Multiplatform profiling of pancreatic neuroendocrine tumors (PanNETs) identifies novel co-occurring pathogenic alterations and associations with clinicopathologic factors.

Abstract

211 Background: We correlated genomic, proteomic, and molecular pathway alterations with clinicopathologic factors and identified novel co-occurring pathogenic alterations of potential clinical relevance to PanNET management through multi-omic profiling. Methods: Perthera, Inc. deploys an IRB-approved registry that was utilized in partnership with PanCAN’s “Know Your Tumor” program. PanNETs having undergone molecular profiling for precision matched therapeutic purposes were screened. We performed correlative analyses by pairwise comparisons between pathogenic alterations or altered molecular pathways and clinicopathologic variables. Hierarchical clustering was used to visualize associations. The Kaplan-Meier method was used to estimate overall survival (OS) and survival differences across variables were analyzed by the log-rank test. Results: We included 33 patients with predominantly locally advanced and metastatic PanNETs from 12/2014-1/2018. Chromatin remodeling pathway and MEN1 alterations by next-generation sequencing (NGS) were less associated with having high-grade PanNETs, while MEN1 alterations were also less associated with metastatic disease at diagnosis (all Fisher’s exact two-tailed p ≤ 0.05). Several molecular pathway or pathogenic alterations correlated with worse OS: DNA repair pathway (log-rank p = 0.0022), RB1 alterations by NGS (p = 0.018), and TP53 alterations by NGS (p = 0.01). There were several significant co-occurring alterations (Fisher’s exact p ≤ 0.05): ERCC1 expression by immunohistochemistry (IHC) and DAXX (NGS), RB1 (NGS) and DNA repair pathway (NGS), and TS (IHC) and cyclin-dependent kinase pathway (NGS). Having an altered chromatin remodeling pathway was less associated with having an altered receptor tyrosine kinase (RTK) signaling pathway (Fisher’s exact p ≤ 0.05). Conclusions: We identified several molecular signatures of potential clinical significance for therapeutic targeting and prognostication in PanNETs warranting prospective validation. Our findings are hypothesis generating and can inform larger molecular profiling efforts in PanNETs.