Supplementary Material for: Real-World Comparison of First-Line Atezolizumab plus Bevacizumab and Durvalumab plus Tremelimumab for Unresectable Hepatocellular Carcinoma

Introduction: Atezolizumab plus bevacizumab (Atez/Bev) was the first immune checkpoint inhibitor (ICI) regimen approved in 2020 for unresectable hepatocellular carcinoma (uHCC), followed by durvalumab plus tremelimumab (Dur/Tre) in 2023. There is very little data available comparing the efficacy and safety of Atez/Bev with those of Dur/Tre. This study aimed to clarify the therapeutic outcomes and safety of Atez/Bev and Dur/Tre.
Methods: We retrospectively analyzed patients with uHCC (BCLC-B/C and Child-Pugh class A) treated with Atez/Bev (n=302) or Dur/Tre (n=129) as first-line systemic therapy across multiple institutions between 2023 and 2025. A retrospective comparison of the treatment outcomes and safety of Atez/Bev and Dur/Tre was conducted.
Results: The objective response rate (ORR) and disease control rate (DCR) were comparable between the Atez/Bev and Dur/Tre groups (31.8%/71.9% vs. 28.7%/63.6%, p=0.570/p=0.110, respectively). Median progression-free survival (PFS) was longer with Atez/Bev (9.1 vs. 5.0 months, p=0.002), while OS was comparable (25.6 vs. 22.1 months, p=0.172). Same results were shown after adjusting with inverse probability weighting (IPW). Grade 5 immune-related adverse events occurred in 0.3% (n=1, interstitial pneumonia) of the patients receiving Atez/Bev and 0.8% (n=1, colitis) of those receiving Dur/Tre. In the Cox hazard analysis adjusted with IPW, Dur/Tre demonstrated favorable trend of OS (hazard ratio (HR) <0.75) in patients with elevated alpha-fetoprotein (AFP) (≥100 ng/mL) (HR 0.72, interaction p=0.006) or double positive elevation of tumor marker (AFP and des-Gamma-Carboxy Prothrombin (DCP) (≥100 mAU/mL)) (HR 0.66, interaction p=0.005).
Conclusion: Although Atez/Bev was associated with a longer PFS, OS did not differ significantly between Atez/Bev and Dur/Tre. This dissociation between PFS and OS may reflect differences in disease biology, treatment sequencing, and post-progression management rather than intrinsic superiority of either regimen. These findings highlight the importance of individualized treatment selection and careful consideration of tumor characteristics and hepatic reserve when choosing first-line immunotherapy for uHCC.