Personal genomics and rat poison

Warfarin, a commonly used anti-clotting drug, sold under the brand name of Coumadin, has a been a poster child for the promise of pharmacogenomics and personalized medicine.

The excitement has come from the idea that knowing a patient's genotype, in this case for the VKORC1 and CYP2C9 genes, would allow physicians to tailor the dose of the drug and get patients the correct dose more quickly.

And it seems obvious that a test that would allow doctors to predict your ability to metabolize warfarin, would be a great thing, right?

i-80e2eb3f7fe75e4dd952ce9a4753939d-warfarin-and-cytochrome.png
Figure 1. Human Cytochrome P450 Cyp2c9 bound to Warfarin


According an article in today's The New York Times:

Determining the proper dose of warfarin is notoriously tricky. Even a slight change in dosage can mean the difference between too little, which would not be effective in preventing blood clots, and too much, which can cause dangerous internal bleeding. Tens of thousands of people end up in the hospital each year with complications from warfarin, which is also sold under the brand name Coumadin.

But just having a shiny new test doesn't mean that we're ready to benefit. Medicare has decided that the benefits of testing don't justify paying for the test.

Again, from the NYT:

...Medicare said there was little evidence that use of the genetic test led to better outcomes for patients compared with the existing procedure, in which doctors estimate an initial dose based on a patient's age, weight and other factors. Then they test the blood's clotting propensity every few days and adjust the dose accordingly.

Why aren't we seeing a clear benefit from using the new tests?

One issue is cost. The NYT article puts the cost between $50 to $500. This is kind of a wide difference in cost, but Coumadin is much cheaper.

It could also be that the tests don't really save much in treatment costs, since those costs might come from monitoring the patient for blood clotting behavior. I can't imagine anyone saying, "yeah! we can skip that step," simply because they have genotype information.

Another reason is the timing. The NYT reported that the test results are obtained too slowly to provide information for an initial dosing regime.

Finally, even with cool web-based Warfarin dosing calculators, doctors may not know how to use the genotype information for calculating the right dose.

The test may be beneficial in the long-run, but maybe not until the market catches up.

More like this

Late last week saw two announcements from the US FDA on genetic issues in drug safety. The first of these addresses the prescribing guidelines for Coumadin, or warfarin. Coumadin is a "blood-thinner" (or anticoagulant) prescribed for conditions from heart valve and hip/knee joint replacements to…
Warfarin (a.k.a. Coumadin, Jantoven, Marevan, or Waran) is the most widely-prescribed blood-thinning agent on the market. It's also (in the words of Howard McLeod) a "terrible drug" - it has a very narrow therapeutic window, meaning that the minimal useful dose and the maximal safe dose are very…
I'm passionate about medicine. I love practicing it, teaching it, learning it...everything about it. It's also damned scary sometimes. Knowledge changes; patients don't walk in with textbook diseases; and you have to be able to doubt yourself without drowning in indecision. Every time a patient…
If youre a long time reader of SciBlogs, you probably remember when James Watson was on the advisory board of SEED. You also probably remember he was removed from that position after he made some idiotic racist (and sexist) remarks. In a wonderful bit of scientific lulz, Craig Venter recently…

I also think all the data is in adults. There's no relaible data on using the genotype information in pediatrics.

By Rob Pyatt (not verified) on 06 May 2009 #permalink

Rob,

Would kids need this sort of thing?

I thought it mostly for people who'd had heart attacks.