Real-world experience with atezolizumab plus bevacizumab (A+/-B) in Hispanic patients with advanced hepatocellular carcinoma (HCC) at a Hispanic-majority cancer center.

Abstract

549 Background: Combination of atezolizumab (A), a programmed death-ligand 1 (PD-L1) inhibitor, and bevacizumab (B), a vascular endothelial growth factor (VEGF) inhibitor, has demonstrated efficacy for first-line treatment for HCC. However, real-world data on the outcomes of this combination therapy in diverse patient populations, particularly in Hispanic U.S. patients, remains limited. This study aims to evaluate the clinical outcomes of patients intended to treat with A+/-B at a Hispanic-majority U.S. cancer center. Methods: This retrospective study of patients diagnosed with advanced unresectable HCC who received treatment with A+B at a Hispanic-serving NCI-designated cancer center. Patients were included if they received at least one dose of A, either as monotherapy or in combination with B, between January 2019 and April 2022. Demographic and clinical data, including age, gender, ethnicity, liver disease etiology, Child-Pugh (CP) classification, comorbidities, and treatment details, were collected. The primary endpoints were the overall response rate (ORR) and complete response (CR) rate. Statistical analyses were conducted to evaluate the association between baseline characteristics and clinical outcomes. The median overall survival (mOS) evaluated by Kaplan-Meier Method. Results: 38 patients received at least A (84% A+B, 6 patients had B held due to bleeding risk after consent). Median age 65 years (45-90). 32 (84%) patients were male. 22 (58%) patients were Hispanic. 45% had HCV-related liver disease. 32% had CP B with no statistically significant association between CP Class and ethnicity. 53% had concurrent diabetes. 26% patients had obesity. The median time on treatment was 7.5 mo (1- 24 mo) with no statistical difference between Hispanics and non-Hispanics (9 vs 6 mo, p 0.70). CR 37% with ORR of 42%. Using Fisher’s exact test, no association was noted between obesity, ethnicity, treatment option or CP class to occurrence of complete or partial response. mOS was 19 mo. There was no statistical difference in mOS between CP A and CP B (19 vs 10 mo, p value 0.6) and no difference between Hispanics (14 mo) vs non-Hispanics (19 mo) (p 0.5), using log rank test. Conclusions: This real-world analysis of a Hispanic-majority U.S. cohort of patients with advanced HCC shows 42% ORR and 37% CR. The findings suggest that this regimen is effective across a diverse patient population, including those with comorbid conditions, such as liver dysfunction (CP B 32%), HCV (45%), diabetes (53%, and obesity (26%). Limitations are the small study population and retrospective nature of the study. Evaluation of toxicity data is ongoing. Further prospective studies are needed to identify biomarkers of outcomes in Hispanic patients with HCC, which is ongoing at our cancer center (NCT03894917).