Abstract Regorafenib is a small molecule multi-kinase inhibitor approved for the treatment of patients with hepatocellular carcinoma (HCC) who have previously received sorafenib based on an improvement in overall survival versus placebo in the phase 3 randomized, double-blind RESORCE trial (NCT01774344; Bruix et al., Lancet, 2017). Expression of 776 plasma miRNAs was quantified by qPCR in baseline plasma samples collected from 343 of 573 patients (60%) from RESORCE (Teufel et al., ESMO2017, abs 705P). Predictive and prognostic effects of miRNAs were evaluated using a Cox proportional hazards model. Nine miRNAs showed significant predictive association with overall survival. A survival benefit in favor of regorafenib was associated with increased plasma levels of miR-30a, miR-122, miR-125b, miR-200a and miR-374b, and decreased plasma levels of miR-15b, miR-107, miR-320 and miR-645. Here, we focus on the biological interpretation of these statistical results. Expression data from The Cancer Genome Atlas was used to bridge from miRNAs via mRNAs to biological function. Gene set enrichment analysis of mRNAs based on correlation with expression of predictive miRNAs identified a list of the top 142 gene sets concordantly found for at least 6 of the 9 predictive miRNAs at a FDR of 0.01. Top gene sets showed consistent patterns related to HCC subtypes and several processes including liver cancer progression, metabolism of lipids, amino acids, bile acids and xenobiotics, glucuronidation, and doxorubicin resistance. Results suggested that response to regorafenib was improved in G4/G5/G6 versus G1/G2/G3 subtypes of the Boyault classification and in the CTNNB1 and Polysom7 subtypes of the Chiang classification which both coincide with the S3 subtype of the Hoshida classification (Hoshida et al., Seminars in liver disease, 30:35, 2010), a tumor subclass that is generally characterized by a retained hepatocyte-like phenotype, with well-differentiated and smaller tumors, and better survival. This is consistent with known roles of the predictive miRNAs in epithelial-mesenchymal transition, fibrogenesis, glycolysis, proliferation, and enhancement of apoptosis via several mechanisms. For example, miR-122 targets IGF1R, PDK4, LDHA and GALNT10, thereby regulating RAS/RAF/ERK signaling and glycolysis, which has been reported to overcome resistance to sorafenib. In summary, our results suggest that response to regorafenib is favorable in tumors with properties that are consistent with the S3 subtype of the Hoshida classification. Conclusions from this hypothesis-generating investigation require confirmation in follow-up studies with molecular characterization of tumor biopsies and regorafenib response data. Citation Format: Henrik Seidel, Karl Köchert, Michael Teufel. Biological interpretation of circulating miRNA biomarkers predicting regorafenib clinical benefit in patients with hepatocellular carcinoma (HCC) in the RESORCE trial [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 2606.
Read full abstract