Pharmacogenomics: Podcast interview with Medco’s Felix Frueh (transcript)

This is the transcript of yesterday's podcast interview with Medco's Felix Frueh.David Williams: This is David Williams, co-founder of MedPharma Partners and author of the Health Business Blog.  I'm speaking today with Felix Frueh.  He is VP of Personalized Medicine Research and Development at Medco Health Solutions.  Felix, thanks for speaking with me today.Felix Frueh: Thank you for having me.Williams: Felix, we’ll talk about pharmacogenomics today. So first what is pharmacogenomics?Frueh: Pharmacogenomics in the broadest sense is the science of looking at your genetic profile and the influence of this profile on pharmaco therapy; so how do you metabolize drugs, how your body reacts to drug therapy.  You can expand that into a variety of other areas that people sometimes refer to as pharmacogenomics, including personalized medicine in really a much broader sense. You're measuring biomarkers to guide therapy perhaps a little bit beyond genetics, but often the terms are used interchangeably.  So I would restrict it to measuring genetics: conducting a genetic test and using that information to guide your drug therapy.Williams: I understand that Medco and the American Medical Association sponsored a survey of physicians on the topic of pharmacogenomics. Can you tell me a little bit about the background on that survey and what some of the key findings were?Frueh:   At the end of last year, we conducted a survey with the American Medical Association. We faxed a series of questions to about 300,000 physicians in the United States. We wanted to know about  their knowledge and use of pharmacogenomics.So we asked questions such as: “Do you think that genetics or genomics is an important component for drug therapy?” , “Are you using genetic testing?” , “Do you think you will be using genetic testing?” , “Why are you using it?” , “What is the information that you would like to know for using it?”It was really interesting that 98% of physicians believe that genetic information is important for drug therapy.  That provides a really excellent base for what we want to do, namely bringing pharmacogenomic testing into clinical practice.  We have to contrast that number with the only about 10% of respondents that indicated that they're actually using pharmaco-genetic testing today.  So there is a wide gap between what physicians believe is really important and the number of physicians who actually are using that information.That gap is largely due to what we've learned from that survey: the lack of education and the lack of adequate information provided to physicians about the genetic tests.  If we take that survey and parse out some of the key responses, we can learn a lot about where we have to take the next steps for making pharmacogenomics medicine into reality.  It's interesting to note that about 25% of respondents indicated that they will be using genetic testing in the near future; that's more than twice the number of physicians that are using such tests today.  To me that indicates that the awareness is out there and physicians really want to use it, but we need to find a way to provide information better to that community.Williams: What can we learn from the low percentage of physicians that actually use pharmaco-genomic tests in their practice today?  How are they different from the physicians who think that genetics are important but aren't using the tests?Frueh: For example, whether the physicians had formal training during medical school makes a big difference.  The therapeutic area makes somewhat of a difference, too.  So if you're talking to oncologists, they're more aware of the potential of genetic guided therapy than a dermatologist.  One of the other interesting points in this context was that if you ask whether or not they believe that their patient actually benefited from the use of a genetic test, you also get a little bit of an idea of who the doctors are who are perhaps somewhat more reluctant.For example if you ask if they believe that the drug has an improved effectiveness or reduced toxicity depending on what the specialty is and what the application is you see different answers.  As we move forward and we find a way to provide more information to the physician, I think many of the physicians will move into the well-informed group. or a not informed group would move from the latter to the former.  So the most distinguishing factor really is the level of education.Williams: Obviously most physicians have been out of medical school for a while, but they do continuing medical education, so I'm wondering what the specific mechanisms are that you're thinking about to educate some of the physicians that are perhaps less knowledgeable about the application of such tests in their practice.Frueh: There are many venues that I believe need to be explored.  Some of them have already been set up.  Others we're going to see over the next couple of years. For example we have developed brochures and online materials together with the Personalized Medical Coalition and with the American Medical Association to help physicians get up to speed on “pharmacogenomics 101.”Prior to starting working at Medco, I was working at the U.S. Food and Drug Administration. While in that capacity we also collaborated with the American Medical Association and with the American College of Clinical Pharmacology on creating two online courses that are available for free over the Internet to help physicians and practitioners get up to speed on why genetic information is useful in pharmaco-therapy.  So there are many different venues that we need to explore, but I think most important is to look at the vehicles that physicians are mostly using to obtain their information from to make or change their clinical practice directions.What we found was that providing information to clinical guidelines in discussions, peer discussion and through product labeling and peer reviewed literature are the four most important venues that physicians look for.  