LEADING EDGE Does Race Have a Place In Biotechnological Research? BY JACK MCCAIN Senior Contributing Editor O n the basis of a clinical trial in which the race of heart failure patients was self-reported (as black), the U.S. Food and Drug Administration this year granted a small Massachusetts biotech company, NitroMed, approval to market the first race-based drug, BiDil. Here’s what the label for BiDil, a combination product of isosorbide dinitrate and hydralazine, says: BiDil is indicated for the treatment of heart failure as an adjunct to standard therapy in self-identified black patients to improve survival, to prolong time to hospitalization for heart failure, and to improve patient-reported functional status (emphasis added). If other pharmaceutical companies follow NitroMed’s lead, could additional products be developed with the supposedly special needs of specific racial and ethnic groups in mind? Is race a variable that deserves a place in biotechnological research? Or should the idea be tossed in the trash, fast? And does the BiDil saga (see page 56) point to a flaw in the drug approval process? The FDA has adopted the same race and ethnicity categories employed by the Office of Management and Budget. These categories are used to collect census data or monitor housing discrimination, 54 BIOTECHNOLOGY HEALTHCARE · DECEMBER 2005 PHOTOGRAPH BY STEVE WOIT Now that we have the first race-based FDA indication, more may follow. Is diversity necessarily a function of biology – and thus a cause of health-outcomes disparities – or are we fixing molecules at the expense of fixing society? among myriad other purposes. The FDA’s guidance for the pharmaceutical industry on the collection of race and ethnicity data in clinical trials, issued in September, notes that these categories “were not anthropologic or scientifically based designations.” On the other hand, the guidance points to the clinical relevance of various studies in racially and ethnically distinct subgroups in which differences in response to medical products have been observed, mentioning BiDil specifically. In other words, it would seem that the FDA endorses the use of nonscientific categories in the service of a science-based drugdevelopment process. Eventually, routine genetic tests may indicate whether a patient is a good candidate for a given drug. Genotyping tests already can identify patients whose genotype for various cytochrome P450 enzymes makes it likely that they are ultrarapid metabolizers of a drug (and therefore unlikely to achieve a therapeutic level at a normal dose) Poverty, not race, is the critical factor in disparities in health outcomes between blacks and whites, says Jonathan Kahn, JD, PhD, of Hamline University School of Law. Kahn told an FDA committee considering approval of BiDil for black patients that a race-specific label would imply that its use is inappropriate in nonblacks. or poor metabolizers (and therefore more likely to experience adverse effects). But, for the moment, there’s no inexpensive way to tell in advance whether or how a patient will respond to a drug, and the costeffectiveness of the available tests has yet to be ascertained. From the perspective of an individual physician confronting an individual patient, race might be one of many factors taken into consideration. Peter Schulman, MD, a cardiologist at the University of Connecticut Health Center, says the availability of BiDil doesn’t represent a new way of thinking about matching drugs and patients, because physicians already take race into consideration when contemplating therapies. As examples of evidence that might support race-based clinical decisions, he cites the Beta-blocker Evaluation of Survival Trial (BEST), which raised questions about the benefits of beta blockade, at least with bucindolol, in black patients with class III/IV heart failure, and a study showing that, compared with whites, some Asian subpopulations retain higher blood levels of rosuvastatin (Crestor). In fact, its manufacturer, AstraZeneca, has launched three phase 4 trials to determine the statin’s efficacy and safety in populations designated by race or ethnicity: ARIES (African-American Rosuvastatin Investigation of Efficacy and Safety), IRIS (Investigation of Rosuvastatin In South Asian Subjects), and STARSHIP (STudy Assessing RosuvaStatin in the HIspanic Population). “I have no problem with using race to help tailor therapy for an individual patient,” says Schulman. “You want to provide optimal therapy for all your patients,” he says. To that end, Schulman says he would consider using BiDil in a white patient if the circumstances warranted, though so