The US Food and Drug Administration was busy in 2010 dealing with an unprecedented number of regulatory issues pertinent to the personalized medicine field.
In the past year, the agency lifted its longstanding practice of "enforcement discretion" over laboratory-developed tests and began the process of carving out a regulatory framework for overseeing all diagnostic tests; informed several companies that their direct-to-consumer genomic analysis services needed regulatory approval; testified before Congress on companies that sell genomic data directly to consumers over the Internet; held meetings to update the 510(k) device clearance process; weighed in on discussions to develop a genetic testing registry being developed by the National Institutes of Health; and updated labeling for drugs such as warfarin, Plavix, and irinotecan to inform doctors of genetic abnormalities that might impact patients' abilities to respond to these treatments.
Still, some personalized medicine stakeholders feel that the agency hasn't done enough. Many industry players have been waiting to hear from the FDA on topics they feel the agency should weigh in on in order to advance the field, such as issuing guidelines on regulatory strategies for simultaneously developing a drug and a companion test. The agency has said it is working on this, and has decided to issue two separate guidance documents: one on companion diagnostics development slated for release in the first half of the year, and a drug/diagnostic codevelopment guidance that would be released later in 2011 or the following year (PGx Reporter 12/08/10).
Someone close to many of FDA's efforts in the personalized medicine space is Issam Zineh, associate director of genomics at the Office of Clinical Pharmacology in FDA's Center for Drug Evaluation and Research. For example, Zineh has been intimately involved with many of the pharmacogenomics programs underway at the FDA, particularly the Voluntary Exploratory Data Submissions program, which has increased interactions between the agency and non-FDA scientists on genomic biomarkers.
Zineh recently answered PGx Reporter's questions about the agency's ongoing and future efforts in personalized medicine via e-mail. Below is an edited version of the interview.
To what extent is the agency seeing more genomics data in regulatory submissions in the past year? What sort of growth are you seeing in terms of genomics data as part of New Drug Applications?
We are seeing an incredible rate of growth for genomics and biomarkers in regulatory submissions. When we look at the number of [Investigational New Drugs, New Drug Applications, and Biologics License Applications] that our Genomics Group in the Office of Clinical Pharmacology has been asked to review, we see a dramatic increase. We went from about 60 submissions in 2008, to 140 submissions in 2009, to 210 submissions in 2010. We're convinced that not only will these numbers continue to increase, but so will the awareness of pharmacogenomics throughout the regulatory review groups within the agency.
At what stage are these submissions coming? Mostly through the VXDS program, or in actual product NDAs?
Most of these submissions are clearly coming through the regulatory review path (i.e., INDs, NDAs, and BLAs). VXDS was never intended to be a high-volume mechanism to bring in a lot of genomics submissions. Rather, the VXDS program is intended to serve as a platform for non-regulatory scientific exchange between agency scientists and external scientists (mostly drug-development scientists, but others as well) on a variety of issues. It is clear to us that VXDS remains an important way for external scientists from various settings to engage with FDA scientists on genomics- and biomarker-related topics outside of the regulatory review stream.
Can you provide the latest numbers as to the number of VXDS submissions you saw in 2010 versus 2009?
Since its start, the average number of VXDS submissions over the years has been holding at about five per year. In 2009, we had five submissions, and in 2010, we had four. Of note, we have seen an increase in the number of biomarker qualification (BQ) submissions to the CDER BQ Program. For example, in 2008 we had one BQ submission; in 2009 and 2010, we had four and six, respectively. There is a lot of buzz around BQ. Given that BQ submissions typically come from consortia and VXDS submissions come from individual companies or scientists, we want to encourage a healthy balance of both types of submissions. They serve different purposes.
What have been some changes that have taken place at the agency to accommodate the influx of genomics data you are seeing in submissions?
There are several groups within FDA who deal with genomics-related issues. From our perspective as CDER Genomics Group reviewers, our biggest change has been development of best practices, review templates, and other tools to enhance the efficiency of our review workload. This has allowed us to increase our ability to influence clinical trial designs and methodologies at the IND stage, so that genomics data are less likely to be too exploratory to be useful from a regulatory decision-making standpoint at the end of the NDA or BLA review cycle. This has the potential for tremendous benefit to drug development. In addition, we have worked with our colleagues in CDRH to enhance communication and exchange on several genomics policy issues.