So perhaps the education piece needs to be integrated in these venues rather than in standalone education efforts that say you need to know something about genetics.  If there is a new drug coming out for which there is a genetic test available and drive that information flow through clinical guidelines or peer-reviewed literature, that would probably be the more effective way to do it.Williams: When you have a new drug and it has a companion diagnostic, what is the role of the drug manufacturer relative to the genetic test maker?  Do they tend to work hand in hand?  Do they tend to be the same companies?  What's the dynamic like between the two?Frueh: That varies depending on the product.  You have very close interactions; for example there is a drug Herceptin for breast cancer that is very closely tied to the performance of a test –the HerceptTest or FISH test--  to see whether the profile of the patient fits the pharmaco-therapy with Herceptin.  Then you have situations where there's literally no connection between the drug manufacturer and the provider of a genetic test, like warfarin, where the drug has been on the market for 50 years while the genetics of the drug therapy and metabolism only were discovered over the last decade and so the test came to the market much, much later.  The drug obviously is generic.  There are now five or six different FDA approved tests out there.So you have both extremes. I believe that over the next five to ten years it's very, very likely that we'll see many more close drug/test combinations coming to the market. I wouldn't be surprised if we also --because of that-- see tighter interaction between the pharma industry and the developers of the tests.I think that's a really good thing because the sooner we know that a particular biomarker or genetic test is going to be useful for drug therapy, the better the uptake of that test will be in the market. Again looking at the two extremes that I described, when a Herceptin came to the market at the same time as the HercepTest came to the market the combination of the two really led to an immediate uptake of both.  So the oncologist who is prescribing Herceptin never was really asking the question if the HercepTest should be done.  It was just part of what he would do when prescribing Herceptin.Bringing a test to the market after a drug already has been used and faces an established clinical practice is a very tricky thing, but within that context I think you see examples where evidence accumulates in such a compelling way that clinical practice actually adapts fairly quickly.  So if a drug comes to the market without a test, the first reports come out that testing could really help prevent adverse events and after a clinical trial was performed that clearly showed a significant reduction in adverse events.  Everybody started to use genetic testing.  So it is also important is that we conduct appropriate studies that further our understanding in what these biomarkers can do and how they can be deployed into the clinical practice. If we can do that I have very little doubt that clinical practice will change and adapt and we'll see many more of these tests being utilized.Williams: Why is pharmacogenomics specifically of interest for Medco?Frueh: If you look at Medco as a pharmacy benefit provider and the mission of our core business and the fact that we are a pharmacy, it's probably not a stretch to imagine that in order to practice good pharmacy, you want to make sure the drugs we're either filling prescriptions for or managing the benefits for are going to be utilized in the best possible way. Therefore providing access to genetic testing, brokering the access to genetic testing and being involved in the benefit management of the genetic testing is really another piece of providing good pharmacy practice.We want to make sure that whoever gets the pharmacy benefits through us and gets their prescriptions filled through Medco is also having access to the latest technology that helps guide pharmaco-therapy.  It's really the next logical step to a personalized medicine environment, making sure that the pharmacy, the drug, is used in an appropriate way.Williams: I know that some of your competitors are taking an interest as well. I think it was actually just a couple of days ago that CVS Caremark announced that it had invested in Generation Health, which bills itself as a genetic benefit management company. The founder of that company used to be CEO of Medco.  So I’m curious: is Medco planning something along those lines as well?Frueh: First of all I think that Caremark's approach is a great approach. It certainly validates what we've been doing over the last two years at Medco with both our research and our clinical programs in pharmacogenomics.  It really underlines the importance of pharmacy benefit management companies in personalized medicine.  So the fact that Caremark and Generation Health have been teaming up is a very, very positive development.From our perspective as I mentioned we actually are doing this already for two years.  We already have commercial programs for certain drugs that you'll hear about later this year.  We definitely are very innovative in this space.  We're leading the personalized medicine field with respect to the translation of that science into clinical practice.  We have commercial programs.  We have 200 or so clients signed up already for these programs and we are actively involved in the translation of this.Williams: I've been speaking today with Felix Frueh.  He is Vice President of Personalized Medicine Research and Development at Medco Health Solutions.  Felix, thanks so much.Frueh: Thank you very much.