In a commentary published in Science Translational Medicine this week, cardiologist Eric Topol bemoans physician reluctance for adopting pharmacogenetic testing, but applauds pharmacy benefit managers Medco and CVS Caremark for stepping in to conduct clinical utility studies and introduce genetic testing in a portion of their client base to gauge the impact on patient outcomes and healthcare savings.
The involvement of Medco and CVS Caremark in the pharmacogenetic testing space "is certainly a unique trend in the history of US medicine, at a time when prescription drug costs collectively reach $300 billion," Topol writes in the article. "Whereas one might have anticipated that the medical community would, on its own initiative, have ushered in the era of pharmacogenetics, it has been left to the PBMs to get the ball rolling."
In the article, Topol criticizes medical professionals for being "exceptionally resistant to change," and says that their demands for "suitable evidence" supporting PGx testing are a "cover for [their] profound lack of plasticity." As the chief academic officer of Scripps Health, Topol was an early adopter of PGx testing to dose Plavix at Scripps Green Hospital in San Diego (PGx Reporter 10/28/09). The move was criticized by other physicians and researchers for moving too quickly in the face of limited clinical utility data.
As an example of the resistance to PGx among physicians, Topol notes that "the medical community takes no initiative in routinely genotyping patients who are taking clopidogrel (Plavix)," even though CYP2C19 mutations that impair the ability to respond to the drug are prevalent in 30 percent of European patients and 50 of Asian patients; point-of-care platelet function tests can confirm response to Plavix; and higher doses of Plavix or an alternative treatment, such as Effient, can be given to non-responders of Plavix.
While he lauds PBMs for their efforts to drive adoption of PGx testing, Topol recommends that these firms be "transparent about their genotyping strategies and drug recommendations."
"Under the pretext of personalized medicine, these companies potentially may charge patients or insurance companies for genotyping services … while at the same time also profiting from the drugs prescribed and sold," Topol writes in the article. "This may represent a conflict of interest or at least the potential perception of double or triple dipping."
Indeed, neither Medco nor CVS Caremark have revealed what they are charging their clients for providing genotyping services. Also, although both Medco and CVS Caremark have said they use expert, independent advisory bodies for picking which drugs are ripe for PGx testing, they haven't yet revealed in great detail the specific criteria they consider in picking which drugs to study or to offer in their personalized medicine programs.
Topol ─ who is doing his part to educate physicians in genomics by launching an online course called the "College of Genomic Medicine" later this year ─ discussed the commentary and the role of PBMs in driving genetic testing adoption with PGx Reporter. Below is an edited transcript of the interview.
Some have noted that the involvement of PBMs in pharmacogenomically guided medicine puts doctors in a box by increasing the role of pharmacists in recommending genetic testing on one end and insurers asking docs for justification as to why they may or may not have conducted genetic testing for a particular patient on the other end. In your interactions with physicians, how do they view their role in this emerging paradigm? Do they feel boxed in? Or do they welcome the additional guidance?
Unfortunately, on the physician side there hasn't been anything like enthusiasm for pharmacogenetic testing. So, it's been left open for other forces to handle it, as we're now seeing with pharmacy-benefit managers.
The difference here is taking initiative. Physicians may not be the ones who start the process. PBMs, when they do genotyping of willing individuals who they are covering in their services, would of course interact with physicians, give them the data, and also suggest potential strategies to deal with the information on any given patient. So, that may help propel the process forward. They may serve as a trigger [to adopt PGx testing], because it isn't happening by and large by physicians taking their own initiative.
[ pagebreak ]
But with the increased role of pharmacists in the PBM-led PGx programs, do you think there may be some in the physician community that may be resistant to that guidance, because they feel it's their decision, based on their relationship with their patients, whether or not to provide PGx testing?
It's a good question. We don't fully know the answer. It will probably vary considerably between physicians depending on the drug that is being looked at for potential adverse effects or for dosing, or deciding whether to even use that drug. Some physicians will welcome this added data and some will think that maybe it is crowding them or interfering with their direct relationship [with their patients.]
But this is something [doctors] are interested to see, because the data for these drug/gene interactions are particularly compelling. So, it has to start somewhere. It didn't start with physicians for the most part, and PBMs provide a new way to catapult this forward.
I don't know whether there will be a negative response from physicians, but I know that consumers are enthusiastic about PGx testing. They were not so gung-ho and confident about [genetic testing for] disease susceptibility. The interest in the [drug response] space is much more intense and positive. There's hardly a consumer I've spoken to who doesn't think that this would be of value for them. So, if they're also helping to drive this, between the consumers and the PBMs, well that leaves it up to the physicians to support this. Over time, [genetic testing programs through PBMs] will become a physician-supported phenomena.
At Scripps, you have your own genetics program. So, to what extent have Scripps doctors worked with any of these PBM-led PGx programs?
