Interview: Chris Austin on FDA's Wish List for Regulatory Science and Crowd-Sourcing Development (Part 2)

Posted 08 May 2013 | By Alexander Gaffney, RAC 

What comes to mind when you think about federal agencies that affect healthcare product regulation? If you're like many, the first agency to come to mind is the US Food and Drug Administration (FDA), followed by others like the Drug Enforcement Administration (DEA) and the Federal Trade Commission (FTC).

But a lesser-known agency called the National Center for Advancing Translational Sciences (NCATS) may soon be as integral to the regulatory process as FDA is, thanks to a large amount of funding and a new mission to help solve regulatory problems where no incentive yet exists. Formed in 2011 by Francis Collins, director of the National Institutes of Health (NIH), NCATS only recently got its first permanent director, Chris Austin, who is charged with running the 250-person agency and overseeing its $575 million in funding as it attempts to take on some of the most intractable problems in drug development.

With this in mind, we recently sat down with Chris to talk about his role at the organization, the promise of NCATS, regulatory science, translating discoveries, teaching the public about the drug development process, and what the future of regulatory looks like.

Miss part one of this interview? Find it here: (Part One), and check part three here: (Part Three)


Regulatory Focus: NCATS has been touting the use of standardized forms to accelerate drug development. So on one hand, you have these large and intractable problems, but at its core, there's a basic problem: Getting research started is tough. How big of a problem was this in the past that this required specialized, targeted attention right from the start?

Chris Austin: You raise an interesting question. You're referring, I think, to the template agreements that were worked out as part of the new uses for existing molecules program that involved around eight pharmaceutical companies and 58 compounds?

RF: Correct.

CA: I'll give you an anecdote which I've heard repeated many, many times.

When I was at Merck, the biggest problem was in drawing up a collaborative research agreement with an academic organization with which we wanted to work. Sometimes it would take two years to develop a collaborative research agreement.

No kidding, by the time the agreement got signed, the program was very frequently no longer of interest to either party, and many of the people were no longer at those institutions anymore. So the lawyers would come to us very proudly-'We've finally got the agreement signed!'-and there's nobody to do the program because science moves on, the question is no longer relevant, or the people are no longer there! There's a huge amount of wheel-spinning that happened, and continues to happen, at these organizations because of not scientific issues, but collaborative problems.

We view this as a classic NCATS problem because it is nobody's problem in particular, but it is everybody's problem. Nobody really focuses on this problem because they're just trying to get the stupid agreement done. Nobody focused on the question of whether there was a better way, couldn't we have template agreements which allow people to trade information. Couldn't we have, to cite an Internet analogy, an HTML-type situation? The Internet would not be around if there wasn't a common agreement about how information is to be communicated.

And yet we haven't been able to get people to agree to these things before now. It's also typically NCATS-ian because it is not "sexy." A lot of the problems in the translational space are not things that are going to get you on the cover of Nature. And therein lies one of the problems. Because the incentives, of course, are to get those "sexy" things done.

"A lot of the problems in the translational space are not things that are going to get you on the cover of Nature."

And I'm not knocking being on the cover of Nature-I've been on the cover of Nature-but what's very clear in the translational space is that we need an organization that is focused on the really important issues which are not in themselves going to result in a product being developed.

Another analogy that I sometimes use is that NCATS is a bit like the interstate highway system. Think about a problem in which you are trying to go from the East Coast to the West Coast. Now try to imagine that as a product development program.

The situation we encounter now is people buying and developing cars. They get down the road, to Pittsburgh or St. Louis or wherever, and the car dies. But instead of trying to develop a better car, or an intercontinental railroad, or an interstate highway system, people just go back to the East Coast and start again. They buy themselves another car and hope things work out better next time.

Not surprisingly, it doesn't work. Someone needs to work on vehicle technology, on road technology, or think up completely new methods of transport like railroads.

That's what NCATS does.

RF: Along that line, you've been working for a few months already with some of these larger pharmaceutical companies, all of which have approached you in one manner or another to get projects done. What are some of these projects, and how are you working together with industry? Have you have any noteworthy successes so far? Any surprises that caused you to go back to the drawing board?

CA: Realize that everything we do at NCATS is an experiment, including the nature of the collaborations themselves, so we gather information on how well the experiment goes.

I'll give you two examples, neither of which is published yet because it hasn't been long enough yet. But I can tell you what's happened so far.

In this collaboration we're doing now with Eli Lilly where we took the drug collection that NCATS put together and combined it with this very sophisticated set of assays from the Phenotypic Drug Discovery (PD2) program, where all the drugs are being screened across all those assays, and all the data is going to be made public. This is going to be, we think, an unmatched database for systems pharmacology to allow people to understand and we think predict the effects of drug-like compounds in living systems, something that will be beneficial to anybody trying to develop a drug.

