Acting Deputy Commissioner for Operations
Name: Janet Woodcock
Position: Acting Deputy Commissioner for Operations, FDA; Director of the Center for Drug Evaluation and Research, FDA, 1994-present
Background: Director of the Office of Therapeutics Research and Review, Center for Biologics Evaluation and Research, FDA 1992-1994
Medical Officer, Division of Biological Investigational New Drugs, CBER, 1986-1992
Education: MD, Northwestern University Medical School, 1977
Janet Woodcock has been instrumental in ushering in the Critical Path plan, the goal of which is to streamline drug development and evaluation, partly through the adoption of new technologies by industry and regulatory agencies. One field of these technologies is pharmacogenomics Woodcock speaks often about the use of genomic, proteomic, imaging, and other technologies in drug development, and how to get them there.
Woodcock is giving a keynote address at the PharmaDiscovery 2005 conference in Washington, DC, this week entitled "FDA Perspective on Challenges and Opportunities on the Critical Path to New Medicinal Products," and we phoned her up to ask for her, um, guidance.
The description of your talk mentions a 'product development toolkit' to help the pharma industry 'close the productivity gap' to create safer and more effective medical treatments. Can you elaborate on what this toolkit is?
The tools we're talking about are evaluative tools; they're what the industry uses, what FDA uses, and frequently, what we hope the clinical community would use, after approval to monitor or select patients. And the tools are everything from new animal models to new ways of predicting toxicity, to things like pharmacogenetic tests, pharmacogenomic tests, proteomic tests, [and] imaging technologies. All of these, hopefully, will give us better ways to design the trials, and then execute the trials and see whether the products are safe and effective.
Are you trying to finalize a collection of various techniques?
I don't think 'finalize' is a good description we see this as an ongoing process where we update the toolkit. We believe, in general, that the toolkit is everything from late-19th century to 20th century toolkit, and we're using that to evaluate 21st century products, and we think the toolkit has to be updated.
How are you recommending that the pharmaceutical industry conduct drug development. What's different about this effort in diagnostics?
The real difference is that everyone wants to move from what has basically been, after the discovery phase, almost a trial-and-error approach, to a more mechanistic approach, where we have much more predictive information that we can generate at each step of the way. And everything from predictive in vitro or animal-model testing all the way to much more informative human testing.
For example, we recently released a draft guidance that is part of this toolkit. It's called the Exploratory IND, and what it's about is getting into people before you actually get into formal drug development Phases I, II, and III. And people call these studies Phase 0 studies sometimes, and what they are is, they're either microdose or lower dose, or short duration exposure. But the main difference is, the objective of these studies is to study the biology of the product or the pharmacology of the product, not dose-limiting toxicity, like you might do in Phase I trials. And so, these studies there are now promising ways to do microdose with new types of mass spectrometry, so that you can actually do metabolism and pharmacokinetics. You can study multiple compounds in the same IND at these low doses and pick the one you want to carry forward as a candidate based on favorable receptor binding, favorable kinetics, or favorable metabolism whatever you're interested in. You could do imaging with these compounds and these low doses. You could do proof-of-mechanism studies for example, if you think your compound needs to localize at a tumor or certain anatomic site, or get into the brain, then you could do imaging studies, say, with a molecular probe to see how the compound interacts with that probe in the central nervous system.
So, this is an attempt to study, very early, the properties of a drug in a relevant species us.
Will the toolkit ever lead to requirements?
Well, these will be recommendations until as per our pharmacogenetics guidance if any of these become validated biomarkers, then we may require them to be used. For example, you have to have some idea of the metabolism, if it's possible, now in drug development. But that's because we understand the utility of that. But while these are still early, exploratory types of markers, we want to get them developed and qualified.
Is there somewhere to take a look at the toolkit?
We will be publishing a list of opportunities in a month or so. The opportunities list is more like the research that should be done to come up with this new toolkit. It's a huge list of projects that need to be done in various ways.
What we're doing is, we're collaborating with a large number of partners to get some of this work done, but by no means can we do all the work with our given resources. So we hope to stimulate interest in developing some of these things.
Who are you working with?
Well, we're working with the industry, with academia, with groups like the different institues in the [US] National Institutes of Health, for example.
When do you expect a toolkit to be ready?
We think that certain new tools may be available in a year or so, and then we would hope to build on that. It's like any other applied science it's going to take a while to get some of these things out there, but then we can steadily build on them.
What do you expect to be in some of the first versions?
Probably some of the first versions would be our computer models of disease and so forth they would be available next year, I believe. That's a software tool for modeling and simulation use in early trials.
What sort of impact to you expect this all to have?
Right now there's kind of a bottleneck in moving a lot of the new science through the process, and over the next decade, we hope to have an incremental impact in making the process more predictable. Therefore, making drug development more attractive to more players, and actually having better-understood products at the end of the day.
If you work in the in vitro area, what we want to see is more validated diagnostics coming out that can be used not just in drug development, but used to select patients afterwards people who are likely to respond, as well as people who should not take the drug because they're at high risk for an adverse event.
This should aid the development of new diagnostics?
We hope so. It's a huge focus of our effort, in diagnostics.
How will it do that?
One of the problems with diagnostics is that there isn't a really good business model. And what we're doing is getting the pharmaceutical industry interested in getting their muscle behind us, and also working with the NIH and other parties. We're telling everyone that we can tell that diagnostics are excruciatingly important they're actually the foundation of medicine, because you have to know what the person has. And personalized medicine subsetting patients and everything it's going to be the future of medicine. We can't do that without good diagnostics.
Is it your feeling that diagnostics have been ignored?
Does the pharma industry seem interested in going this direction, in changing this model?
Yes, they're very interested. We've talked to the research directors of all the big companies, and to bio[tech], and the small companies. There's tremendous promise here, it's really frustrating.
How is it frustrating?
Well, because some of these things have been developed within companies, and there's a lot knowledge within companies, but this knowledge has not been shared.
I've heard mention about encouraging industry to share knowledge are you encouraging them?
Well, we're not encouraging it we need to set up structures and organizations where that can be done where people can feel comfortable sharing information. We could have appropriate protection of intellectual property.
Sharing with the FDA or with each other?
I think the FDA may have to broker this, but at the end of the day, the information has to be made public for it to be useful.
So this is beyond the IPRG.
Yeah that's for sharing information with the FDA. What we're talking about here is that companies have a lot of biomarker information, a lot of stuff like that that they have not shared beyond their walls, and yet, except for the particular compound that they're developing, it's not that useful to them.
What sort of structure do you have in mind?
We need to have public-private partnerships consortia, like the SNP consortium. That's one example where that happened, where there was appropriate attention to intellectual property, but also public sharing of information.
Is there a group within the FDA that is specifically working on this?
Me and my group we're working on it.
When do you hope to have a prototype?
I don't know it takes a while to set these things up. We're talking to all the different parties. It would probably not be run directly out of the FDA, because of our regulatory role. But we're trying to make it happen.
Where would it be run out of?
There are a variety of places, especially non-profits, and so on, that are willing to set up consortia, that we're talking to.
Are you willing to name any of them?
I can't right now, because we haven't announced anything, but we're in active discussion with a number of groups.
When should we expect an announcement?
I don't know say, whenever the lawyers get finished. You can never predict how long the lawyers are going to take. But as soon as we can, we'll announce that we're in the process of getting things set up.