Atezolizumab/bevacizumab is a first-line therapy for unresectable hepatocellular carcinoma (HCC). Among several adverse events, grade ≥2 proteinuria is considered a significant adverse event that may cause bevacizumab interruption. Studies have shown that proteinuria might predict improved prognosis, although data are scarce and the association remains controversial, and the mechanisms and predictive factors remain unclear. We aimed to clarify these. In this multicenter retrospective study, we screened patients with HCC treated with atezolizumab/bevacizumab. The prognostic impact of grade ≥2 proteinuria was examined in patients with proper clinical data and preserved serum for growth factor analysis. For biomarker analysis predicting proteinuria, baseline serum vascular endothelial growth factor (VEGF)-A, VEGF-C, and VEGF-D levels were analyzed. This study included 75 patients, and 32 (42.7%) experienced grade ≥2 proteinuria. No significant differences were observed between those with or without proteinuria, except for aspartate transaminase and alanine transaminase levels. Time-dependent Cox proportional hazards analysis revealed that grade ≥2 proteinuria was significantly associated with better prognosis (hazard ratio 0.221; 95% confidence interval 0.082-0.592; p=0.003). In biomarker analysis, low baseline serum VEGF-C and VEGF-D levels were significantly associated with proteinuria, and multivariate analysis demonstrated that baseline serum VEGF-D level was significantly associated with grade ≥2 proteinuria (hazard ratio 0.101; 95% confidence interval 0.029-0.357; p<0.001). Grade ≥2 proteinuria in patients with unresectable HCC treated with atezolizumab/bevacizumab indicates a better prognosis, and baseline serum VEGF-D levels can help predict its occurrence. These findings can help in managing adverse events and prognosis in advanced HCC treated with atezolizumab/bevacizumab.
Read full abstract