far he has yet to prescribe it for anyone. IT’S ABOUT POVERTY Jonathan Kahn, JD, PhD, a lawyer from Hamline University School of Law, in St. Paul, Minn., has a different perspective. Kahn, who has followed BiDil very closely (Kahn 2003, 2004), spoke before the FDA’s cardiovascular advisory committee in June when BiDil was being considered for approval. Kahn told the committee there is a difference between a physician’s use of probabilistic correlations with race or ethnicity to guide dosage and administration (as Schulman does) and the establishment of a racespecific label for a drug. Kahn says that although physicians may have been trained to take race into consideration and feel comfortable using it to inform prescribing decisions, it would be more helpful to ask about a patient’s ZIP code or income, because that information is likely to be more useful DECEMBER 2005 · BIOTECHNOLOGY HEALTHCARE 55 LEADING EDGE Evolution of BiDiL, the first race-based drug 1980–1985 — Vasodilator Heart Failure Trial (V-HeFT I) finds that, in CHF patients receiving background therapy of digoxin and a diuretic, the alpha-blocker prazosin is no better than placebo in reducing mortality, whereas the combination of hydralazine and isosorbide dinitrate does reduce mortality. 1986–1991 — V-HeFT II compares hydralazine/isosorbide dinitrate against the ACE inhibitor enalapril and establishes ACE inhibitors as first-line treatment for heart failure; this combination is reserved for patients intolerant of ACE inhibitors. 1987 — Jay Cohn, a principal investigator in the V-HeFT trials, applies for a patent for the use of hydralazine/isosorbide dinitrate to reduce mortality associated with congestive heart failure. 1989 — Cohn is awarded a patent, which will expire in 2007. 1992 — Trademark application filed for a single pill combining hydralazine and isosorbide dinitrate, BiDil. 1995 — Trademark registered to Medco Research, which acquired intellectual property rights to BiDil from Cohn. 1996 — BiDil new drug application submitted to FDA for approval. 1997 — FDA rejects BiDil application, on the grounds that a retrospective analysis of V-HeFT data was inadequate statistically; Medco allows intellectual property rights to revert to Cohn. 1997 — FDA Modernization Act calls for the development of guidance on inclusion of women and minorities in clinical trials. 1999 — Cohn and colleagues publish a retrospective analysis of V-HeFT data (Carson 1999). 1999 — NitroMed obtains intellectual property rights to the fixed-dose combination of isosorbide dinitrate and hydralazine, then announces its intention to seek FDA approval to market this combination as treatment for heart failure in blacks. 2000 — Cohn and Peter Carson file a patent application for the use of hydralazine/isosorbide to treat heart failure in African-Americans. 2001 — FDA tells NitroMed that a successful trial of combination product in blacks with heart failure probably would lead to marketing approval; African-American Heart Failure Trial (AHeFT) begins enrolling patients to investigate hydralazine/isosorbide dinitrate as a treatment for heart failure in African-Americans. 2002 — U.S. patent number 6,465,463 awarded to NitroMed for using hydralazine/isosorbide to treat heart failure in African-Americans, the first granted to an existing drug for a racespecific use. Patent expires 2020. 2004 (July) — A-HeFT is terminated prematurely after a 43 percent improvement survival was seen in BiDil group relative to placebo group. 2004 (Nov.) — Results of A-HeFT published (Taylor 2004). 2005 (June)— FDA approves BiDil for the treatment of heart failure as an adjunct to standard therapy in self-identified black patients. SOURCE: KAHN 2004 than asking (or making assumptions) about the patient’s race. ZIP codes and income levels provide clues about affluence or, more importantly, the lack thereof. Poverty, he says, not race, is the critical factor in disparities in, for example, survival rates between black and white cancer patients (Bach 2002). Harold P. Freeman, MD, the director of the National Cancer Institute’s Center to Reduce Cancer Health Disparities, points in his writings to the convergence of poverty and culture, not race itself, as the explanation for excess mortality in certain groups. He points out that health outcomes are simi- 56 BIOTECHNOLOGY HEALTHCARE · DECEMBER 2005 lar in predominantly black Harlem, N.Y., and predominantly white Harlan, Ky. In both settings, Freeman says, the combination of poverty and culture are factors in overconsumption of fatty and salty foods, alcohol, and tobacco. Race need not be invoked. But if poverty is associated with LEADING