Investigator sites are very important to clinical studies because they make sure that new drugs are safe and work well. In clinical trials, which are long and complicated, picking the right places is very important for success. Investing time and effort in these trials can lead to diminishing returns. Estimates suggests that it costs in the region of $3B to bring a new drug to market [1], with daily revenue losses in the range of $1M to $8M due to delayed market entry.

Choosing high-performing sites that are right for running certain studies is key to avoiding budget overruns and study delays, which often feed the growing ‘study rescue’ business. In the past, however, the selection process was often done by hand, which was time-consuming and led to a lot of wild speculation about how true detectives' claims of abilities were.

Forget the hyperbole, the numbers tell a sobering story. According to research from the Tufts Center for the Study of Drug Development (CSDD), almost 40% of clinical sites selected for trials fail to enrol their target numbers, and 11% didn’t recruit a single subject. Most (approx. 90%) studies eventually meet their enrolment objectives, but often achieve it at the expense of extending the study (often doubling initial estimates). The predictable response of contract research organisations (CROs) involved in delivery of these studies is to recruit more sites than estimates originally suggested would be needed, in anticipation that several will underperform (and may eventually be dropped) impacting on delivery objectives.

This knee-jerk response has fuelled a lack of trust and transparency between sites and sponsors/CROs, potentially causing needless increases in costs and study delivery timelines. The 2016 Tufts CSDD report notes that each year about 40% of investigators decide not to get involved with any future clinical trials [2], Just as typical multicentre studies require 30% new investigative sites [3], and the clinical research industry continues to experience a concerning global shortage of experienced clinical researchers, the lifeblood professionals whose main function is to monitor clinical trials [4].

Research also cites how slow patient enrolment is the top reason clinical trials fall behind their delivery schedules. Overall, poor site selection, choosing sites that are unable to meet predicted rates of enrolment, and the subsequent need for study rescue activities can markedly increase the delivery costs of trials. Most disturbingly, despite this understanding and the investment of considerable time and efforts to address these issues, cycle times have not changed in more than 20+ years [5].

Sponsors and CROs, often lack a transparent, evidence-based site selection strategy. Even now, with the advent of algorithms and network-based search strategies, many recruitment plans still rely on spreadsheets. At the Sponsor level, the process often relies on data provided by CROs as part of feasibility assessments conducted when the CRO was first involved in the bidding process. Clearly, there is a significant gap between the CRO’s assessment of feasibility and in identifying sites that are going to deliver the study. More recently, CROs and specialist service providers have begun to offer interrogative (some say intensive) software solutions that scour the internet and various online resources to identify and, in some cases, validate top recruiting sites, potentially increasing return (in terms of speed and completeness of recruitment) on Sponsor investment.

The issue the industry is facing in part derives from the fact that site selection (being able to identify good sites) has been an often-a poorly understood discipline, with the consequences of it being done badly not being appreciated. The result – it has become a critical factor in the time taken and cost incurred in bringing new drugs to market. The application of sophisticated search strategies to overcome these limitations and deliver a swift resolution is certainly appealing. Globally, various databases can give insights into site performance. These include commercially available databases such as DrugDev (IQVIA, USA), Citeline (Informa Plc., London, UK), Global Data (GlobalData Plc., London, UK) or public repositories (e.g., National Library of Medicine Clinical Trials Registry, www.clinicaltrials.gov) [6]. However, the service is not inexpensive and you have to ask, how wise is it to replace one system you don’t understand with another? A smaller scale approach is to investigate the publication track record of individual investigators on PubMed.

Regardless of the sophistication and intelligence (artificial or otherwise) of any system you employ to identify potential sites to include in your study, the problem remains a very human one. At the end of the day, your success will depend on the relationships you build (with the site staff and the investigators) and your understanding of the problem at hand. Experience and your own network will be invaluable – perhaps better suited to providing you with the objective answers to somewhat subjective questions. And yet, Sponsors frequently rely on archaic and totally underpowered paper-based or spreadsheet methods to identify sites across the globe that may offer what they hope will be a reasonable chance of enrolling the required number of patients on schedule, and generate the quality data they need.

