pathtalk.org is a weblog about pathology.

Testing-testing: drug dosing based on SNPs?

Warfarin is one of the most commonly prescribed anticoagulants for prophylaxis and treatment of venous and arterial thromboembolic disorders or those at high risk for such disorders. Warfarin also has an FDA-issued “Black Box” warning for a high risk of fatal bleeding, most commonly from the gastrointestinal tract and in the brain. It acts by inhibiting the synthesis of the vitamin K dependant clotting factors (II, VII, IX, X) and the anticoagulants proteins C and S. Vitamin K promotes the synthesis of gamma-carboxyglutamic acid residues in these vitamin K-dependant proteins. Warfarin acts to reduce the regeneration of vitamin K from vitamin K epoxide in the vitamin K cycle through inhibition of vitamin K reductase.¹

The molecular target of warfarin is the protein product of the Vitamin K epoxide Reductase Complex, subunit 1 gene (VKORC1). This gene product is inhibited by warfarin. Polymorphisms of this gene product have been extensively studied over the past several years, and variant genotypes are associated with variable response to warfarin.¹ Compared to patients with a wild-type genotype, patients with at least one variant allele had an increased risk for elevated INR (HR = 1.4) and required more time to achieve a stable dosing (median = 95 days).¹

Warfarin is eliminated via metabolism in the liver. It is stereoselectively metabolized by hepatic microsomal enzymes (cytochrome P-450).⁴ The warfarin breakdown products have minimal anticoagulant activity.⁴ The metabolites are principally excreted into the urine, and to a lesser extent into the bile.⁴ The drug is comprised of a mix of the R and S stereoisomers. The S isomer is 3 to 5 times more potent than the R isomer, and is metabolized by the cytochrome P450 family. The cytochrome P-450 isozymes involved in the metabolism of warfarin include 2C9, 2C19, 2C8, 2C18, 1A2, and 3A4.⁴

CYP2C9 is reported to be the main liver P-450 which modulates the in vivo anticoagulant activity of warfarin.⁴ Two variants of CYP2CP have been identified.¹ These variants ( the *2 allele -R144C and the *3 allele – I359L) are reportedly associated with a decrease in enzymatic activity of 30% and 80%.¹

To summarize, identified SNPs (single nucleotide polymorphisms) are associated with altered warfarin metabolism and utilization. The CYP2CP*2 and CYP2CP*3 genetic variations are associated with a need for a lower dose of warfarin. The VKORC1 genetic variation is also associated with a greater variation in warfarin dosing to maintain a stable INR.

In August 2007, the FDA issued a release in which it recommended genetic testing for these genetic variations in warfarin metabolism, and altered the drug label for warfarin to reflect this recommendation. This recommendation is a part of the Critical Path Initiative which seeks to harness new scientific knowledge in the areas of gene expression, bioinformatics, and new analytic methods in the service of product development while also emphasizing its commitment to personalized medicine. ⁵ Some institutions are beginning to implement testing for the identified polymorphisms using new testing modalities, such as methods utilizing nanoparticles for “NAT” testing or invader chemistry.

Specific dosing recommendations are not yet determined for patients with these known genetic variations. This is currently under study, but the hope is to give doctors a dosing regimen based on the presence or absence of these variations.^6-7 There are a myriad of other influences of dosing requirements for individual patients such as age, sex, weight, and diet, calling into question the utility of these tests. However, the FDA label change to reflect the push towards personalized medicine, along with the risk of fatal bleeding with overdosing and stroke/clotting with underdosing, the additional information given by the test will be helpful according to the studies published on this topic over the past few years.

1. The Pharmacogenomics Journal (2004) 4, 40–48. doi:10.1038/sj.tpj.6500220 (Published online 16 December 2003)
2. http://www.nanosphere.us/VerigeneWarfarin
MetabolismNucleicAcidTest_4472.aspx
3.http://www.fda.gov/bbs/topics/NEWS/2007/NEW01701.html
4. http://www.rxlist.com/cgi/generic/warfarin_cp.htm
5. http://www.fda.gov/cder/drug/infopage/warfarin/qa.htm
6. J Thromb Thrombolysis. 2007 Jul 29; 7. Clin Pharmacol Ther. 2006 Oct;80(4):346-55

Gretchen – I was waiting for someone to touch on warfarin testing – and I had a feeling it would be you. So to summarize the data (at least as I understand it), if you have a polymorphism in CYP2c9 that has less activity you are a warfarin slow metabolizer. As a whole this population of slow metabolizers will require on average a lower dose of warfarin. In addition, if you are a slow metabolizer you have an increased chance of having a significant bleed if you are taking warfarin (the major life threatining side effect of warfarin).

