Background: The randomized phase 3 REFLECT trial (NCT01761266) demonstrated that lenvatinib was non-inferior to sorafenib in OS in 1L uHCC (HR, 0.92; 95% CI, 0.79-1.06). PFS (HR, 0.64; 95% CI, 0.55-0.75; p<0.0001) and ORR (odds ratio, 5.01; 95% CI 3.59-7.01; p<0.0001) by IIR per mRECIST favored lenvatinib versus sorafenib. Recent data suggest that viral/nonviral etiology may impact treatment outcomes. This post-hoc analysis evaluated patients with nonviral etiology in REFLECT.
Methods: In REFLECT, 1L uHCC patients were randomized to lenvatinib (12 mg/day, bodyweight ≥60 kg; 8 mg/day, bodyweight <60 kg) or sorafenib (400 mg twice-daily) in 28-day cycles. This post-hoc analysis included patients without hepatitis B/C (medical history) who were randomized to receive lenvatinib or sorafenib. PFS, ORR (by IIR per mRECIST), and OS were analyzed.
Results: 127 Patients randomized to lenvatinib and 108 patients randomized to sorafenib had nonviral etiology. Among these patients, mOS was 13.8 months (95% CI, 10.5-18.7) with lenvatinib and 13.9 months (95% CI, 11.7-17.5) with sorafenib (HR, 1.03; 95% CI, 0.75-1.43). mPFS was 7.4 months (95% CI, 5.5-8.7) in the lenvatinib arm and 4.0 months (95% CI, 3.6-5.5) in the sorafenib arm (HR, 0.60; 95% CI, 0.42-0.87). ORR was 39.4% (95% CI 30.9-47.9) in the lenvatinib arm and 20.4% (95% CI 12.8-28.0) in the sorafenib arm. Fewer (n=34 [26.8%]) patients with nonviral etiology randomized to lenvatinib received anticancer medication during survival follow-up than those randomized to sorafenib (n=46 [42.6%]).
Conclusion: In this post-hoc analysis, OS, PFS, and ORR in patients randomized to lenvatinib with nonviral etiology were consistent with the overall population of patients randomized to lenvatinib. These results, along with the results of the primary analysis of REFLECT, demonstrate the efficacy of lenvatinib regardless of etiology in uHCC, and support lenvatinib as a standard-of-care-option for patients with 1L uHCC. Prospective trials by etiology are still needed.