Treatment patterns post atezolizumab plus bevacizumab for advanced hepatocellular carcinoma: Real-world analysis at Veterans Health Administration comparing lenvatinib vs. sorafenib.

Abstract

e16195 Background: Atezolizumab plus Bevacizumab (A+B) has been the standard first-line therapy for advanced hepatocellular carcinoma (HCC) patients. There needs to be more data on the efficacy and selection of optimal sequences following resistance to A+B. Our study aims to review treatment patterns for disease progression following A+B, focusing on the most common 2nd line treatment utilized by the Veterans Health Administration (VHA). Methods: Patients with advanced HCC receiving line A+B at the VHA between Dec 1, 2019, to Mar 1, 2022, were selected electronically using ICD-9 and ICD-10 codes. Abstractors reviewed EMR following each patient from their index date of A+B initiation, sequential therapies, until death, or their last VHA visit, with the study period ending on Jan 31, 2023. Results: Three hundred thirty-two patients received A+B during our study period, and 1/3rd (n = 107) of these patients went on second-line treatments. 87 % started Tyrosine Kinase Inhibitors with 52, 29,12, and 1 on Lenvatinib, Sorafenib, Cabozantinib, and Regorafenib, respectively. Two second-line cohorts were selected, A (52 on Lenvatinib) and B (29 on sorafenib). The median age was 66 yrs vs. 65 yrs in cohorts A and B, respectively. In both cohorts, 60% non-Hispanic White and 90% with ECOG ≤1, there were no statistically significant differences between both cohorts in overall and progression-free survival. The outcomes are shown. Conclusions: Despite minimal improvement in PFS and a lower discontinuation rate due to toxicity favoring Lenvatinib over Sorafenib, Sorafenib had slightly better overall survival. This study is one of the most significant projects that describes 2nd line treatment patterns post-A+B failure. Further studies are warranted to evaluate larger cohorts to study treatment sequencing in the second and third lines.[Table: see text]