OBJECTIVES: EC/GEJC has a poor prognosis with < 50% of patients with advanced disease surviving at 1 year. Our objective was to determine the cost-effectiveness of adjuvant nivolumab versus current clinical practice (surveillance) in patients with EC/GEJC who have residual disease after neoadjuvant chemoradiotherapy followed by complete resection.

METHODS: Compliant with French Commission for Economic Evaluations and Public Health (CEESP) guidelines, a simplified four-state semi-Markov model was applied with four states: pre-recurrence, locoregional-recurrence, distant-recurrence, and death. Pre- to post-recurrence transition was informed by the time-to-recurrence (TTR) distribution of CheckMate 577 patients with locoregional and distant recurrence. Registry data informed overall survival (OS) after each recurrence type because CheckMate 577 lacks mature OS. A 15-year time horizon simulated population-matched patients included in the trial. Costs and outcomes were discounted 2.5% per year. Robustness of results was evaluated with probabilistic sensitivity analyses (PSA) and scenarios.

RESULTS: Nivolumab was associated with an incremental cost of 48,634€ and incremental QALY gain of 0.98 compared with current clinical practice, resulting in an incremental cost-utility ratio (ICUR) of 49,572€/QALY. Alternative assumptions had limited impact: a maximum 26% ICUR increase at 10-year time horizon; 21% ICUR decrease with a TTR extrapolation simulating pseudo-cure after 6 years; and negligeable impact (−0.1%) using a 3 health-state semi-Markov structure (pre-recurrence, post-recurrence, and death). PSA confirmed base-case ICUR with 80% probability of nivolumab being cost-effective at a willingness-to-pay threshold of €75,000/QALY.

CONCLUSIONS: CEESP accepted this CEA methodology, only criticizing the choice of a constant hazard ratio for post-recurrence survival. Compared with previously evaluated and accepted CEA for immune-checkpoint inhibitors in a metastatic setting, ICUR appears particularly low because of capped treatment duration for nivolumab (1-year stopping rule) and longer pre-recurrence survival associated with quality of life-benefits.