Abstract
BACKGROUND: Although patients with Eastern Cooperative Oncology Group performance status (PS) of 2 constitute a significant proportion of the cancer population, they are often excluded from pivotal clinical trials owing to presumed higher risks of treatment effect dilution, toxicity, and lower compliance. Here, we conducted a systematic review and meta-analysis to evaluate the impact of including PS 2 participants on efficacy and safety outcomes in pivotal cancer clinical trials. MATERIALS AND METHODS: We searched the 'Oncology/Hematologic Malignancies Approval Notifications' and 'Drugs@FDA' databases for clinical trials supporting 'Food and Drug Administration' anticancer drug approvals from 1 January 2009 to 31 December 2024. Eligible studies were randomized phase III clinical trials of systemic therapies for metastatic solid tumors permitting the inclusion of PS 2 participants. We assessed efficacy outcomes [progression-free survival (PFS) and overall survival (OS)] and safety outcomes [occurrence of any-grade adverse events (AEs), high-grade AEs, serious AE (SAEs), AE-related deaths, and treatment modifications] in the included studies. RESULTS: Thirty-six trials were included. In subgroup analyses, no statistically significant differences were found between PS 2 and PS ≤1 participants for PFS [hazard ratio (HR) 0.45, 95% confidence interval (CI) 0.30-0.69 versus HR 0.52, 95% CI 0.41-0.66, P = 0.59] and OS (HR 0.81, 95% CI 0.68-0.97 versus HR 0.71, 95% CI 0.66-0.77, P = 0.18). In meta-regression analyses, no significant associations were found for efficacy outcomes. However, a higher proportion of PS 2 participants was significantly associated with an increased risk of SAEs, AE-related deaths, and treatment discontinuations. CONCLUSIONS: Although PS 2 participants showed a greater propensity to serious toxicity, no significant differences in efficacy outcomes were observed compared with those with PS ≤1. Our results support the inclusion of PS 2 participants in clinical trials, as their exclusion limits the generalizability of results.