EDGE poor health, and if poverty is inequitably distributed among racial and ethnic groups, then statistical portraits of racial and ethnic groups will capture the resulting differences in disease prevalence and mortality. Gregory Dorr, PhD, a historian with an interest in the intersection of race and medicine, notes that the median wealth of blacks in the United States is one tenth that of whites. “Would anyone seriously suggest that a genetic difference of 0.02 percent accounts for such a gross economic disparity?” asks Dorr, a postdoctoral associate at MIT’s Center for the Study of Diversity in Science, Technology, and Medicine. “No one wants to hear this, because it smacks of redistribution of wealth. Our culture prefers to seek technological solutions for social problems, but that approach comes with unintended consequences.” Foremost among these, Dorr fears, is that it bestows government sanction on racial categories — giving implicit legitimacy to racial categories that have no genetic basis. “The last time we labeled something for blacks only it was water coolers and bus seats,” he says. Although the genetic differences between races are very small, the social differences are immense, Dorr maintains. Unfortunately, there is a long, inglorious history in the United States of ascribing differences in mortality rates between whites and blacks to inherent biological differences, all the while invoking the name of science. The seminal document was Race Traits and Tendencies of the American Negro (1896), by the highly influential statistician and actuary Frederick L. Hoffman (1865–1946). After analyzing data from the 1890 census, Disparities between blacks and whites in age-adjusted death rates for heart disease, cancer, and diabetes Annually, per 100,000 population 600 Heart disease 500 Black White 400 300 Black Cancer 200 White 100 Diabetes 0 Black White 1950 1960 1970 1980 1990 2000 SOURCE: HEALTH, UNITED STATES, 2004 (TABLE 29), NATIONAL CENTER FOR HEALTH STATISTICS Hoffman claimed the statistics demonstrated that blacks were overly susceptible to illness, constitutionally unfit for survival — and destined for extinction. “From this perspective,” Dorr says, “slavery was construed as salubrious, because it helped blacks who could not survive as well on their own. Absent slavery, Southerners justified segregation as necessary to prevent race war and black extinction through disease.” Why do humans attach such importance to race? Biologist Richard Dawkins, author of The Selfish Gene and other books, says our insistence in assigning people to rigid racial and ethnic categories is an exam- ple of the “tyranny of the discontinuous mind” — a mind that is uncomfortable when grappling with continuous variables. Dawkins speculates that because humans associated the external appearance of their mates with appealing cultural characteristics, sexual selection gave rise to widely varying superficial features among various populations, much as dog breeders selected for appealing characteristics of domesticated wolves that led, rather rapidly, to the emergence of breeds as visually distinct as Dachshunds and Great Danes. But under the fur, a dog is a dog, biologically speaking, and under the skin of whatever hue, a person is a person. DECEMBER 2005 · BIOTECHNOLOGY HEALTHCARE 57 LEADING EDGE “Even though, if you take the totality of genes into account, we are a very uniform species, we are astonishingly variable in superficial features that are trivial but conspicuous,” Dawkins has written. The very conspicuity of trivial racial features has had nontrivial biological consequences, however, with respect to patterns of disease prevalence and mortality, because race often is the reason for socioeconomic inequities and patterns of behavior that provide different populations with varying levels of access to health care. As the NCI’s Freeman(2003) has written: becomes invisible. So do lots of black patients in the context of the common knowledge that blacks don’t respond well to ACE inhibitors. It may be true that, in general, blacks don’t respond as well to beta blockers and ACE inhibitors as do whites, and blacks in general respond better than whites to diuretics and calcium channel blockers, as a meta-analysis by Seghal (2004) has shown. But the same meta-analysis also shows that 80 to 90 percent of blacks and whites respond similarly to these agents. Seghal’s conclusion: “Race has little value in predicting antihypertensive drug response.” At this time, it appears that the single most important, and perhaps only valid, reason for continuing to use racial classification is to measure and monitor the effects