Systematic Review and Meta-Analysis of Correlation of Progression-Free Survival-2 and Overall Survival in Solid Tumors

Background: The use of progression-free survival (PFS)2, time from randomization to 2nd disease progression or death from any cause, is proposed as a surrogate endpoint for overall survival (OS) in oncology clinical trials. We used published data from solid tumor clinical trials to assess whether PFS2 and OS are correlated. Methods: A literature search identified studies that included data for PFS, PFS2 and OS. Two reviewers screened results for eligibility, and documented PFS2, PFS or time from 1st to 2nd disease progression or death and OS. Correlation between PFS2 and OS was assessed and verified using: (1) Kendall’s Tau statistics + Pearson’s correlation coefficient in randomized controlled trials (RCTs); (2) Meta-analysis with the random effects model to compute the pooled correlation of PFS2 and OS. Results: Overall, 133 studies met search criteria, 15 had complete PFS2 and OS data, yielding 28 individual arms in ovarian, gastric, colorectal, prostate, lung, renal and breast cancer tumors. A positive correlation for PFS2 and OS (Kendall’s Tau = 0.7 (95% CIs 0.54, 0.78) was found using all 15 studies. Data from 10 RCTs showed a Pearson’s correlation coefficient of 0.86. Meta-analysis of data from 7 trials with ≥1 study identified a pooled Spearman’s correlation coefficient of 0.84 (p=0.0001; 95% CIs 0.71, 0.96). Conclusions: In this retrospective analysis in solid tumors, PFS2 strongly correlates with OS. The strong correlation allows PFS2 to be confidently used to show long term clinical benefit ahead of mature OS or when OS cannot be assessed.