Safety and efficacy of atezolizumab plus bevacizumab in patients with unresectable hepatocellular carcinoma in early clinical practice: A multicenter analysis

Hepatol Res. 2022 Mar;52(3):269-280. doi: 10.1111/hepr.13732. Epub 2021 Nov 30.

Abstract

Purpose: To assess the impact of clinical factors on the safety and efficacy of atezolizumab plus bevacizumab (ATZ + BV) treatment in patients with unresectable hepatocellular carcinoma (u-HCC).

Method: Ninety-four u-HCC patients who were treated with ATZ + BV at multiple centers were enrolled. We defined Child-Pugh (CP)-A patients who received ATZ + BV treatment as a first line therapy as the 'meets the broad sense of the IMbrave150 criteria' group (B-IMbrave150-in, n = 46), and patients who received ATZ + BV treatment as a later line therapy or CP-B patients (regardless of whether ATZ + BV was a first line or later line therapy) as the B-IMbrave150-out group (n = 48). Patients were retrospectively analyzed for adverse events (AEs) and treatment outcomes according to their clinical characteristics, including neutrophil lymphocyte ratio (NLR) at baseline.

Results: The overall incidence of AEs was 87.2% (82/94 patients). The frequency of interruption of ATZ + BV treatment due to fatigue was higher in CP-B than CP-A patients (p = 0.030). Objective response (OR) rates of the B-IMbrave150-in group (28.3%, 39.1%) were significantly higher than those of the B-IMbrave150-out group (8.3%, 18.8%; p = 0.0157, 0.0401) using Response Evaluation Criteria in Solid Tumors (RECIST) and modified RECIST, respectively. In multivariate analysis, NLR (hazard ratio (HR), 4.591; p = 0.0160) and B-IMbrave150 criteria (HR, 4.108; p = 0.0261) were independent factors associated with the OR of ATZ + BV treatment using RECIST.

Conclusion: In real-world practice, ATZ + BV treatment might offer significant benefits in patients who meet B-IMbrave150 criteria or have low NLR.

Keywords: adverse events; atezolizumab plus bevacizumab; hepatocellular carcinoma; immune checkpoint inhibitor; molecular targeted agents; response.