Criteria to help optimise site selection include:

1. Confirm site infrastructure: Does the site have the required infrastructure to meet the requirements specified in your study protocol?
2. Establish site staff experience: What staff does the site have available that will conduct the trial and what is their trial experience?
3. Capture past performance: Experience is relative and a site’s history of delivering similarly sized and complex studies must be considered, ensuring that any recent performance reflects past performance (don’t forget to check whether the site or investigator appears on any blacklists).
4. Identify disease scope: Sites that specialize in managing the disease under study are less likely to experience difficulties in recruiting and so should be set up to manage the needs of their patients as well as implement all aspects of the trial protocol (unless they are getting involved in too many studies).
5. Measure proximity to target population: Travel is a major consideration for patient engagement.Selecting sites that are ‘close’ to a population catchment of sufficient size (for the numbers you need to recruit) is a factor to consider and should influence how quickly the site anticipates completing enrolment.
6. Review enrolment history: Has the site met enrolment targets previously? This would provide some indication (and potential validation) of a site’s enrolment ambitions.
7. Determine start-up cycle times: Does the site measure their cycle times and if so will recruit appropriately? For example, does the site have data on ‘site activated to enrolment cut-off’ (a measurement of time to full enrolment of subjects, that the site has committed too), ‘enrolment cut-off to last-patient-in’ (a measurement of the trials completeness at the site with subjects)? Are the site data on par or better than industry benchmarks?
8. Detail access to patients in target disease: It seems obvious to confirm whether the site currently treats patients for the disease and are they the right specific disease type? If so, how many do they have access to? Are they prepared to enrol these patients in the trial?
9. Define the investigator track record: has your investigator published on the target condition under investigation? A ‘thought leader’ in the therapeutic area with connections to colleagues in the field increases the likelihood of timely engagements and study commitment, and may be a champion for recruiting other potential investigators and/or sites.
10. Seek continual innovation: You need to engage sites that are up-to-date in terms of best practice and trial conduct? Does the site perform post-mortems on studies to gauge areas for process improvements? What new processes/systems have they recently adopted? Why? And what was the result?

There is no denying that site engagement is a critical success factor in clinical trials. These tips should be considered when conducting site feasibility assessments and the pre-study visit, a critical event that sets the stage for an open, collaborative relationship to last throughout the study – imperative to the overall success of a study as many crucial tasks are accomplished during these visits. Engaging with the suggestions here could eliminate or reduce the need for extensive surveys and pre-study visits.

Optimising site selection results in an efficient operating model, where study teams come together to adopt an united front with shared goals and aligned structures and processes. Getting this right will lead to better use of resources, lower costs, and, in the end, more successful studies. Want to know more? Read our Insider’s Insight on site selection [7].

References

1. Wouters OJ, McKee M, Luyten J. Estimated Research and Development Investment Needed to Bring a New Medicine to Market, 2009-2018. JAMA. 2020;323(9):844–853.
2. Getz K. Study Assessing Practices and Inefficiencies Associated with Site Selection, Study Start Up, and Site Activation. Proprietary presentation 2016, citing data from the Tufts Center for the Study of Drug Development.
3. Getz K. Uncovering the drivers of R&D costs. Proprietary presentation 2015, citing data from the Tufts Center for the Study of Drug Development.
4. 21st Century Clinical Monitors - How Will Industry Address Significant Shortages in Experienced, Well- trained Talent? Lisa Feeney, MBA, DIA 2015, June 14- 18, 2015
5. Getz K. Assessing and addressing site identification and activation inefficiencies. Tufts Center for the Study of Drug Development. March 2016.
6. Laaksonen N, Bengtström M, Axelin A, Blomster J, Scheinin M, Huupponen R. Clinical trial site identification practices and the use of electronic health records in feasibility evaluations: An interview study in the Nordic countries. Clin Trials. 2021;18(6):724-731.
7. Site selection - identifying high performing clinical sites: An Insider’s Insight.