755P - Alpha-fetoprotein (AFP) response in patients with unresectable hepatocellular carcinoma (HCC) in the phase III RESORCE trial

Abstract

BackgroundA decrease in AFP level during treatment has been associated with improved overall survival (OS) in patients with HCC. In the phase 3 RESORCE trial (NCT01774344), the multikinase inhibitor regorafenib significantly improved OS versus placebo in patients with HCC and disease progression on sorafenib. We analyzed OS in patients who had an AFP response in RESORCE. MethodsPatients with Barcelona Clinic Liver Cancer stage B or C HCC, Child -Pugh A liver function, ECOG performance status 0−1, and documented radiologic progression on prior sorafenib were randomized 2:1 to regorafenib 160mg/day (n=379) or placebo (n=194) for Weeks 1-3 of each 4-week cycle. Serum AFP levels were assessed at baseline, every 4 weeks during treatment, and at the end of treatment. Patients who had a baseline AFP ≥20ng/mL and an AFP measurement at the start of Cycle 3 were evaluable for AFP response, defined as a decrease of≥20% in AFP level from baseline at the start of Cycle 3. ResultsOverall, 232 of the 573 randomized patients fulfilled the above criteria and were included in the analysis. Patients with an AFP response had a median OS from randomization of 13.8 months vs 8.9 months in those without an AFP response (Table). The rate of AFP response was 46% (77/168) in the regorafenib group and 11% (7/64) in the placebo group. Among regorafenib-treated patients, the median OS was 13.8 months in patients with an AFP response vs 9.8 months in those without an AFP response. A landmark analysis of OS from the start of Cycle 3 showed that the hazard ratios for AFP response vs no response were 0.57 (95% CI 0.40, 0.82) in the regorafenib and placebo arms combined and 0.72 (95% CI 0.48, 1.08) for patients in the regorafenib arm.Table755PTableRegorafenib + /placebo arms combinedRegorafenib arm(n=232)(n=168)AFP response (n=84)No AFP response (n=148)AFP response (n=77)No AFP response (n=91)Median OS (95% CI) from randomization, months13.8 (11.8, 16.5)8.9 (8.0, 9.7)13.8 (11.7, 16.5)9.8 (8.2, 14.2)HR (95% CI) AFP response yes/no0.57 (0.40, 0.82)0.72 (0.48, 1.07)CI, confidence interval; HR, hazard ratio. ConclusionsIn this exploratory analysis of RESORCE, an AFP response with regorafenib was associated with improved OS. Clinical trial identificationNCT01774344. Editorial acknowledgementJennifer Tobin of OPEN Health Medical Communications (London, UK), with financial support from Bayer. Legal entity responsible for the studyBayer. FundingBayer. DisclosureJ. Bruix: Honoraria (self): Bayer; BTG; Eisai; Ipsen; Sirtex; Terumo; Advisory / Consultancy: AbbVie; Adaptimmune; Arqule; Astra-Medimmune; Basilea; Bayer; Bio-Alliance; Bristol-Myers Squibb; BTG; Eisai; Gilead; Incyte; Ipsen; Kowa; Lilly; Merck Sharp & Dohme; Nerviano; Novartis; Quirem; Roche; Sanofi Aventis; Sirtex; Terumo; Research grant / Funding (institution): Bayer; BTG. M. Reig: Honoraria (self): AstraZeneca; Bayer; Bristol-Myers Squibb; BTG; Gilead; Ipsen; Lilly; Advisory / Consultancy: AstraZeneca; Bayer; Bristol-Myers Squibb; Ipsen; Lilly; Roche; Research grant / Funding (institution): Bayer; Travel / Accommodation / Expenses: Bayer; Bristol-Myers Squibb; Gilead; Ipsen; Lilly. P. Merle: Advisory / Consultancy: AstraZeneca; Bayer; Bristol-Myers Squibb; Eisai; Exelixis; Ipsen; Merck Sharp & Dohme; Onxeo; Roche. M. Kudo: Honoraria (self): Bayer; Eisai; Merck Sharp & Dohme; Advisory / Consultancy: Bayer; Bristol-Myers Squibb; Eisai; Merck Sharp & Dohme; Ono Pharmaceutical; Research grant / Funding (institution): AbbVie; Astellas Pharma; Bristol-Myers Squibb; Chugai; Daiichi Sankyo; EA Pharma; Eisai; Gilead; Medico’s Hirata; Otsuka; Taiho; Takeda. G. Meinhardt: Shareholder / Stockholder / Stock options: Bayer; Full / Part-time employment: Bayer. M. Zhang: Shareholder / Stockholder / Stock options: Bayer; Full / Part-time employment: Bayer. K. Ozgurdal: Full / Part-time employment: Bayer.