Surrogate endpoints in metastatic breast cancer treated with targeted therapies: an analysis of the first-line phase III trials

Progression-free survival (PFS) has not yet been established as a surrogate endpoint for overall survival (OS) in metastatic breast cancer (MBC). In particular, surrogacy has not been investigated with modern molecular agents. Randomized phase III trials for MBC were identified using a PubMed and EMBASE search. Correlations between PFS/time to progression (TTP), post-progression survival (PPS), response rate (RR) and OS were evaluated through a linear regression analysis. We also evaluated the potential of PFS to serve as a surrogate for OS in first-line trials exploring new drugs. Twenty trials with a total of 32 treatment arms and 10,138 patients were included. Spearman rank correlation coefficients (r(s)) between median PFS/TTP, RR and PPS with OS were 0.81 (95 % CI 0.58-0.92), 0.61 (95 % CI 0.59-0.63) and 0.73 (95 % CI 0.71-0.74). The correlation coefficient between hazard ratios in PFS/TTP and OS (HRPFS/TTP/HROS) was 0.73 (95 % CI 0.719-0.738; p < 0.00001); the slope of the regression line was 0.56 +/- 0.0034, indicating that an agent producing a 10 % risk reduction for PFS will provide a 5.6 +/- 0.34 % risk reduction for OS. In particular, the HRPFS/TTP/HROS correlation is stronger in an HER2-positive setting versus HER2-negative (r(s) = 0.91 vs. 0.67; p for difference <0.0001). These results suggest that improvements in PFS/TTP in MBC strongly correlate with improvements in OS with molecular agents, especially anti-HER2 therapeutics. Further analyses at patient-level data are required to confirm the surrogacy of PFS for OS with new agents.