Taking The Guesswork Out Of Site Selection

By Anna Rose Welch, Editorial & Community Director, Advancing RNA

As the industry continues to be shaped by globalization, regulatory pressures, and a growing demand for trial participants, clinical trial site selection inherently becomes a more difficult task.

Clinical Leader turned to Ramita Tandon, SVP and General Manager, inVentiv Clinical Trial Recruitment Solutions (iCTRS) to gain insight into current trends, challenges, and best practices to follow when selecting a site for a clinical trial — especially in an age when inaccurate patient population estimates abound and threaten trial efficiency.

Anna Rose Welch: What are some of the changes in the industry that are affecting site selection?

Ramita Tandon: The industry itself has been challenged by increasing regulatory pressures, which are creating more demand for patients. At the same time, the number of available investigators has grown very little. The clinical trial enrollment process has notoriously been a significant limiter in bringing new treatments to market, and trials are consistently delayed.

Sponsors can accurately forecast enrollment timelines and identify variables that can be adjusted to influence those timelines by employing data, analytics, and technologies to bring more certainty into the equation. However, these solutions are often overlooked, and critical planning decisions are often based on unreliable estimations from the onset of the trial, making it near impossible to adhere to study timelines.

Welch: Are these unreliable estimations a consequence of inefficiencies in current selection procedures in the industry?

Ramita Tandon, SVP and General Manager, inVentiv Clinical Trial Recruitment Solutions

Tandon: Yes, trial planners simply do not have adequate information about research centers and locations where centers operate.  This is probably one of the biggest inefficiencies in the industry today.  Sponsors often put blind faith in selecting a site without thorough research.

To combat this, it is common practice for sites to be surveyed with questionnaires for information about their sites, such as local patient population, research activity, infrastructure, personnel, and timelines. Theoretically, this should be a simple solution — but it’s not the reality.

Most physicians do not actually run queries against a patient database to answer the questionnaire with any precision; they simply provide a rough estimate of their current patient population. Patient counts gathered directly from investigators should be taken for what they are: a best-guess estimate provided by a busy professional eager to do the best thing for their patients.

Similarly, feasibility surveys frequently go unanswered at investigator sites. The questionnaires are time-consuming, and the ones that are answered are repetitive and often paper-based, which results in inaccurate information. Therefore, it is impossible to know how many potentially productive sites are overlooked due to inaccurate or incomplete feasibility surveys.

Welch: In light of some of these challenges, have any best practices emerged that have helped streamline site selection?

Tandon: Overall, we take a data and analytics driven approach to selection. This allows us to overcome common challenges associated with study planning decisions that are made based on inaccurate estimations. EHR, integrated claims, and pharmacy transactional data all help pinpoint sites located closest to a high concentration of patients who meet the study protocol’s inclusion/exclusion criteria. Reviewing historical information on sites is also a tactic we have used to understand past experience in a disease state as well as current study activity. Lastly, we use proprietary scoring algorithms that generate a composite performance score for each site. Data and analytics play a key role in ensuring that high performing sites with access to an abundant number of patients are selected for a study.

Welch: Have you found any deal breakers in terms of deciding whether a site will or will not be feasible for a trial?

Tandon: For us, the biggest deal breaker is if a site does not have access to an ample number of patients. Similarly, if supplementing site recruitment activities is necessary, it is important to consider where these patients are in relation to the study location. They must be within a reasonable distance from the site location.

Welch: What kinds of trials/indications are the most difficult for which to find sites, and why do you think this is?

Tandon: It is usually more complex and time-consuming to find sites for rare disease trials. One reason for this is that there is usually not enough documented experience in trial intelligence databases to draw from. As a result, more thorough investigative research is needed to locate the right sites to support these types of studies. This type of research is often more art than science, and may involve interactions with advocacy groups and relevant disease networks. Once a few key investigators have been identified, however, they often serve as good referral sources for other investigators experienced in the particular rare disease being studied, as these physicians tend to form close, informal networks.

Welch: Is there anything else that could be done to make this process smoother?

Tandon: There are several industry initiatives in progress to address the need for information on site capabilities to support trials in the various diseases areas. Some of these initiatives involve collection and aggregation of trial experience data from public and proprietary sources, while others rely on experience updates from individual sites.

Welch: As trials begin to span different continents and cultures, what are some things to keep in mind to ensure successful site selection?

Tandon: In any state, country, or culture, it is critical to remember the patient throughout the site selection process. To pick the right site, you need to have a deep understanding of the patient. Conducting behavioral research on a global scale can help uncover patient insights specific to the disease state population, as well as potential barriers and motivators, which vary by country and culture.

In addition to using global data sources and predictive modeling, we have also found it helpful to leverage our global footprint of staff across 70 countries worldwide in our site selection efforts. Performing this kind of global outreach to gain knowledge of local standard of care, disease incidence and prevalence, and local knowledge of clinical research best practices will benefit a study.