Background

The treatment landscape for patients with unresectable hepatocellular carcinoma (uHCC) has recently expanded. However, there is an unmet need for effective treatment options in patients with Child-Pugh (CP) B liver function. Lenvatinib is approved first-line in uHCC based on the REFLECT study results, which evaluated lenvatinib versus sorafenib (patients with CPA liver function were allowed, per inclusion criteria). In a prior analysis of REFLECT, patients treated with lenvatinib benefited irrespective of baseline liver function (ALBI grade 1 or 2; CP score 5 or 6). To determine outcomes in patients with reduced liver function, we conducted a post hoc analysis of patients from REFLECT who progressed to CPB (“CPB patients”) and those who did not (“CPA patients”) within the first 8 weeks of treatment.

Methods

In REFLECT, patients with uHCC were randomly assigned 1:1 to lenvatinib (per bodyweight: 12 mg/day for ≥60 kg; 8 mg/day for < 60 kg) or sorafenib (400 mg twice daily) in 28-day cycles. Here, we assessed objective response rate (ORR) along with landmark analyses (starting at week 8) of progession-free survival (PFS) and overall survival (OS) in CPB patients and in those who remained CPA, 8 weeks postrandomization. Tumors were assessed by mRECIST by independent imaging review. Safety was assessed from baseline, and adjusted by treatment duration.

Results

This analysis included patients in the lenvatinib arm (CPB, n=60 and CPA, n=413) and the sorafenib arm (CPB, n=47 and CPA, n=427). In the lenvatinib arm, ORR was 28.3% (95% CI 16.9–39.7) for CPB patients and 42.9% (95% CI 38.1–47.6) for CPA patients; in the sorafenib arm, ORR was 8.5% (95% CI 0.5–16.5) for CPB patients and 12.9% (95% CI 9.7–16.1) for CPA patients. A landmark analysis (starting at week 8) in the lenvatinib arm showed a median PFS of 3.7 months (95% CI 1.8–7.4) and 6.5 months (95% CI 5.6–7.4) for CPB and CPA patients, respectively; in the sorafenib arm, median PFS was 0.5 months (95% CI 0.1–3.6) and 3.6 months (95% CI 2.7–3.7) for CPB and CPA patients, respectively. In a landmark OS analysis, median OS was 6.8 months (95% CI 2.6–10.3) for CPB patients and 13.3 months (95% CI 11.6–16.1) for CPA patients in the lenvatinib arm; in the sorafenib arm, median OS was 4.5 months (95% CI 2.9–6.1) for CPB patients and 12.0 months (95% CI 10.2–14.0) for CPA patients. In CPB and CPA patients, median durations of treatment were 3.2 months and 6.9 months in the lenvatinib arm, respectively; and 1.9 months and 3.7 months in the sorafenib arm, respectively. Among CPB and CPA patients, the adverse event (AE) rates (number of AE episodes per patient-year) for grade ≥3 treatment-related AE episodes in the lenvatinib arm were 3.65 and 1.41, respectively; and 3.38 and 1.71, respectively, in the sorafenib arm.

Conclusions

This post hoc analysis is limited by its descriptive nature; however, the results indicate that patients with uHCC whose liver function deteriorates to CPB after initiation of therapy may continue to receive lenvatinib, and further study of lenvatinib in CPB patients with uHCC is warranted.