If you don't like the word "racial," just substitute "population." In any case, Many African-Americans Have A Gene That Prolongs Life After Heart Failure:
About 40 percent of African-Americans have a genetic variant that can protect them after heart failure and prolong their lives, according to research conducted at Washington University School of Medicine in St. Louis and collaborating institutions. ... "By mimicking the effect of beta blockers, the genetic variant makes it appear as if beta blockers aren't effective in these patients," he explains. "But although beta blockers have no additional benefit in heart failure patients with the variant, they are equally effective in Caucasian and African-American patients without the variant." ... The researchers...found that 41 percent of African-Americans have a variant GRK5 gene that more effectively suppresses the action of adrenaline than the more common version of the gene. People with the variant gene could be said to have a natural beta blocker, Dorn says. The variant is extremely rare in Caucasians, accounting for its predominant effects in African-Americans.
Here's the original paper, A GRK5 polymorphism that inhibits bold β-adrenergic receptor signaling is protective in heart failure:
β-adrenergic receptor (βAR) blockade is a standard therapy for cardiac failure and ischemia. G protein-coupled receptor kinases (GRKs) desensitize βARs, suggesting that genetic GRK variants might modify outcomes in these syndromes. Re-sequencing of GRK2 and GRK5 revealed a nonsynonymous polymorphism of GRK5, common in African Americans, in which leucine is substituted for glutamine at position 41. GRK5-Leu41 uncoupled isoproterenol-stimulated responses more effectively than did GRK5-Gln41 in transfected cells and transgenic mice, and, like pharmacological βAR blockade, GRK5-Leu41 protected against experimental catecholamine-induced cardiomyopathy. Human association studies showed a pharmacogenomic interaction between GRK5-Leu41 and β-blocker treatment, in which the presence of the GRK5-Leu41 polymorphism was associated with decreased mortality in African Americans with heart failure or cardiac ischemia. In 375 prospectively followed African-American subjects with heart failure, GRK5-Leu41 protected against death or cardiac transplantation. Enhanced βAR desensitization of excessive catecholamine signaling by GRK5-Leu41 provides a 'genetic β-blockade' that improves survival in African Americans with heart failure, suggesting a reason for conflicting results of β-blocker clinical trials in this population.
The key here is that you can't assume all genetic backgrounds are the same; something that should be obvious to you if you've been reading for the past generation about how Africans exhibit more extant genetic variation than all other human populations combined. The medical regime which encouraged the prescription of beta blockers presupposed a white treatment population, which makes sense since most Americans are white (greatest good for the greatest number and all). Against that genetic background beta blockers exhibit a great deal of efficacy; but against an African American genetic background far less so. Why? In this case the answer seems rather clear in that there is an allele of this locus whose polymorphism has a direct causal relationship to a biochemical pathway implicated in the production of the relevant enzyme. These sorts of data are the rationales for attempting to make sure you have a diverse populations against which you perform medical trials; humans are mostly the same, but differences on the margins are not trivial (you won't think so when you get sick!). Varied genetic backgrounds and their biomedical importance are one of the primary reasons behind research endeavours such as the International HapMap Project. Obviously I'm really interested in evolution and humans, and how the two relate. But research with basic science yield is funded ultimately due to interest in proximate human well being; in other words, evolutionary biological implications are a side effect of the quest for medically salient data. Why establish a rigorous taxonomy of human populations? Why develop genetic analyses which can smoke out cryptic population substructure? Why map out as many QTLs as possible for a host of assorted molecular genetic processes? Knowledge is power, and hopefully power which those in medical disciplines will use to craft the regime most well suited to the host of priors any given individual brings to the table. Of course, I would also make the case that an acknowledgment of the ubiquity of evolution in shaping where we start out in life (or end up) is essential to extracting maximum insight from these data. Note:Take a closer look at the SNP yourself. "A" codes for the non-protective variant.