Molecular characteristics of pancreatic neuroendocrine tumors, do they correlate with metastases?

Document Type

Conference Proceeding

Publication Date


Publication Title

Modern Pathology


Background: Pancreatic neuroendocrine tumors (PNET) have an unpredictable biological behavior that cannot be reliably predicted by histological and clinical manifestations. We performed next generation sequencing (NGS) on PNET to understand the molecular pathogenesis and to identify potential biomarkers correlating with metastases and survival. Design: Hybrid capture-based comprehensive genomic profiling was performed on both primary and metastatic PNET from 28 patients. Metastatic sites were grouped as lymph nodes and distant metastases (consisted of liver followed by bone). NGS was performed on genomic DNA & RNA isolated from formalin-fixed paraffin embedded tissue using the whole exome NGS assay covering 20,000 genes. Average number of DNA reads was 52 million at 3:1 somatic to germline. All variants were detected with >99% confidence based on allele frequency and amplicon coverage, with an average sequencing depth of coverage >500x and an analytical sensitivity of 5%. Tumor mutation burden and MSI status were compared using the two-sample t-test between tumor types. Frequency of mutations in each gene was compared between tumor types using the Fisher's exact test. To account for multiple comparisons, the False Discovery Rate was estimated using Benjamini-Hochberg procedure. R package "maftools" was used to summarize and visualize the findings. Results: Mutation burden was higher in primary tumors with metastases versus primary tumors without metastases (p=0.034) while no significance was found in MSI status between these two groups. Among the 20 mutated genes, MUC4 and MEN1 were the most frequently mutated genes in both primary and metastatic tumors (fig1&2). At a p-value threshold of 0.2, two distant metastatic tumors had mutation on PCNX3 gene, while none were found in lymph node metastases (p=0.057). In addition, more mutations on TGFBR1 in distant metastases compared to lymph node metastases (p=0.154). We also found that three primary tumors with metastases had mutations on E2F4, HRNR, LRP1B genes while none were found in primary tumors with no metastases (p=0.2). (Figure presented) Conclusions: Decoding the complexity and unpredictable nature of PNET has been perplexing and the subject of many several research projects. The current project, the first ever done on the subject, with an attempt to distinguish the metastatic vs non-metastatic tumors try to underlie the difference of molecular signature between the two. Further studies are needed to consolidate the findings and bring them to clinical practice.





First Page


Last Page


This document is currently not available here.