Chapter 8 170 We have so far examined various use cases of leveraging data from expanded access programs for regulatory, reimbursement, or scientific purposes. We caution, however, to see expanded access as a panacea to generate useful evidence. In fact, the intensified burden on patients and physicians to engage in expanded access research should carefully be weighted against the potential outcomes of said research. Rozenberg and Greenbaum had published an opinion piece in the American Journal of Bioethics highlighting the underutilization of expanded access data. While we agree that this could have negative long-term health consequences, it is of importance to note that some data already has been used as shown through our previous work, and that data collection comes with both benefits and drawbacks. In their open peer commentary: ‘Making It Count: Extracting Real World Data from Compassionate Use and Expanded Access Programs’,52 Rozenberg and Greenbaum (R&G) discuss important matters concerning the added value of extraction of real-world data from expanded access programs. R&G address practical concerns and provide recommendations for the collection, analysis, and interpretation of such data. However, we feel four points warrant further clarification. First, extracting real-world data from expanded access programs is not as novel as R&G presumed. In our recent review of drug approvals that incorporated real-world data from expanded access programs, we identified 49 approvals relying on these data.50 Remarkably, some approvals date back to the 1990s. Back then, they were simply called observational, epidemiological studies, or ‘expanded access studies’: far less fancy than ‘real-world data’, yet perhaps equally effective. Second, the authors seem to underestimate the conflicting interest expanded access brings to clinical trial recruitment. Similarly alluding to SARS-CoV-2 as the authors do, the Mayo Clinic authors of the convalescent plasma expanded access program point out that ‘Physicians, hospitals and patients have the choices of this program (red: the expanded access program) versus a RCT. It is clear that over 90,000 patients and over 10,000 physicians elected to participate in the pragmatic, real-world evidence study design’.89 Hence, the expanded access program was actually competing with ongoing clinical trials. Randomizing only a fraction of the patients that participated in the expanded access program would have yielded far more actionable insights – urgently needed in times of a pandemic. Third, at first sight it seems natural to extract real-world data from expanded access programs. It might be more appropriate to ask the question why that data is not being collected. Indeed, other scholars have argued it is ethically imperative to collect data when patients are treated with investigational medicine.11 Taking a closer look, data collection requires infrastructure, such as well-designed and validated databases, but also legislation that facilitates data collection. It is for example unclear in what regulatory light expanded access data collection, or subsequent
RkJQdWJsaXNoZXIy MTk4NDMw