of past and present social injustice. If people are grouped in a certain category and treated differently, then being in the category itself is the problem. EVOLUTION OF BIDIL BiDil combines fixed doses of two older vasodilators, hydralazine 37.5 mg and isosorbide dinitrate 20 mg, both of which are available as inexpensive generics and long used to treat heart failure patients who can’t tolerate ACE inhibitors. BiDil started out following the conventional route to approval — generating data from the V-HeFT (Vasodilator Heart Failure Trial) studies in the general population — but when its initial NDA was rejected, the investigators revisited their data and saw that the hydralazine/isosorbide dinitrate combination seemed to be more effective in a small subpopulation of black subjects. In the meantime, there had been calls for greater involvement of women and minority groups in clinical trials, and the FDA gave the goahead for a trial of BiDil in a population of self-described black patients with heart failure. About 60 percent of the patients were male, with a mean age of 57; 95 percent had New York Heart Association class III heart failure, with a mean Sickle cell disease is commonly held up as an example of how race is associated with certain diseases. Everybody knows sickle cell disease is a black disease, right? But in fact, sickle cell disease isn’t a function of having black African ancestors; it’s a function of having ancestors from regions where malaria was endemic. So although the sickle cell gene is most commonly found in people with sub-Saharan ancestry (but not in blacks from southern Africa), it’s also found in Mediterranean, Middle Eastern, and Indian populations — anywhere malariainduced evolutionary pressures have come into play. The problem is that when everybody knows sickle cell disease is a black disease, the nonblack patient 58 BIOTECHNOLOGY HEALTHCARE · DECEMBER 2005 ejection fraction of 24 percent. The trial was stopped prematurely after a 43 percent reduction in mortality was observed among patients receiving BiDil in comparison with those receiving placebo. The approval of BiDil was heralded as a therapeutic advance for black patients with heart failure, and perhaps it is; few have suggested otherwise. But soon after its approval, NitroMed was quick to suggest that BiDil was likely to be effective in other populations, too. Does it really work better in blacks than in whites? No one knows, because the African-American Heart Failure Trial (A-HeFT, a subsequent trial of the product in blacks with heart failure) didn’t examine that question. Treatment of heart failure has changed substantially since the VHeFT trials were conducted in the 1980s, with ACE inhibitors and beta blockers now being employed routinely in addition to diuretics and digitalis. Something else that has changed in recent years is that mortality rates for heart failure have declined in general, but most notably among black men and women. U.S. Centers for Disease Control and Prevention data show that between 1980 and 1995, total annual deaths attributed to heart failure increased from 27,415 to 46,484, with 94 percent of the 1995 deaths occurring in patients who were 65 or older. (The median age of A-HeFT subjects was 57, so even if the benefits of BiDil extend beyond self-described blacks, A-HeFT doesn’t speak to the age at which the vast majority of heart failure mortality occurs.) In the 65-and-up age group, ageadjusted death rates increased between 1980 and 1988 but then began falling, by annual rates of 3.0 LEADING EDGE percent in black men, 2.2 percent in black women, 1.7 percent in white men, and 0.5 percent in white women. As a result, the black:white mortality ratio for men fell from 1.3:1 to 1.1:1 between 1980 and 1995; for women, from 1.4:1 to 1.1:1. The CDC speculates that improvements in detection and treatment of hypertension, myocardial infarction, and heart failure are responsible for declining death rates in black adults and white men, and that the diminishing disparity in the black-to-white mortality ratio stems from improved control of hypertension and access to medical care among older blacks. In his remarks to the FDA advisory committee, Kahn argued that BiDil should be approved for use in the general population without regard to race. “Most drugs on the market today were approved by the FDA based on trials conducted almost exclusively in white patients,” he said, “but those drugs are not designated as white drugs, and rightly so.... When approving drugs tested in white populations, the proper assumption of the FDA was that the category white people did not differ in any meaningful