Dosage Optimization in Drug Development: an FDA Project Optimus Analysis of Post-Marketing Requirements Issued to Repair the Cracks

1598 Background: Dosage optimization (DO) is a critical aspect of oncology drug development. Because of this, the US FDA, through Project Optimus, has focused on reforming the DO and dosage selection paradigm for oncology products. When approving products, FDA may determine that although the recommended dosage possesses a favorable benefit:risk (BR) profile, there are significant safety concerns that require evaluation of alternative dosages to potentially improve the BR profile. Such postmarketing requirements (PMRs)—issued as a condition of approval—are examined here to better understand the characteristics and potential limitations of DO studies in the postmarketing setting. Methods: Between Jan. 1, 2010 to Dec. 31, 2022, oncology new molecular entities (NMEs) and original applications for biologics, hereafter referred to as new drugs, approved by CDER with at least one PMR issued for DO were identified in the FDA electronic records database. The designs and results of the associated trials or analyses for these PMRs, along with the status of the PMRs, were collected and evaluated. Results: There were 24 PMRs issued for 21 new drugs (15% of 138 total approved new drugs). The most common drug classes were: tyrosine kinase inhibitors (TKIs) (10), antibody drug conjugates (ADCs) (3), or GTPase Inhibitors (2). The following trials or analyses were planned to address these PMRs: randomized dose-comparison trials (18), non-randomized dose-finding trials (2), food effect trials (2), exposure-response analysis (1), and pending (1). The planned sample sizes for the 18 dose-comparison trials were 3 with <100 patients, 4 with 100 to 199 patients, 7 with 200 to 500 patients, and 4 with >500 patients. The dose-comparison trials used the following (co)primary endpoints: response rate (12), safety (7), OS (3), and PFS (2). Fourteen PMRs were either fulfilled (11) or released (3). Of the 11 fulfilled PMRs, 8 led to dosage changes for 5 new drugs (cabazitaxel, carfilzomib, ponatinib, ceritinib, and pexidartinib), and 3 did not lead to dosage changes. Ten PMRs are ongoing. The median time from issuance to fulfillment or release (not including ongoing PMRs) was 6 years (range: 0.7 to 8.3). Conclusions: A high proportion (15%) of new drugs required a PMR, which generally involved conducting a randomized trial with a substantial patient enrollment to evaluate an alternative dosage(s). This process was slow with a median of 6 years to fulfill a PMR. DO in the premarket setting has the potential to rapidly maximize BR, avoiding both patient exposure to unnecessary toxicity and large, resource-intensive, multi-year postmarketing trials. Project Optimus is working with stakeholders to advance a DO paradigm that occurs early and throughout premarket development to ensure an optimized recommended dosage at the time of approval.