Abstract LB-061: Systematic, network-based, comparative analysis of differential mechanisms of action of lanreotide and octreotide in neuroendocrine tumor derived cell lines

Abstract

Abstract Lanreotide and octreotide are somatostatin receptor (SSTR) agonists thought to exhibit overlapping Mechanism of Action (MoA), despite small differences in their SSTRs affinity and differential PK response in vivo [1]. To systematically and quantitatively assess both concentration- and time-dependent MoA differences at the molecular level —by characterizing the full protein repertoire that mediates the compound's pharmacological effects— we assessed the protein-specific activity of lanreotide and octreotide in two gastroenteropancreatic neuroendocrine tumor (GEP-NET) derived cell lines (H-STS and KRJ-1). We performed RNASeq profiling at 4 different time points (24h -96h), following cell perturbation at 4 non-toxic doses (2 nM - 1 μM). RNASeq profiles were used to assess compound activity on 6,276 regulatory proteins, using VIPER, an experimentally-validated algorithm to infer protein activity based on the expression of its targets [2]. The analysis was performed using a GEP-NET-specific regulatory model inferred from a collection of 212 primary and metastatic human samples. Concentrations above the EC50 of the drugs were selected to (a) assess potential off-target effects and (b) discount potential effects from differential compound degradation. Confirming the robust nature of our approach, short-term (24h and 48h) transcriptional and protein activity profiles, following compound perturbation, were virtually identical across both cell lines and drug concentration (FET p-value < 10-12). However, the sensitivity and reproducibility of VIPER analysis, which represents the foundation for two NYS CLIA approved tests, allowed identification of 367 proteins affected differentially by these drugs (FDR < 0.05), which were enriched in cancer-relevant functional categories, including apoptosis, immune response, cell differentiation, cell proliferation and DNA-repair (FDR < 10-5). Most critically, while lanreotide-modulated proteins were virtually unchanged over time (from 24h to 96h, p < 0.05), the effect of octreotide on the same proteins was transient and virtually lost at 72h. This discrepancy cannot be attributed to concentration-related effects, since it was manifested identically across all tested concentrations. This work constitutes the first systematic, genome-wide characterization of the MoA of two clinically relevant SSTR agonists, showing both common as well as differentially affected pathways. Two key elements which may account for differential clinical response in patients emerged from the analysis: (1) Differences in the repertoire of affected proteins (MoA) by each compound, and (2) different response kinetics, with sustained lanreotide response from 24h to 96h vs. transient octreotide response with subsequent adaptive cell response after 48h. [1] Enzler et.al (2017) Sem. Oncol. 44:141.[2] Alvarez et.al (2016) Nat. Genet. 48:838. Citation Format: Mariano J. Alvarez, Gideon Bosker, Roswitha Pfragner, Mark Kidd, Irvin Modlin, Andrea Califano. Systematic, network-based, comparative analysis of differential mechanisms of action of lanreotide and octreotide in neuroendocrine tumor derived cell lines [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr LB-061.