C-reactive protein-to-albumin ratio predicts survival in hepatocellular carcinoma on atezolizumab plus bevacizumab.

Background Systemic inflammation and nutritional status influence tumor progression and response to systemic therapy. The C-reactive protein-to-albumin ratio (CAR), a simple composite of these factors, has been associated with prognosis in hepatocellular carcinoma (HCC); however, its clinical significance in patients treated with atezolizumab plus bevacizumab (Atez/Bev) remains unclear. We aimed to evaluate the prognostic value of baseline CAR in patients with unresectable HCC receiving Atez/Bev. Methods We retrospectively analyzed 289 patients with unresectable HCC treated with Atez/Bev across three institutions. The optimal CAR cutoff was determined using time-dependent receiver operating characteristic analysis at the median overall survival (OS) time point. Survival outcomes were evaluated using Kaplan-Meier analysis and Cox proportional hazards models. Results The median progression-free survival (PFS) and OS were 6.8 and 26.5 months, respectively. The optimal CAR cutoff was 0.056. Patients with high CAR showed significantly poorer radiological response, PFS, and OS than those with low CAR. In multivariate analysis, CAR ≥0.056 was independently associated with shorter PFS (HR, 1.752; 95% CI, 1.296-2.367; p < 0.001) and OS (HR, 2.175; 95% CI, 1.323-2.948; p < 0.001). Time-dependent ROC analysis showed that CAR and neutrophil-to-lymphocyte ratio had comparable prognostic performance up to 18 months, whereas CAR showed significantly higher AUROC values from 24 months onward. Conclusion Baseline CAR is an independent prognostic biomarker in patients with unresectable HCC treated with Atez/Bev and may be a useful biomarker for risk stratification and prediction of long-term outcomes.