Jonathan Kahn, The Troubling Persistence of Race in Pharmacogenomics, 40 Journal of Law, Medicine & Ethics 873 (Winter 2012) (80 Footnotes)
“The state is not abolished, it withers away”
-Friedrich Engels, Anti-Durhing
“I'm gonna tell you how it's gonna be.”
-Buddy Holly and the Crickets, Not Fade Away
In 1878, Friedrich Engels famously wrote that on the road to realizing the communist utopia, “the state is not abolished, it withers away.” In a similar manner, biomedical researchers telling us that come the promised land of individualized genomic medicine, the need for using race will also “wither away” in the face of scientific progress. Such millennial hopes are, no doubt, sincere, but they enable the continued casual proliferation of racial categories throughout biomedical research, product development, marketing, and clinical practice. My contrasting quotation to frame this article is drawn from the 20th century pioneer of rock and roll, Buddy Holly (né Charles Hardin Holley) whose 1957 hit “Not Fade Away” begins with the line, “I'm gonna tell you how it's gonna be” --the point being that far from withering away, race is persisting even as genomic milestones are being reached and passed. In short, despite biomedicine's promises to the contrary, race will “not fade away” of its own accord even as the science of genetics progresses.
This article is concerned about what may be happening to race and medicine in the “meantime” between today's clinical realities and the promised land of pharmacogenomics. It argues that previous debates over the use of race in medicine are being side-stepped as race is being reconfigured from a “crude surrogate” for genetic variation into a purportedly viable placeholder for variable drug response--to be used here and now until the specific genetic underpinnings of drug response are more fully understood. Embracing the trope of “promise” in pharmacogenomics alongside the idea of using race as a useful interim proxy for genetic variation raises concerns that new diagnostic and therapeutic interventions may reflect or be mapped upon existing social categories of race, class, gender, and ethnicity in a harmful or dangerous manner. At the most basic level, the politics of the meantime in pharmacogenomics may be promoting what sociologist Troy Duster has termed, the ““molecular reinscription of race”--the scientifically unjustified and socially dangerous recasting of race as a social and historical construct into a reified genetic category.
As race gains renewed legitimacy as an interim measure on the road to individualized medicine, a curious thing is happening: race is persisting even as genetic milestones are reached and passed. This article begins with a brief general examination of how race is persisting in biomedical research with particular consideration of how the concept of the “unknown” is used to create a space for race as a genetic construct. After marking the power of the concept of race as a “stepping stone” in biomedical research, it moves on to explore the case of the widely prescribed blood-thinning drug warfarin (marketed by Bristol-Meyers Squib under the trade name Coumadin®), which involves the persistence of race even as specific genetic variations directly tied to warfarin response are being identified.
I argue here that race is persisting for two additional reasons. First, race is evolving into a “residual category” that is being used to explain any variation in drug response that is not captured by genetics. The irony here is that genetics will never explain 100% of variable drug response. Many complex environmental, dietary, and behavioral factors also affect drug response. In short, an unknown aspect to drug response will always exist; therefore, a potential use for race will always exist as a category to catch this residuum of unexplained variation. Second, there is an inertial force to race in biomedicine. Once introduced into a conceptual system for evaluating biological differences, it is very difficult to dislodge race. It becomes part of the common sense of biomedical practice and continues to be used almost reflexively--not least because the array of federal mandates require biomedical research to gather and classify data by race.
Why Racial Profiling Persists in Medical Research
In August, 2009, Time magazine published a story titled, “Why Racial Profiling Persists in Medical Research.” The article is not remarkable for any particularly new insights but stands as prominent example of a type of reasoning that typifies a powerful strand of current discourse about race, medicine and genetics. Curiously, the article began with a reference to Harvard professor, Henry Louis Gates, Jr., who had been arrested on his front porch the previous month by a Cambridge police officer. Gates is one of the country's most prominent African American intellectuals. The officer had responded to a call from a passer-by who had seen Gates and another black man seeming to force their way into the house. It later turned out that Gates had just returned from a trip to China and found the front door to be stuck and so asked the driver to help him force it open. The officer arrived on the scene after Gates had gained entry. He showed the officer his identification and explained that this was his home. Gates took umbrage at being questioned in his home, allegedly shouting, “This is what happens to black men in America.”A loud argument ensued which continued onto the front porch, and finally the officer arrested Gates for disorderly conduct. Shortly thereafter, President Obama caused a furor when he referred to the Cambridge Police Department's handling of the affair as ““stupid.” The famed “beer summit” ensued, with President Obama, flanked by Vicepresident Biden, hosting Gates and the arresting officer for a reconciliatory beer at the White House.
