Faculty Peer Reviewed
Given the traumatic and often criminal role that medicine and the larger scientific community played in some of the most shameful acts of the 20th century, it is natural that the consequences of these collaborations have continued to reverberate to the present day. The chills sent down our spines can be sparked from reading treatises purporting to demonstrate the undeniable genetic evidence of racial superiority, or from the revulsion towards the degrees of complicity needed for the horrific Tuskegee experiment to have such devastating success. It is easy to see why as a society we are distrustful of the study of race in our current century.
It consequently should not be so surprising that an innocently titled manuscript, “The African-American Heart Failure Trial: Background, Rationale, and Significance,” was published in 2002 . In a remarkable departure from typical manuscripts detailing the findings of randomized control trials, this was an article focused solely on the justification for a drug trial in a specific study cohort. The trial agent was “BiDil,” a combination pill of isosorbide dinitrate and hydralazine, and the study population was African American. But it almost does not matter what the agent being discussed is. BiDil serves as a useful lens through which to view two phenomena: (a) the role that pharmacogenetics will play in the future of drug therapy for illness; and (b) the degree in which ancestry and its association (or lack thereof) with race and genetics can influence the heritable risk factors of illness . The latter phenomenon will be (partially) addressed, not only in African Americans, but in other communities with complex ancestries in the US, such as Hispanics/ Latinos.
The aforementioned 2002 communication was framed as anticipated results, but in reality, this was not the case. BiDil was an agent that initially failed to receive Food and Drug Administration (FDA) approval earlier because of its lack of statistically significant success in the initial study cohort, except for one group: African Americans . Not surprisingly, when the drug made its way again through the approval process as therapy explicitly for African Americans, the federal regulatory authorities eventually were called upon to enter the fray. The FDA then, in 2005, approved for the first time in its history a pharmacotherapy for a specific racial group: “black patients” . In justifying its rationale, the FDA made sure to emphasize the undue burden of heart failure on African American patients, and why such unconventional approaches needed to be taken given the statistically significant improvement made with the regimen in this group. This did not allay the controversy surrounding the decision, which became further magnified both within the popular press and academia. Articles were penned ranging from the measured “What’s right (and wrong) with racially stratified research and therapies”  to the more inflammatory “There is no scientific rationale for race-based research” . One consistent criticism was the conflation of race and genetics: if the drug is targeted for African Americans, how do we know who are African Americans to whom we should be prescribing BiDil?
An exhaustive analysis of the complexities surrounding the formation of the modern African American community is beyond the scope of this discussion. Yet the term “African American” and the racial category of “black” are inherently social inventions that are not descriptive of inherent traits. They are definitions derived from the tormented relationship the United States had with slavery as the first modern republic based on democratic rule that excluded officially until the 1960s one segment of the demos: people of West African ancestry. A complex pigmentocracy eventually became distilled into a bi-racial hierarchy determined by the rule of the famed “one drop:” if you were of part African ancestry, you were black or African-American. Naturally, one can see the dilemma posed when such a social construct is used within scientific discourse.
Proof of this incoherent relationship between race and genetics was yielded on whole-genome analysis of African Americans in comparison to African populations . When looking at markers derived from single nucleotide polymorphisms (SNPs) that could differentiate continental ancestry, it was found that not only was the average European contribution to the African American genome about 18.5%, but that there was notable variability in this contribution. This phenomenon is not limited to African Americans. For example, Hispanics/ Latinos have posed such a conundrum to the US Census that obtuse racial categories are conjured up like “White Non-Hispanic” that treat origin from Latin America as an ethnicity. When you compare individuals from communities with origins in continental or Caribbean Latin America to Native American, European, and African populations, you easily see the inadequacy of treating Hispanics/ Latinos as one homogenous group . The country-specific designations themselves are problematic, with some significant differences notable even within Puerto Rico, a small island comparable in size to Yellowstone National Park.
This begs a natural question: what medical benefit is derived from studying such differences in disease severity and outcome in specific communities that are inherently social inventions? As BiDil demonstrated to us, even though the days of Jim Crow and Tuskegee seem consigned to dusty history books, the collective trauma they have inflicted on society’s psyche is still evident. Such a legacy must be addressed as the study of the heritable risk for non-Mendelian illnesses like cancer or hypertension or diabetes flourishes as high density genomic data becomes increasingly available. Undoubtedly, an aspect of this inheritance will continue to bear the same genetic markers that can also differentiate people based on place of origin, in what we today assign the label “race.” BiDil demonstrates that without moving past the darkness of the 20th century, we will not be able to explore the full promise of genomics in the 21st century and its application to better disease prevention, management, and treatment. Perhaps fulfilling such a promise will be a first act of atonement for the horrors that some medical practices have inflicted on humanity and a way to demonstrate the true aspirations of our profession: caring with compassion and expertise for all sick persons through the rational and unbiased application of new knowledge, thus contributing to the advancement of our society.
Dr. Christopher David Velez is a 2011 graduate of NYU School of Medicine
Peer Reviewed by Antonella Surbone, MD, Ethics Editor, Clinical Correlations
Image courtesy of Wikimedia Commons
1. Taylor AL, Cohn JN, Worcel M, et al. The African-American Heart Failure Trial: background, rationale, and significance. Journal of the National Medical Association. 2002; 93 (9): 762-769 http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2594160/
2. Kahn JD. Pharmacogenetics and ethnically targeted therapies. BMJ. 2005; 350: 1508
3. Bloche MG. Race-based therapeutics. New England Journal of Medicine. 2004; 351(20): 2035-2038 http://www.nejm.org/doi/full/10.1056/NEJMp048271
4. FDA Consumer Magazine September-October 2005 Issue. FDA approves Heart Drug for Black Patients. http://permanent.access.gpo.gov/lps1609/www.fda.gov/fdac/features/2005/505_BiDil.html Accessed February 24, 2011.
5. Sade RM. What’s right (and wrong) with racially stratified research and therapies. Journal of the National Medical Association. 2007; 99(6): 693-696
6. Hoover EL. There is no scientific rationale for race-based research. Journal of the National Medical Association. 2007; 99(6): 690-692. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2574368/
7. Bryc K, Auton A, Nelson MR, et al. Genome-wide patters of population structure and admixture in West Africans and African Americans. PNAS. 2010; 107(2): 786-791 http://www.pnas.org/content/107/2/786.full
8. Bryc K, Velez CD, Karafet T, et al. Genome-wide patterns of population structure and admixture among Hispanic/Latino populations. PNAS. 2010 107: 8954–8961 http://www.pnas.org/content/107/suppl.2/8954.full