PCORnet Blog

Extensive Aspirin Study Uses Stakeholder Input to Work Out Protocol

Date: December 10, 2015
Author: Matthew Roe, MD, ADAPTABLE Principal Investigator, Duke University

A clinical trial that proposes to enroll 20,000 patients with a budget of less than $1,000 per participant is sure to raise eyebrows in the research world. But even before it enrolled its first participant, the ADAPTABLE trial has broken new ground in clinical research—by drawing on the wisdom of patients, doctors, and researchers in designing the trial itself.

The primary aim of ADAPTABLE, or Aspirin Dosing: A Patient-Centric Trial Assessing Benefits and Long-term Effectiveness, is to determine the optimal dose of aspirin for patients who have coronary heart disease. An estimated 15 million people in the United States have the disease, and many of them take aspirin for prevention of heart attack and stroke. But the ideal dose of the over-the-counter medicine, which carries risk of bleeding as a side effect, has never been determined.

Engagement with patients and other stakeholders has been a priority of ADAPTABLE beginning with the planning stages. We seek patient engagement at every critical step, from protocol development and consent design through dissemination of final study results.
When the ADAPTABLE Steering Committee met to finalize the study’s protocol, some questions kept recurring regarding eligibility criteria for trial participants. The committee wanted the study’s eligibility criteria to reflect the best judgment of practicing clinicians to achieve the most generalizable study population. Therefore, we posted our eligibility questions in a public forum, the PCORnet website. This approach is in keeping with PCORI’s and PCORnet’s goals of stakeholder input, open science, and transparency. We decided to go one step further and post the complete draft protocol for review and comment.

We used an online survey tool and invited patients, clinicians, research staff, and the general public to access the survey through the Duke Clinical Research institute (DCRI) and PCORnet professional and social media channels. The DCRI site community is made up of site investigators and study coordinators who partner with us to address clinical research challenges and opportunities that will shape the future of medical care. The community spans more than 35,000 sites worldwide. Patient members of the study’s steering committee were also invited to comment. Within three weeks, we had more than 100 comments from investigators, study coordinators, and patients with heart disease. This input helped us formulate the eligibility criteria.

There were four key areas where we sought input:

1. Should the ADAPTABLE trial include only patients already taking aspirin at the time of screening?
We assumed most patients recruited for trial participation would be on aspirin, but we wondered whether to exclude aspirin-naive patients. The majority (57.8 percent) of survey respondents suggested we should not exclude patients who were not already taking aspirin at screening. As a result, we decided to include those currently taking and not taking aspirin.

2. Should patients taking ticagrelor at the time of screening be allowed to enroll in the ADAPTABLE trial?
After much discussion, we decoded to exclude patients taking ticagrelor, a platelet aggregation inhibitor that is often prescribed to heart disease patients to reduce the risk of heart attack or stroke. Because of its bleeding side effect, ticagrelor carries a warning against its use along with high-dose aspirin (more than 100 mg). In our survey, 49.8 percent of respondents advised us to include patients taking ticagrelor; 36.2 percent advised us not to include these patients. However, some representatives of Clinical Data Research Networks (CDRNs) stated that their sites would exclude patients taking ticagrelor or voiced concern that their IRB or ethics committees would not approve the protocol. As a result of these and other concerns, we decided to exclude patients taking ticagrelor.

3. Should patients with a clear indication for an oral anticoagulant be excluded from the trial, even if they are not currently taking an oral anticoagulant?
Our concern with this criterion was the risk of bleeding and trial drop out if we included patients who may require treatment with an oral anticoagulant after randomization, such as patients with a history of atrial fibrillation, deep venous thrombosis, or pulmonary embolism. Polling results were fairly equal, with 48.2 percent voting to exclude these patients. In the final protocol, we opted to exclude both patients taking oral anticoagulants and those likely to require an oral anticoagulant.

4. Will clinicians agree to keep patients on their assigned aspirin dose after an event such as hospitalization for myocardial infarction, stroke, or a revascularization procedure?
This question was purely hypothetical to gauge interest and support from clinicians. We were encouraged that 60 percent of survey respondents said that clinicians will agree to maintain patients in the study on their assigned aspirin dose after a cardiovascular event or procedure. However, we recognize that ongoing educational efforts will be needed to educate clinicians about the ADAPTABLE trial and secure their support to keep enrolled patients on their randomized aspirin dose throughout the trial follow-up period.

In an era of open science and transparency, we believe that by incorporating input into the ADAPTABLE protocol and sharing the resulting plan, we have set the stage for other clinical trial stakeholders—including academic organizations, government organizations, and pharmaceutical industry sponsors—to adopt a pragmatic and community-centered approach for future randomized clinical trials.