Taming The Competitive Enrollment Stampede

By Norman M. Goldfarb, executive director, Site Council

Study sponsors naturally want to enroll patients in their clinical studies as quickly as possible. They certainly do not want slow sites to delay the completion of enrollment. Thus, competitive enrollment was born.

In its purest form, competitive enrollment sets up a free-for-all race for sites to enroll study participants. The starter drops the flag, sites race to enroll patients, and enrollment ends when the study is fully enrolled.

If the enrollment goals are challenging or the enrollment period lengthy, competitive enrollment is unnecessary because it is relatively easy for the study sponsor to adjust individual site targets and, if necessary, add additional sites. However, if sites collectively have excess capacity to enroll patients or the enrollment period is short, competitive enrollment can create significant problems for even high-performing sites:

• Some sites may get off to a slow start, perhaps through no fault of their own. For example, they may be too small to justify a master clinical trial agreement. Or the study sponsor was slow to review their documents. Or they raised an important issue that other sites did not notice. Or their site initiation visit was delayed. Or their test article was delayed. Or translation of the informed consent form was delayed. The possibilities are endless.
• Some sites may be more careful and deliberate than others. For example, instead of asking patients to sign the informed consent form during the initial consent visit, they may ask patients to carefully review the informed consent form at home before signing it.

“Our site group can enroll patients very quickly, but the uncertainty associated with competitive enrollment can make it very difficult to schedule study participants and assign personnel,” said Jesse Hoffman, chief business officer at AMR.

Competitive enrollment tends to disfavor the following types of sites:

• Sites within hospitals or health systems
• Sites with inexperienced investigators or staff
• Sites with disfavored patient populations
• Sites in inconvenient or foreign locations

“Our health system can enroll large numbers of diverse patients, but, given our complex workflows, it takes time to get rolling,” said Mary Klein, senior director of operations at Jefferson Clinical Research Institute, Jefferson Health.

Competitive enrollment may have the following negative impacts on sites:

• Sites may be forced to accept disadvantageous contract and budget terms.
• Sites may lose some or all of their investment in study start-up, patient recruitment, or study staffing.
• Sites may disappoint patients who get excited about the study but are then unable to enroll, especially after multiple reschedules. Sites may lose such patients for the study and for any future studies.
• Site personnel, patients, or study participants may feel pressured to take risky shortcuts.

Sites that get burned by competitive enrollment may lose interest in future studies of that type, in that sponsor’s studies, or in any clinical studies at all.

In general, competitive enrollment tends to strengthen a study sponsor’s relationships with sites that perform well in them and weaken its relationships with other sites. This effect may work to its disadvantage by, for example, narrowing its site options for future studies. “Just as our sponsors want predictability, I can’t overstress the need that sites have for predictability,” said Kurt Mussina, CEO of Paradigm Clinical Research. Numerous authors have published articles on predicting patient enrollment, but not so much on making enrollment more predictable.^1-6

Study sponsors could guarantee to each site a minimum number of enrollments, but these guarantees could become problematic if a site enrolls slowly or other sites enroll quickly, not to mention the possibility of the study terminating early, e.g., based on the decision of a data and safety monitoring board.

Implementing Managed Competitive Enrollment

Fortunately, there is a system that works well for everyone: managed competitive enrollment.

Managed competitive enrollment employs a system by which a wide variety of sites are both motivated to enroll patients quickly and have confidence that their investment in the study will be worthwhile. In any managed competitive enrollment system, three things must be true:

• The sponsor must be satisfied with the patient enrollment timeline.
• Study sites must have excess enrollment capacity available on a timely basis.
• All sites must have a fair chance to achieve their respective enrollment goals.

