TR: Bidil is a high-profile example of race-based medicine. What do you think of the clinical trials of that drug and the FDA's decision to approve Bidil with race-specific labeling? SL: The Bidil study is a very interesting case. There was actually very little pharmacogenomics in that story. It is a combination therapy of two substances already available as treatment for heart disease. In early trials, scientists found they weren't able to produce significant results that the FDA required for approval. When they went back to the original data and pulled out all the information on African-American patients enrolled in the trial, they found a significant treatment difference. Based on that information, the FDA told Nitromed, the drug maker, to do another study exclusively on African Americans. That trial ultimately showed a significant reduction in mortality for those on Bidil. Whether the FDA should have labeled that drug for African Americans is an open question. Since the approval, Nitromed has sponsored its own study to identify the genetic variation associated with [the drug's effectiveness]. According to initial reports, it is quite prevalent in African Americans--about 60 percent. But it's also fairly prevalent in those of European descent--about 40 percent. That begs the question of whether it should really be labeled only for African Americans. It's a cautionary tale for us to carefully consider. TR: Do you think scientists should stop using race in pharmacogenomics studies? SL: We should think carefully about what the implications might be when thinking about race embedded in genetics. So I think we should ask researchers why they are including race when they do it. Race is often described as merely a first step on the road towards individualized therapies--a way station. However, it is often the case that racial findings are reported without any further interrogation of how race is serving as a proxy for other factors. The focus remains on race, which further reifies the notion that race is somehow genetic. For example, if a researcher were to conduct a pharmacogenomic study and found that one racially identified group had a significantly better drug response than another, one would hope that further studies on the reason for this difference would be conducted to see whether there were underlying factors at play. With Bidil, the FDA could have done a genetic study before they put a race-specific label on the drug. Then we could think about the results with much more complexity. TR: One of the opposing arguments for this idea is that requiring such studies will lengthen the time frame to get a drug approved. Do you think the societal dangers of approving race-based drugs override slowing the approval process? SL: There is no question that drugs that prove more effective than standard therapies should be made available to those who could benefit from them. However, rapid developments in genomic technologies have created new questions that should be addressed carefully. Institutions such as the FDA are in the important position of providing some guidance on how issues of race, genetics, and medicine should be addressed. At a minimum, we should explore what social costs may be involved. An interdisciplinary dialogue on the implications of continuing down the road of racially targeted medicines might help us avoid some of the mistakes we have made in the past where race and genes were conflated. |
Audio »
Share »
Digg this
Add to del.icio.us
Add to Reddit
Add to Facebook
Slashdot It!
Stumble It!
Add to Newsvine
Add to Connotea
Add to CiteUlike
Add to Furl
Googlize this
Add to Rojo
Add to MyWeb
Favorite
Print
E-mail
Genetics' Role in Health Disparities
03/21/2008
Massachusetts Institute of Technology
Comments
gabrielg01 on 12/20/2006 at 3:06 PM
313
Unreal...can anyone get more unscientific than this? If race is not embedded in the genome, then where is it coming from? The stork decides what race the newborns will be?...
Then they bring up a very old, fallacious argument: "more genetic variation exists within race-based populations than between populations"
This statement in itself is true, yet the conclusion is completely wrong. The fact is that you can make the same comparison (and argument) not only with races, but with species, as well. More genetic variation could (and does) exist within species-based populations than between species. Yet no biologist in their right mind would conclude that species are just a social construct.
The key of course, is not to cherry pick the results, like these quacks are doing. The variations across genes are not uniform. Some genes can have a lot of variability, without visibly impacting the phenotype. And some genes have very little variability.
If you pick a bunch of genetic markers from the highly variable group that show no biological phenotype, then it's easy to see how one can come to the fake conclusion that genetics and race are not related.
If our genome is 99% common to that of the chimpanzee, then we can easily imagine that the genomes of different human races are infinitesimally close. Maybe 99.99999% identical. And that would mean that race differences are contained in that 0.00001% genetic difference.
Did these guys search for and compared genetic markers in the 0.00001% pool, or in the 99.99999% pool?
Some common sense please!!!!
tcaruso on 12/21/2006 at 11:14 AM
6
Populations can be defined based on the genes they have in common. As has been shown, particular gene variants, for instance single nucleotide polymorphisms or SNPs, are widely distributed throughout the human population independent of skin color, facial traits, geography or some other artificial means of discriminating among humans. In other words, any particular SNP can generally be found in multiple human groupings that have been inappropriately labeled as races. One means of correctly identifying a specific population is as the set of all humans with that particular SNP.
One threat is, as Dr. Lee expressed, that these populations will be generalized to fit particular artificial segments of the population, like race, primarily for purposes of marketing. Companies find it a lot easier to promote their product to segments that have similar demographic characteristics. But how do we focus on the population and not the segments of these populations?
A second threat is that once we have identified SNP populations, companies, governments and individuals could then discriminate based on which SNP population an individual belongs. Visualize that, without government regulation, a health insurance company that knew that a particular SNP increased the likelihood for Alzheimers disease, would either deny health insurance for individuals in this population, or charge much higher premiums for these individuals. Gattaca, a recent movie about an individual in the future who falsifies information about their genetics, shows how this type of discrimination might develop.
In my opinion, both threats will be difficult to address as we learn more about the human genome and develop low cost means of characterizing our individual instances of the human genome.
gabrielg01 on 12/21/2006 at 12:36 PM
313
There are SNPs, and there are SNPs, and then there are even more SNPs...all different categories. Some SNPs have been linked to diseases, so they can be used as genetic markers, useful in healthcare. There are some race-linked SNPs as well. These can be used in forensic analyses.
But the vast majority of SNPs have no bearing whatsoever on the diversity of our phenotype. These are part of the 99.99999% common genome that we all share.
And by the way, the movie GATTACA was released in 1997. It's almost 2007 now, so I wouldn't call that movie "recent". Nevertheless, the message of the movie on the dangers of genetic discrimination is very important.