Does this follow?: If you give a slow metabolizer a lower dose of warfarin, they are less likely to bleed. Is there data to show this? Keep in mind, in the US, there are approximately 20 million warfarin prescriptions written each year with 2 million people taking warfarin at any one time. Lets say this test is performed 5 million times / year at a cost of 250 dollars/ test. That is 1.25 BILLION dollars of addition health care costs. What if it is performed for every new presciption of warfarin – that 5,000,000,000 dolars. That’s a lot of zeros.

very interesting point!!!

In the parlance of our times, this is a “hot” topic indeed of which this post touches the surface. I hope there will be more posts to display the full scope of this issue.

the main issues here are :

1. pharmacogenomics and the pathologists role as consultant
2. personalized medicine
3. SNP testing and the automation of these tests with new chemistry techniques
4. a lot more …:)

To me an equally interesting aspect… besides the implications of the Hollywood sleuth movie popularized saying “follow the money” is that we are seeing an evolution of a new technology or a paradigm shift or whatever buzz phrase you want to use in lab testing. (Which we both previously discussed in the Two SNPs for $200 post.) A shift from interpreting a panel of lab values in comparison to a referance range of “normals”(vis a vis the CBC) and drawing conclusions based on known pathophysiology ( ie a patient with basophilia, marked leukocytosis with mature and immature granulocytes and thrombocytosis may have a myeloproliferative proces) to a panel of yes/no values from a series of genetic tests.

these test results are a yes/no punch- yes this patient has this polymorphism or no they dont, and the interpretation of these tests are currently being studied.

But the warfarin “sensitivity” genes assays are interesting because having these variations do not diagnose a patient with a pathophysiologic disorder as we think of disease currently-

these tests are studied and plan to be implemented as a means of Preventative Medicine-

-a concept which seems logical but the implementation has previously been unclear

Gretchen – One final point from me (that I think I mentioned previously). The role of the Molecular Pathologist (at least in my practice) is to determine if a test SHOULD be performed on a patient or patient population. This decision is often followed by a great deal of physician and patient education. It comes down to a very simple question: does a test add value to a patient’s care? Value can be interpreted in a very broad sense: give them a better chance of getting better, prevent them from having a side effect, help arrive at the correct diagnosis, etc. If a test sits at a decision tree this is an easy question to answer (imatinib resistant mutation present – don’t treat with imatinib)

So, I return to my question about this testing in particular: If you give a slow metabolizer a lower dose of warfarin, are they less likely to bleed?

BTW – If your interest is molecular pathology, you should check out the QOD at http://www.pathologypics.com (Sorry all for the shameless plug, but anyone who has followed this thread to this point might be interested as well).

the short answer is

“?”

i guess we need more prospective studies

the blood article below followed warfarin clearance

the clin. pharm article says knowing the genotype makes dosing more efficient

the thrombosis article states that there were less minor bleeding events

all signs point to too early to tell right now

1. Thromb Haemost. 2005 Apr;93(4):700-5
2. Clin Pharmacol Ther. 2007 Sep 12;
3. Blood. 2005 Oct 1;106(7):2329-33.

[...] Testing-testing: drug dosing based on SNPs? (5) [...]

Gretchen – a prospective study would certainly be in order… OR. From the retrospective studies that exist. Efficacy of testing could be inferred by knowing a couple of simple facts for each patient.

a. What was their dose of warfarin when they had their adverse event?

b. What was their INR when they had their adverse event.

c. What was their INR and dose at their previous visit.

I’m surprised that this wasn’t discussed in any publications on the topic (I may have missed it – so much to read, so little time)

If the slow metabolizers were already on a lower dose and were stable at their previous visit. Dosing based on genotype will not help. One could argue that a slow metabolizer has a propensity for a LESS STABLE blood warfarin level. Outside challenges (grape fruit juice one saturday morning?) to CYP2c9 could serious disrupt serum levels of warfarin. Just a thought.

[...] Testing-testing: drug dosing based on SNPs? (Pathtalk): A perfect explanation of FDA’s Warfarin story. [...]

Great post. This is the first time I have seen your blog. Some great stuff here…Keep it up.
-Steve
http://www.thegenesherpa.blogspot.com
http://www.helixhealth.org/index.htm

[...] Testing-testing: drug dosing based on SNPs? (Pathtalk): A perfect explanation of FDA’s Warfarin story. [...]

Thanks for the comments!! and discussion!!! Interesting sites for both Pathdoc15 (love the QOD) and Dr. Murphy!! (Subscribed to your feed-looking forward to reading more!)

