OBJECTIVES: New treatments’ benefit is evaluated, when they are marketed, by health authorities based on clinical trials results at individual patient level. However, treatments outcomes are rarely re-assessed in real-world practice and/or at the population level. The objectives are to present a methodology to assess population-based clinical benefit of treatments.

METHODS: In France, the Haute Autorité de Santé (HAS) publishes medico-technical and cost-effectiveness (CE) assessments of all new indication of drugs. In CE assessments, extrapolated Kaplan-Meier survival curves over a lifetime horizon and utility values are available. The Hospital Reimbursement database (PMSI) allows to identify the number of patients initiating a given drug and to follow their use over time. The date of availability of each drug by indication is published in the Official gazette, as well as if it has benefited of early access. Once consolidated, this public dataset gives an opportunity to generate retrospective nationwide population-based estimation of drugs outcomes, expressed as deaths prevented (DP), life years (LY) and quality-adjusted life years (QALYs), by monthly incident cohorts.

RESULTS: Between 2014 and 2021, 5 immunotherapies (avelumab, atezolizumab, durvalumab, nivolumab and pembrolizumab) were used in France in 21 indications in 8 tumor localizations. Based on the PMSI reports, we identified 132,924 patients who initiated an immunotherapy during the period. Using the extrapolated curves, we estimated 16,173 [13,803–17,141] deaths were prevented compared to previous standard of care by December 31^st of 2021. Compared to standard treatments, immunotherapies allowed 37,318 [33,583–41,053] LYG and 27,710 [23,785–30,451] QALYs. We were able to calculate the results per drug, tumor localization and line.

CONCLUSIONS: This methodology is an opportunity to estimate the clinical benefit of a drug (or a therapeutic class) at a nationwide population level.