Biom-13. association of hand-foot skin reactions with survival in recurrent glioblastoma patients treated with regorafenib

Abstract INTRODUCTION The phase 2 REGOMA trial suggested a survival benefit of the multikinase inhibitor regorafenib for treating patients with glioblastoma at first recurrence. Hand-foot skin reactions (HFSR) following multikinase inhibitor therapy have been reported to predict prolonged overall survival (OS) in patients with solid tumors. The predictive value of HFSR in patients with recurrent glioblastoma undergoing regorafenib therapy remains to be investigated. METHODS Fifty-seven patients (age range, 35-77 years) with recurrent IDH-wildtype glioblastomas were retrospectively identified from five centers in Germany and Switzerland. Regorafenib was administered at 160 mg/d for the first three weeks of each 4-week cycle, with individual dose adjustments according to toxicity. Patients were stratified into two groups based on the occurrence of HFSR following regorafenib. Group differences were investigated using the Mann-Whitney rank-sum and Fisher's exact tests. Kaplan-Meier analysis was used for univariate analyses, and the Cox proportional hazards regression model for multivariate analyses. RESULTS Patients received a median of two regorafenib cycles (range, 1-14 cycles). Median follow-up after recurrence was 7.0 months (range, 0.6-35.9 months). The median number of previous treatment lines was 2 (range, 1-6 lines). At the time of data analysis, 47 patients (82%) had died. HFSR were observed in 14 out of 57 patients (25%) and associated with a significantly longer OS (10.6 vs. 6.0 months; P = 0.040). No significant group differences (HFSR compared to no HFSR) were observed regarding age, the extent of resection, MGMT promoter methylation, Karnofsky performance status, or the number of completed regorafenib cycles (all P > 0.05). Cox proportional hazards regression analysis confirmed that HFSR were associated with longer OS (P = 0.039; HR, 0.438), independent of age, Karnofsky performance status, and the number of previous treatment lines (all P > 0.05). CONCLUSION Our data suggest that the occurrence of HFSR following regorafenib is associated with longer OS in recurrent glioblastoma patients.