Speakers
Stefan James
Professor of Cardiology
Dept. of Medical Sciences and Uppsala Clinical Research Center
Uppsala University
Uppsala, Sweden
Jonas Oldgren
Professor of Cardiology
Dept. of Medical Sciences and Uppsala Clinical Research Center
Uppsala University
Uppsala, Sweden
Keywords
DAPA-MI, R-RCT, myocardial infarction, regulatory evaluation, SGLT-2 inhibitors, pharmacological, pragmatic
Key Points
- Previously observed improvements in post-Myocardial Infarction (MI) prognosis have reached a plateau in recent years. Given that SGLT-2 inhibitors have positive effects on almost all cardiometabolic parameters, investigators at the Uppsala Clinical Research Center (UCR) hypothesized that dapagliflozin could reduce the risk of heart failure following MI. SGLT-2 inhibition could become the first new beneficial pharmacological treatment concept for patients with MI in over a decade. Dapagliflozin is already known to be beneficial to patients with diabetes or chronic heart failure, and has a known safety profile.
- The DAPA-MI team conducted a pragmatic, registry-based, double-blind randomized clinical trial using national registry data, also known as an R-RCT, from 103 sites in the U.K. and Sweden. Their objective was to evaluate the effect of dapagliflozin on cardiometabolic outcomes in patients with acute myocardial infarction without diabetes or chronic heart failure. Data was collected as part of routine clinical care and transferred to the UCR web application, which helped the PIs to identify and enroll patients. That data was then transferred to the DAPA-MI database.
- An R-RCT is a prospective randomized controlled trial that uses a clinical quality registry for one or several major functions for trial conduct and outcomes reporting. The DAPA-MI study team saw this approach as combining the best of two worlds: real-world registry data with the features of a randomized trial.
- The registry would allow them to enroll as many patients as possible, in a pragmatic setting, at a lower cost, while the RCT would allow them to draw causal inference and understand the efficacy of the agent. The team’s other objective in conducting this trial was to evaluate the possibility to conduct a registrational clinical trial in a very new way for a marketed product with a known safety profile.
- Their initial primary endpoint was a composite of hospitalization for heart failure and cardiovascular death, the traditional standard outcome for heart failure trials. During the course of the trial, however, it became evident that the number of collected primary composite outcomes in the DAPA-MI trial was substantially lower than anticipated. In February 2023, the trial was modified from an event-driven time-to-event approach to a hierarchical composite outcome approach including clinically relevant cardiometabolic outcomes.
- They found that DAPA-MI met the primary endpoint and dapagliflozin treatment demonstrated significant clinical benefit compared to the placebo. With a Win ratio of 1.34, the likelihood for a better cardiometabolic outcome with dapagliflozin is 34% higher compared to placebo.
- The feasibility of the R-RCT design was confirmed with the recruitment of 4017 patients with acute MI without diabetes and chronic HF in only two countries and 103 sites. Collaboration between UCR and NICOR enabled the success of the trial; excellent data quality confirmed the regulatory-grade of R-RCT design; and the registry-based approach was appreciated and attracted Investigators all over Sweden and the UK representing both large and small hospitals.
Discussion Themes
-Why did you reduce the power of the study when you shifted to the hierarchical endpoint? What would’ve been the power of the hierarchical endpoint if you’d stayed at 6,000 patients? The trial enrolled well, but we realized we wouldn’t be able to double the number of patients from 6,000 to 12,000, which might have been required to get power on the first primary endpoint. I think many of us would like to have continued to enroll patients above 6,000 or 6,500, but there with multiple issues with that. If we had a sufficient power to stop at 4,000 patients, it was difficult to convince the sponsor, perhaps the trial organization to continue enrolling if it wasn’t absolutely necessary.
–What were your interactions with the regulators like? It was interesting to have these regulatory discussions with the different agencies, with very different comments and requests on our trial for the proposal. I would frankly say that the FDA was the easiest regulatory agency to discuss with, very supportive of a new way of doing trials, very supportive of not adjudicating events, finding pragmatic solutions. The U.K, they requested that we do specific, cumbersome safety testing with blood tests, renal function tests. They were afraid that dapagliflozin would hurt the kidneys. The journals were even more difficult. They wanted us to not do a pragmatic trial at all; they wanted us to understand the mechanisms of the drug with repeated echo measurements. So very different ideas.
–Did you have any feedback that was different across the different agencies around safety reporting? Yes, safety reporting was also viewed differently. The FDA were also supportive of not doing specific safety reporting more than we do in regular care. The EMA and the other European agencies were more concerned about safety reporting, but we were able to get support on doing minimal safety reporting since the agent and the safety profile were so well known. We also set up a study in which we are now retrospectively comparing safety reporting with what we can get from registries. So that’s an ongoing process to compare and see if the safety reporting that we can achieve from registries allows us to understand the safety profile as good as or better than the safety reporting.
Tags
#pctGR, @Collaboratory1