Clinical research monitors serve as the linchpins of clinical trials, ensuring operational integrity across research sites. Acting as representatives of the sponsoring organization, they coordinate with onsite principal investigators to maintain protocol adherence and smooth operations.
“I used to be at a site. I kind of thought of him as a little policeman coming and making sure that I did everything right,” said Pamela Tenaerts, Chief Medical Officer at Medable, a health technology company specializing in clinical trial administration software.
Pamela Tenaerts is the Chief Medical Officer at Medable.
Medable
These monitors evaluate patient enrollment, drug supply levels, medication changes indicating potential adverse events, and numerous other variables. Recognizing the sheer volume of data these professionals must track, Tenaerts and her colleagues at Medable developed an AI tool to assist them, reasoning it would be invaluable.
Their creation, Medable’s Clinical Monitoring Agent, checks documentation for missing fields, identifies discrepancies in trial paperwork, and tracks variables across multiple sites. While the team hoped these functions would save time and foster stronger relationships with onsite teams, they needed empirical evidence to quantify the tool’s actual impact.
“We always wanted to lead with evidence and understand how these tools are changing the trajectory, hopefully for the better,” said Tenaerts. “It felt like it would make sense that this would be a benefit, but until you test it, you don’t know.”
Medable partnered with the Tufts Center for the Study of Drug Development (CSDD) for an independent analysis. Using Medable’s data benchmarked against recent real-world oncology drug development data, Tufts CSDD researchers modeled the tool’s impact. Although not yet peer-reviewed, their findings indicated that the AI agent enhanced monitor efficiency and could save drug companies millions in development costs.
Modeling Benchmarked Clinical Trial Data and the AI Agent
To evaluate the agent, the Tufts team assessed it against five years of benchmarked oncology clinical trial data. “We chose oncology because it’s the most active area in drug development and reflective of broader R&D pipeline shifts,” said Ken Getz, Executive Director of Tufts CSDD. He noted that programs in oncology, rare disease, and immunology increasingly target narrowly defined patient populations.
Ken Getz is the Executive Director of the Tufts Center for the Study of Drug Development.
Tufts CSDD
However, Getz noted that this analysis could be applied to other disease areas, such as metabolic or cardiovascular development, by using benchmarked data from the respective target patient populations in their models.
In their initial analysis, Getz and his team evaluated the basic return on investment for a single trial. They found that the AI agent reduced costs by minimizing the number of site visits required, thereby lowering travel-related expenses.
When analyzing data by phase, the team found that savings from reduced site visits amounted to $4.4 million per Phase 2 trial and $5.6 million per Phase 3 trial. Tenaerts explained that Phase 3 trials carry lower failure risk than Phase 2, resulting in higher predicted savings. “There’s still risk, but it’s decreasing as you move through the funnel,” she said.
Financial Gains with the AI Clinical Monitoring Agent
The Tufts CSDD team also applied the Expected Net Present Value (eNPV) model, which factors in multiple risk variables to determine financial value. “We look at an entire development program, adjusting for technical, regulatory, and operational challenges, as well as commercial performance, and discount it to present value,” explained Getz. “It’s expressed as net present value creation, comparing the financial value of a program with versus without the AI monitoring tool by phase.”
Overall, the researchers found that Medable’s AI agent generated eNPV gains of $7.5 million for Phase 2, $11.3 million for Phase 2/3, and $21 million for Phase 3. The agent also accelerated trial enrollment by 18 weeks, shortening the clinical development timeline.
Medable created an AI Clinical Monitoring Agent to assist clinical research monitors.
Medable
Getz added that certain unmodeled aspects of drug development could affect financial returns, such as a drug gaining additional indications or expanding into new patient populations. “There’s a lot of inherent intangible value that might be realized in subsequent programs,” he explained. The team is preparing a manuscript for submission to a peer-reviewed journal by mid-October.
For Tenaerts and Getz, the most rewarding aspect was obtaining independent evidence that the AI clinical monitoring agent is as useful as they had hoped.
“For the first time there’s actually an empirical study that shows this can help,” Tenaerts said. Getz agreed. “It took so much time to set up this model the right way and derive these conservative estimates, so that part has been really exciting,” he said. “We know that this is something the industry has been waiting for.”