Predictors of good outcomes in patients with hepatocellular carcinoma (HCC) treated with transarterial radioembolization (TARE).

e16219 Background: No reliable biomarkers or risk factors guide treatment selection in HCC patients. Management is typically institution-specific, and compliance with the available guidelines is low. Our retrospective review aims to identify HCC patients suitable for TARE. Methods: HCC patients treated with TARE between 1/1/2015 and 8/30/22 at Ohio State University were included in this study. Patient-related information was extracted from chart review. Various risk factors (related to HCC and TARE) were compared between good (GR) and poor (PR) responders (progression-free survival (PFS) ≤ vs. > median (m)), and good (GSv) and poor (PSv) survivors (overall survival (OS) ≤ vs. > m). Chi-square test and logistic regression were performed using JMP Pro 17 (SAS Institute Inc., Cary, NC). Results: Our cohort included 141 patients. The median age was 65 years (range 49-88), 80% were males and 80% were White (16% were African American and 4% were other). The median OS and PFS were 9 months (m) (95% confidence interval [CI], 5-22.5) and 4m (95% CI, 2-9), respectively. Compared to PSv (OS ≤ 9m, N = 71), the GSv group (OS > 9m, N = 70) had a high alcohol history (hx) (64% vs. 44%, p = 0.01); less baseline (BL) encephalopathy (Enc) (21% vs. 37%, p = 0.04) and post-TARE ascites (As) (31% vs. 63%, p = 0.001); were able to receive immune-checkpoint inhibitor (ICI) (44% vs. 6%, p < 0.001) before (6% vs. 1%) or after (39% vs. 10%) TARE. Other key BL features with a trend to significance (p = 0.06-0.08) were less BL cirrhosis (69% vs. 87%) and As (19% vs. 34%), hx of HCC treatment (systemic or local, 43% vs, 28%), and ability to get second TARE (19% vs. 8%). In worse survivors (PFS ≤ 5m, N = 40), post-TARE As (65% vs. 41%, p = 0.008) was significantly higher. Only 2/40 patients were able to get ICI (95% vs. 37%, p = 0.0002) in this group. Compared to PR, the GR group (PFS > 5m, N = 64) had a higher rate of diabetes mellitus (DM) (45% vs. 26%, p = 0.02) and less post-TARE As (34% vs. 58%, p = 0.004). BL and HCC-related factors prominently NOT featured in these comparisons were hepatitis (B or C) or smoking hx and macrovascular invasion. Conclusions: TARE suits patients with hx of alcohol use or DM, no BL Enc, As, or cirrhosis, and failed prior lines. Post-TARE As is poor prognostic markers. ICI use will improve the outcomes.