Presentation (CTI)

OBJECTIVES: Progression-free survival (PFS) is a common primary endpoint in first-line epithelial ovarian cancer (EOC) trials; however, its relationship with overall survival (OS) remains uncertain. This study evaluated the patient-level PFS-OS association among individuals with BRCA wild-type (wt) advanced (FIGO-stage III/IV) EOC treated with first-line platinum-based chemotherapy (PBC).
METHODS: This retrospective analysis pooled patient-level data from first-line advanced EOC trials sourced from the Medidata Enterprise Data Store. Eligible patients had BRCAwt disease and initiated PBC (index date) after primary debulking surgery (PDS) or as neoadjuvant therapy before planned interval debulking surgery (IDS). The PFS-OS association was evaluated using (1) Spearman's correlation (ρ) and (2) landmark analyses at pre-specified timepoints (9, 12, 15, and 18 months).
RESULTS: Among 487 patients (50.1% aged ≥65 years; 90.1% white; 63.7% FIGO-stage III; 54.6% ECOG-PS 0; 37.3% PDS; 53.0% IDS; 9.7% planned IDS but did not undergo; median follow-up 3.4 years [IQR, 2.7]), the median PFS and OS were 15.6 months (95% CI, 13.8-16.8) and 42.1 months (95% CI 37.8-45.7), respectively. The PFS-OS correlation (ρ) was 0.7 (95% CI 0.6-0.8). Across all landmark timepoints, patients who remained progression-free at landmark experienced longer subsequent OS; at the 15-month landmark (highest concordance index), the unadjusted HR was 0.29 (95% CI 0.22-0.37). Associations remained after covariate adjustment and did not vary by best overall response, bevacizumab use, or surgical type/outcomes.
CONCLUSIONS: In this patient population, PFS demonstrated a consistent association with OS across multiple analytical approaches. However, patient-level associations alone are insufficient to validate PFS as a surrogate for OS; additional research across a larger number of RCTs is warranted to assess trial-level surrogacy.