way from the category human being. The same assumption should apply to a drug tested in an African-American population.” Kahn pointed out that A-HeFT enrolled only self-identified blacks, so there is no scientific basis to claim that BiDil works differently in other racial or ethnic groups. Rather than guide physicians in the selection of drugs, he argued, a racespecific label for BiDil would direct them to use it only in blacks and would imply that its use is inappropriate in nonblacks. Charles L. Curry, MD, chief of the Notable notions “While attempting to provide medical benefit, or market products, scientists and the pharmaceutical industry may reinvigorate the very notions of biological difference that may have resulted in racially disparate treatment and racially disparate health” (Ossorio and Duster 2005). “The continued use of ‘race’ in biomedical research is contrary to the goals of personalized medicine” (Shields 2005). “Every human being is genetically unique and so must be treated as an individual, not an example of a group defined by geography or race” (Bamshad 2005). “Geographic origin, patterns of intermating, and migration are strong determinants of population characteristics; these populations are not equivalent to race” (Freeman 2003). “The frequency of variant alleles and their encoded proteins varies among populations according to race and ethnic background. However, for a patient, the critical determinant is the genotype of the particular enzyme and not race or ethnic group, which is assigned by subjective criteria” (Wilkinson 2005). “The publication of the first draft of the sequence of the human genome should force an end to medical research that is arbitrarily based on race. The Human Genome Project now gives us the power to uncover the true origins of genetic variations; linking them to race has become passé” (Schwartz 2001). “[Race-based] research mistakenly assumes an inherent biologic difference between black-skinned and white-skinned people. It falls into error by attributing a complex physiological or clinical phenomenon to arbitrary aspects of external appearance” (Schwartz 2001). Division of Cardiovascular Diseases at Howard University College of Medicine and president of the board of trustees of the International Society of Hypertension in Blacks, spoke immediately after Kahn. “I could almost say ‘Amen’ to what Dr. Kahn just said and sit right down,” Curry told the panel. Endorsing the approval of BiDil for any patient without respect to race, he noted that American cardiologists readily applied the results of the Scandinavian Simvastatin Survival Study broadly, instead of restricting statin therapy to Scandinavians. Three experts from the National Human Genome Center (NHGC) at Howard University also spoke, including Charles Rotimi, PhD. In 2003, NHGC hosted a workshop, “Human Genome Variation and ‘Race,’” that was the basis for a supplement to Nature Genetics last year. The workshop and supplement were sponsored by the U.S. Department of Energy’s Office of Science, whose director wrote in the intro- DECEMBER 2005 · BIOTECHNOLOGY HEALTHCARE 59 LEADING EDGE duction to the supplement: “The solid science presented in this issue … is providing proof that oversimplified concepts of race simply don’t work in any objective realm. It’s bad medicine, and it’s bad science.” During the FDA meeting, Rotimi said, “It would be tragic not to approve [BiDil] and it would be even more tragic just to approve it for African-Americans.” Rotimi, an epidemiologist who came to the United States from Nigeria in 1981, asked why blacks are at elevated risk for numerous conditions. He said, “Are they just selectively acquiring bad genes? I don’t think so. There must be something in our social environment that drives people toward poor health, and only by addressing that can we reduce health disparity.” He showed them data from his own research depicting how environment, not race, is the driving force behind hypertension in blacks.1 He also showed them data from Seghal’s aforementioned metaanalysis (2004) indicating that the majority of whites and blacks respond similarly to commonly used antihypertensive agents. He closed his remarks by asking what good is a drug that reduces heart failure mortality if it leads biomedical research down the wrong path by suggesting, without proper scientific justification, that the so-called racial categories are biological? Two of Rotimi’s colleagues at NHGC, Charmaine Royal, PhD, and Vera Ota Wang, PhD, a voting 1 In rural Nigeria, the frequency of a genetic variation associated with hypertension is around 90 percent, and it is nearly that high among