Comparing Lenvatinib/Pembrolizumab with Atezolizumab/Bevacizumab in Unresectable Hepatocellular Carcinoma: A Real-World Experience with Propensity Score Matching Analysis

Yu-Chun Hsu,Po-Ting Lin,Wei Teng,Yi-Chung Hsieh,Wei-Ting Chen,Chung-Wei Su,Ching-Ting Wang,Pei-Mei Chai,Chen-Chun Lin,Chun-Yen Lin,Shi-Ming Lin
DOI: https://doi.org/10.3390/cancers16203458
2024-10-13
Cancers
Abstract:Background: The combination of anti-angiogenic therapy and immune checkpoint inhibitors has revolutionized the management of unresectable hepatocellular carcinoma (uHCC). While an early-phase study demonstrated promising outcomes for lenvatinib plus pembrolizumab (L+P) in treating uHCC, the LEAP-002 trial did not meet its primary endpoint. However, the comparative efficacy between L+P and atezolizumab plus bevacizumab (A+B) as first-line treatment remains a topic of uncertainty. This study aimed to assess the effectiveness and safety of L+P in contrast to A+B among patients diagnosed with uHCC. Methods: We conducted a retrospective analysis of enrolled patients with uHCC who received L+P or A+B as initial systemic treatment at Chang Gung Memorial Hospital from June 2019 to December 2022. The overall survival (OS), progression-free survival (PFS), objective response rate (ORR), and disease control rate (DCR) by modified RECIST were compared. Results: 121 patients were recruited, with 37 receiving L+P and 84 receiving A+B. Among them, 95 (78.5%) patients were BCLC stage C, and 99 (81.8%) patients had viral etiology for HCC, predominantly chronic HBV (68.6%). Both the L+P and the A+B groups demonstrated comparable OS (18.2 months versus 14.6 months, p = 0.35) and PFS (7.3 months versus 8.9 months, p = 0.75). The ORR and DCR were similar. After propensity score matching, the results remained consistent between the matched patients. Treatment-related adverse events of any grade occurred in 30 (81.1%) in the L+P group and 62 (73.8%) in the A+B group. Conclusions: Our findings suggest that L+P and A+B exhibit comparable efficacy and safety profiles in real-world settings.
oncology
What problem does this paper attempt to address?