Background
REFLECT: open-label, phase 3 non-inferiority trial in 954 patients with unresectable hepatocellular carcinoma, Child-Pugh A, BCLC stage B or C, randomized 1:1 to lenvatinib or sorafenib.
Interventions and follow up
Arm A: Lenvatinib 12 mg/day (body weight ≥60 kg) or 8 mg/day (<60 kg)
Arm B: Sorafenib 400 mg PO BID
Primary endpoint: OS (non-inferiority)
mFollow up: NR
Arm B: Sorafenib 400 mg PO BID
Primary endpoint: OS (non-inferiority)
mFollow up: NR
Results
OS: 13.6mo vs 12.3mo; HR 0.92, 95% CI 0.79-1.06 (met non-inferiority)
PFS: 7.4mo vs 3.7mo; HR 0.66, 95% CI 0.57-0.77; P<.0001
PFS: 7.4mo vs 3.7mo; HR 0.66, 95% CI 0.57-0.77; P<.0001
Adverse events
Overall: Treatment-related grade ≥3 events 57% vs 49% (Arm A vs B)
Vascular/dermatologic: Hypertension 42% vs 30%; palmar-plantar erythrodysesthesia 27% vs 52% (Arm A vs B)
Constitutional/GI: Decreased appetite 34% vs 27%; weight loss 31% vs 22%; fatigue 30% vs 25%; diarrhea 39% vs 46% (Arm A vs B)
Vascular/dermatologic: Hypertension 42% vs 30%; palmar-plantar erythrodysesthesia 27% vs 52% (Arm A vs B)
Constitutional/GI: Decreased appetite 34% vs 27%; weight loss 31% vs 22%; fatigue 30% vs 25%; diarrhea 39% vs 46% (Arm A vs B)
Conclusions
Lenvatinib was non-inferior to sorafenib for overall survival in untreated advanced hepatocellular carcinoma, with superior PFS and response rate.
Key Limitations
Non-inferiority (not superiority) design; main portal vein invasion and >50% liver involvement excluded; limited to Child-Pugh A; open-label assessment.
Clinical Context
REFLECT led to FDA and EMA approval of lenvatinib for first-line unresectable HCC (2018). ASCO/ESMO list lenvatinib as a preferred first-line TKI alternative to atezolizumab/bevacizumab, particularly when immunotherapy is contraindicated.
References
Kuda M et al, Lancet, 2018, PMID: 29433850
Llovet JM et al, NEJM, 2008 (SHARP), PMID: 18650514
Llovet JM et al, NEJM, 2008 (SHARP), PMID: 18650514