Exposure-response (ER) relationship of regorafenib (REG) in patients with hepatocellular carcinoma (HCC).

Abstract

374 Background: In the phase 3 RESORCE trial (N = 573), REG (160 mg QD, 3 weeks on/1 week off) significantly (P < 0.001) improved overall survival (OS) versus placebo in patients with HCC who progressed during sorafenib treatment (HR 0.63; 95% CI 0.50‒0.79). This exploratory analysis evaluated the ER relationship between REG and both OS and time-to-progression (TTP). Methods: Exposure estimates were available for 327 and 315 patients for the OS and TTP analysis, respectively. A multivariate Cox proportional hazard analysis was used to evaluate differences between exposure quartiles and efficacy, taking into consideration all preselected baseline covariates. Subsequently, the functional form of the ER relationship within the REG group was evaluated. Results: The baseline risk factors ECOG status, AFP category, and AST/ALT baseline levels were significantly associated with OS (P < 0.01), with higher HRs for ECOG category ≥ 1, AFP levels ≥ 400 ng/mL, and AST/ALT levels > 3 x ULN. For TTP, a significant correlation (P < 0.01) with age was observed, with a higher HR for patients < 65 years of age. After adjusting for these baseline risk factors, the average exposure over the first 28 days grouped in exposure quartiles for REG was significantly associated with OS and TTP (P < 0.01), i.e. the HR in all exposure quartiles was < 1 versus placebo. By contrast, the difference in HR between exposure quartiles was found to be small and not significant. While no statistically significant ER relationship could be identified between (continuous) individual exposure and efficacy in the REG group, the analysis of the functional form of the ER relationship indicates that OS and TTP improved with an increase in exposure, under the assumption of a linear and a non-linear ER relationship. Conclusions: In addition to a significant difference in OS and TTP between REG and placebo, a trend towards improved efficacy was observed with an increase in REG exposure in HCC patients, which was not statistically significant when taking into consideration ECOG status, AFP level, and AST/ALT risk factors (OS) or age (TTP). Therefore, similar efficacy was observed in patients with HCC who started on REG 160 mg QD 3 weeks on/1 week off over the entire exposure range. Clinical trial information: NCT01774344.