BAY 1163877 is a potent and selective, oral, small molecule pan-FGFR inhibitor with anti-tumor activity in a wide range of cancer types. In vivo profiling in xenograft models identified tumor FGFR mRNA levels as a predictor of drug efficacy. Anti-tumor efficacy was largely independent of the tumor type surveyed or the FGFR isoform being overexpressed. Preclinical models included a patient-derived squamous head and neck cancer (HNSCC) xenograft (PDx) model overexpressing FGFR3 mRNA, a bladder cancer PDx model overexpressing FGFR2 mRNA, and a squamous esophageal cancer PDx model overexpressing FGFR1 mRNA. Against this background, clinical proof-of-concept was assessed by enrolling patients into the stratified expansion cohort of a Phase 1 study of BAY 1163877 (NCT01976741) based on tumor FGFR1-3 mRNA levels. FGFR1-3 mRNA was quantified in archival or newly obtained FFPE tumor biopsies by RNA in situ hybridization (RNAscope, ACD) and digital transcript counting (NanoString). In total, biopsies from > 500 patients were studied for FGFR1-3 mRNA expression levels. Based on preclinical xenograft experiments showing that low to moderate FGFR1-3 mRNA overexpression was not sufficient for a robust drug response, only patients with an RNAscope score of 3 or 4 (range 0-4) or a normalized Nanostring signal of 800 counts were eligible for enrollment. By applying these stringent criteria, FGFR1-3 mRNA positivity was on average observed with a 2- to 3-fold higher prevalence than published data for genetic aberrations of FGFRs in the respective tumor types (including amplifications, translocations, mutations). We further identified FGFR mRNA(+) tumor types in which genetic FGFR aberrations have not been previously reported. FGFR1-3 mRNA positivity ranged from 10% in lung adenocarcinoma to 45% in squamous non-small cell lung cancer (sqNSCLC) and 54% in HNSCC. As of August 2016, 57 FGFR mRNA(+) patients were enrolled and treated with BAY 1163877. Six patients experienced a partial remission (PR) by RECIST v1.1 criteria; including an FGFR3 mRNA(+) HNSCC patient without FGFR3 amplification or translocation, an FGFR1 mRNA(+) adenoid cystic carcinoma patient, and an FGFR1 mRNA(+) sqNSCLC patient (both without FGFR1 gene amplification), as well as an FGFR3 mRNA(+) bladder cancer patient without FGFR3 amplification, mutation or translocation. These results suggest that high FGFR1-3 mRNA expression identifies patients sensitive to FGFR inhibition. This population includes patients with and without genetic aberrations of FGFR1-3 encoding genes. In summary, an mRNA expression-based selection approach may identify a broader patient population with potential benefit from BAY 1163877, including tumor types not previously associated with altered FGFR signaling.

Citation Format: Peter Ellinghaus, Matthias Ocker, Sebastian Bender, Christoph Kneip, Stuart Ince, Markus Joerger, Martin Schuler. Use of tumor mRNA expression for patient selection in a phase I study of the pan-fibroblast growth factor receptor inhibitor BAY 1163877 [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 3738. doi:10.1158/1538-7445.AM2017-3738