P14.19 Regorafenib in recurrent glioblastoma patients: a large real-life experience

Abstract BACKGROUND Regorafenib (REG), an oral multikinase inhibitor of angiogenic, stromal, and oncogenic receptor tyrosine kinases showed encouraging benefit in recurrent GBM patients enrolled in the randomized, phase 2 REGOMA trial. We investigated the clinical outcome and safety of REG in a real-life population of recurrent glioblastoma patients treated at Veneto Institute of Oncology as off-label use. MATERIAL AND METHODS Patients receiving REG at Veneto Institute of Oncology (Padua, Italy) were entered prospectively on a clinical database. Data were retrospectively analyzed. The primary endpoints of the study were overall survival (OS) and safety. The major inclusion criteria were: histologically confirmed diagnosis of GBM, disease progression as defined by RANO criteria after surgery followed by radiochemotherapy with temozolomide, ECOG PS ≤ 2; PTS with ≥ 2 prior lines of therapy were excluded. According to original schedule, patients received REG 160 mg once daily for the first 3 weeks of each 4-week cycle until disease progression, death, unacceptable toxicity, or consent withdrawal. Kaplan-Meier method was used to estimate the survival curves, RANO criteria for radiological assessment, CTCAE v5.0 for drug related adverse events. RESULTS From February 2018 to September 2020, 54 consecutive patients were treated with REG and enrolled in this study: median age was 56, ECOG PS 0–1 in 91% of patients, MGMTmet in 53%, second surgery at the time of relapse were performed in 30% of enrolled patient, 41% of patients underwent steroids at baseline. At the time of analysis, median follow-up was 11.1 ms, 30 PTS (56%) had died and 50 PTS (93%) had progressed. Median OS was 10.2 ms (95%CI, 6.4–13.9), 12m-OS was 43%; median PFS was 2.3ms (95%CI, 1.3–3.3) and 6m-PFS was 18%. All patients were evaluable for response: disease control rate (DCR) was 46.3%; stable disease was reported in 38.8% and partial response in 7.4%. Age, MGMT status and corticosteroid use at baseline were not statistically significant on multivariate analysis for OS. Grade 3 drug-related adverse events (AEs) occurred in 10 patients (18%) and the most frequent were hand-foot skin reaction, asthenia and increased lipase and transaminases; 1 PT (2%) reported a grade 4 AE (rash maculo-papular). AEs led to REG dose reductions in 37% of patients and, it was permanently discontinued in 5%. No death was considered to be drug-related. CONCLUSION We reported a large, mono-institutional “real world” experience of REG in recurrent glioblastoma patients. Overall, results are close to those reported in REGOMA trial although, we showed a longer OS. Toxicity was moderate and manageable. Encouraging clinical benefits of REG in recurrent GBM population were confirmed.