The Acceptance of Overall Survival Extrapolation Methods in Solid Tumor Treatments by Health Technology Assessment Agencies in England, France, and Australia between 2017 and 2022

Background Survival extrapolation is used to predict patients’ overall survival beyond clinical trial follow-up. It can significantly affect the results of a cost-effectiveness analysis and subsequent pricing and reimbursement decisions. However, selecting an appropriate model involves subjectivity, leading to discrepancies between methods submitted by manufacturers and those accepted by health technology assessment (HTA) agencies. This review describes the acceptance and criticisms of overall survival extrapolation methods by HTA agencies in England, France, and Australia. Methods Electronic searches conducted on September 4, 2022, identified HTA evaluations for oncology therapies indicated for the treatment of solid tumors from the National Institute for Health and Care Excellence (NICE) in England, the Haute Autorité de Santé (HAS) in France, and the Pharmaceutical Benefits Advisory Committee (PBAC) in Australia, published between August 2017 and August 2022. Information on the overall survival extrapolation model submitted by the manufacturer was extracted. The acceptance decision of the HTA agency and the key criticisms were also recorded. Results A total of 140 HTA evaluations were identified. The initial overall survival extrapolation method was accepted in 21% of cases. The most frequently cited criticisms related to a lack of or inappropriate incorporation of treatment effect waning over time (31%). Other criticisms included choice of parametric distribution, in which multiple distributions were often considered plausible (24%); immaturity of survival data (15%); and concerns about the proportional hazards assumption, which lacked clinical validity (8%). Conclusion This review highlights the low acceptance of survival extrapolation methods and the areas of discordance between manufacturers and HTA agencies in England, France, and Australia. Low acceptance rates of survival extrapolation methods by HTA bodies can affect pricing and reimbursement decisions of new therapies, delaying patient access. Highlights This review found that the survival extrapolation methods initially submitted by companies were accepted in only 21% of cases. The most common areas of discordance between companies and HTA agencies were inappropriate modeling of treatment effect over time, choice of parametric distribution, immaturity of survival data, and concerns about the proportional hazards assumption. There is a need for more consistent guidance on the selection of an appropriate extrapolation method to limit the inherent subjectivity surrounding survival curve selection.