Comparative cost-effectiveness of nivolumab first-line and second-line therapy for advanced esophageal cancer in Japan.

Abstract

A model-based cost-effectiveness analysis comparing first-line and second-line nivolumab therapy for advanced esophageal cancer was performed to support public healthcare in Japan. A partitioned survival model was developed to predict costs and outcomes. Survival data were obtained from two phase 3 clinical trials (Attraction-3 and Checkmate-648), and direct medical costs were estimated from the perspective of the Japanese National Health Insurance payer. The time horizon for the model was set to 20years. Health outcomes were calculated and defined as quality-adjusted life years (QALYs), and incremental cost-effectiveness ratios (ICER) were compared to those of control therapy. A sensitivity analysis was performed based on parameter settings and model uncertainties. A willingness-to-pay threshold of 15 million Japanese yen (JPY) was established. Compared to that of each control therapy, the ICER for nivolumab per QALY gained was 15,712,265 JPY (143,099 USD) for first-line combination therapy with chemotherapy in the overall population, 10,657,085 JPY (97,059 USD) in the population with ≥ 1% Programmed Death-Ligand 1 (PD-L1) expression, and 41,184,322 JPY (375,085 USD) for second-line nivolumab monotherapy. A probabilistic sensitivity analysis estimated that nivolumab was cost-effective as a first-line therapy for the overall population (61.5%) and for the population with ≥ 1% PD-L1 expression (76.5%), but not as second-line monotherapy (32.3%). Nivolumab is recommended as a first-line therapy in combination with chemotherapy owing to its cost-effectiveness, but not as a second-line monotherapy. Patient selection based on PD-L1 expression may help to improve the cost-effectiveness of using nivolumab as a first-line treatment.