Racist Medicine

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 compared his genome with that of James Watson and made some predictably astute observations about race, genes, and drugs.

'Individual Genomes Instead of Race for Personalized Medicine' (its open access right now!)

Lets say you have a headache, so you decide to take some Tylenol. It dissolves in your stomach/intestines/etc, your headache goes away... But the Tylenol doesnt stay in your bloodstream forever! There is a family of enzymes, cytochrome P450, in all of your liver cells, breaking down the acetaminophen until it can be filtered out by your kidneys.

For a drug to work 24/7 (a heart medication, anti-depressant, etc), you need the drug to stay at a constant, therapeutic concentration in your bloodstream. You need to replace metabolized drug as fast as its being degraded, so you take doses every 6/8/12/whatever hours. BUT, you dont want to take too much, or else the drug might shoot past 'therapeutic concentration' and into 'toxic concentration'. Everyone has heard of the grapefruit juice example-- Grapefruit juice can inhibit some of the CYP450 enzymes, and you can accidentally OD if you take some drugs with grapefruit juice.

Things get trickier. Depending on genetic variations you have in your CYP450 genes, you might metabolize a drug too fast, or not be able to metabolize it at all. For example 5-10% of Caucasians have non-functional CYP2D6, so codeine has absolutely no effect on them (they cant metabolize codeine into morphine, SOOO glad Im not them). Some African Americans dont respond to first-line hypertension drugs, so a group has invented a 'race based' alternative drug.


Here is where we get back to Venter *grin*

Venter compared his CYP450 genes to James 'Africans are dumb' Watsons CYP450 genes.

Two white guys.

Venter has the 'white guy' alleles.

Watson has, ahem, 'not white guy' alleles. heh.

Venters point? Prescribing drugs based on 'appearance' of race is stupid. A 'white' person like James Watson can have 'Asian' CYP450 genes. 'Black' people from Ethiopia, Tanzania, and Zimbabwe can have different CYP450 profiles.

Venters solution?


In one case study, a patient with heart problems was prescribed warfarin. Over the next 5 years, the man visited the hospital at least 20 times to have his blood samples taken as his doctor tried to establish the correct dosage. During this time, besides the inconvenience, the patient was at risk for hemorrhaging. Finally, in year 4, his DNA was genotyped, and he was discovered to be a poor metabolizer based on the sequence of his CYP2C9 gene, which is involved in metabolizing warfarin. Soon thereafter, the patient's dosage was stabilized.
Because the cost of genotyping and sequencing has decreased dramatically in the past few years, we no longer need to guess the genetic makeup of an individual. With current technology we can definitively determine a million genotypes for an individual in a matter of days, and this will only improve as technology advances. Revisiting the patient on warfarin who was a poor metabolizer, we have no idea what color his skin was, and it does not matter--it was his genotype that ultimately mattered.

This is why I get so angry with people like Francis Collins, scared of everyone being sequenced. Sequencing everyones CYP450 profile will not only save lives, but improve the quality of life for all of us!


More like this

Compounds in grapefruit juice inhibits an enzyme required for metabolism of nearly half of prescription drugs on the market. If you inhibit drug metabolism, would that allow you to take a lower (and cheaper) dose of one of those drugs, especially an expensive drug? That is the proposition of a…
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…
As noted in the previous post on the anticoagulant, Coumadin (warfarin), last week demonstrated how pharmacogenetic variations in drug metabolism and drug responses are giving rise to what is increasingly known as "personalized medicine." In their second such warning last week, the US FDA alerted…
Roche Molecular Diagnostics offers a test that can determine which type of genes a person has for enzymes that metabolize antidepressant medication.  The test costs $ 300 to $400, and can be ordered by healthcare professionals, or by consumers. The idea is that it might be possible to predict which…

Small type: "codine" should be "codeine".

By Alex Besogonov (not verified) on 06 Sep 2008 #permalink

I actually agree with Collins on this point (and I am pro-genomic sequencing and personalized medicine). In order for it to be sensible for everyone to be sequenced there is a requirement for proper healthcare and treatment options to be available (its not just the CYP450 variants that sequencing would throw up, its things like genes that link to early onset Alzheimers which has no current therapies).
So its partly political (get a proper universal healthcare system like the rest of the developed world), and partly scientific (develop treatments that prevent Alzheimers etc in individuals with the high risk alleles).