Since the agency is seeing a lot more genomics data in regulatory submissions, what are some challenges for the FDA in the coming year?
I don't see challenges so much as I see opportunities. With the increase of pharmacogenomics submissions, we have a real opportunity to further refine our thinking on the use of pharmacogenomics in drug development and its role in helping us make informed public health decisions. If I had to pick one challenge, I would say that it would be ideal if we could capture the collective agency experience with respect to genomics and personalized medicine in order to best inform policy. That's a difficult thing to do since every review division and every center may have its own experience that is not centrally captured in a knowledge management system. Our continued informatics work will hopefully go a long way, but that is likely to be a longer-term project.
What do you think will be the main disease areas for personalized product development, i.e., Rx/Dx development, in the coming year?
Cancer will of course continue to be a major area for personalized product development. The oncology community has been, in a sense, personalizing medicine for decades. Evaluation of oncology drugs or biologics products represents about half of the Genomics Group's review business. I will say, though, that there are a few other areas where I think we will see a lot more consideration of genomics. These include cardiology, immunology, and hepatitis C drug development.
When you have discussions with drug developers, what do they identify as barriers or challenges to developing PGx-guided personalized medicine products?
There are probably two major challenges. First, under the current paradigm, most drug developers collect DNA in a voluntary way to be later analyzed in an exploratory fashion. The voluntary nature of this collection usually means we have incomplete sample acquisition, which of course creates real challenges in interpretation of data from a regulatory standpoint. We're hearing that this heterogeneity in sample acquisition is a function of variability in how local or regional [institutional review boards] and ethics committees allow drug developers to consent their patients to do pharmacogenomics substudies in global clinical development programs. Another challenge is simply one of certainty. Many times, drug developers feel they have insufficient data at the end of phase 2 in drug development to confidently select out a target population to be tested in the definitive efficacy studies. These challenges are surmountable to my mind, but we certainly can't underestimate them.
A lot of industry players are waiting for the codevelopment guidance. What are the areas within the proposed guidance that the various centers at the agency have to work through most carefully?
I can't comment specifically on the companion diagnostic guidance. I will say, however, that the issues are important ones and we spent a significant amount of time discussing some of them at a public FDA-[Drug Industry Association] workshop earlier in 2010. In my opinion, I don't think any one guidance is going to be the definitive "how to" on personalized drug development. It will likely be a combination of guidances combined with face-to-face interactions between developers and FDA. For example, we have a guidance on pharmacogenomics in early clinical trials and another on enrichment in the works. Both of these are likely to be important guidances that we hope will enable more efficient drug development.
[There has been some discussion of enacting] joint FDA/CMS review for sponsors that want it. One of the main concerns for developers of personalized medicine products is that the evidence requirements for FDA and payors aren't aligned. Do you think this type of joint review might be a good option for personalized medicine developers?
We did have a recent open docket in the federal register soliciting feedback from the public on such a mechanism. It will be valuable to review the comments once they have been collated. At this point, it's too early to tell what the implications of a joint review will be. It will probably be different for true co-development (e.g., when the drug and test are simultaneously developed in the context of a given development program) vs. tests cleared by FDA for the intended use of guiding an already approved therapy based on retrospectively generated data.
There are a few legislative efforts underway that would create a separate government body for dealing with personalized medicine products. Do you think such a specially dedicated body under HHS is necessary for the advancement of the field?
It's always a good thing when you can catalyze advances in the development of personalized therapeutics. Coordinating personalized medicine efforts throughout HHS, however, may be a bit challenging given the different charges the different agencies have. For example, we (FDA) are largely charged with protecting the public health, whereas our sister agency NIH has a very different research-oriented mission and goals. True advancement in the field is likely to come as a product not only of HHS activities, but also engagement by clinicians, payers, patients, and others.
From the FDA's perspective, what should the personalized medicine community focus on in 2011?
This is not really a regulatory opinion, but more of a clinical one: I'd like to see some innovation in the pharmacogenomics research that's being done. We have a significant amount of technology out there that can help us define the genetic architecture underlying variable drug responses. That's great from a genomic discovery side, but I'd like to see more research on implementation. Use of N-of-1 trials or cluster randomization designs to really get at whether pharmacogenomics can be done at a clinical practice level would be welcome. Also, safety pharmacogenomics is progressing along quite nicely. I'd like to see some advances in the use of genomic markers to enrich for clinical responses (predictive enrichment).