urban African-Americans. However, whereas the rate of hypertension is about 34 percent among the urban African-Americans, in Nigeria the rate of hypertension is only 7 percent. member of the FDA advisory committee, also spoke. Royal asked how blacks would be identified in the clinical setting. Would there be standardized national criteria to determine who is eligible for BiDil? Would patients be allowed to selfidentify (as was done in A-HeFT), or would the physician decide who is or is not black? Ota Wang questioned the rationale of the A-HeFT researchers for using a self-reported racial category as an inclusion criterion instead of a more direct, quantitative biological trait. “There is a presumption here that, somehow, this selfidentified social identifier is somewhat equivalent or representative of a biological process, and I am not sure it really is.... We need to look carefully at the issue of selfidentified racial categories, because if the assumption is that these population differences are biological, the self-identified population is a social and political construct.” But if comments like these elicited a response at all, it was to dismiss them. Said the advisory committee’s chairman, Steven Nissen, MD, the oft-quoted Cleveland Clinic cardiologist: Drugs aren’t racist; people are.... [In this committee] we try to identify whether a drug has efficacy and the population that the drug will benefit. When the overwhelming evidence that leads me to believe that this drug is effective comes from a population we can define by some characteristic — self-identified race, in this case — that is very unusual. It is precedent-setting.... We are moving toward the era of genomic-based medicine, [where] we are going to have the ability to look for an SNP [single nucleotide polymorphism] that tells us that this group of people will benefit from x 60 BIOTECHNOLOGY HEALTHCARE · DECEMBER 2005 drug or will be harmed by y drug.... We already know that there are differences [in people of African origin]. We know, for example that ACE inhibitors don’t seem to work as well in African-Americans. We know that certain diseases are more or less prevalent.... We are using self-identified race as a surrogate for genomic-based medicine and I don’t think that is unreasonable.... I wish we had the genetic markers ... to decide who is going to respond to what drug. But in the absence of that, we have to use the best available evidence, and that evidence was used in this trial and it worked. Said an FDA representative, “What are the choices one has? This is sort of what you work with because it is the best you have and everyone agrees it is inadequate.” In other words, pragmatism trumps science. CONNECTING RACE, HEALTH As President Bush spent part of 2005 attempting to build support for his idea to let workers invest some of their Social Security withholdings in private accounts, he advanced an interesting argument: “African-American males die sooner than other males do, which means the system is inherently unfair to a certain group of people.” Bush suggested that because the average age of death for an AfricanAmerican man is 69, versus 75 for a white man, black men have fewer years to enjoy Social Security benefits after they retire. What about being AfricanAmerican leads to death at an earlier age? Was Bush perhaps referring to certain inherent biological differences between black and white men that naturally result in shorter life expectancies for blacks? LEADING EDGE The statistics Bush cited were life expectancies at birth for males born in 2001. For the purpose of the Social Security debate, it might be more useful to look at life expectancies for males approaching retirement age. According to the CDC, black and white males born in 1950 could expect to live an average of 59 and 66 years, respectively. With respect to Social Security retirement benefits, however, an even more pertinent statistic would be years of life remaining for males who are 65 years old. As of 2002, black males at age 65 had 14.6 years of life remaining, not much fewer than the 16.6 years remaining for white males. (In 1950, years of life remaining for 65-year-old black and white males averaged 12.9 and 12.8 years, respectively.) That is, once most men make it to 65, they’re still looking forward to about 15 more years of life, on average.2 Close inspection of the table above shows that blacks generally have higher death rates for nearly all the causes listed — cancer, cardiovascular disease, diabetes. Could it be that blacks have certain biological characteristics that make them more susceptible than members of other racial groups to the