Clinical Trials: Beyond the Plenary Stage

The thought of engaging with clinical trials can be daunting. It often seems trials are meant only for AUA plenary sessions and high-impact journal publications, but clinical trials should be for everyone. As we look to improve care for our patients, we should be aiming to increase the availability and accessibility of clinical trials by making clinical trials easier to run and participate in for all patients, providers, and practices, for multiple reasons.

Clinical trials have massive benefits to science, society, and individuals. For some conditions like cancer, clinical trials can be considered standard of care. For example, the National Comprehensive Cancer Network clearly states in all its clinical practice guidelines “the best management for any patient with cancer is in a clinical trial.”¹ But historically clinical trials have been difficult to implement and improve, creating hurdles preventing patient and provider participation, with some groups more impacted than others. The resulting gaps in clinical knowledge and equity have real implications for patients and our practice. We must find ways to overcome these barriers to make clinical trials more efficient and equitable.

There has been an increasing effort to address these issues and engage urologists and patients in clinical trials. Sessions at our AUA and Society of Urologic Oncology meetings focused on engaging with clinical trials are a good start. However, expanding our approach to how we design and implement trials—that is, the science of clinical trials—holds promise as an opportunity to improve clinical trials themselves. This article will give an overview of how we are applying implementation science concepts to facilitate this process, and ultimately aim to make it easier for all to engage in clinical trials.

Is There Really a Problem With Clinical Trials Now?

In addition to anecdotal difficulties, clinical trials suffer from high documented failure rates. Urologic oncology trials struggle with both enrollment and completion, with 1 in 6 trials failing to reach the primary end point, mostly due to poor enrollment, and one-third of even “completed” trials failing to get close to anticipated end points.^2,3 Similar issues are faced by other urological subspecialties, and in other cancer types.^4,5 Thousands of patients are enrolled in trials that ultimately fall short, with loss of the promised benefits to science promised as part of the trials consent process. Further, these trial failures and inefficiencies contribute to, and waste part of, the over $200 billion spent annually on clinical trials.⁶

Considering these shortcomings, the extant science of trial improvement is lacking. For example, existing strategies to improve trial enrollment have limited evidence, small effect sizes, and uncertain methods of scaling for broad application.⁷ Further, these approaches are generally not theory informed, making it difficult to compare approaches and adapt existing methods to more efficiently develop clinical trials science and generalizability of findings.

Figure 1. Reasons for urologic oncology trial termination. PI indicates principal investigator. Reprinted with permission from Stensland KD et al. Urol Oncol. 2021;39:154-160.³

Interfacing Implementation Science and Trial Improvement

To address these problems, we have proposed using implementation science to improve clinical trials by considering clinical trials per se as evidence-based interventions.⁸ Similar to other evidence-based interventions, like smoking cessation or vaccines, clinical trials have huge benefits but often suffer from suboptimal implementation. By applying techniques including rigorous context assessment before and during trial implementation, implementation outcome evaluation for a given trial, and targeted intervention development to improve trial implementation, we can build new trial improvement science on the platform of existing implementation and behavior change science.

A major advantage of this approach is that it emphasizes the importance of making trials easier (ie, more feasible and acceptable) for physicians and other providers in addition to patients. While improving the scientific value of trials is important, ensuring trials can be delivered effectively and applied in real-world settings is critical and well supported by the principles of implementation science. For example, barriers to urological cancer trials in rural communities have been explored using these approaches.⁹ In our own work, we have incorporated qualitative methods to ensure our concepts and approaches are acceptable, applicable, and well understood by physicians and other stakeholders.¹⁰

Figure 2. Adapted Implementation Research Logic Model applied to the clinical trial-side outcome of poor enrollment. CFIR indicates Consolidated Framework for Implementation Research; ERIC, Expert Recommendation for Implementing Change. Reprinted with permission from Stensland KD et al. Implement Sci Commun. 2022;3(1):109.¹¹

Applying the Science of Implementation to the Trials Context

The basic implementation science approach is to define outcomes to evaluate how well an evidence-based practice is being implemented, identify barriers and facilitators to the uptake of the practice, and then design implementation strategies to overcome the identified barriers. In other words: why isn’t something being used, how do we measure how and why people aren’t using it, and what can we do to get people to use it?

Figure 3. Tool identifying areas with many prostate cancer (PCa) cases but few clinical trials. Adapted with permission from Stensland et al. Contemp Clin Trials. 2021;111:106600.¹⁶

We adapted existing frameworks to structure this approach specifically for clinical trials.¹¹ We used the Consolidated Framework for Implementation Research to identify barriers and facilitators to trial uptake, adapted Proctor’s implementation outcomes to evaluate how well trials are being implemented, and linked these to implementation strategies from the Expert Recommendation for Implementing Change compilation, all through an adaptation of the Implementation Research Logic Model.^12-15 This process allows for linking root causes of problems to targeted improvement interventions with a higher chance of working to improve problems like poor enrollment or representation in trials. Laying this approach out also could explain why some trial improvement interventions are unsuccessful: they target the wrong problem.

For example, a prostate cancer trial struggling to enroll patients has multiple options for improvement. Hiring more research staff or developing an electronic medical record system alert could enhance the penetration to eligible prostate cancer patients. However, if there are only 10 eligible prostate cancer patients presenting to a clinic every year, there is no amount of research staff hiring that can increase enrollment to 100 patients annually. Instead, in this case identifying new trial sites would be a more rational enrollment improvement intervention.

Similarly, this approach can highlight new research directions. Continuing the question of available eligible patients (ie, trial feasibility), we developed a tool to identify areas with many incident prostate cancer cases but few available trials, as we found trials were more likely to be successful in areas of higher cancer incidence.¹⁶ This tool, or similar analyses, could be helpful in selecting future prostate cancer clinical trial sites, especially for trials struggling with enrollment specifically due to low local prostate cancer incidence or competing trials.

The Path Forward: A Trail to Trials

Moving forward, we hope to expand efforts to improve clinical trials, and reduce barriers to trial implementation and participation through feasible, acceptable interventions. We encourage participation in clinical trials when possible, and to consider applying implementation science approaches to make and measure outcomes of targeted improvements to ongoing trials adding to trial success and generalizable knowledge.