IMbrave150: Exploratory analysis to examine the association between bevacizumab (bev) ever being skipped and bev never being skipped in patients with unresectable hepatocellular carcinoma (HCC) treated with atezolizumab (atezo) plus bev in a global phase 3 study.

538 Background: Treatment with atezo + bev has been approved globally for patients with unresectable HCC who have not received prior systemic therapy, based on results from the Phase 3 IMbrave150 study (NCT03434379, Finn NEJM 2020 and Cheng J Hepatol 2022). According to Phase 1b study data (Lee Lancet Oncol 2020), when combined with atezo, bev has a clinically relevant contribution in terms of immune enhancement and normalization of tumor blood vessels. However, it is unknown if bev being skipped due to bev adverse events of special interest (AESIs) has an impact on the efficacy of atezo + bev. Here, we conducted an exploratory efficacy analysis examining patients who had skipped bev vs those who never skipped bev using IMbrave150 study data. Methods: In Arm A of IMbrave150, patients were assigned to receive atezo + bev. Group A-1 included patients whose bev was ever skipped due to bev AESIs. Group A-2 included the patients who were not included in group A-1. Landmark analyses of overall survival (OS) and progression-free survival (PFS) were performed in patients who received atezo + bev for ≥6 months to minimize immortal time bias. Results: Of 210 patients who received ≥6 months of atezo + bev, 69 were assigned to group A-1 and 141 were assigned to group A-2. No obvious differences between groups were observed in the distribution of baseline characteristics. Of the group A-1 patients, 75.4% were BCLC stage C and 76.8% were Child-Pugh A5, while group A-2 patients were 80.1% BCLC stage C and 77.0% were Child-Pugh A5. At the data cutoff (Aug 20, 2020), median OS was 25.8 vs 26.2 months (HR, 1.04; 95% CI: 0.64, 1.69) and median PFS per independent review facility-assessed RECIST 1.1 was 15.5 vs 10.0 months (HR, 1.07; 95% CI: 0.74, 1.55) for groups A-1 and A-2, respectively. Conclusions: The patients who had ever skipped bev did not show different efficacy compared with those who had never skipped bev in this post hoc analysis. Although limitations due to the non-randomized and exploratory nature of this comparison should be acknowledged, the results suggest that skipping bev did not have considerable impact on the efficacy of atezo + bev. Clinical trial information: NCT03434379 .