A prospective trial to read on this topic is
Y Caraco, S Blotnik, and M Muskat Clinical and Pharmacology and Therapeutics Sep 12 2007- Pubmed still calls this epub ahead of print
if you have any issues let me know if I can help

They gave a control group the normal dosing algorithm and a control group a genotype based algorithm (for CYP2C9 only). Both groups had similar proportions of patients with a wild type (non-variant) allele and homozygous and heterozygous CYP2C9 variant alleles.

The genotype based dosing was associated with a longer time to reach therapeutic INR but less time outside the therapeutic range ie more stable dose.

During the study, there were 15 bleeding episodes (14 minor events, 1 major event) with a 12.2% incidence in the control group versus 3.2% incidence in the genotype based dosing group. The major bleeding event occurred at an INR of 1.74 (nontherapeutic) and had angiodysplasia found in a colonoscopy which was felt to be the cause of the bleeding event.

The three patients in the genotype based group who had an adverse event were wild type genotype (ie did not have the variant allele associated with bleeding). Seven of the control group (normal algorithm) carried a variant allele. Bleeding for control patients coincided with INR above the therapeutic range.

OK this was long winded but I was attempting to address some of PathDoc15’s questions–the answer to them all are probably still “?”-this is still under study. Still need more information from trials-
I am still looking for papers on this any other suggestions?

But what this trial seems to point to is that genotype based dosing does not prevent all bleeding episodes–but it does seem to help with achieving a stable INR in patients prone to instability and thus curtailing/controlling the presence of very high INR which is associated with bleeding.

Gretchin – Fascinating. I hadn’t seen this article yet. I am much less the cranky skeptic today. Nothing like a little data…

http://www.nature.com/clpt/journal/vaop/ncurrent/pdf/6100316a.pdf

To my knowledge this is the only prospective study yet to be published. (Of course, I didn’t see this one come out… On my really fast read, I didn’t see that they referenced any prospective studies) It is a pretty strong publication. Unfortunately, it was underpowered to determine if dosing based on genotype gives a decrease in MAJOR bleeding events (they only had one and it was secondary to angiodysplasia – as you point out). I’m not sure that the minor bleeds even reached significance -it certainly wasn’t presented as such: 12% vs 3% out of 90 ish. What’s that? 3 vs 11? Seems like it may be close

HOWEVER, it uses an INR > 3 as a surrogate and the numbers look really good. If an individual is dosed based on genotype, then they spend a lot less time with an INR > 3 (and less time with an INR 3 and time to therapeutic INR are good indicators of propensity for major bleed – then all is well.

Somehow pathtalk ate the middle of my post?

Anyway, I think it looks good.

Time with INR>3 as a surrogate – seems very sound.

All is well.

Now if I could just figure out the magic dosing regimen from the publication…

I just got some scoop

results should be coming down the pipeline in the next few months including preliminary results from a pilot RCT for a genotype based dosing regimen vs clinically based dosing.

for now- a decrease in major events has not been shown but trials are being developed to study this

My wife has a leaflet valve (St. Jude Medical) in the mitral position. Though she has been on coumadin for years she now seems to be more sensitive to dosage and bounces between extremes for INR. I am concerned about both the risk of stroke for high INR and threat to the valve for low INR. I heard that supplemental vitamin K has been used to elevate the sensitivity threshold. Can anyone direct me to further info on this?

I am sorry about the dosing issues with your wife and Coumadin. This can be a stressful aspect about the drug. This website mainly focuses on pathology and laboratory medicine- that is we are discussing the aspects of laboratory testing and not treatment or dosing related issues.

I can tell you that many things can alter the metabolism of coumadin such as diet and other medications. Coumadin blocks the action of enzymes which utilize vitamin K in the vitamin K cycle which is responsible for generating clotting factors. The blood becomes “thinner” because there are less of these functional clotting factors due to the action of coumadin.

Therefore, we must assume that vitamin K supplements will also interact with coumadin. It is best for your wife to talk to her physician that is prescribing coumadin about any supplements before starting on them.

There is a website I often use when I want to read about a medication- http://www.rxlist.com. It lists some technical details about the medicine but also gives information on interactions and warnings. There is also the very dense package insert for the drug- which if you can make out the very tiny letters can give some additional information about the drug.

Thanks much, I suspected I was in the wrong forum but we are getting a bit desperate as this seems beyond the cardiologist’s ability to control. We’ve read all the inserts and websites and are hoping to raise the threshold above sensitivity levels with a constant input of supplemental K. I am trying NIH now. Again, thanks for your response.