Evolution of Randomized Trials in Advanced/Metastatic Soft Tissue Sarcoma: Endpoint Selection, Surrogacy, and Quality of Reporting
A partir d'une revue systématique de la littérature (52 essais identifiés entre 1974 et 2014, 9 762 patients inclus), cette étude évalue l'intérêt d'utiliser divers critères de jugement intermédiaire dans les essais randomisés de traitements chez des patients atteints d'un sarcome des tissus mous de stade avancé ou métastatique
Résumé en anglais
Purpose Randomized controlled trials (RCTs) in soft tissue sarcoma (STS) have used varying endpoints. The surrogacy of intermediate endpoints, such as progression-free survival (PFS), response rate (RR), and 3-month and 6-month PFS (3moPFS and 6moPFS) with overall survival (OS), remains unknown. The quality of efficacy and toxicity reporting in these studies is also uncertain.
Methods A systematic review of systemic therapy RCTs in STS was performed. Surrogacy between intermediate endpoints and OS was explored using weighted linear regression for the hazard ratio for OS with the hazard ratio for PFS or the odds ratio for RR, 3moPFS, and 6moPFS. The quality of reporting for efficacy and toxicity was also evaluated.
Results Fifty-two RCTs published between 1974 and 2014, comprising 9,762 patients, met the inclusion criteria. There were significant correlations between PFS and OS (R = 0.61) and between RR and OS (R = 0.51). Conversely, there were nonsignificant correlations between 3moPFS and 6moPFS with OS. A reduction in the use of RR as the primary endpoint was observed over time, favoring time-based events (P for trend = .02). In 14% of RCTs, the primary endpoint was not met, but the study was reported as being positive. Toxicity was comprehensively reported in 47% of RCTs, whereas 14% inadequately reported toxicity.
Conclusion In advanced STS, PFS and RR seem to be appropriate surrogates for OS. There is poor correlation between OS and both 3moPFS and 6moPFS. As such, caution is urged with the use of these as primary endpoints in randomized STS trials. The quality of toxicity reporting and interpretation of results is suboptimal.
