By Irene Yeh 

March 13, 2024 | Finding the right trial site with access to a diverse patient population is challenging. While the clinical trial and research fields have made great efforts to overcome these challenges, there are still obstacles that slow down the process and frustrate researchers. 

During February’s SCOPE 2024, Anju, in partnership with Incyte Corporation, presented their intelligence tool, TA SCAN, that is designed to help determine site capacity and vet population diversity and demographic data. The presentation was given by Elke Ydnes, Associate Director of Business Solutions at Anju Software; Stacy Eckstein, Manager of Trial Informatics at Incyte Corporation; and Luke Moyer, Head of Global Clinical Supply Chain at Incyte Corporation.  

What to Consider When Finding the Right Trial Site 

Site selection must consider indication, access to a diverse patient pool, good enrollment rate, site capacity, production of quality data, relationship with the sponsor, being within the sponsor’s footprint, interest in study drug, no competing trials, fast study start-up timelines, and more.  

These factors are the starting points for the strategy researchers want to set in place, but to effectively go about it, data are needed—and gathering data can start from the inside. 

“When it comes to our internal enrollment data and our internal performance data, it’s important that we also ask our CRAs and our CTMs, the individuals that have these personal site relationships with the site,” comments Eckstein. “They will be able to tell you a lot of information about our sites that you may not even glean out of a feasibility questionnaire.”  

Internal Data and External Data 

But to get a full picture, both internal and external data need to be collected. Internal data provides information to help researchers understand site contact information, gauge the historical performance of these sites, have PI and site score cards, and access quality metrics. Other internal data include supply forecast, costs and budgets for clinical costs, and financial impact. Internal data are also specific to a company and provides transparency, which is “definitely a big plus in trusting the data sources,” according to Ydens. 

However, there are drawbacks to internal data. While they are a great starting point, internal data cannot address every single factor and question considered, Ydens continues.  

External data give more information by providing epidemiology data to understand the patient profile, as well as serial intelligence on site selection and the costs associated with it. External information can also tell a researcher about what is happening in the space, country recruitment, industry benchmarks, the competitive landscape, current on-going studies, and whether a site has access to a diverse patient population. However, having all this information may not necessarily be ideal.  

“None of these data sources can give you 100% certainty into what would be the perfect site for your upcoming study,” Ydens warns. “But [they] are pieces of the puzzle that could contribute.” 

TA SCAN Intelligence Tool 

Anju’s TA Scan helps put these puzzle pieces together by analyzing data points of the site capacity. Ydens demonstrated by showcasing the tool during the presentation. TA Scan presents different site points across the country as dots, and when a researcher hovers over the dot, the tool provides site capacity. There are graphs used to indicate if a site can or cannot take on an additional study.  

In addition, TA SCAN records population diversity and demographic data of the location. For instance, TA SCAN can help you see if there is a large enough African American population for your study, thus allowing researchers to set realistic targets and open the conversation with the sites they are considering.  

The tool also allows researchers to assess if an investigator has interest in the indication and has published or presented on a specific indication or even a subpopulation. TA SCAN provides full profiles of these investigators and an unbiased scoring system that allows easy navigation to determine if investigators have a scientific or clinical footprint within the target indication. The tool can help researchers understand if these investigators have a real interest and past experiences with the indication. 

Optimizing Overwhelming Data 

Though TA Scan can provide useful information, this results in another challenge: too much data to organize. This is nothing new in the industry. In fact, it is a chronic issue. But there is a “work-in-progress” solution that Anju and Incyte are developing.  

This project is in the works with a “soon-to-be-named partner” and is a currently unnamed optimization tool designed to take and integrate all internal and external sources, which will then allow a user to get a similar result in a consistent, repeatable, and reproducible way, according to Moyer. 

“We are in the process of developing the data lake,” says Moyer. “Once that's ready, then we'll sort of be able to transport this over to them. And this will enable us to sort of reach into that deep well of data to come up with something.” 

For instance, if a researcher selects a certain country for their trial site and assesses if it meets their diversity plan, the optimization tool will be able to analyze several different scenarios. The tool can then analyze and compare other factors, such as the speed of enrollment vs. the cost of enrollment. Looking at these selected scenarios, it will help with the process of site identification, feasibility questionnaires, and site selection. Once the sites are selected, the output of the tool will be a site-specific enrollment forecast that will then be integrated with other systems, such as the supply chain forecasting and budget forecasting, thus allowing researchers to know when they’re falling off track and revaluate what changes need to be made. 

Anju and Incyte plan to unveil more on their optimization tool for next year’s SCOPE.  

“It's important that we start by knowing what questions we want our data to answer,” says Moyer. “This is the vision that we integrate these systems with the optimization tool. We start less wrong, and then we get increasingly less wrong throughout the life of the study by integrating the real-time data to bring it in, continue to re-forecast, update all those dependent systems at the same time, and track it to the time it takes to meet the milestone. That’s the vision.”