No CrossRef data available.
Published online by Cambridge University Press: 10 July 2025
Background: Glioma trials may use selective criteria, limiting their generalizability to real-world patients. This systematic review and meta-analysis quantifies the prevalence of these criteria and evaluates their impact on trial outcomes, assessing whether reducing selectivity to improve generalizability and applicability is feasible without compromising safety or efficacy. Methods: 51 glioma trials were extracted from the National Clinical Trial (NCT) database on June 1st, 2024. Eligibility criteria were classified as selective—defined as likely to exclude patients who could benefit, or generalizable—justified due to potential harm or trial focus. The selective criteria were analyzed for correlation with median overall survival (mOS). Results: The average number of selective criteria per study was 6.8 (range: 0–14, median: 7). The most common were “No prior malignancy with a specified disease-free period” (N=29), “Exclusion based on Karnofsky score” (N=27), and “No prior brain radiotherapy” (N=16). Meta-analysis showed no significant correlation between the number of selective criteria and mOS (p = .327). Conclusions: Selective criteria are common in glioma trials, particularly exclusions based on prior malignancies, performance status, and past treatments. However, their lack of correlation with mOS indicates minimal impact on outcomes. These findings suggest reducing selectivity in trial criteria may improve generalizability and applicability without compromising safety or efficacy.