P-264 - HGCSG1401: A retrospective cohort study evaluating the safety and efficacy of regorafenib in patients with metastatic colorectal cancer: Analysis of risk factors for liver dysfunction

Abstract

Introduction: Regorafenib (REG) prolongs overall survival (OS) and progression-free survival (PFS) for patients (pts) with metastatic colorectal cancer (mCRC) in RCTs. However, there is a problem with REG-induced severe liver dysfunction (≥Grade 3) which occurs in 5-11% of treated Japanese pts. Therefore, we analyzed the incidence, therapeutic efficacy, and potential risk factors for REG-induced severe liver dysfunction in a community-based retrospective study (HGCSG1401) of pts treated with REG. Methods: 173 pts treated with REG were retrospectively registered from 22 centers in Japan. Survival analyses were performed with Kaplan-Meier method. Log-rank test and Cox-proportional hazard model were used to compare Grade 0-2 and 3-5. To identify risk factors for REG-induced severe liver dysfunction, a multivariate analysis was performed using the logistic regression analysis with backward elimination for variables with p < 0.10 in univariate analysis. Results: In 173 eligible pts, 24 (13.9%) experienced REG-induced severe liver dysfunction. The median PFS of Grade 0-2 and 3-5 were 2.2 and 1.6 months, respectively. The median OS of Grade 0-2 and 3-5 were 6.9 and 3.4 months, respectively. In the analysis of PFS and OS, there were significant difference between Grade 0-2 and 3-5 (PFS: HR 1.578, p = 0.045, OS: HR 1.799, p = 0.010). Univariate analyses showed that high AST, LDH, ALP, and GGT level at baseline were associated with REG-induced severe liver dysfunction. In multivariate analyses, high baseline AST level (≥ 50 U/L) was significantly associated with increased risk of REG-induced severe liver dysfunction (odds ratio=3.458, p = 0.032). Conclusion: Compared with previous reports, this analysis showed the slightly higher incidence of REG-induced severe liver dysfunction (13.9%). In multivariate analysis, it was inferred that high AST level (≥ 50 U/L) at baseline might be an independent risk factor. REG-induced severe liver dysfunction population significantly showed shorter OS than mild liver dysfunction. It is presumed that this patient population could not receive TAS-102 as post regorafenib therapy. Since this is an exploratory analysis, we consider it necessary to verify in large data set.