Measures to Improve Trial Enrollment

Clinical trials generate the bedrock of evidence from which our patient care directives are derived. Although there has been significant evolution in the design and execution of cardiovascular clinical trials over the past several decades, shifting demographic trends, operational inefficiencies, and rising costs still pose threats to the current trial ecosystem. From 2008 to 2018, the estimated cost for developing a new cardiovascular drug, from the nonclinical through postmarketing phases, was $153.5 million.¹ In 9 large randomized cardiovascular drug trials that enrolled at North American research sites, the median time from study protocol delivery to the first participant enrollment was 9 months.² In addition, the increasing globalization of cardiovascular drug trials, often with higher enrollment rates outside the U.S., has resulted in a more severe underrepresentation of real-world U.S. populations.³ Analogous data regarding cardiovascular device trial processes are sparse, but we suspect similar limitations because recent evidence has shown that the demographics of enrolled populations in device trials may influence which patients receive the intervention in the real world.⁴ It is clear that strategies to accelerate trial enrollment, improve efficiency, and ensure generalizability and representativeness of target populations to increase the rate of new data generation and reduce trial duration and costs are needed. Herein, we discuss innovative strategies to achieve these goals in the context of current clinical trial infrastructures.

GAMIFICATION TO INCENTIVIZE STUDY PERFORMANCE

Recently there has been investigation of improving start up time and enrollment rate with the use of gamification, a technique based on behavioral science principles. At each site within a clinical trial, a dashboard displays to the investigative team its site’s performance in key metrics of trial performance, relative to targets and to those of other sites. It is expected that the ensuing competitive spirit might enhance performance at each site. In this issue of JACC: Heart Failure, Bhatt et al⁵ report on the impact of a novel gamification intervention within a large, global, randomized clinical trial on increasing trial enrollment. In this intervention deployed at 92 U.S. sites, when >70% of target U.S. patients were already enrolled, the investigators delivered a gamified electronic dashboard to sites in the FINEARTS-HF (Finerenone Trial to Investigate Efficacy and Safety Superior to Placebo in Patients With Heart Failure; NCT04435626)⁶ trial every 2–4 weeks. Effectiveness was assessed by: 1) comparing trial enrollments before and after the intervention rollout; and 2) comparing U.S. sties with non-U.S. sites, serving as nonrandomized control sites where no intervention was deployed. Although the intervention was not associated with a significant change in screening or enrollment rates compared with predeployment in either the U.S. or the non-U.S. sites, the authors provide several possible explanations which further inspire the exploration of this type of intervention to improve clinical trial performance. They also clarify that they considered this investigation to be a pilot study. Behavioral economics principles deployed within trial designs, like those employed by Bhatt et al,⁵ merit further exploration. Gamification has been shown to accelerate performance and increase enjoyment of the trial workforce during the trial startup phase,⁷ and its potential impact on other trial execution metrics and enrollment should be evaluated in dedicated and randomized studies. As the authors summarize, several issues, including timing of gamified intervention within the trial, heterogeneity of global trial sites, and lack of ability to objectively measure engagement with the intervention, all present opportunities for future investigation.

The concept of a trial enrollment dashboard, as piloted in this study, introduces the potential for a more expansive enrollment tracking platform that can be used by site principal investigators and tracked by trial leadership in real time. One such platform, the Accrual Quality Improvement Program, has been deployed in cancer prevention studies through the National Cancer Institute.⁸ That program provides a web-based centralized infrastructure for collecting, analyzing, displaying, and saving participant-, site-, and study-level data at each stage of the recruitment and enrollment process. Achieved metrics, such as enrollment rates, can be compared with projected rates and displayed along with recommendations for modifications to recruitment strategies.

STREAMLINING DATA COLLECTION AND CASE REPORT FORMS

The HFC (Heart Failure Collaboratory), a multidisciplinary group of investigators, statisticians, regulators, industry representatives, and patients, has been working diligently to address each of the roadblocks to delivering effective drugs and devices to patients more efficiently and rapidly. One major focus has been the basics of clinical trial design, working, eg, to standardize a “lean” case report form (CRF), free of procedures and measurements that are not part of the pathway to regulatory approval.⁹ The HFC has also devised specialized modules that can be added to the lean CRF as adjunctive information to be collected when appropriate for a particular disease state, population, or intervention. It is hoped that adoption of these standardized tools will reduce resistance to enrollment and expedite trial conduct.

