Background The phase 3 REFLECT study (NCT01761266) established lenvatinib's noninferiority to sorafenib for unresectable hepatocellular carcinoma (uHCC). This subanalysis examines patients treated with lenvatinib according to the depth of their tumor response and their status as an alpha-fetoprotein (AFP) responder/nonresponder. Methods Of 478 lenvatinib-treated patients, 194 achieved an objective response by mRECIST per independent imaging review. Patients with tumor response were further categorized by their maximal tumor reduction (assessed every 8 weeks). Patients were also assessed according to their AFP response (≥20% reduction from baseline by week 8) status, including AFP responders (n = 227) and nonresponders (n = 73). Results Median duration of response in patients with ≥75% (n = 59)/≥50% to <75% (n = 72)/≥30% to <50% (n = 63) tumor reduction was 9.1 months (95% CI:7.4-9.3)/7.3 months (95% CI:5.5-7.4)/3.7 months (95% CI:3.7-5.6), respectively. Median PFS/OS were 11.0 months (95% CI:9.3-12.9)/23.4 months (95% CI:14.3-30.1) for ≥75% tumor reduction, 9.2 months (95% CI:7.4-11.1)/19.8 months (95% CI:14.1-23.1) for ≥50% to <75% tumor reduction, and 7.4 months (95% CI:5.5-9.2)/14.4 months (95% CI:13.1-19.1) for ≥30% to <50% tumor reduction, respectively. Efficacy was improved among AFP responders versus AFP nonresponders (objective response rate: 48.0% versus 13.7%; median PFS, 7.4 months 95% CI: 5.6-7.8 versus 3.5 months 95% CI: 1.9-3.7; median OS, 13.4 months 95% CI: 11.5-14.3 versus 8.3 months 95% CI: 6.5-10.7). Patients with baseline AFP levels <400 ng/mL had longer median OS (19.0 months; 95% CI: 14.6-22.5) versus those with ≥400 ng/mL (10.1 months 95% CI: 8.5-11.7). Conclusions This analysis highlights the importance of tumor reduction and AFP response as predictors of survival outcomes in lenvatinib-treated patients with uHCC. These data continue to support lenvatinib as an effective first-line treatment option.
Mahipal et al. (2026) studied this question.