United BioSource (UBC) has been built one acquisition at a time from a 9-figure investment war chest. The company provides clinical trial research services, but is not a contract research organization (CRO) in any traditional sense. It counts some CROs as customers and calls itself a specialty services organization. With 1,000 employees, UBC reports that more than 30 percent have advanced degrees. The company supplies plenty of technology, but is not a typical vendor. (The company partners with firms like Phase Forward and Medidata for electronic data capture.)

“We have a high focus on late-stage clinical trials. We do a lot of registry work. We do IVR. And we do data management,” says United BioSource’s general manager Mike Borkowski.

Based in San Francisco, Borkowski says traditional CROs are in a bit of a quandary about adaptive trials. They can’t leave a minor trend untouched. They want market share. “The large CROs have to jump in,” says Borkowski. “The industry is moving that way.”

Fighting the Trend

But there’s a risk. The whole value proposition of adaptive designs is that more trials of ineffective drugs will be stopped early. That’s good for the industry as a whole. But such cancellations potentially threaten the steady nature of traditional CRO revenue streams. Says Borkowski: “If you are able to cancel it early, you save millions of dollars. Those are millions of dollars that would otherwise have gone to a CRO.”

Another issue is the biostatistical expertise in Bayesian and other high mathematical arts as they are applied to the clinical research realm. That expertise is not necessarily easy to locate and retain. “Those people are maxed out,” he says. “Their plates are full.”

In United BioSource’s case, Borkowski notes, one of the field’s top biostatisticians has an established relationship with the company. “We work with Don Berry. He’s a good friend of ours.”

Core Strengths

United BioSource, Borkowski says, senses an opportunity for adaptive trials for which its in-house expertise and technologies are well-matched, not only in biostatistics but also data management, technology and clinical operations. “We have a team of people. Those things are our core businesses and they’re things that are core to adaptive trials,” says Borkowski. “That’s why you find us aggressively in that space. We can handle all the key elements that relate to an adaptive trial.”

Perhaps forty people at UBC work on the nuances of such studies. “What we do is provide the supply management solutions, the adaptive randomization hosting and randomization calculation.” The company can also help with in silico trial simulations that lead to changes in the trial design.

An Adaptive Surge

Are adaptive trials under way in significant numbers? Or are they part of the industry’s future?  With the statistical community on board, and FDA concerns widely understood, Borkowski says the next 18 months will see a more dramatic increase of actual adaptive trials starting. A handful of big-pharma leaders—Wyeth, Merck, Novartis—will not be the only companies doing such studies, Borkowski says.

“Over the next year to 18 months, you’ll see a real ramp up in the number of adaptive trials being run,” he says. Finding external or internal resources to run such trials may be challenging throughout the industry, Borkowski says. “There are capacity issues in the industry with regard to how much people can do.”

The Right Rand

Borkowski believes that one technology platform to run adaptive trials is something that the industry would like. But it’s by no means clear when that platform will be ready.

But there’s one thing that may not be included in that platform: the randomization engine. He believes the rand will be something that some sponsors will want to retain tight control of. “Most of the players don’t know how to do the adaptive randomization. People want to keep IVR outside of their CRO. The randomization of the trial is the centerpiece of the trial. They don’t want that in the hands of the people running the trial. In an adaptive trial, there is even more reason to keep it separate. We have customers tell us that. You cannot do randomization and run the trial.”

Of course, the CRO industry remains in a thriving condition. It may well find ways to adapt to adaptive trials just as it has adapted to electronic data capture and other theoretically disruptive technologies. If adaptive trials in any quantity remain rare, special projects, there will hardly be any measurable impact on CROs. On the other hand, if all trials routinely use adaptive techniques at some distant time in the future, CROs could well be obliged to accept a less predictable revenue stream.