We haven't worked with them on actual patient management. We have talked with both Medco and CVS Caremark about what their plans are, because obviously they have big plans. But we haven't yet encountered [PBM-based programs] at Scripps, because they haven't introduced their genotyping strategies throughout their 100 million-people customer base. They're introducing it in a staged, pilot way, and we've not experienced it at Scripps. We have a large employee base of 13,000 people. They've not offered PGx testing in our particular employee base.
You are developing a genetics course for doctors. Where are you in the development process, and are you working with PBMs in any way to develop this course, since PBMs may have valuable information about what drives genetic testing adoption among doctors?
We call it the College of Genomic Medicine, and it's led by a group of experts in genomics and pharmacogenomics. A module in pharmacogenomics is one of the principle parts of the curriculum that we're developing. We'll hopefully have this Web-based course ready by year end. It really intends to provide doctors with a background in genomics, but also builds up to why the integration of genomic knowledge makes sense in … prescribing drugs today.
PBMs have a financial incentive for promoting PGx testing for certain generic drugs. Do you think that for greater adoption of genetic testing by docs, the financial incentives will need to change for the physician community also?
Physicians should realize they have a financial incentive. It may not be a direct incentive, but, as pointed out in the article, there is a $300 billion prescription drug annual consumption. Obviously, we are in a crisis of increasing healthcare cost. So, we can use PGx testing to make healthcare more efficient, by avoiding very serious side effects that require hospitalization or consumption of additional medical resources. If we can optimize the right drug and right dose in patients, and avoid unnecessary prescriptions that are doing nothing, because it's the wrong dose, this would be an incentive to make healthcare economics more palatable in cutting costs. [However,] physicians are unlikely to have direct financial incentives for providing PGx testing.
In your article you note that since PBM-led PGx programs are promoting both certain drugs and companion genetic testing, they will have to be "transparent about their genotyping strategies and drug recommendations." However, neither CVS Caremark nor Medco have revealed what they are charging customers for genetic testing. In what ways do you think that the era of personalized medicine will require PBMs to operate their businesses differently in terms of transparency?
This is a tricky one to answer, because they have been transparent to date about the fact that they are planning on rolling out these programs, and are planning to make pharmacogenomics one of their key initiatives. I guess what we don't know is how the genotyping is going to be charged. I would assume that no consumers are going to have obligatory genotyping — that this is going to be their option. And certainly, employers, because they hire PBMs to administer their plans, will also have the option of not participating in pharmacogenetic work.
There are a lot of unknowns here. I'm not privy to the details of these programs that the PBMs are initiating. Certainly … there is a charge for the genotyping in addition to the prescription itself. Genotyping is actually very inexpensive. So, the question is, is it being done with a profit motive, or rather is it being done at cost to make the administration of prescription drugs much more efficient? That's what's unclear here: whether this represents an additional revenue stream or if it is embedded into the efficiency goal.
[ pagebreak ]
Do you think that PBMs have been transparent enough about how they're picking various drugs for placement in their personalized medicine pilot programs?
Medco worked with Mayo to conduct a large study on PGx testing for warfarin dosing. So we've seen that Medco can initiate its own studies when there is a large topic for which the question of testing is not settled. Still today, warfarin remains not fully resolved, and there is a large [National Institutes of Health]-funded trial to look at that.
So, I think what you're asking is when is there consensus that a particular drug is ready for this kind of routine genotyping? In part, some of their pilot studies are also providing data … on the optimal use of a drug. That data is a positive feedback loop to do more and expand the program … I think one that is controversial right now is tamoxifen. That one is not settled. And I think Medco is doing pilot studies on that, and my guess is that it's that pilot study work that would then be looked at to see whether it further substantiates expansion.
You touch on the advent of whole-genome sequencing in your paper, and discuss the need for education — especially in light of the increasing amount of information that will bring. What do you think whole-genome sequencing holds for these PBM-led personalized medicine programs?
I don’t see that PBMs are going to start doing whole-genome sequencing. On the other hand, I think those studies are really important for commonly prescribed drugs, to determine what explains the biological variability and the propensity for significant side effects. The academic research community and big pharma could execute whole-genome sequencing studies to further explain variable response for drugs for which we know some of the story today, like Plavix, but also others. We currently only know part of the story.
For example, there is a whole-genome sequencing study coming out on metformin, the most commonly used drug for type 2 diabetes, which hopefully will be useful in guiding appropriate use of that drug. And that needs to be done for all the commonly prescribed drugs. If PBMs continue to embrace pharmacogenomics then I see them selecting specific genotypes from the whole-genome sequencing studies to incorporate into their routine testing. I don't see them getting involved in using sequencing platforms. That seems highly unlikely. But of course, who could have predicted where we are today?