The article's brief reference to this event was used to juxtapose racial profiling as a “social” practice vs. racial profiling as a “scientific” practice. The former was clearly understood as “bad,” while the latter was presented as problematic but potentially useful. This established the frame for the entire article, which focused on the publication that July of a study by Kathy Albain et al. in the Journal of the National Cancer Institute (JNCI) involving a meta-analysis of data concerning more than 19,000 patients who participated in clinical trials involving treatments for a variety of cancers. As presented by Time, the major finding of the study was “that all other/actors being equal, black patients had on average a significantly lower cancer survival rate than whites.” The article takes this finding as emblematic of a strain of biomedical research whose use of racial categories is “borne out in studies that attribute health disparities between blacks and whites not to socioeconomics or access to health care alone but also to genetic differences between the races--a concept that implies that a biological category of race exists.”
Several things are going on here. First is the idea that when “all other factors” are held equal, observed racial differences in cancer are likely due to genetics. Second, being genetic, these “differences” are not “disparities” implicating socioeconomic forces. By implication, such purported genetic differences require biomedical interventions at the molecular level but do not require policy interventions to address questions of equity or social justice.
What then are these “other factors”? If we look at the actual article published by Albain et al., we see that its conception of “other factors” is encompassed by “[e]stimates ... derived from education category and income level as assessed by the linkage between patient zip code and the US census data.” As subsequent critical letters to the JNCI pointed out, Albain et al. presented a remarkably thin and fundamentally flawed approach to controlling for non-genetic factors. One letter from epidemiologists at the Centers for Disease Control, Duke University, and McGill University pointed out several particularly glaring flaws “rendering [Albain et al.'s] adjustments inadequate and their conclusions therefore unsupported.” First, they noted that as long ago as 1998, University of Michigan professor of Health Behavior & Health Education Arline Geronimus had shown that using “zip code-level socioeconomic status proxies ... for individual-level socioeconomic status” was inappropriate and misleading (essentially conflating group statistics with individual attributes). Second, the article Albain et al. cite as a source for using such census data actually “proposes that aggregated statistics be used for monitoring disease trends and not that they be used for individual-level control in racial disparity studies.” And third, even by Albain et al.'s own account, “The socioeconomic status data were missing for between 27% and 79% of subjects depending on the clinical trial.” Another letter to the editors noted that “Albain et al. provide no information about how they measured the central variable: race. This omission makes it impossible to disentangle the many behavioral, environmental, or genetic influences on cancer mortality that may be associated with race.”
Albain et al.'s flawed approach to controlling for “all other things” was essential to creating the space for genetics to enter into the characterization of racial difference. As Albain told Time,“Something big is going on among people who are getting equal care,” and that something, the authors concluded, “must be some unknown biological or genetic factor that differs by race.” The “unknown” is central to the dynamic of geneticizing racial difference. The logic here is clear, and clearly flawed. First, observe a biomedical racial difference (response to a drug or experience of a disease). Second, “control” for non-genetic factors (conveniently reducing these to income and education or some even more flawed identification with census data). Third, attribute any residual racial difference to genetics (even though you have no specific genes identified). Voila! A biological basis for race.
This approach was echoed in some responses to the JNCI article. For example, Dr. Lisa A. Newman, Director of the University of Michigan Breast Care Center, said, “There seems to be something associated with racial and ethnic identity that seems to confer a worse survival rate for African Americans. I think it's likely to be hereditary and genetic factors.” Yet, as Harvard Professor of Public Health David Williams noted in commenting on the Albain et al. study, “The biology is a fall-back black box that many researchers use when they find racial differences .... It is knee-jerk reaction. It is not based on science, but on a deeply held, cultural belief about race that the medical field has a hard time giving up.”
Albain, however, claimed that race was merely a surrogate for unknown genes: “When we find out what the [genetic] ‘it’ is,” she asserted, “we will be able to test everyone for ‘it’ and we will find some Caucasians who have it and some blacks who don't and we won't be talking about black and white anymore.”Albain here adopted the familiar trope of race as merely an interim measure, to be used as a proxy for genetics in the “meantime” until we reach the promised land of genomic medicine where race will then “wither away.” In this approach, however, rather than questioning their use of socioeconomic controls, the authors use race as a residual category to capture and geneticize all ““unknown” causes of racial difference. As epidemiologist, Jay Kaufman noted in criticizing the JNCI study, “If you are trying to make the argument that [different health outcomes] must be genetic by exhausting other possibilities and saying what is left over must be genes, well, that's never going to work. There are a million things that affect people's lives. If you think it's genes, then measure genes.”