major killers? If so, should research be aimed at developing drugs that specifically target these biological differences? A longitudinal view shows a different picture. The age-adjusted mortality rates for heart disease, cancer, and diabetes in 1950 and 1960 suggest that at that time blacks were just as healthy (or not) as whites (see chart on page 57). But 2 Higher rates of infant mortality, HIV infection, and homicide among black males account largely for the difference between blacks and whites in total life expectancy. Age-adjusted death rates for selected conditions United States, 2002, per 100,000 population All causes Diseases of heart Ischemic heart disease Cerebrovascular diseases Malignant neoplasms Trachea, bronchus, and lung Colon, rectum, and anus Prostate Breast Chronic lower respiratory diseases Influenza and pneumonia Chronic liver disease and cirrhosis Diabetes mellitus Black or African-American White 1,083.3 308.4 203.0 76.3 238.8 61.9 26.8 62.0 34.0 31.2 24.0 8.5 49.5 837.5 239.2 171.0 54.6 195.6 57.5 19.5 25.8 25.6 46.9 22.6 9.0 22.2 SOURCE: HEALTH, UNITED STATES, 2004 (TABLE 29), NATIONAL CENTER FOR HEALTH STATISTICS since then, mortality rates for blacks and white have diverged. Whereas in 1950, the diabetes death rate was essentially the same for blacks and whites, by 2000 it was more than twice as high among blacks. Even though rates of death from heart disease have declined in both groups, by 2000 the mortality rate among blacks was 30 percent greater than that among whites. So was the mortality rate from cancer. What happened over the course of 50 years? A lot of new drugs were developed during that interval, so do blacks just happen to respond more poorly to all or most of them, owing to genetic differences in drug metabolism? Are blacks more susceptible genetically to a host of environmental risk factors for these diseases that have emerged just during the last half century? Perhaps there’s just something about the biology of self-described African- Americans that is different. As the chairman of the FDA advisory committee contemplating BiDil said in June, “The road to hell is paved with biological plausibility but there is a biologically plausible explanation.” Once you start considering races to be sufficiently distinctive genetically to explain health disparities between them, there’s no telling where it will lead you. This slippery slope worries sociologist Troy Duster, PhD, immediate past president of the American Sociological Society. In recent publications (Duster 2005, Ossorio 2005), Duster has noted that in 1950, the incarcerated percentages of the black and white populations were 0.38 and 0.09 percent, respectively, a 4:1 ratio; by 1995, the shares respectively were 1.65 and 0.21, or 8:1. To apply the logic that “genetic differences between whites and blacks explains widening variances DECEMBER 2005 · BIOTECHNOLOGY HEALTHCARE 61 LEADING EDGE in health outcomes” to Duster’s point about incarceration, perhaps you would have to advance an argument that some genetically based propensity for criminal behavior also was emerging amidst American blacks, too. Of course, over an interval as short as 50 or 60 years, the emergence of any evolutionary changes in black and white biology that might account for disparities in health outcomes or incarceration rates is highly implausible. At the same time, it is likely that DNA data compiled from people who have been incarcerated would identify certain SNP profiles that are found at statistically significant higher rates in blacks — simply because black inmates outnumber white inmates. Duster warns that if race is used as the phenotype to distinguish populations subjected to SNP profiling, then once it becomes widely available, “the likelihood of committing the fallacy of misplaced concreteness, in science, is nearly overwhelming” (Duster 2005). In the wake of the approval of BiDil, it’s likely that drug companies far and wide are scurrying to reanalyze data from past studies that might support a race-based indication that would extend the patentprotected life of a product. What if, for example, Merck decided to pursue a race-based indication for losartan (Cozaar), which was mentioned during the FDA advisory board meeting on BiDil as an example of a drug presenting a “huge hazard” in the black population but not in the white population? If NitroMed could do a retrospective analysis of BiDil data among black patients, gain a racebased patent from the U.S. Patent Office, and get FDA approval to run a new clinical trial of BiDil in selfdescribed blacks, who could fault Merck for doing a retrospective analysis of its losartan