Evaluating the safety and efficacy of the addition of liver radiation (RT) to atezolizumab and bevacizumab (A/B) in advanced hepatocellular carcinoma (HCC): A retrospective cohort analysis.

Abstract

481 Background: Recently, the addition of stereotactic body RT to sorafenib in patients (pts) with advanced HCC demonstrated improved overall survival. However, A/B is the current standard of care for pts with advanced HCC based on IMBrave150. For these pts, RT may potentiate the antitumor immune response of A/B but carries the risk of bowel toxicity and impaired liver function. To date, there are limited data regarding safety and outcomes with the addition of liver RT to A/B in pts with advanced HCC. Here, we describe our single-center experience of adding RT to A/B for pts with advanced HCC. Methods: This single-center retrospective cohort study was conducted at the University of Pennsylvania. We identified all pts with HCC naïve to systemic therapy who received infusions of A/B and those who received RT to the liver within 30 days of starting A/B until discontinuation from 1/1/2020 until 2/2/2023. We assessed the real-world response rate (rwRR) determined from clinician impression after initiating A/B, safety outcomes, overall survival (OS), and time-to-progression (TTP) from initiation of A/B. Time-to-event outcomes were analyzed by the Kaplan-Meier method with log-rank test and Cox proportional hazards modeling. Results: We identified 49 pts (n=34 control, n=15 RT) with advanced HCC receiving first-line A/B. Those who received RT were more likely to have alcoholic liver disease (p=0.013) and were less likely to have ascites (p=0.04) but were otherwise balanced regarding demographics, prior locoregional therapies, presence of macrovascular invasion and extrahepatic spread, and baseline laboratories. The median RT dose was 63 Gy. The rwRR in the RT group was 73.3% (11/15) compared to 17.6% (6/34) in the control group (Risk Ratio (RR) 3.9, 95% CI 1.7-8.6, p<0.001). Two pts in the control group (5.8%) and one patient in the RT group (6.7%) experienced clinically significant bleeding (RR=1.1 95% CI 0.1-11.6). In the RT group, three pts (20%) developed worsening liver dysfunction after completing RT, concerning for possible radiation-induced liver disease. Six pts in the control group and one patient in the RT group developed immune-related adverse events leading to discontinuation of A/B. The median OS in the RT group was 14.4 mos and 10.8 mos in the control group (Hazard Ratio (HR) death 0.49 (95% CI 0.20-1.22, p=0.12)). Median TTP was 6.4 months (mos) with RT compared to 5.8 mos in the control group (HR progression 0.72 (95%CI 0.35-1.46, p=0.36. Conclusions: In our retrospective cohort, the addition of liver RT to standard A/B resulted in increased responses and a non-significant improvement in outcomes without an increased risk of bleeding. An assessment of changes in liver dysfunction by Child-Pugh and ALBI score is ongoing. Overall, these data support prospective trials to evaluate the addition of RT to A/B.