PCN188 - ASSESSING THE ROBUSTNESS OF A COST-EFFECTIVENESS ANALYSIS OF LENVATINIB VERSUS SORAFENIB IN UNRESECTABLE HEPATOCELLULAR CARCINOMA IN JAPAN

Abstract

In the REFLECT trial, lenvatinib demonstrated non-inferiority in overall survival (OS) compared with sorafenib in the first-line treatment of unresectable hepatocellular carcinoma (uHCC). At baseline, a greater proportion of lenvatinib patients had an α-fetoprotein (AFP) level of ≥ 200 ng/mL, an adverse prognostic factor. We evaluated the cost-effectiveness of lenvatinib versus sorafenib in uHCC in Japan over a lifetime horizon, adjusting for this imbalance in patient AFP levels. A partitioned survival model was developed with 3 health states: progression-free, post-progression, and death. Clinical outcomes, adverse events, and health state utilities were obtained from the REFLECT trial. Direct medical costs were based on Japanese data. We carried out deterministic sensitivity analysis (DSA), probabilistic sensitivity analysis (PSA), and scenario analyses to explore the potential influences of the different parameters in our analysis. The estimated quality-adjusted life years (QALYs) for lenvatinib were 1.46 QALY, and 1.23 QALY for sorafenib, respectively. The total costs of lenvatinib therapy were estimated at 5,088,957 JPY, while the sorafenib costs were 5,495,264 JPY. Therefore, treatment with lenvatinib led to 0.23 QALY improvement and lowered costs by 406,307 JPY. In DSA, the three most significant model parameters were OS curves, progression-free survival curves, and progression-free utility; varying these parameters still resulted in lenvatinib being favourable over sorafenib. In PSA, lenvatinib was favourable to sorafenib in 81% of 1,000 simulations, at the 5 million JPY willingness-to-pay per QALY threshold and 77% at the 10 million JPY per QALY threshold. Our conclusions were robust in scenario analyses. Results of this analysis suggested lenvatinib offered improved health outcomes (i.e., QALYs) with lower costs for uHCC patients compared with sorafenib. Our results were robust to sensitivity analyses and scenario analysis.