Q&A: Sponsor's perspective on site selection and patient recruitment

Precision recruitment modeling is current trend

[Editor's note: Joshua Schultz, vice president of worldwide patient recruitment, clinical research services of PAREXEL International of Boston, MA, answers questions from Clinical Trials Administrator about how biopharmaceutical companies are responding to current pressures regarding patient recruitment and selecting the best clinical trial sites for studies. PAREXEL is a global biopharmaceutical services organization.]

CTA: The biopharmaceutical industry is increasingly challenged to meet patient recruitment goals. Given your experience in this area, would you please describe what data-driven patient recruitment means and how your company employs it?

Schultz: Data-driven patient recruitment isn't new as a concept. In fact, investigators and sponsors have always used data to recruit patients and develop their estimates of likely recruitment speed. However, as the challenge of patient recruitment has continued to grow, the ability to refine these estimates through the use of modeling tools, investigator-specific performance data, and site-level patient data has become increasingly important.

The use of sophisticated recruitment modeling tools is replacing the reliance upon straight-line estimates of recruitment that have been traditionally used. Previous estimates that neglected to include key factors such as seasonal variations, variable site activation periods by country or type of site, and site fatigue were often off by 25% or more.

Combining more precise modeling, enabled by widely available technologies, with improved recruitment rate estimates allows for further improvements. PAREXEL relies upon a dedicated feasibility and evaluation group to survey investigators and compare this information to previous studies run by PAREXEL as well as published literature. This approach provides additional data to make more effective decisions regarding the number of sites, the length of the trial, and necessary recruitment support.

Site-level data, used properly, can be a critical component of successful recruitment. Sites with searchable patient databases can be much more effective at identifying and contacting potentially relevant patients for trials, while also being more accurate in their predication of the number of patients they are likely to enroll.

CTA: How might sponsor companies and sites best use technology to better model recruitment during the planning stage?

Schultz: The ability to more accurately model patient recruitment is substantially enabled by technology, either through relatively simple calculation tools (e.g., Excel) or custom developed computer systems. Regardless of the technology platform that is chosen, common issues must be addressed to model recruitment from a sponsor's perspective, including:

  • number of sites;
  • start-up timelines by country;
  • start-up timelines within a country for 20%, 40%, 60%, 80%, and 100% of sites;
  • first month of enrollment at each site that is unlikely to be a full 30 days;
  • gap between site initiation and patient enrollment; and
  • miscellaneous factors such as site fatigue, seasonal issues, direct-to-patient outreach campaigns, and database "boosts."

From a site's perspective, accurate modeling is predicated upon having good data on the prospective patients—names, visit schedules, potential interest in the study, and other sources of patients such as referrals or patient outreach. While some of these factors, such as the likelihood of subjects consenting, are more art than science, many of them can be more accurately predicted through database searches and site staff analysis.

CTA: In your experience, do sponsor companies and sites typically spend enough time in the planning stage of patient recruitment, and how can they improve this process? For instance, you've discussed creating more accurate feasibility estimates, and how might they do so?

Schultz: While sites and sponsors routinely engage in feasibility activities, given the enormous impact an accurate feasibility process can have on major budgeting and operational decisions, it is PAREXEL's belief that the industry would be well-served by spending more time on robust feasibility activities.

Feasibility should be driven by previous recruitment results from similar trials, modified by input from internal and external experts to address trial-specific issues. In addition, internal and external experts can identify other important factors such as standard of care by country, patient concerns, or site staff issues. Literature reviews can also provide insight into how others have solved similar problems and what has been achievable in comparable situations.

Through the combination of these elements, a reasonably tight range of most likely enrollment rates can be developed. This information should be compared to results of a blinded survey of potential investigators in each relevant country regarding likely enrollment rates, Institutional Review Board (IRB)/Ethics Committee (EC) issues, and other factors that may influence the conduct of the trial. This process plays a critical role in refining the analysis (described above) and identifying potential high enrollers.

This entire process should be coordinated by a central group separate from the trial team, allowing for process efficiency and objectivity, which are critical to accurate planning.

CTA: Does PAREXEL identify the high-quality and high-performing sites based on data, and what do you do with this information? Also, why don't more biopharmaceutical companies follow suit?

Schultz: To improve patient recruitment performance, the industry must rethink its use of investigator databases, which typically contain little more than contact information, previous trial participation, and areas of therapeutic specialty. Choosing the right sites is particularly critical given PAREXEL analyses showing 80% of patients typically come from the top 20%-30% of investigators across a range of therapeutic areas and countries.

By using a more data-driven approach to choosing investigators that includes previous enrollment performance (relative to peers), external data, and key metrics, it is possible to more accurately identify "high potential" investigators that are much more likely to be top sites that will drive recruitment for the study.

The industry has not managed this issue as aggressively as required for two main reasons. The first reason is that there is an ongoing misperception that the patient volume coming from most investigators is roughly equal; rather, the vast majority of patients comes from a small minority of investigators on any given trial. The second reason is that the industry has experienced difficulty in collecting and processing all of the relevant data from disparate systems. Fixing this issue often requires dedicated resources to implement and integrate systems for more efficient data management.

CTA: How can sponsor companies apply contingency-based escalation principles to manage recruitment during a study?

Schultz: PAREXEL has implemented a process of Predictive Management that focuses on using improved data assets (such as those described above) to plan for Last Patient In (LPI) and staging the appropriate tactics to manage this milestone despite changes in the environment. Predictive Management builds upon many of the traditional tools used to determine recruitment timelines, such as self-reported data from physicians and estimates from previous trials, and augments them with ongoing analysis of recruitment data gathered from systems such as Interactive Voice Response Systems.

For this contingency planning to be maximally effective, it is critical to develop early warning analytics and "triggers." These analyses should be driven not only by recruitment results, but also by variances to planned study start-up timelines, screening/enrollment rates, and average speed for sites to begin enrolling their first patients. These factors can provide early evidence of delays long before they are typically spotted using only recruitment results.

The recruitment plan should incorporate both the current situation as well as multiple levels of contingency activities based upon the potential need. Resources required, lead times, and expected impact should be part of this plan to allow for thoughtful and timely implementation of contingency activities. Developing these types of explicit escalation plans allows for thoughtful preparation to begin, such as getting patient outreach materials approved as part of the initial IRB/EC submission, which allows for substantial time and cost savings if enrollment results lag.

CTA: What is your advice to clinical trial sites that hope to survive in the increasingly competitive, global research world?

Schultz: With the right data and tools, patient recruitment can be more accurately predicted, monitored, and managed, while avoiding expensive and risky trial delays.

Clinical trial sites will need to partner with contract service providers and study sponsors to take a more strategic approach to patient recruitment execution, which can reduce costs and bring new products to market faster—improvements that are invaluable in today's highly competitive and increasingly global biopharmaceutical marketplace.