Multiple endocrine neoplasia type 1 was first described as an autosomal dominant trait by Dr. Wermer in 1954 [8]. However, concomitant hyperplasia and/or neoplasia in the classic triad of the pancreas, anterior pituitary and parathyroid glands had been described in the medical literature prior to this [9]. In addition to pancreatic, pituitary, and parathyroid lesions, affected individuals may develop other tumors including thymic carcinoids, thyroid adenomas and adrenocortical tumors [10, 11]. In 1988, the MEN1 gene was localized by deletion mapping to the long arm of chromosome 11, and it was demonstrated that the unaffected allele is lost in pancreatic lesions from affected kindred [12]. In 1990, MEN1 was further determined to be located in a narrow region on chromosome 11q13 [13]. In 1997 the MEN1 gene was cloned [14]. Subsequent to this, numerous deleterious mutations (the majority leading to a truncated protein) have been found in different affected families [15]. The penetrance of the disease is nearly complete [16]. PNETs occur in about 40 % of patients with MEN1 [10]. A hallmark of MEN1-associated PNET is multiplicity; the pancreata of MEN1 patients often show multiple microadenomas, with one or a few larger NETs [17]. This can be contrasted with sporadic cases which present with a solitary lesion [7].

Von Hippel-Lindau disease (VHL) is a highly penetrant autosomal dominant trait with a prevalence of 1/36,000 live births [18]. Affected individuals may develop hemangioblastomas, pheochromocytomas and endolymphatic sack tumors, in addition to various forms of pancreatic cysts and pancreatic neuroendocrine tumors [18]. PNETs have been determined to develop in 12–17 % of patients affected by VHL [19, 20]. The tumors are most frequently non-functional [21]. The VHL gene was localized and cloned in 1993 [22]. The protein functions as part of an E3 ubiquitin ligase [23] that is required for the degradation of hypoxia-inducible factor (HIF) [24]. In the absence of the functional protein, a pseudo-hypoxic state occurs, which drives cellular survival and proliferation. Thus, it plays the role of a tumor suppressor. Like MEN1, VHL follows Knudson’s two-hit hypothesis, with the inherited wild-type allele being deleted in the tumors [6].

Tuberous sclerosis complex (TSC) is an autosomal dominant genetic syndrome which may involve development of multiple benign tumors in various organ systems as well as behavioural abnormalities, seizures and developmental delay [25]. The syndrome is caused by inactivating mutations in either TSC1 on chromosome 9 [26] or TSC2 on chromosome 16 [27]. The two genes encode the proteins hamartin and tuberin, respectively, which form a heterodimer that acts as a negative regulator of mTOR signalling [28]. The first report of a PNET in the context of tuberous sclerosis complex was published in 1959 [29]. Since then, a number of case reports describing PNETs in patients with TSC have been published. A recent review of the association between neuroendocrine tumors pinpointed PNETs as a likely rare feature of TSC [30]. The association is further corroborated by the presence of TSC2 mutations in a subset of sporadic PNETs [31].

Neurofibromatosis type 1 (also known as von Recklinghausen’s disease) is a common autosomal dominant disorder affecting 1 in 2500–3000 individuals [32]. Those affected have a characteristic appearance with a large numbers of café-au-lait spots and multiple cutaneous neurofibromas [33]. Other manifestations of the disease may include cognitive deficits [34] and several forms of malignant tumors including gastrointestinal stromal tumors [35], gliomas [36] and pheochromocytomas [37]. The NF1 gene locates to chromosome 17 and functions as a tumor suppressor [38]. Neurofibromin, the gene product, has GAP activity [39, 40] and functions as a negative regulator of the mitogenic Ras signalling pathways [41]. PNETs producing somatostatin or insulin have been described in patients with the NF1 syndrome [42]. Single case reports have described loss of NF1 immunoreactivity in tumor tissue from NF1 patients with pancreatic NETs, supporting a causal relation between the two entities [43, 44].

