Baseline liver function and outcomes in the phase III REFLECT study in patients with unresectable hepatocellular carcinoma (uHCC).

524 Background: Lenvatinib (LEN) is approved for first-line treatment of uHCC. Baseline (BL) liver function (Child-Pugh score [CPS] and albumin-bilirubin grade [ALBI]) was prognostic in uHCC patients (pts) who received sorafenib (SOR) but has not been assessed with LEN in uHCC. Here, we report post hoc analysis of BL liver function and efficacy/safety outcomes from the phase 3 REFLECT study. Methods: Overall survival (OS), progression-free survival (PFS), objective response rate (ORR), and safety were stratified by BL ALBI or CPS. OS and PFS were estimated by Kaplan–Meier method. Independent radiologic review utilized mRECIST criteria for ORR. Safety was assessed using NCI-CTCAE, version 4.0. Results: Liver function measured by ALBI and CPS seemed to be prognostic for OS and ORR. Median OS was longer in ALBI grade 1 (ALBI-1) vs grade 2 (ALBI-2) pts or for CPS-5 vs CPS-6 on either treatment arm and was longer for LEN vs SOR. ORR was higher in pts with better ALBI or CPS and for LEN vs SOR. Rates of treatment-emergent adverse events grade ≥3 were lower with better BL liver function (ALBI-1 vs ALBI-2: 70% vs 86%; CPS-5 vs CPS-6: 72% vs 86%). Study-drug withdrawal, dose reduction, and dose interruption occurred more often in pts with worse BL liver function. Conclusions: This post hoc analysis suggests ALBI (by OS, PFS and ORR) and CPS (by ORR) may be prognostic in uHCC pts and that BL liver function may be linked with efficacy/safety outcomes. This analysis also found that LEN provided benefit vs SOR for uHCC, regardless of BL liver function. The benefit of LEN may be underestimated, as more ALBI-2 pts and fewer ALBI-1 pts received LEN vs SOR. Clinical trial information: NCT01761266. [Table: see text]