The Site Council Managed Competitive Enrollment System operates in a hypothetical study as follows:

• The study sponsor contracts with 10 sites to enroll 100 patients each for a total enrollment goal of 1,000 patients over 10 weeks.
• The 10-week enrollment period is divided into 10 one-week periods, with an average enrollment goal of 100 patients during each week. In other words, if every site enrolls 10 patients per week for 10 weeks, the enrollment goal of 1,000 patients in 10 weeks will be achieved.
• Any site may enroll up to 20 patients each week, starting in the week in which it starts patient enrollment.
• Different sites may have different enrollment goals, based on initiating at different times and having different enrollment capacities, which may vary over the course of the enrollment period based on pre-specified rules.
• If enrollment is nearing completion, the study sponsor notifies sites that they each may enroll up to five additional patients with whom (a) a consent visit has been scheduled and (b) consent and screening can be obtained within the next seven days.
• Any site that has the opportunity but does not enroll more than a minimum number of patients in a full week loses its right to enroll further patients, with its share of enrollment goals reassigned to other sites that hit their enrollment goals for the week and have additional capacity.
• If the study sponsor delays enrollment at a site and enrollment is nearing completion, the site will be allowed a minimum of one week to enroll patients.
• Enrollment goals are divided to the extent required to achieve gender, age, racial, regional, or other diversity objectives.
• Separate goals may be created for certain sites, e.g., rescue sites and those with key opinion leaders.

In this system:

• Sites that initiate early or enroll quickly can achieve relatively high enrollment goals.
• Sites that initiate late or enroll slowly can achieve relatively low, but not zero, enrollment goals.
• Sites that join late, e.g., as rescue sites, can achieve reasonable enrollment goals.

Study sponsors can adjust the parameters of (and make exceptions to) the above system to meet the needs of each study. However, unplanned midcourse changes may reduce site trust in the study sponsor.

Before employing the Site Council Managed Competitive Enrollment System — or any other competitive enrollment system — for a given study, the study sponsor should consult with representative sites.

Higher Quality and Predictability Without Sacrificing Speed

Study sponsors that employ free-for-all competitive enrollment may achieve very fast enrollment, but they may also suffer negative consequences, as discussed above. A well-designed and managed competitive enrollment system will be more predictable for all parties, no slower, and may deliver superior quality. “Managed competitive enrollment makes a lot of sense, provided enrollment expectations are realistic,” said Vatche Bartekian, president of Vantage BioTrials. In addition, sponsors that employ managed competitive enrollment can build stronger relationships with sites that are more highly motivated. It also smooths out the workload and supply chain demands for study sponsors.

References:

1. “3 Benefits of Clinical Trial Enrollment Forecasting, https://blog.onestudyteam.com/clinical-trial-enrollment-forecasting,” OneStudyTeam, 2023, https://blog.onestudyteam.com/clinical-trial-enrollment-forecasting.
2. “Data Doesn’t Lie: Clinical Trials Enrollment Forecasting—Is It Worth It?” Nick Karrasch, Trialfacts, https://trialfacts.com/articles/data-doesnt-lie-part-l-clinical-trials-enrollment-forecasting-worth-it/.
3. “Enrollment Forecast for Clinical Trials at the Portfolio Planning Phase Based on Site-Level Historical Data,” Sheng Zhong, Yunzhao Xing, Mengjia Yu, and Li Wang, 2023, arXiv, https://arxiv.org/abs/2301.01351
4. “Enrollment Rate Prediction in Clinical Trials based on CDF Sketching and Tensor Factorization tools,” Magda Amiridi, Cheng Qian, Nicholas D. Sidiropoulos, and Lucas M. Glass, IEEE Xplore, 2023, https://ieeexplore.ieee.org/document/10096026.
5. “Prediction of clinical trial enrollment rates,” Cameron Bieganek, PLOS ONE, 2022, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8870517/.
6. “Unlocking Tomorrow's Cures: Predicting Clinical Trial Enrollment with Data Science Modeling,” Abhinav Chaturvedi and Amar Vaish, 2023, Calance, https://blog.calanceus.com/tech-studio/predicting-clinical-trial-enrollment-with-data-science.

About the Author:

Norman M. Goldfarb is executive director of the Site Council and of the Clinical Research Interoperability Standards Initiative (CRISI). Previously, he was chief collaboration officer of WCG Clinical, founded and led the MAGI conferences, and published the Journal of Clinical Research Best Practices. Those wanting to ask questions or make suggestions can contact Norman on LinkedIn.