There are reports of PNETs occurring in other genetic syndromes than those discussed here. Since PNETs do occur sporadically, it is difficult to determine whether there is a true causal relationship or whether co-occurrence between two conditions arises by mere chance. Of interest is a recent report of a PNET in a patient with paraganglioma syndrome due to a germline mutation in SDHD. The wild-type allele was lost in the tumor tissue, and the tumor lacked immunoreactivity for SDHB, compatible with a causative role of the SDHD mutation [45, 46]. Inactivating mutations in SDHD lead to oncogenesis by causing accumulation of succinate in the cell. Succinate functions as an inhibitor of alfa-ketoglutarate-dependent enzymes that degrade hypoxia-inducible factor, thus causing a pseudo-hypoxic state [47, 48].

The MEN1 gene, which when mutated in constitutional DNA gives rise to multiple endocrine neoplasia type 1, has been found to be mutated in sporadic PNETs not associated with MEN1 [52]. Estimates of the prevalence of MEN1 mutations vary. A whole-exome sequencing study of non-functioning PNETs found somatic MEN1 mutations in 44 % of tumors [31]. The prevalence of MEN1 mutations is apparently lower in insulinomas than in other PNETs. One study found a mutation in one out of seven investigated tumors [50], while another found two mutated tumors among twelve investigated tumors [53], suggesting a MEN1 mutation prevalence of around 15 % in sporadic insulinomas.

mTOR, the mechanistic (formerly mammalian) target of rapamycin is a serine/threonine kinase which plays an integral role in coordinating several cellular processes, including protein synthesis, cell division, cellular metabolism and mRNA synthesis. Since the discovery of the mTOR protein, it has, together with its regulators, increasingly become recognized as a key player in several human diseases, including various cancers [54]. In cancers, inactivating mutations are frequently found in negative regulators of mTOR signalling, such as PTEN [55], TSC1 and TSC2. Activating mutations are found in the positive mTOR regulator PIK3CA [56] but seldom in the MTOR [57] gene itself. In 2011 a whole-exome sequencing study of ten non-functioning pancreatic neuroendocrine tumors (with a verification cohort of an additional 58 PNETs) found frequent mutations in several genes in the mTOR pathway: PTEN, PIK3CA and TSC2 [31]. In total these mutations appear to occur in 14 % of sporadic pancreatic neuroendocrine tumors. Consistent with the role of the mTOR pathway in PNETs, mTOR inhibitors such as everolimus have shown efficacy in treatment of pancreatic neuroendocrine tumors [58].

The same study that found mutations affecting the mTOR pathway in PNETs also identified mutually exclusive mutations in ATRX and DAXX in 42.6 % of the analysed tumors. Several of the mutations were frameshift insertions/deletions, and the affected tumors showed loss of immunoreactivity for the affected protein, indicating that they act as tumor suppressors. The proteins encoded by ATRX and DAXX interact and participate in chromatin remodelling [59]. Subsequent studies have coupled these mutations to an alternative lengthening of telomeres (ALT) phenotype [60, 61]. Similar mutations are also found in other tumors, including pheochromocytomas [62] and glioblastomas [63]. The prognostic value of ATRX/DAXX mutations in PNETs is unknown, with different studies reporting discrepant results [31, 60, 64].

Three separate studies have identified a hotspot mutation (p.T327R) in the YY1 gene in sporadic insulinomas [49–51]. YY1 codes for the Yin Yang 1 protein, which is a zinc finger transcription factor. The mutated Yin Yang protein has a different DNA-binding motif than the wild-type protein and causes alterations in gene expression [50]. ADCY1 and CACNA2D2 which are implicated in the regulation of insulin secretion and cell proliferation were dramatically overexpressed in YY1-mutated tumors compared to YY1-wild-type tumors and were found to neighbour binding sites for mutant YY1 [50]. Clinically, patients with YY1-mutated insulinomas present at a later age than those with YY1-wild-type tumors in two of the above-mentioned studies [49, 51]. While one study found an association between YY1 mutation and female gender [51], the other two studies report no such association [49, 50].