RWE: Cost and Time Savings Entice Sponsors as Pitfalls Linger
The New England Journal of Medicine published an article earlier this week raising questions about the use of observational real world evidence (RWE), and how part of this drive toward using nonrandomized studies to assess the effects of treatments is due to the expensive and complex way that randomized clinical trials are being conducted.The timing of the article coincides with the US Food and Drug Administration (FDA) Commissioner Stephen Hahn recently pointing to the importance of RWE, saying, “I believe there is great promise in the effective use and integration of patient-level data or real-world evidence such as electronic health records, clinical trials, medical studies, and patient registries.”
And FDA now has guidance on both the submission of RWE on the drugs and biologics side, as well as the use of RWE on the device side, and released an RWE framework in December 2018, part of the work of the 21st Century Cures Act from 2016. And although the agency has only used RWE in limited amounts so far (at least eight new indications have been approved for previously approved treatments), the promise of RWE is clearly at the forefront of FDA’s and drug developers' agendas.
And it’s not just FDA. Even the Institute for Clinical and Economic Review (ICER) said that as part of its 2020-2023 Framework update, the drug cost-effectiveness researchers will conduct a pilot program to generate new RWE to inform updated assessments of therapies approved under FDA's accelerated approval pathway.
The goal of ICER's pilot will be to supplement the comparatively limited evidence base that often accompanies these approvals, which rely on surrogate endpoints, with RWE to provide stakeholders with a more comprehensive understanding of the early impacts of these therapies.
Rebecca Wood, former chief counsel of the FDA and current leader of the law firm Sidley Austin’s Food, Drug and Medical Device Regulatory practice, explained to Focus that this is the beginning of the adoption of RWE for different premarket and postmarket uses.
She said RWE may lead to fewer time-consuming trials, “if you can get the certainty you need, but whether the science and the legal framework will keep up is a question.”
As far as potential uses, she pointed to using datasets to better design clinical trials, especially for novel therapies in smaller populations, and to ease the process of recruiting the right patients.
The NEJM article, meanwhile, titled “The Magic of Randomization versus the Myth of Real-World Evidence,” and authored by University of Oxford professor Rory Collins and others, notes that “because of the potential biases inherent in observational studies, such studies cannot generally be trusted when — as is often the case — the effects of the treatment of interest are actually null or only moderate (i.e., less than a twofold difference in the incidence of the health outcome between using and not using the treatment).
“In those circumstances, large observational studies may yield misleading associations of a treatment with health outcomes that are statistically significant but noncausal, or that are mistakenly null when the treatment really does have clinically important effects,” they write.
The authors further said that the solution to the problems caused by “the bureaucratic burdens” imposed on randomized trials over the past 25 years “is not to replace randomization with unreliable nonrandomized database analyses.”