Understanding the efficacy of frontline therapies for hepatocellular carcinoma through etiologic stratification.

Abstract

491 Background: Despite significant advancements in recent years, hepatocellular carcinoma (HCC) treatment options remain limited for patients who progress on locoregional therapy. Given increasing interest in personalized cancer care for HCC, analysis was done to understand whether the etiology of HCC influenced the efficacy of four common first-line FDA approved systemic therapy regimens. Methods: This study used the nationwide Flatiron Health electronic health record (EHR)-derived de-identified database from 1/1/2003 to 8/1/2021 with 3519 HCC patients who have received either atezolizumab-bevacizumab (AB), nivolumab, sorafenib, or lenvatinib as a first-line systemic therapy. Information on demographics (age, gender, sex race, ethnicity, ECOG, SES Index), cancer stage, and treatment type was collected. In addition, comorbidities of hepatitis B, hepatitis C, alcohol use, diabetes, and obesity were collected as surrogates for disease etiology: viral hepatic disease, alcohol induced hepatic dysfunction or non-alcoholic fatty liver disease (NAFLD). Patients with missing data in any of the population stratifying variables were excluded. Overall survival (OS), Progression-free-survival (PFS) and time to treatment discontinuation (TTTD) were analyzed for each systemic therapy stratified by etiology as endpoints. Kaplan-Meier (KM) univariate analysis was performed along with the log-rank test for the primary outcome. Multivariate analysis (MVA) was conducted on all three endpoints via cox proportional hazard model to adjust for possible covariates in the patient population. Results: After exclusion criteria, 548 patients with complete data were included (AB N=124, Sorafenib N=324, Lenvatinib N=61, Nivolumab N=39) with variable etiologic presence (hepatitis B N=14, hepatitis C N=50, alcohol use N=41, diabetes N=50, Obesity N=44). When comparing patients across all recorded systemic therapies stratified by the etiologic variables, there was no statistically significant difference in OS, PFS, or TTTD. When stratified by presence of Hepatitis B, patients treated with atezolizumab-bevacizumab trended towards improved outcomes but did not reach statistical significance (P = 0.064). Other treatments showed mixed results in relation to specific etiologies, but no clear trend in efficacy emerged. Conclusions: This early retrospective study suggests that systemic therapy for HCC is not significantly impacted by factors associated with the etiology of disease, such as hepatitis C, alcohol use, and obesity. However, small population sizes and potential nonrandom covariate missingness in this study may obscure effects. Future studies are needed to further elucidate this trend along with subpopulation effects of other systemic therapies.