Progression-free survival as a potential surrogate end point for overall survival in advanced cervical carcinoma

This systematic review and meta-analysis was performed to study the potential of progression-free survival as a surrogate end point for trials evaluating the use of systemic therapy in advanced cervical cancer.

We performed a systematic review selecting phase II and III randomized trials including locally advanced, recurrent, and metastatic cervical cancer, with mature progression-free survival and overall survival data. Our study adhered to the Reporting Items for Systematic Reviews and Meta-Analyses checklist and was registered in International Prospective Register of Systematic Reviews (CRD42023405604). Exclusions comprised trials involving adjuvant treatment as the primary end point. The magnitude of progression-free survival and overall survival was assessed using standardized z-scores. Subgroup analyses were conducted based on treatment type, line of treatment, and prior radiotherapy exposure. Surrogacy was evaluated according to the recommendations by the Institut für Qualität und Wirtschaftlichkeit im Gesundheitswesen.

A total of 20 studies were included in the final analysis. In the overall population, a moderate correlation between progression-free survival and overall survival was observed (r = 0.75, p < .001). The surrogate threshold effect (STE) indicated a threshold for progression-free survival z-score (z-progression-free survival of 2.53), intersecting with null overall survival outcome (z = 1.96). Subgroup analysis revealed a weak and nonsignificant correlation in chemotherapy trials (r = 0.5, p = .12) with a higher STE (z-progression-free survival = 2.83). Conversely, chemoimmunotherapy combinations exhibited a robust and statistically significant correlation (r = 0.99, p = 0.01) with a lower STE (z-progression-free survival = 2.39). Trials exploring advanced therapy lines demonstrated a higher and more significant correlation (r = 0.98, p = .02) with a lower STE at 2.08, whereas upfront-line therapy trials showed a moderate correlation (r = 0.67, p = 0.01) with an STE magnitude of 2.58.

Progression-free survival exhibits a moderate correlation with a modest STE. For chemoimmunotherapy combinations, there is a strong correlation between overall survival and progression-free survival, with a notably lower STE. This suggests that the relationship between progression-free survival and overall survival may vary significantly based on the treatment strategy.