I agree, my anthropology professor cited the "black" hypertension drugs as an example of a failure to understand the non-biological nature of race (it's socially construed). She pointed out that there is no reason that this drug should be restricted to a single racial group when it may have further reaching benefits.

If you like Collins, you'd Love Leon Kass.

1) Spending money, whether we realize it or not, can always be seen as political. There are a lot of important things that could be done with the money it would take to do that kind of sequencing *cough*mosquito bednets*cough* .
2) I also think we need to make sure the HIPPA laws catch up with sequencing info. Unless we get everyone covered by health insurance, I personally might hesitate to get long-term-risk alleles sequenced (which is different from drug-metabolism alleles).
3) Venter's solution is always HACK THE PLANET SEQUENCE EVERYTHING!!!11ELEVENTY!!!
4) All that being said, I still lust over the data it would produce. Mmmmmmmmmmm forbidden data.

I agree, my anthropology professor cited the "black" hypertension drugs as an example of a failure to understand the non-biological nature of race (it's socially construed).

Yeah I agree, "black" isn't necessarily a useful ethnic category, much like "white" as it suggests an extremely broad groups of people. For instance using the group "Asian" would complex in determing whether an individual can hold their grog or not. Several different alleles for ethanol dehydrogenase (ADH) which are commonly associated with different Asian populations can result in a higher or much lower than average tolerance to alcohol.

However if you do look at broad ethnic backgrounds you can see how they might be more likely to have different alleles for specific genes. This is especially true if you have non-random mating for instance the rate of sickle cell anemia is higher in African-American populations compared to the rest of the U.S. because traditionally blacks/whites only (or more frequently) married/had kids with people of their own ethnic group.

Making or ruling out a diagnosis solely on someones ethnic appearance probably isn't a good idea because alleles can obviously make their way through any population, but it might be a useful starting point.

Getting clues about probabilies of things occuring based on racial background isn't such a bad thing. If having a certain ancestry means you have a 60% or an 0.1% chance of having a disease, so therefore the doctor runs that test first, instead of fourth or fifth.

Obviously when we can instantaneously and cheaply sequence an entire persons genome and know the phenotypic consequences of every allele (like in Gattaca), racial constructs and ancestry will become obsolete to medicine.

I should also point out that personal genetic counseling i.e. asking about the history of disease in someones family, is probably more useful for determing the probablity of a disease/drug reaction occuring than just making assumptions by someones ethnicity.

I'm not completely in love with the idea of sequencing everyone until we have completely socialized medicine. Otherwise this could backfire against people who are genetically predisposed to certain diseases and their insurance premiums.

I think we need to do these two things in the correct order.

By Hamsterpoop (not verified) on 06 Sep 2008 #permalink

Wait, James Watson the co-discoverer of the structure of DNA? He says shit like that?

Now I'm all sad.

By Terry Small (not verified) on 07 Sep 2008 #permalink

I think this is a step in the right direction. To pretend that a physician's assignment of race in deciding on what treatment would or would not be effective in a particular patient is scientifically flawed concept. Controlled studies on the "correct" assignment of race by health care providers have shown this subjective assignment of race to be very unreliable. This finding is probably the principle reason for the universal screening of newborns for sickle cell disease using detection of hemoglobin S, as opposed to selective screening of newborns of parents which visually appear to be of a racial group at risk for sickle cell disease. Individual genetic testing does seem to be much more objective and scientifically sound. If the costs and ethical/privacy/access to healthcare insurance issues are worked out, I feel this will be way to do things in the near future and we will no longer hear physicians proudly proclaim: "I practice racial profiling".

Your anger with Francis Collins is ill-informed.

By Sean Eddy (not verified) on 07 Sep 2008 #permalink

fwiw, it's a lot easier to sequence particular genes of interest than full sequencing. It's still a huge undertaking for sequencing human genomes.

--Illumina/Solexa junkie

I want to be sequenced. I have a rare genetic disorder, Ehlers-Danlos Syndrome, Fibromyalgia, (which I think is a symptom of EDS and not it's own disorder, but that just my layman's opinion),constant migraines, and all kinds of annoying health issues. I don't metabolize drugs well. Depo-Provera almost killed me, seriously. I metabolize the "caines" so quickly that I can't get through a root canal without feeling the last half of it, not fun. I hate taking meds because I never know how they will affect me or what unpleasant side effects I'll get. I would love for someone to be able to look at my code and figure out the best way to treat me I don't understand why some people are afraid of being sequenced.

