KEY POINTS:
- There was no significant difference in primary composite endpoint of all-cause mortality, hospitalization for myocardial infarction, or hospitalization for stroke in patients with coronary artery disease taking 81 mg versus 325 mg of aspirin every day.
- There was no significant difference in the safety endpoint of hospitalization for major bleeding with blood product transfusion in patients with coronary artery disease taking 81 mg versus 325 mg of aspirin every day.
- Conducting pragmatic study in a real world, patient-centric approach will likely continue to emerge as an appealing, cost effective trial design in the field of cardiology.
Cardiovascular disease is costly and a major cause of death worldwide. Aspirin has been the mainstay of treatment in secondary prevention of coronary artery disease for over thirty years. More recently, several analyses have suggested varying doses aspirin (81 mg vs 325 mg) as well as the use of other, newer antiplatelet agents for the treatment of coronary artery disease.
Aspirin is an accessible, affordable, over the counter, household medication. Given its enduring presence in the medical community and our patients’ medicine cabinets, a high powered, expensive, large randomized control trial examining aspirin dosing for secondary prevention is something unlikely to be sponsored by industry. In order to circumvent this issue, Dr. Schuyler Jones, Associate Professor of Medicine, Duke University and Duke Clinical Research Institute, developed an innovative, pragmatic trial design. This type of trial is more generalizable to our patients and also significantly more cost efficient.
Presented as a Late Breaking Clinical Trial on May 15, the opening day of ACC 2021, and concurrently published in the New England Journal of Medicine , ADAPTABLE is a large, pragmatic, intention to treat study in which Dr. Jones and his colleagues used electronic health records (EHR) and claims data to find eligible patients with a history of coronary artery disease and an enrichment risk factor (age, smoking status, among other characteristics) using International Classification of Diseases, Ninth Revision (ICD-9) and Tenth Revision (ICD-10) codes.
Over 15,000 patients were enrolled online from 40 US centers from 2016-2019, and in an open label fashion after randomization, purchased their own high (325 mg) or low (81 mg) dose aspirin to complete the study. Median follow up was 26 months, and both groups were balanced with similar rates of other comorbidities. Primary endpoints included composite all-cause mortality (cardiovascular mortality was not included as specific cause of death as this data is not usually reported into EHR), hospitalization for myocardial infarction, and hospitalization for stroke. The safety endpoint included hospitalization for major bleeding with blood product transfusion.
What were the results? Low dose aspirin was shown to have similar outcomes when compared to high dose aspirin in meeting both the primary composite endpoint (hazard ratio of 1.02 with confidence interval of 0.91-1.14) and safety endpoint (hazard ratio of 1.18 with confidence interval of 0.79-1.77). Interestingly, the primary safety endpoint of bleeding occurred with relatively low frequency, happening in only 53 patients in the low dose aspirin group and in 44 patients in the high dose aspirin group.
Why wasn’t there more bleeding in the higher dose aspirin group? According to Dr. Jones, several hypotheses can be drawn. Due to the open label trial design, several patients randomized to high dose aspirin switched over to the lower dose group on their own. In addition, the safety endpoint requiring hospitalization for a blood transfusion was a very strict endpoint. Other clinically significant bleeding outcomes (like upper gastrointestinal bleeding that did not require hospitalization) were excluded from the safety endpoint.
While pragmatic trials have several advantages including their cost effectiveness and streamlined trial approach, they also come with several limitations. As mentioned previously, due to the open label real world approach, many patients taking high dose aspirin self-transitioned to low dose aspirin. While this is unlikely to happen in a double-blind, randomized, effectiveness trial, Dr. Jones reminds us this is something that happens in the real world; our patients don’t adhere to prescribed doses of their medication all the time.
The ADAPTABLE trial is an innovative, pragmatic study that examined the most appropriate dosage of aspirin in the treatment of patients with established coronary artery disease and was found to be a neutral study. How does this late breaking trial translate into clinical practice? Given the neutral trial result, Dr Jones anticipates that clinicians will likely continue to default to their standard practice prescribing 81 mg of aspirin daily.
When asked more about the utility of pragmatic trials in the field of cardiology going forward, Dr. Jones stated, “Because the ADAPTABLE trial question was so simple, it lent itself well to a pragmatic study design.” Dr. Jones also mentioned “pragmatism doesn’t have to be perfect,” suggesting that when thinking about trial design in the future, aspects from both pragmatic trials and effectiveness trials can used to create a hybrid study. “If any field can conduct a pragmatic trial, it’s cardiology. We are so used to conducting large studies and really do believe that evidence guides us to the right answer.”
ADAPTABLE was funded by the Patient-Centered Outcomes Research Institute. ADAPTABLE ClinicalTrials.gov number, NCT02697916.
Additional background and context about ADAPTABLE here