Discrepancies in therapeutic indications for the same oncology drugs among FDA, EMA, and Brazilian agency.

Abstract

e23035 Background: Divergent marketing approvals for oncology drugs among prominent agencies can prompt questions, especially when they are grounded in the same evidence. The aim of this study was to evaluate and compare the assessment and approval processes for novel products and indications by the FDA, EMA, and the Brazilian Health Regulatory Agency (ANVISA). Methods: Therapeutic indications for FDA-approved anticancer drugs from January 2019 to December 2023 were identified. Data, including approval dates, details on authorization procedures and evidence presented, were extracted from the databases of the FDA, EMA, and ANVISA. Only approvals for non-hematologic solid tumors were considered. All data were publicly available, and ethical approval was not required. No patient involvement occurred in this study. Results: Between 2019 and 2023, the FDA granted approval for 154 therapeutic indications for marketing authorization, resulting from 148 pivotal trials. The predominant cancer types during this period were non-small cell lung cancer (31 approvals [20.1%]). Most trials were in phase III (66.8%) and randomized (68.1%). Additionally, 41 (26.6%) received accelerated approvals, with 6 subsequently obtaining regular approvals. Among the therapies, 66 trials (42.8%) involved active comparators, while 53 (34.4%) and 35 (22.7%) were single-arm and placebo-controlled trials, respectively. Approximately half (74 [48%]) of the therapies were approved for metastatic first-line treatment, 54 (35%) for second-line treatment, and the rest for neoadjuvant and/or adjuvant treatment. Regarding endpoint characteristics supporting approval, the most prevalent was a combination of objective response rate (ORR) + duration of response (DoR) (38 [24.6%]), followed by progression-free survival (PFS) + overall survival (OS) (31 [20.1%]). When comparing the 154 FDA-approved indications, 50 (32.4%) were not approved by EMA, and 60 (38.9%) were not approved by ANVISA. Furthermore, in 33 cases (21.4%), there were disparities in clinical indications between the FDA and EMA, primarily related to the target population specification, with 10 cases (6.4%) attributed to programmed death ligand-1 (PD-L1) expression. Notably, in these instances, the FDA approved the drug regardless of PD-L1 expression, while the EMA required a percentage of cells with PD-L1 expression. In all these cases, ANVISA aligned with the FDA's assessment. Conclusions: Significant clinical disparities were identified in the outcomes of the approval process for oncology products among the FDA, EMA, and ANVISA. Additional efforts are needed to standardize decision-making across regulatory systems.