Abstract 560: Retrospective evaluation, of the randomized Phase 3 MARQUEE trial of tivantinib (T) + erlotinib (E) versus placebo (P) + erlotinib (E) using VeriStrat in patients with previously treated nonsquamous NSCLC

Abstract

Abstract Background: MET, the high-affinity receptor for hepatocyte growth factor, is frequently deregulated in human cancer and is thought to be involved in resistance to EGFR TKIs. Tivantinib is an orally administered MET inhibitor currently under development as a cancer therapy. The Phase 3 MARQUEE trial enrolled 1048 subjects with locally-advanced or metastatic non-squamous NSCLC who had received one or two lines of prior systemic therapy. Patients were randomized 1:1 to T+E (N = 526) or P+E (N = 522) until disease progression. In the ITT population, addition of tivantinib to erlotinib significantly improved PFS and ORR but did not improve OS, the primary endpoint. VeriStrat® (VS) is a multivariate serum protein test that utilizes MALDI-TOF mass spectrometry to assign a “GOOD” (VS-G) or “POOR” (VS-P) classification and has demonstrated broad prognostic and predictive utility for EGFR TKI treatment regimens. Methods: Pre-treatment serum samples from the MARQUEE study were blinded to clinical outcome and analyzed by VeriStrat. Clinical benefit of T+E vs. P+E was analyzed as a function of VS status, in the ITT population as well as in pre-defined subgroups. Results: A total of 996 subjects were classified: 718 as VS-G and 278 as VS-P. Subjects testing VS-G had superior outcomes to those testing VS-P in both the T+E arm (OS HR = 0.32, CI: 0.26-0.41, p<0.0001; PFS HR = 0.55, CI: 0.44-0.69, p<0.0001) and P+E arm (OS HR = 0.43, CI: 0.34-0.55, p<0.0001; PFS HR = 0.70, CI: 0.55-0.88, p<0.028). VeriStrat status was a significant, independent predictor of OS, in these erlotinib treated patients, after adjustment for EGFR mutation, KRAS mutation, or MET IHC status. However, VeriStrat did not predict OS or PFS benefit for T+E vs. P+E (treatment * VS interaction p-value for OS of p = 0.09 and PFS of p = 0.19). VS was neither predictive nor prognostic for tivantinib effect in any molecular subgroup. Conclusion: These data support the previously demonstrated ability of VS to be both prognostic and predictive for survival in erlotinib treated patients, but not for tivantinib treated patients. Citation Format: Giorgio Scagliotti, Wallace Akerley, Joachim von Pawel, Joanna Roder, Dale Shuster, Brian Schwartz, Dominic Spinella. Retrospective evaluation, of the randomized Phase 3 MARQUEE trial of tivantinib (T) + erlotinib (E) versus placebo (P) + erlotinib (E) using VeriStrat in patients with previously treated nonsquamous NSCLC. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 560. doi:10.1158/1538-7445.AM2015-560