Predicting the CAROLINA trial with real-world evidence: Three takeaways for pharma
Kristen Deken, M.B.A.
Director of Strategic Partnerships, Aetion
Jeremy Rassen, Sc.D.
Co-founder, President, and Chief Science Officer, Aetion
The U.S. Food and Drug Administration has long relied on randomized controlled trials (RCTs) as the gold standard for evaluating drug effectiveness and safety. But with increasing momentum to expand the role of real-world evidence (RWE) in drug regulation, catalyzed by federal mandates, FDA is now also exploring ways to apply RWE in regulatory decision-making.
The CAROLINA study: A new frontier
To develop guidance for RWE-based submissions, FDA has invested in several demonstration projects. One in particular has attracted much interest: RCT DUPLICATE, led by Jessica Franklin, Ph.D., and Sebastian Schneeweiss, M.D., Sc.D., at Brigham and Women’s Hospital. This project uses RWE to replicate findings from completed RCTs, as well as to predict findings of ongoing Phase III and IV trials. The work seeks to build empirical evidence on when and how to use RWE in drug regulation—and provides a first glimpse into a new paradigm in regulatory approvals.
In the latest phase of RCT DUPLICATE, results from its predictive study were released on Friday, June 7, at the American Diabetes Association’s 79th Scientific Sessions, while results from the RCT were released the following Monday at the same conference. The RWE study results have now been published in Diabetes Care.
The RWE prediction study used commercially available administrative claims data to generate RWE that would predict the findings from the ongoing CAROLINA® trial (CARdiovascular Outcome study of LINAgliptin versus glimepiride in patients with type 2 diabetes). The trial comes as the result of 2008 FDA guidance which mandated long-term cardiovascular outcome trials (CVOTs) to assess safety for new antidiabetic drugs used to treat type 2 diabetes.
Accelerating time to results at lower cost
This demonstration work exemplifies FDA’s intention to shift the dialogue in drug regulation. It asks whether there are cases where RWE can be used to augment or even take the place of RCTs, and draws consideration for use of RWE to support regulatory decisions, under appropriate circumstances.
In doing so, it brings to light two driving factors where RWE delivers significant benefits over RCTs: time to results and cost.
The CAROLINA trial took nearly eight years to complete and required active patient recruitment, in addition to the slew of other resources needed from medical staff and study teams. In contrast, the RWE study only required a few months with a small team of epidemiologists and medical experts, clinical data that already exists, and an RWE analytics platform.
Confidence is king
Establishing RWE as the new standard in certain drug approval settings is not without its challenges. The success of this new standard will come down to instilling confidence in the results of RWE studies. This is the edge RCTs currently have over RWE, as RCTs have long been in practice and relied upon by regulators for the most critical decisions regarding drug effectiveness and safety.
Concerns around RWE stem from uncertainty of the results. There are numerous examples of the “same” RWE study being conducted while reaching different conclusions. The FDA voices concern in its framework that, hypothetically, RWE studies could be conducted numerous times with only the most favorable study being submitted.
How can we be sure that the submitted study reached the right conclusion? And if it didn’t, how will we know?
Confidence in RWE, therefore, will be absolutely necessary in order to adopt this new tool in drug regulation. This confidence will come from transparency and the ability to demonstrate robustness and reliability of findings from RWE studies via analytics platforms. This is the bar FDA intends to set through RCT DUPLICATE and other demonstration projects exploring where RWE can be reliably implemented—and where there may be limitations—to craft guidance on the standards and processes required for RWE submissions that drive regulatory decisions.
How do decision-makers in pharma prepare for a fundamental change in FDA regulatory processes?
Steps to take now
If we are at the brink of a paradigm shift, what now? How do decision-makers in pharma prepare for a fundamental change in FDA regulatory processes?
- Pay attention to demonstration work and FDA guidance. FDA is investing in demonstration projects as part of its evaluation for using RWE in regulatory decision-making. This work will inform FDA’s 2021 draft guidance. Therefore, look for any early insights into the uses and limitations of RWE arising from this work. While CAROLINA is a single study, it could inform the design of future type 2 diabetes studies that use RWE. More studies will follow: The RCT DUPLICATE program involves seven more trial predictions, alongside the 30 trial replications.
- Understand RWE’s ideal use cases. The future is never as far away as we might think. Understand where and how RWE use cases can be applied across your organization, both now and in the future. Identify synergies that can be achieved across functions with respect to data, infrastructure, technology, and resources. And build experience with advanced RWE use cases today, through studies to inform drug development, synthetic control arms, or RWE-driven post-market safety programs.
- Prepare your organization to take advantage of the strategic and financial benefits that may emerge. Good data is a prerequisite for any good RWE study. The CAROLINA trial was chosen for demonstration, says Franklin, because the real-world study could be conducted with commercially available claims data. Those same data sources may work for other CVOTs but be less relevant—not fit for purpose—for other therapeutic areas. Early identification of fit-for-purpose data sources for your studies is critical.
Preparation for potential new FDA guidance will require exploring what data currently exists, whether it’s reliable and fit for purpose, and assessing what actions or investments—including possible demonstration cases to gain familiarity with the data—need to be taken to begin employing real-world evidence across your organization.