Post-market evaluation is a critical stage of the drug development process. After FDA approval, there is often still a lack of knowledge of long-term safety and/or rare safety risks, especially in populations not well-represented in pre-approval trials. Regulators often require such evidence to be demonstrated through Post-marketing Commitments and Requirements (PMC/Rs) which are crucial to understanding drug performance in routine care settings where the product’s benefit/risk profile continues to evolve.
PMC/Rs are common in oncology therapeutics, given the volume of drug development and the frequency of accelerated approvals based primarily on efficacy and unmet need. In fact, Flatiron Health's analysis of the FDA PMC/R database reveals that 24% (130/535) of the PMC/Rs issued between 2022-24 are related to oncology products and 42% (55/130) of these PMC/Rs mention safety as one of the endpoints.
Among the large number of PMC/Rs recently issued for oncology products, 35% (19/54) of those with final reports due between 2020-2023 did not meet the deadlines assigned by the agency, leading to further delay in filling crucial evidence gaps1. This is why Flatiron Health, as part of its mission, remains committed to supporting sponsors fulfilling PMC/Rs through both retrospective and prospective data collection approaches, using cutting-edge digital tools.
Current Landscape and Challenges of PMC/Rs
Our analysis of the FDA's PMC/R database revealed several interesting trends among the 55 oncology PMC/Rs requiring safety data issued between 2022-2024. These trends underscore the FDA’s focus on fast tracking the drug approval process through providing multiple pathways for collecting post-market data:
- Long-term safety data: 25% (14/55) of the PMC/Rs request long-term safety data, defined as a trial with a follow up period of at least 15 months or more.
- Safety vs. Multiple endpoints: 22% (12/55) request only safety data, and the remaining 78% (37/55) request one or more of the following endpoints or populations: efficacy, diversity, pharmacokinetic profile, pediatric patients and patients with renal or hepatic impairment.
- Prospective vs. Retrospective: 49% (27/55) request prospective data, 20% (11/55) request retrospective data, and 22% (12/55) request integrated analysis combining both retrospective and prospective data. FDA did not state a method of data collection for 9% (5/55) PMC/Rs.
- Interventional vs. Observational: FDA explicitly requests interventional design for 42% (23/55) PMC/Rs and observational design for 7% (4/55) PMC/Rs.
Therapy class also shapes the types of safety commitments required. Kinase inhibitors, for example, often demand long-term monitoring for adverse events such as cutaneous reactions, thromboembolic events, ocular toxicities, or bleeding risks. 34 of the 55 safety PMC/Rs are related to kinase inhibitors, while the rest are divided amongst other drug classes such as- cellular therapy, hormone therapy, immunotherapy, bi-specifics etc.
For cell therapies like CAR-T, the FDA has almost universally been requiring fifteen years of follow-up to evaluate the risk of secondary malignancies. While data requests for such specific adverse events make up more than half of the PMC/Rs (29/55), a sizable portion of these studies also require general safety data, depending on the drug class or safety profile. These realities highlight the unique requirements and challenges associated with different drug classes, especially novel therapies such as CAR-T.
The safety PMC/Rs associated with different drug classes and different study designs present distinct challenges for safety data collection. For example, retrospective data provides a fast and efficient way of collecting general safety data from real-world sources, but it may not be the best resource to assess a specific adverse event’s characteristics such as- prognosis, management methods, and confounding factors.
Similarly, prospective data collection offers a tailored research approach with controllable variables in a real-world setting, but it may be considered expensive, time-consuming and operationally difficult to execute. Due to the small number of people meeting criteria for specific special populations, such as pediatric patients, racial and ethnic minority groups, or patients with renal or hepatic impairment, patient enrollment remains a major challenge in fulfilling many safety PMC/Rs.
How Technology can Enhance Prospective and Retrospective Evidence Generation for PMC/Rs
The evolution of digital health technologies offers new ways to meet the challenges in evidence generation to fulfill PMC/Rs. Large scale data curation using ML and LLMs allows vastly larger patient cohorts and expanded clinical variables to power regulatory-grade retrospective analyses. The results from these analyses confirmed the feasibility of using artificial intelligence–based technology to support extraction of adverse event data from source documents such as laboratory records and physician narratives.
Additionally, prospective clinical trials are benefitting from many operational approaches that also leverage digital tools such as tech-enabled patient screening, data-driven site selection, and automated electronic health record (EHR) to electronic data capture system (EDC) transfer applications. Tech-enabled patient screening and data-driven site selection allow researchers to identify eligible populations more efficiently while broadening access to underrepresented communities, while connection between the EHR and EDC systems streamline data collection.
These integrations are suitable for the vast majority of safety PMC/Rs, supporting prospective studies and hybrid study designs where both prospective and retrospective data are combined for integrated analyses. These methods bring efficiency to trial execution by reducing the site burden and error frequency associated with manual data entry, and solve many of the major roadblocks associated with timely completion of these PMC/Rs.
Moving from Obligation to Opportunity
PMC/Rs may seem to be another regulatory obligation to be checked off from a list in the drug approval process; however, these obligations allow drug manufacturers to establish trust and confidence among regulators and the broader public. As the drug discovery process has seen a tremendous acceleration with new potential therapies entering the development path at increasing rates, the drug approval process has also adapted to bring these novel therapies to market as fast as possible to allow patients access.
As regulators across the globe face increased pressure and scrutiny to bring safe and effective drugs to market in a timely manner, the evidence generation community must overcome the challenges associated with timely completion of the PMC/Rs. The tech enhancements discussed in this blog have already been used in multiple trials to help generate evidence for PMC/Rs.
Flatiron Health proudly continues to provide the data and technology for retrospective PMC/R studies. In the prospective study space, Flatiron has demonstrated feasibility in multiple PMC/Rs, where digital enhancements and pragmatic trial elements provide benefits in terms of efficiency. Flatiron believes that the future of cancer care depends on timely execution of clinical research, including PMC/Rs, that help build the foundation of the evidence needed to deliver safe and effective novel treatments. Together with technology, the future of evidence generation we envision is fast, efficient and operationally seamless.
- Meadows, E.S., Chen, A., Reynolds, M.W., Collins, J., Ghanim, M.T., Altomare, I., Serra, C.A. & Liaw, N. Feasibility of prospective real-world studies to fulfill regulatory commitments and requirements in oncology within the United States. Poster presented at the 40th Int. Conf. Pharmacoepidemiol. Ther. Risk Manag., Berlin, Germany, 24–28 Aug 2024.
