Trial, not error

INC Research forecasts patient enrollment in clinical trials

Clinical trials contribute heavily to the cost of bringing new medicines to market. INC Research is helping its customers control those costs by using analytics to forecast patient enrollment in trials.

“Understanding patient enrollment is key to forecasting how long a trial’s going to run, but the uncertainty of site start-up timelines and enrollment rates make this challenging,’’ says John Whitaker, Senior Vice President at INC. “It is important for us to be able to develop solid plans for our customers and to detect, as early as possible, when things are not going as planned.”

That’s an important consideration. Phase III trials, for instance, can eat up a whopping 40 percent of the $5.8 billion dollars it costs to bring a drug to market, according to Forbes magazine.

There are some standalone trial-modeling products that some of our competitors charge extra for. Ours is integral to the services we provide.

John Whitaker
Senior Vice President

INC Research is a therapeutically focused global contract research organization (CRO) that helps biopharmaceutical customers develop new medicines by conducting Phase I through Phase IV clinical trials. The company’s experience spans more than 100 countries and its expertise is aligned with the largest and fastest-growing therapeutic areas.

Enhancing predictability

Accurately estimating the duration of a clinical trial has been a challenge in the past, says Whitaker. “We would look at historical data from multiple systems, ask managers of similar trials, and look at published papers.” Enrollment tracking was manual, with data coming from multiple sources.
INC now has a better approach. Using analytics, the company’s model forecasts trial startup and enrollment using data from multiple internal systems and third-party vendors. The model was prototyped in a week and checked against 10 historical studies. “We found phenomenal fits on nine. The 10th study was an infectious disease study where monthly enrollment was far more seasonal.’’

This forecasting model has been used for more than 200 trials and has become a standard practice for proposals and trial start-up. Feedback has been overwhelmingly positive. “Customers like the disciplined methodology and that we can better quantify enrollment risk. Our project teams like it because it saves them time and provides actionable information.”

INC Research is in the process of rolling out the next phase of functionality to automate study tracking. “When project managers log in, they will quickly see how their trial is performing against plan and revised estimates to complete enrollment,’’ Whitaker says. From there, INC Research staffers can discuss corrective actions, if needed, with the sponsor for shutting down inactive sites, adding more sites or working with existing sites to find more patients.

Trials in existence before modeling began have not been transitioned over, though Whitaker reports that staff on those trials are eager to work with the forecasting and automated tracking.

Expanding the role of analytics

INC is now looking at the historic enrollment performance of sites to better understand which sites are particularly good for which types of drug trials, and which physicians and practices are particularly engaged in finding patients for trials. This is critical because it takes a great deal of time, money and regulatory hurdles to get a site up and running for a clinical trial. “Eventually we hope to not just forecast how long a trial will take but also figure out how to help shorten that timeline through better site selection,’’ Whitaker says.

“This is just the tip of the iceberg. Being able to better predict patient enrollment not only allows us to better manage timelines but also model when patient data will be available, which in turn would allow us to better manage our workforce allocations. We are looking at analytics to help us better manage our risk-based monitoring efforts and eventually we’d like to look at network analysis software for fraud detection. Further, we wanted a standard offering that comes automatically when a customer signs on with INC. There are some standalone trial-modeling products that some of our competitors charge extra for. Ours is integral to the services we provide and was built with upstream and downstream integration in mind.’’

INC’s ability to offer solutions like this one continues to create new opportunities. It’s also opened some interesting – and sometimes challenging – conversations with customers. “We’ve had customers who’ve told us they think the trial can be completed in 18 months and we come back and tell them based on our forecast it looks more like 28 months. That was a hard message to deliver in the past when you were using a spreadsheet. But when you can walk them through the data-driven modeling methodology it becomes a very concise argument,’’ Whitaker says.

Challenge

Forecast the time it takes to complete a clinical trial.

Solution

SAS® Business Intelligence

Benefits

  • More accurate forecasts compared to previous method of forecasting.
  • Modeling is a key selling point with new and existing customers.
The results illustrated in this article are specific to the particular situations, business models, data input, and computing environments described herein. Each SAS customer’s experience is unique based on business and technical variables and all statements must be considered non-typical. Actual savings, results, and performance characteristics will vary depending on individual customer configurations and conditions. SAS does not guarantee or represent that every customer will achieve similar results. The only warranties for SAS products and services are those that are set forth in the express warranty statements in the written agreement for such products and services. Nothing herein should be construed as constituting an additional warranty. Customers have shared their successes with SAS as part of an agreed-upon contractual exchange or project success summarization following a successful implementation of SAS software. Brand and product names are trademarks of their respective companies.

Back to Top