinology and Metabolism2015100E776–E782. (doi:10.1210/jc.2015-1100).
JonkersYMClaessenSMPerrenASchmittAMHoflandLJde HerderWde KrijgerRRVerhofstadAAHermusARKummerJA. DNA copy number status is a powerful predictor of poor survival in endocrine pancreatic tumor patients. Endocrine-Related Cancer200714769–779. (doi:10.1677/ERC-07-0111).
DebelenkoLVEmmert-BuckMRZhuangZEpshteynEMoskalukCAJensenRTLiottaLALubenskyIA. The multiple endocrine neoplasia type I gene locus is involved in the pathogenesis of type II gastric carcinoids. Gastroenterology1997113773–781. (doi:10.1016/S0016-5085(97)70171-9).
AgrawalNJiaoYSausenMLearyRBettegowdaCRobertsNJBhanSHoASKhanZBishopJ. Exomic sequencing of medullary thyroid cancer reveals dominant and mutually exclusive oncogenic mutations in RET and RAS. Journal of Clinical Endocrinology and Metabolism201398E364–E369. (doi:10.1210/jc.2012-2703).
WelanderJAndreassonAJuhlinCCWisemanRWBackdahlMHoogALarssonCGimmOSoderkvistP. Rare germline mutations identified by targeted next-generation sequencing of susceptibility genes in pheochromocytoma and paraganglioma. Journal of Clinical Endocrinology and Metabolism201499E1352–E1360. (doi:10.1210/jc.2013-4375).
GaalJBurnichonNKorpershoekERoncelinIBertheratJPlouinPFde KrijgerRRGimenez-RoqueploAPDinjensWN. Isocitrate dehydrogenase mutations are rare in pheochromocytomas and paragangliomas. Journal of Clinical Endocrinology and Metabolism2010951274–1278. (doi:10.1210/jc.2009-2170).
YangCZhuangZFliednerSMShankavaramUSunMGBullovaPZhuRElkahlounAGKourlasPJMerinoM. Germ-line PHD1 and PHD2 mutations detected in patients with pheochromocytoma/paraganglioma-polycythemia. Journal of Molecular Medicine20159393–104. (doi:10.1007/s00109-014-1205-7).
CronaJDelgado VerdugoAMaharjanRStalbergPGranbergDHellmanPBjorklundP. Somatic mutations in H-RAS in sporadic pheochromocytoma and paraganglioma identified by exome sequencing. Journal of Clinical Endocrinology and Metabolism201398E1266–E1271. (doi:10.1210/jc.2012-4257).
OudijkLde KrijgerRRRapaIBeuschleinFde CubasAADei TosAPDinjensWNKorpershoekEMancikovaVMannelliM. H-RAS mutations are restricted to sporadic pheochromocytomas lacking specific clinical or pathological features: data from a multi-institutional series. Journal of Clinical Endocrinology and Metabolism201499E1376–E1380. (doi:10.1210/jc.2013-3879).
JuhlinCCStenmanAHaglundFClarkVEBrownTCBaranoskiJBilguvarKGohGWelanderJSvahnF. Whole-exome sequencing defines the mutational landscape of pheochromocytoma and identifies KMT2D as a recurrently mutated gene. Genes Chromosomes & Cancer201554542–554. (doi:10.1002/gcc.22267).
Travis WD Müller-Hermelink HK Harris CC Brambilla E. Pathology and Genetics. In WHO Classification of Tumours. Lyon: IARC Press 2004
CronaJGustavssonTNorlenOEdfeldtKAkerstromTWestinGHellmanPBjorklundPStalbergP. Somatic mutations and genetic heterogeneity at the CDKN1B locus in small intestinal neuroendocrine tumors. Annals of Surgical Oncology201522 (Suppl 3) 1428–1435. (doi:10.1245/s10434-014-4351-9).
CronaJBackmanSMaharjanRMayrhoferMStalbergPIsaksonAHellmanPBjorklundP. Spatio-temporal heterogeneity characterizes the genetic landscape of pheochromocytoma and defines early events in tumourigenesis. Clinical Cancer Research2015214451–4460. (doi:10.1158/1078-0432.CCR-14-2854).
WagleNGrabinerBCVan AllenEMHodisEJacobusSSupkoJGStewartMChoueiriTKGandhiLClearyJM. Activating mTOR mutations in a patient with an extraordinary response on a phase I trial of everolimus and pazopanib. Cancer Discovery20144546–553. (doi:10.1158/2159-8290.CD-13-0353).
KhanMSKirkwoodAATsiganiTLoweHGoldsteinRHartleyJACaplinMEMeyerT. Early changes in circulating tumor cells are associated with response and survival following treatment of metastatic neuroendocrine neoplasms. Clinical Cancer Research20162279–85. (doi:10.1158/1078-0432.CCR-15-1008).