Review of All Solid Tumor Drug Approvals From 2019 to 2024 by US Food and Drug Administration, European Medicines Agency, and Brazilian Health Regulatory Agency.

Purpose: Regulatory agencies play a pivotal role in evaluating clinical evidence for oncology drug approvals. This study aimed to compare approved indications across the US Food and Drug Administration (FDA), European Medicines Agency (EMA), and Brazilian Health Regulatory Agency (ANVISA), with a focus on the robustness of the supporting evidence and consistency between labeling and clinical trial data.

Methods: A systematic review identified all new drugs and indications for solid tumors approved by the FDA from January 1, 2019, to December 31, 2024. Each approval was cross-referenced with corresponding decisions by EMA and ANVISA. For each indication, the pivotal trial's design, primary end point, study population, and the concordance between the approved label and the clinical evidence were identified.

Results: During the 6-year period, 199 new indications for solid tumors were approved by the FDA; 138 (69.3%) were also approved by EMA and 124 (62.3%) by ANVISA. Discrepancies between approved labels and the primary end point population occurred in 10.0% of FDA, 18.8% of EMA, and 12.9% of ANVISA approvals, most often due to label restrictions to subgroups, indicating a more conservative approach by EMA and ANVISA. Labeling differences were noted in 19 of 139 FDA-EMA and 10 of 124 FDA-ANVISA shared approvals. A total of 190 pivotal trials supported these approvals, including three phase I, 59 phase II, and 128 phase III studies. The most frequent end points were overall response rate and progression-free survival, including in phase III trials and regular approvals, rather than overall survival or quality of life.

Conclusion: Notable differences in timelines, regulatory mechanisms, and evidentiary thresholds across agencies result in divergent labeling. These variations have clinical and policy implications for oncology drug access and adoption worldwide.