IMPROVING TRIAL REPRESENTATION AND DIVERSITY

Historically, trial screening and enrollment strategies have been standardized and uniform for all participants; emerging strategies, however, have shown the potential to increase enrollment of specific populations with tailored outreach approaches. As an example, the pragmatic PRO-HF (Patient-Reported Outcome Measurement in Heart Failure Clinic; NCT04164004) trial used a virtual enrollment strategy via e-mails, text messages, and telephone calls. Asian, Black, Native American, and Pacific Islander patients enrolled in greater proportions by telephone than by e-mail, and those who enrolled by e-mail were likely to be older.¹⁰ Leveraging multimodal communication strategies tailored to sociocultural preferences should be further evaluated as a strategy to improve enrollment of underrepresented individuals. Beyond traditional outreach methods, more contemporary patient-facing platforms such as social media may harbor untapped potential. About one-fourth of adults in U.S. households have reported that they use social media at least weekly to find health-related information, and this proportion is higher among Hispanic, Black, and low-income adults.¹¹ This iteration of direct-to-consumer advertising has shown promise in expediting enrollment and at reduced costs.¹²

The association between diverse trial leadership and research teams and increased enrollment from historically underrepresented populations has been demonstrated in heart failure¹³ and other cardiovascular specialties. Multidisciplinary consortia such as the HFC,¹⁴ Women As One, and the American College of Cardiology are leading the way in creating academic-industry partnerships and skill-development programs to cultivate early career trialists from diverse backgrounds. Leadership positions that can have an impact on trial enrollment may extend beyond executive or steering committees. At a recent HFC training program for data monitoring committee members,¹⁵ participants discussed the role of data safety monitoring boards in recommending temporary pauses on subject accrual if certain populations are disproportionately represented. Furthermore, the HFC has emphasized the role of everyone involved, from sponsors to journal editors, to increase diverse study leadership leading to downstream diverse study cohorts.¹⁴

RETOOLING SITE AND INVESTIGATOR SELECTION

Identifying clinical trial sites and investigators has largely relied on previous trial enrollment performance, site experience in a particular therapeutic area, and other such historic metrics. Recent data from the U.S. Platinum Diversity (NCT02240810) and Promus Element Plus (NCT01589978) studies demonstrated that enrollment of underrepresented populations varies by research site, geographic region, and surrounding population.¹⁶ Now, with the increasing integration of artificial intelligence in health care, we can imagine a future state in which multiple sources of information such as electronic health records, cardiovascular imaging, sponsor and trial databases, and local epidemiology and demographic patterns are inputs to machine learning models used for predictive¹⁷ and prognostic enrichment.¹⁸ As a result, trial sites with access to patients most similar to these targets would be prioritized for site activation. This practice may also serve to expand the number of investigators in heart failure clinical trials, Where there is currently recognized to be an inadequate workforce, whose participation is frequently financially and professionally discouraged.¹⁹

COLLABORATING WITH SPONSORS AND REGULATORY AGENCIES

Strategies to increase the enrollment of diverse populations in heart failure clinical trials have been extensively reviewed elsewhere.^3,20 Despite a relatively increased emphasis on this issue in recent years, there has not yet been substantial progress in the representation of women and underrepresented minorities. In June 2024, the U.S. Food and Drug Administration issued a draft guidance, “Diversity Action Plans to Improve Enrollment of Participants From Underrepresented Populations in Clinical Studies,” intended to guide trial sponsors through the process of submitting Diversity Action Plans to the FDA.²¹ The Food and Drug Omnibus Reform Act has mandated this new requirement in an effort to encourage sponsors and investigators to design trial enrollment rationale and goals to be representative of the affected population. As of this time, a Diversity Action Plan must include the sponsor’s goals for enrollment disaggregated by race, ethnicity, sex, and age group of clinically relevant study populations, rationale for these goals, and explanation of how the sponsor plans to meet these goals. Regulatory agencies can require robust goals and plans in the approval of study protocols and could even incorporate goal achievement into the approval process.

Bhatt et al⁵ should be congratulated for their ingenuity and initiative in deploying a novel method to improve patient enrollment through embedding their behavioral intervention within a large global clinical trial. Although there was no indication of effectiveness within their pilot study, they highlighted several improvements that can be made within future investigation: refining the intervention, randomizing individual matched sites across the geography, and launching the intervention earlier, ideally to include study start-up. Their work highlights the ongoing challenges in executing clinical trials that are representative, agile, and cost-effective, and inspires frameworks to rigorously test innovative strategies to improve trial performance across relevant domains and domains of trial conduct.