Genetics and Epigenetics of Gastroenteropancreatic Neuroendocrine Neoplasms

Abstract

Gastroenteropancreatic (GEP) neuroendocrine neoplasms (NENs) are heterogeneous regarding site of origin, biological behavior, and malignant potential. There has been a rapid increase in data publication during the last 10 years, mainly driven by high-throughput studies on pancreatic and small intestinal neuroendocrine tumors (NETs). This review summarizes the present knowledge on genetic and epigenetic alterations. We integrated the available information from each compartment to give a pathway-based overview. This provided a summary of the critical alterations sustaining neoplastic cells. It also highlighted similarities and differences across anatomical locations and points that need further investigation. GEP-NENs include well-differentiated NETs and poorly differentiated neuroendocrine carcinomas (NECs). NENs are graded as G1, G2, or G3 based on mitotic count and/or Ki-67 labeling index, NECs are G3 by definition. The distinction between NETs and NECs is also linked to their genetic background, as TP53 and RB1 inactivation in NECs set them apart from NETs. A large number of genetic and epigenetic alterations have been reported. Recurrent changes have been traced back to a reduced number of core pathways, including DNA damage repair, cell cycle regulation, and phosphatidylinositol 3-kinase/mammalian target of rapamycin signaling. In pancreatic tumors, chromatin remodeling/histone methylation and telomere alteration are also affected. However, also owing to the paucity of disease models, further research is necessary to fully integrate and functionalize data on deregulated pathways to recapitulate the large heterogeneity of behaviors displayed by these tumors. This is expected to impact diagnostics, prognostic stratification, and planning of personalized therapy.

Figure 1.

Anatomic location of GEP-NENs. Sites where large-scale genetic and epigenetic data are available are in bold; sites with preponderance of poorly differentiated carcinomas are marked with an asterisk.

Figure 2.

Different molecular genesis of NETs and NECs. Main histological and molecular characteristics of NECs are similar throughout the body, and their aggressiveness is comparable to that of adenocarcinomas, whereas NETs are more indolent.

Figure 3.

HR repair of DNA double-strand breaks and its deregulation in ALT. (a) In normal conditions, DNA double-strand breaks trigger the HR complex via ATM and RAD50. The complex repairs broken DNA strands using the complementary strand of the intact sister chromosome as a template. (b) This is normally inhibited on shortening telomeres by the DAXX/ATRX complex. In a fraction of NETs, loss of DAXX/ATRX function causes the HR complex to attempt “repair” of telomeres, resulting in ALT. Proteins whose gene has been reported as altered in neuroendocrine tumors are shaded in red.

Figure 4.

Subtypes of PanNETs and SiNETs according to chromosomal alterations. Four subgroups have been identified in PanNETs and three in SiNETs. Copy gains are shown in red, losses in blue. In each subgroup of tumors, cases with an identical CNV pattern are represented by individual rows, and the height of the row is proportional to the fraction of cases harboring that CNV pattern. chr, chromosome; multiCNV, multiple CNV; PP, polyploid; RPCG, recurrent pattern of chromosomal gains.

Figure 5.

Outline of the main altered pathways in PanNETs. Pathway members whose genetic alteration has been proven are shaded in red, and those inactivated by epigenetics are in green. Approved targeted drugs are shaded in orange. MEN1 interacts and modulates all core pathways acting as a hub gene. DAXX/ATRX also cooperate with the other genes of the chromatin remodeling complexes. DAXX fosters RASSF1A promoter methylation, but also retains RASSF1C in the nucleus, releasing it upon DNA damage; it also modulates PTEN distribution between the nucleus and the cytoplasm whereas PTEN modulates DAXX’s gene expression regulation. MGMT repairs of O⁶-methylguanine but, in case the repair fails, it triggers futile cycling of the mismatch repair complex that leads to cytotoxicity. Both the mTOR pathway and a set of dysregulated miRNAs trigger HIF1/2-dependent gene expression in a large subset of PanNETs. The role of chromatin remodeling genes is complex, still unclear, and under active investigation.

Figure 6.

Outline of the main altered pathways in SiNETs. Pathway members whose genetic alteration has been proven are shaded in red, and those inactivated by epigenetics are in green. Approved targeted drugs are shaded in orange. The mTOR pathway is detected to be altered mainly via gene amplifications (PIK3CD, AKT1, mTOR, SRC, MAP2K2, PDGFR2) and epigenetically by hypermethylation of a large set of genes [represented by the green shade, described in Karpathakis et al. (19, 21)]. RASSF1A downregulation by hypermethylation and SMAD4 loss may foster the Wnt pathway, which is however contrasted by CTNNB1 methylation and miR-196a hyperexpression, with the latter repressing transcription of Wnt pathway genes. The cell cycle is involved in CDKN1B-mutated tumors whereas UCHL1, a TP53 stabilizer, is lost in metastatic tumors by methylation. MUTYH, OGG1, and IPMK mutations involve familial cases, whereas the role of miRNA and chromatin deregulation is currently unclear.