OBJECTIVES : To assess the application of progression-free survival (PFS) as surrogate endpoint for overall survival (OS) within cost-effectiveness models appraised by National institute for Health and Care Excellence (NICE) given the consistent clinical evidence supporting correlation between PFS and OS in advanced breast cancer.

METHODS : Using published NICE reports, 14 unique technology appraisals (TAs) in advanced breast cancer were identified from January 2012 to May 2021. Full-text screening of the committee papers and TA guidance was conducted by a single investigator and the abstracted data were validated by a second investigator. Information on drug of interest, specific indication, model structure, source of PFS and OS, approach for long-term OS projection, and committee commentaries was extracted.

RESULTS : Of the 14 final appraisals, only three used PFS to predict long-term OS beyond trial period. Although this approach has generally been accepted by the committee, given the complex and difficult-to-predict relationship between PFS and OS, the committee supported a partial surrogacy between the two endpoints—that is, an improved PFS is likely to be partially (27.5%–100%) translated into an OS gain. Approximately half of the remaining 11 appraisals acknowledged the OS data from trial were immature and projected long-term OS based on direct extrapolation of trial data using parametric fittings and informed by external data in some cases. Both approaches did not preclude the validity of long-term OS estimates as a key consideration by the committee.

CONCLUSIONS : In the absence of mature trial data on OS, to derive an estimate of long-term OS within the economic analysis surrogate measure provides an alternative to directly projecting OS from trial data in advanced breast cancer; however, it has not been commonly used in the existing models appraised by NICE. Strong evidence is warranted to facilitate the acceptance of this approach.