data, getting a new race-based patent, and conducting a trial that would lead to a new race-based indication for losartan in patients with left ventricular hypertrophy — but just for selfdescribed whites? In fact, lots of people would fault Merck, says Dorr, at MIT, because whites with left ventricular hypertrophy aren’t regarded as a particularly disadvantaged group — and because efforts to develop a whitesonly drug would suggest racism. Then why could BiDil succeed with a blacks-only approach? Dorr says it’s because BiDil is viewed as benign — a tool for narrowing disparities in health outcomes between blacks and whites. “BiDil redirects peoples’ attention from knotty social problems to slick — but flawed — technological fixes,” Dorr says. Kahn holds a similar view. “I’m all in favor of using race to track disparities in health care,” he says, “but the approval of BiDil sends the message that, to the extent that there are disparities, the solution is to fix the molecules, not the underlying problems.” Stepping aside from the pragmatism that drives much of clinical practice, it would appear that racebased medicines offer a quick fix — and a profit for manufacturers willing to take advantage of the current regulatory mindset — for problems that are more than skin deep. BH Based in Durham, Conn., Jack McCain is a freelance medical writer and editor. He holds degrees from Allegheny College and Wesleyan University. 62 BIOTECHNOLOGY HEALTHCARE · DECEMBER 2005 FURTHER READING Bach PB, Schrag D, Brawley OW, et al. Survival of blacks and whites after a cancer diagnosis. JAMA. 2002;287:2106–2113. Bamshad M. Genetic influences on health. Does race matter? JAMA. 2005;294:937–946. Carson P, Ziesche S, Johnson G, Cohn JN. Racial differences in response to therapy for heart failure: analysis of the vasodilator-heart failure trials. Vasodilator-Heart Failure Trial Study Group. J Card Fail. 1999;5:178–187. Cooper RS, Rotimi CN, Ward R. The puzzle of hypertension in African-Americans. Sci Am. 1999;280:56–63. Dawkins R. The Ancestor’s Tale. 2004. Duster T. Race and reification in science. Science. 2005;307:1050–1051. FDA (U.S. Food and Drug Administration). Collection of Race and Ethnicity Data in Clinical Trials. Rockville, Md.: FDA. September 2005. Available at: «http://www.fda.gov/cder/guidance/ 5656fnl.pdf.» Accessed Sept. 30,. 2005. FDA. Cardiovascular and Renal Drugs Advisory Committee [transcript of meeting testimony]. Available at: «http://www.fda.gov/ohrms/ dockets/ac/05/transcripts/2005-4145T2.pdf». Accessed Nov. 9, 2005. Freeman HP. Commentary on the meaning of race in science and society. Cancer Epidemiol Biomarkers Prev. 2003;12:232s–236s. Jorde LB. Genetic variation and cardiovascular care. Manag Care. 2004;13(10 suppl):9–13. Kahn J. Getting the numbers right: statistical mischief and racial profiling in heart failure research. Perspect Biol Med. 2003;46:473–483. Kahn J. How a drug becomes “ethnic”: law, commerce, and the production of racial categories in medicine. Yale J Health Policy Law Ethics. 2004;4:1–46. Kahn J. Misreading race and genomics after BiDil. Nat Genet. 2005;37:655–656. Nerenz D. Health care organizations’ use of race/ethnicity data to address quality disparities. Health Aff (Millwood). 2005;24:409–416. Ossorio P, Duster T. Race and genetics. Controversies in biomedical, behavioral, and forensic sciences. Am Psychol. 2005;60:115–128. Rotimi CN. Are medical and nonmedical uses of largescale genomic markers conflating genetics and ‘race’? Nat Genet. 2004;36(11 suppl):S43–S47. Sankar P, Cho MK, Condit CM, et al. Genetic research and health disparities. JAMA. 2004; 291:2985–2989. Seghal AR. Overlap between whites and blacks in response to antihypertensive drugs. Hypertension. 2004;43:566–572. Shields AE, Fortun M, Hammonds EM, et al. The use of race variables in genetic studies of complex traits and the goal of reducing health disparities: a transdisciplinary perspective. Am Psychol. 2005;60:77–103. Schwartz RS. Racial profiling in medical research [editorial]. N Engl J Med. 2001;344:1393–1394. Taylor AL, Ziesche S, Yancy C, Carson P, et al.; African-American Heart Failure Trial Investigators. Combination of isosorbide dinitrate and hydralazine in blacks with heart failure. N Engl J Med. 2004;351:2049–2057. Erratum in: N Engl J Med. 2005;352:1276. Wang VO, Sue S. In the eye of the storm: race and genomics in research and practice. Am Psychol. 2005;60:37–45. Wilkinson GR. Drug metabolism and variability among patients in drug response. N Engl J Med. 2005;352:2211–2221. Williams DR, Jackson PB. Social sources of racial disparities in health. Health Aff (Millwood). 2005;24:325–334.