Drug Evaluation Committee Flexible Survival Time Modeling Methodology and Extrapolation of Survival Time Using External Information
June 2023
The UK’s NICE (National Institute for Health and Care Excellence), which adopted cost-effectiveness analysis early on, has published 22 Technical Support Documents (TSDs) to date on statistical methods and other topics related to health economic evaluation.These documents serve as valuable references for Japan, which officially adopted cost-effectiveness analysis in 2019.
The Data Science Subcommittee’s FY2022 Continuing Task Force 7 is organizing the topics covered in NICE’s TSDs and providing information deemed useful for conducting cost-effectiveness evaluations in Japan.This report, based on TSD 21, “Flexible methods for survival analysis,” published in 2020, summarizes flexible survival time modeling and methods for extrapolating survival times using external information, along with case examples.
Traditionally, parametric models using standard distributions such as the exponential or Weibull distributions were sufficient for survival time modeling. However, in recent years, with the advent of immunotherapy and other treatments, there has been an increase in cases where survival curves exhibit complex shapes due to delayed responses to treatment and the presence of long-term survivors.Conventional standard models are insufficient to represent such complex shapes, necessitating more flexible survival time models. Furthermore, extrapolation beyond the follow-up period of a clinical trial requires strategies such as incorporating external data. This report introduces these methodologies and provides specific case studies to help readers better visualize the key considerations when performing modeling.
We hope that this report will serve as a useful reference when conducting survival time modeling or extrapolation using external data, assisting you in selecting appropriate methods and performing analyses.
Japan Pharmaceutical Manufacturers Association, Drug Evaluation Committee,
, Data Science Subcommittee, FY2022 Continuing Task Force 7