How do I get sequenced and is it highly accurate?

(Oh and BTW, Watson is a dickhead.)


I'm sorry to hear about the confluence of conditions you seem to be stricken with, but I think one thing you should consider is, "what will come out of having my genome sequenced?"

You've already been diagnosed with a bunch of things; how will knowing your genotype/haplotype affect the care that you're already receiving or contribute to the knowledge you already have?

If there are things that you are concerned about that have not been diagnosed then sequencing those particular genes might be of use (though, again, a SNP chip might be better).

But just a knee-jerk saying "I want to get sequenced" isn't really going to help you, I'm afraid. That, and, I did a back-of-the-napkin calculation and determined that if I were to use the methods we have here to sequence you, it would cost 152,000 just in reagents. Not to mention the time it would cost for someone (e.g. me) to reassemble the content.

Again, sorry to hear and best of luck. But I don't think sequencing is the panacea you're looking for.

Actually, Jason, it could be used to confirm a diagnosis of Ehlers Danlos if she has not had FISH (fluorescence in situ hybridization) analysis for the specific mutant alleles of the disease. This could confirm beyond any doubt the cause and possibly shed light on which treatment method is best. Right now we only have a few treatments for it, but in the future, with sequencing, we could see better procedures for this.

Again, the goal is to get the overall cost of complete genome sequencing to less than $10,000 per person making it easier to do and far better than running dozens of tests.

But to be fair, costs currently are way too pricey for most people to afford.

Rayven, I would like to tell you how to go about getting your DNA sequenced, and that it would do good in the short term, however, only a few places do it and unless you have a few million dollars to throw at it, you probably won't be able to afford it. Anyway, we currently have no treatments for Ehlers Danlos for specific alleles that I know of, so it probably would do no good at this time anyway. But to throw this out, 454 Life Sciences did Watson's at a cost of $2 million...

What you can do is have FISH or specific gene analysis done. These are both much cheaper than DNA sequencing, but still fairly expensive.

Jared - Rayven did not say she's not been diagnosed at all - so why would you encourage FULL GENOME sequencing for something she knows she has? Presumably it's been confirmed if, as she indicates, it was a diagnosis.

Right now Solexa sequencing for 10x coverage (and reasonably SNP calling) would be $152,000
Assumed: 51-mer reads, 10 million reads per lane, cost of $2400 / lane, 3.2 GB genome, accurate SNP calling at 10x coverage

Though, that's for single-end reading, and really paired end would be better (so double the cost).

Also, i don't think FISH is an efficient way for SNP detection, considering also that there are several genes implicated for the disease (depending on which collagen gene is affected). I don't believe the resolution to be that good.

/genomics rant

Jason, I didn't make my reasons clear, sorry. You're right, there is nothing that can be done for EDS, other than pain and symptom management. I would like to be sequenced to find out what drugs I don't respond well to, not for help with EDS, there is none, (at least not yet).

I am severely allergic to Sulfa and had a Steven's Johnson reaction when I was 13, my sons and I also react to sulfates and sulfites, (yes, I know they say the three are supposedly not chemically related, but it sure seems suspicious to me. Perhaps Sulfer is not as hypoallergenic as believed to be, or maybe it has something to do with our defective collagen and it's not an actual *allergy* but a lack of metabolic ability.) We also react to thimerisol (murcury, latex, mangoes, cashews, and I forget what else. We don't metabolize certain types of drugs well. It's made me almost paranoid about taking any new meds. I would love to at least have a hint of what we should stay away from, or what we might not respond to.

Plus, I am a total geek and I think our biological data could benefit a researcher who might be studying EDS. When I die my body will be donated to science so they can study the disorder.

