Prevalence of Immature Survival Data for Anti-Cancer Drugs Presented to the National Institute for Health and Care Excellence and Impact on Decision Making

Abstract

Objectives: This research aims to explore how often the National Institute for Health and Care Excellence (NICE) uses immature overall survival data to inform reimbursement decisions on cancer treatments, and the implications of this for resource allocation decisions.

Methods: NICE cancer technology appraisals published between 2015 and 2017 were reviewed to determine the prevalence of using immature survival data. A case study was used to demonstrate the potential impact of basing decisions on immature data. The economic model submitted by the company was reconstructed and was populated first using survival data available at the time of the appraisal, and then using data from an updated data cut published after the appraisal concluded. The incremental cost-effectiveness ratios (ICERs) obtained using the different data cuts were compared. Probabilistic sensitivity analysis was undertaken and expected value of perfect information estimated.

Results: Forty-one percent of NICE cancer technology appraisals used immature data to inform reimbursement decisions. In the case study, NICE gave a positive recommendation for a limited patient subgroup, with ICERs too high in the complete patient population. ICERs were dramatically lower when the final data cut was used, irrespective of the parametric model used to model survival. Probabilistic sensitivity analysis and expected value of perfect information may not have fully characterized uncertainty, because as they did not account for structural uncertainty.

Conclusion: Analyses of cancer treatments using immature survival data may result in incorrect estimates of survival benefit and cost-effectiveness, potentially leading to inappropriate funding decisions. This research highlights the importance of revisiting past decisions when updated data cuts become available.

Keywords: cost-utility analysis; health technology assessment; immature survival data; oncology; survival analysis; survival extrapolation.