"What I can tell you is that we're finding many, many activities of these drugs which we never anticipated"

That program got-despite the cost of the program and the fact that all the data are being made public-off the ground much more quickly than I expected, and now most of those drugs have been screened through the PD2 battery at Eli Lilly. We're in the midst of analyzing that data, and what I can tell you is that we're finding many, many activities of these drugs which we never anticipated.

That tells us two things. One is that the pharmacopoeia of current drugs-not a particular company's drugs, this collection is the integral of the entire history of drug development since it started, and contains every drug that has ever been approved-holds a lot more promise than we are currently aware, and we're busy analyzing that. Of course all those data will be made public as soon as we can figure out what it all means.

Second, on the new therapeutic uses program, we wondered whether the pharmaceutical industry would be interested in doing this, to be frank. And that was answered by the fact that there were three companies that were present at the time the program was announced a year ago. In the month between when the announcement was made and the RFA was put out, another five companies came to us asking to be a part of the program, and we were willing to make a huge amount of data available. There is more data on these drugs available than I ever thought there would be available in my lifetime. This is all proprietary data on unapproved drugs that have historically been quite secretive.

Another question was whether academics would be interested in this, as it's kind of a different model than what exists right now. And we got about 160 pre-applications to utilize these compounds. So that answered that question.

And then we asked, "Would people be interested in a variety of compounds, or just focus on a choice few?" We got applications for virtually every compound in that list, and in many cases-I think it was about 15 cases-there were four or more different applications for single compounds for different disease applications. That suggested not that there were many applications, but that there were many disease applications.

So we went back to the pharmaceutical companies and asked them, "Had you ever thought of this, and would you have ever done it yourselves?" And the answer to the latter of those two questions is obviously "no," because they would have done it already. But in many cases they hadn't even thought of it. Which is exactly what you'd expect! It's the advantage of crowd-sourcing.

What we're currently doing, to answer your question regarding roadblocks, is evaluating the efficiency and worthwhile-ness of those template agreements. The hypothesis was that this was going to smooth collaboration and make it more efficiency and speedy.

"It's the advantage of crowd-sourcing."

What we're now doing is going back to the investigators who got through the initial peer review and were asked to submit a definitive application. Those applications had to come from the academic and the company together. Not every investigator who was invited to submit a definitive application actually did so. And there could be a lot of reasons for that, but we're most interested in figuring out why. If the company and the academic were interested in the application, what happened? Why was there a fall-off? We're hoping to learn from that to figure out how to overcome the next set of roadblocks.

RF: Two related questions: First, given that these compounds have already in a sense failed-either to launch or in clinical testing-is there only upside here for all parties involved? Second, how does all of this affect the regulatory process going forward? There's one side of NCATS that's very much focused on toxicology and pre-clinical challenges, but is there a regulatory aspect aside from just finding more drugs?

CA: For this particular program, the roadblock that NCATS is focused on is a collaborative one-getting the football from the five-yard line and into the end zone with comparatively little effort. It's the converse of the drug failure problem. That 99.9% of drugs in development fail tells you that for every drug that succeeds, there are almost 1,000 that don't. Some of those compounds fail at later stages than others, but these compounds that are being investigated by NCATS all failed for efficacy reasons, not toxicological ones. Or they were shelved for strictly business reasons, such as because of shifting priorities.

In regards to the regulatory question and working with FDA, it's very clear to us that in order to do anything truly useful in the product development space, we need to identify needs.

Very early after becoming director, I went to talk to and have talked to all of the various parts of FDA, including Margaret Hamburg, Vicki Seyfert-Margolis, Janet Woodcock, Jeffery Shuren, Jesse Goodman, Karen Midthun… all of these folks are currently working on a wish list of FDA-NCATS joint projects.

"All of FDA's Center Directors are currently working on a wish list of FDA-NCATS joint projects."

I should say that we see the world in very much the same way, because we're both disease agnostic, we both want to make the product development translational process work better, and we both work in a world where there's a fair amount of misunderstanding about the whole drug development process and how it works that leads to a whole lot of mischief, which leads to a whole lot of nonproductive research.

There's now an NCATS-FDA working group that just met for the first time and came up with a list of six or seven different things that we're going to work on together.

RF: Can you talk about those things at all?

CA: I can't talk about them right now, but they are a lot of the things you might imagine them to be. They're the same problems that everybody realizes exist.

Just to summarize in no particular order the kinds of things we talked about are better ways to assess toxicity like the Tox21 collaboration and the tissue chip project, better ways to assess efficacy, better ways of understanding the broader effects of exogenous xenobiotics in living systems, new diagnostic criteria for diseases, new endpoint criteria for diseases-which is remarkably not a solved problem-and biomarker qualifications.

You know, everybody talks about biomarker qualification, but we don't really have a clear path of going from a candidate biomarker-a snip that someone discovers-to how it actually gets qualified before the eyes of FDA and its standards to assist in the review process.

These are all the kinds of things that we're talking about that we're going to be working on together.


Enjoying this interview? Check out parts one and three: (Part One) (Part Three)


Regulatory Focus newsletters

All the biggest regulatory news and happenings.


Most Viewed Articles