I haven't had a skin biopsy done, my diagnoses was based on joint hyper-mobility and other symptoms. I've been meaning to get a biopsy done for years but I admit I am chicken. I scar badly, and I am afraid of nerve damage from the biopsy. I had a tiny cyst removed from my armpit and have mild nerve damage in that upper arm, I also have nerve damage from a 34 ft fall I took as a teen. I sustained many injuries in that fall, three of which were compression fractures to the spine, L3 is smushed to about a tenth it's normal size on one side and in a bizarre wedge shape. I have nerve damage in the right thigh due to that injury and it seems every time I get a bad cut I end up with odd nerve damage. It seems my body cannot repair itself worth a shit.

I'm rambling on, sorry. My point was, I am behaving like a big chicken because I dread more nerve damage. It feels funky when your skin is numb all the time. I know my sons and I have EDS, but since I haven't been biopsied, I don't know which type we have or whether it's an undiscovered type.

Jared, I suppose you're right, until the costs come down I would not have it done. There are many other things I'd like to spend that money on.

...but if anyone needs a genetically defective guinea pig and happens to want to sequence them, I'd volunteer! LOL

Jason, I wasn't encouraging it, I was explaining what it could do (and probably wouldn't) right off-the-cuff. My point was that if her doctor had a full DNA sequence for her, even with that information, Ehlers Danlos treatments are the same. I should have also added that many drug interactions are also unknown with various genotypes and some interactions are developmental, although interactions with specific genotypes are known. After thinking about it for a few moments, you're right, FISH would be poor as it works best with large stretches of DNA to hybridize with, but it was the first idea that came to mind (I was reading about VCFS). I'm pretty sure we have a few non-genetic lab tests to confirm EDS, I think type VI has something to do with a protein that modifies collagen. I also think some are tested for via cultured fibroblasts (not sure which types). Again, this is all off-the cuff, feel free to correct me at any time.

Also, Helicos has a pretty cool machine for sequencing, I know it's supposed to reduce the cost of sequencing a genome, not sure by how much, though.

Since it wasn't really about EDS in the first place (misunderstanding or possibly jumping to conclusions on my part) I would have to say that I understand slightly better where you are coming from, but I suspect you should read up on allergies if that is your concern. Allergies, as far as the evidence indicates, are not passed on strictly genetically. In any case, there is no way to determine allergies from a genetic analysis, it has to be tested for by scratch tests or something similar where you look for reactions.

Drug allergies and sensitivities can be genetic, as can the inability to respond to certain drugs. When I had the SJ reaction, I was told to never allow my offspring to take Sulfa drugs, it's not worth the risk. The times they have accidentally been given drugs with sulfites, or sulfates (they are used as preservatives in eye/ear drops),they had reactions. I now double-check every prescription we're given because many doctors are sloppy.

Even food allergies can be genetic. My sister and I both have our lips swell up so large they cover our nostrils if we touch them to even the most minute traces of mango. She once had a bad reaction because her neighbor brought her a slice of cake and had used the knife to also cut a mango. She had wiped the blade off before cutting the cake, but that's all it took for my sister to swell up like a freak. (BTW, Mango is in the same plant family as Poison Ivy, so people who react strongly to P.Ivy might want to avoid mangoes and cashews.)

If there are genetic components to certain allergies or reactions, they will eventually be found through research. That's why this kind of research should move forward and be encouraged. It is invaluable for so many reasons. Genetic research is in it's early stages, I think they will discover amazing things in the future. Truthfully.,I think when we can afford to map everyone and correlate the personal data we will begin to see patterns and begin unraveling many medical mysteries.

Hmn. Country of origin comments notwithstanding (God save ERV's dog), I can't imagine how James Watson made it on the board of an evolander website (sciencblogs.com).

Say it wasn't so.

I love it when racist idiocy is torn apart by genetic truth. My parents' long-time genealogical research long ago disabused me of any notions that people are ever just of one "race"; everyone is more diverse than they think. In particular, the research into Melungeons and the prevalence of certain inherited diseases in unexpected family groups is fascinating. When a bunch of "white" people from Kentucky turn up with Familial Mediterranean Fever, it can make some people rethink what they "know" about race.

Why would an intelligent man of science like James Watson think intelligence to be of such importance over other attributes. You need more than just intelligence to get through life.
If you just inbred all the intelligent people they would become more prone to extinction. You need a diverse genome to have a good chance of surviving. Luckily with increased mobility of the different races and changes in social behavior we are mixing up our genomes which should increase our chances of survival.

By Georges Genovese (not verified) on 17 Sep 2008 #permalink