Research Summary

PFS Shows Moderate Correlation With OS but Limited Surrogacy in Phase 3 Mantle Cell Lymphoma Trials

Edited by:
Ashton L. Stahl

Key Highlights

  • PFS and OS treatment effects showed a moderate association in phase 3 MCL trials.
  • A large PFS benefit was needed to confidently predict an OS benefit.
  • Only 17.6% of evaluable comparisons showed significant benefit in both PFS and OS.
  • Authors recommended caution when using PFS as a standalone surrogate endpoint.

Progression-free survival (PFS) showed only moderate reliability as a surrogate for overall survival (OS) in phase 3 mantle cell lymphoma (MCL) trials, according to a trial-level surrogate endpoint analysis published in Clinical Lymphoma, Myeloma and Leukemia. The authors reported that although PFS and OS treatment effects were correlated, PFS improvements did not consistently translate into statistically significant survival benefits.

The investigators conducted a systematic review of ClinicalTrials.gov and the EU Clinical Trials Register from database inception through December 29, 2025. The eligible studies were phase 3 MCL trials in adults that reported hazard ratios (HRs) for both PFS and OS. The final analysis included 16 trials, 19 treatment comparisons, and 5,996 participants. Researchers used weighted linear regression, with trial size as the weighting factor, to assess the association between treatment effects for PFS and OS.

Study Findings

The weighted regression model showed a correlation coefficient of 0.86 and an R² of 0.75, indicating that 75% of the variance in OS treatment effects was accounted for by changes in PFS treatment effects. Based on the authors’ predefined thresholds, this result represented a moderate association between PFS and OS in phase 3 MCL trials.

However, the surrogate threshold effect suggested that only substantial PFS benefits could reliably predict OS benefit. For a hypothetical average-sized trial of 334 patients, the PFS HR needed to be less than 0.59 to predict a nonzero OS benefit with 95% certainty. For trials at the 25th and 75th percentiles of the sample size distribution, the surrogate threshold effects were 0.51 and 0.63, respectively.

Among 17 treatment comparisons with confidence intervals for both endpoints, only 3 (17.6%) demonstrated statistically significant improvements in both PFS and OS. Ten comparisons (58.8%) showed a statistically significant PFS benefit without a statistically significant OS benefit, and 4 comparisons (23.5%) showed no benefit for either endpoint.

Clinical Implications

According to the study authors, the findings suggest that PFS should be interpreted cautiously as a surrogate endpoint in MCL, particularly when treatment effects are modest. The authors noted that future endpoint selection may require evaluation of alternative intermediate endpoints, high-risk subgroup stratification, and patient-centered measures, such as quality-of-life outcomes. Limitations included the use of aggregate trial-level data rather than individual patient data, heterogeneity across trials, and a limited number of treatment comparisons.

Expert Commentary

“This surrogate endpoint evaluation shows that PFS and OS are moderately correlated in MCL clinical trials. However, the substantial magnitude of PFS benefit required to confidently predict survival improvement limits its reliability as a standalone surrogate endpoint, particularly when treatment effects are modest,” the researchers concluded.

Reference

Gorzewski AM, Raker C, Gill VA, Palladino T, Kota SP, Pelcovits AR. Association between progression-free survival and overall survival in mantle cell lymphoma: a trial-level surrogate endpoint analysis. Clin Lymphoma Myeloma Leuk. Published online May 23, 2026. doi:10.1016/j.clml.2026.05.001