Atezolizumab-bevacizumab in very elderly with hepatocellular carcinoma: Age alone is not a limiting factor except in ALBI grade 3.

BACKGROUND &

aimsEvidence on atezolizumab plus bevacizumab (AtezoBev) in very elderly people with hepatocellular carcinoma (HCC) remains rare in the European ancestry population. We compared outcomes in patients aged ≥75 years with younger patients.

methodsThis multicenter retrospective study included patients treated with first-line AtezoBev for advanced HCC. Patients aged ≥75 years were matched 1:1 with those <75 years using propensity scores. Overall survival (OS), progression-free survival (PFS), tumor response, and adverse events (AEs) were analyzed.

resultsAmong 814 patients, 566 (69.5%) were <75 years (median age 64, 84% male) and 248 (30.5%) ≥75 years (median age 78, 86% male). After matching, 484 were analyzed. After a median follow-up of 28.0 months, OS and PFS were similar in ≥75 and <75 (15.4 vs. 16.07 months; p = 0.936), (7.2 vs. 6.5 months; p = 0.706) respectively. Age ≥ 75 years was not associated with PFS nor OS. In patients aged ≥75 years, modified albumin-bilirubin grade 3 (mALBI) was the only factor associated with disease progression (HR 4.37, 95% CI 2.04,9.37; p <0.001) and mortality (HR 5.62, 95% CI 2.47,12.8; p <0.001). In mALBI 3 median OS and PFS were 5.43 and 2.3 months, respectively. Immune-related AEs were less frequent in ≥75 including (22.1% vs. 36.9%; p <0.001) or excluding dysthyroidism (15.9% vs. 25.9%; p = 0.01). In univariate analysis, OS and PFS were longer in patients ≥75 years who developed hypertension (p = 0.04 and p = 0.09), proteinuria (p <0.0001 and p = 0.015) and Immune-related AEs (p = 0.02 and p = 0.007). Hypertension during treatment was associated with proteinuria (odds ratio = 13.1; 95% CI 4.1-42.4), without difference at baseline (p = 0.35).

conclusionsAtezoBev is effective and safe in patients ≥75 years but mALBI 3 warrants particular caution. IMPACT AND IMPLICATIONS: Very elderly patients (≥75) are underrepresented in immunotherapy trials, leaving uncertainty about the benefit-risk of atezolizumab plus bevacizumab in this subgroup. In real-world practice, European ancestry patients ≥75 with advanced hepatocellular carcinoma are common. Our study indicates that atezolizumab-bevacizumab is generally safe and effective in this population, with mALBI grade 3 as a notable high-risk exception. Although retrospective, the use of propensity score methods mitigates confounding and supports the robustness of our conclusions. CLINICAL TRIALS REGISTRATION NUMBER: NCT06416683.