Albain et al.'s approach is all too common. For example, in their study of genetic research into Type 2 Diabetes, Paradies, Montoya, and Fullerton were particularly concerned by “the offhand manner in which a number of these studies dismiss the relevance of sociocultural variables by failing to measure but nonetheless rejecting lifestyle factors, (e.g. diet and physical activity), as well as environmental and sociocultural characteristics as possible influences on study findings.” Given the complexity of interactions among human biology, behavior, history, society, and the environment, there will always be some causes of biomedical differences correlating with race that remain “unknown” and susceptible to being geneticized. As long as biomedical researchers give short shrift to socioculutral variables, there will continue to be a place for a flawed genetic concept of race long after the promised land of individualized pharmacogenomic medicine is reached.
Stepping Stones and Poster Children
The productive power of the unknown to create a space for race is clearly evident in a 2008 article published by Robert Temple, Director of the Office of Medical Policy and the FDA's Center for Drug Evaluation and Research and his colleague, Shi-Mei Huang of the Office of Clinical Pharmacology. Discussing the role that genes and race play in predicting drug response, they cite a range of currently marketed drugs that include race-specific or genetic information on their labels and note,
Some of the observed racial differences may be explained by the genetic differences listed in the labeling (e.g. warfarin and carbarnazepine). Possible mechanisms for others either have not yet been included in labeling (e.g. rosuvastatin and tacrolimus) or are as yet unknown (e.g. isosorbide dinitrate-hydralizine, which is effective in heart failure in black patients.).”
Race, genes, and drugs are bound together through the concept of the unknown. In the example of warfarin, race is being used to identify differing allele frequencies across population groups. Yet in the case of isosorbide dinitrate-hydralizine, race is simply being used to account for an otherwise unexplained observation of purported differences in drug response across racial groups. All are deemed legitimate references by the FDA. As long as an unknown factor exists, there will be a space for race, or so it seems here.
The isosorbide dinitrate-hydralizine combination referred to by Temple and Huang is BiDil, the first drug ever approved by the FDA with a race-specific indication: to treat heart failure in a black patient. Dr. Steven Nissen, the chair of the FDA panel reviewing BiDil, directly cast race as an acceptable surrogate for genetics:
[W]hat we are doing is 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 gene chip. I wish we could do it on a genetic basis. But, in the absence of that, we have some information that suggests that African Americans--we know that African Americans, self-identified, get a pretty robust response to the drug.”
Nissen related race to “some information,” but its true justification came from its purported ability to stand in for the unknown underlying genetics of drug response. Until adequate genetic technology came along, Nissen deemed race sufficient and therefore to be embraced.
Similarly, on the eve of BiDil's approval, Lawrence Lesko, director of the Office of Clinical Pharmacology at the FDA, and a point man in the FDA's efforts to integrate genomics into the drug approval process, asserted that race-based medicine could be a “stepping stone” to the higher goal of “target treatment.” As one news report put it:
Lesko and other advocates of this approach envision treatment tailored to people according to the results of genetic tests. They say that race-based medicine is just a first step toward discerning people's genetic makeup for the sake of better individual treatments.
The myriad controversies sparked by BiDil are here defused by casting race-based therapies as stepping stones. Any problems created by using race in biological context (e.g., the reification of race as genetic, the perpetuation of racial stereotypes that cast one race as more biologically “fit” than another, or concerns about the misallocation of scarce health care dollars) are minimized as temporary and hence a small price to pay for the ultimate goal of ““better individual treatments.” The advocates of race-based medicine might even concede the point that race is a crude surrogate for genetic or other biological variability and yet still use it. Thus, two senior FDA officials involved in the review and approval of BiDil declared,
Race or ethnicity is clearly a highly imperfect description of the genomic and other physiologic characteristics that cause people to differ, but it can be a useful proxy for those characteristics until the pathophysiologic bases for observed racial differences are better understood.
Similarly, the year after BiDil's approval, an article on “Race and Ethnicity in the Era of Emerging Pharmacogenomics” published in the Journal of Clinical Pharmacology asserted,
As the science of pharmacogenomics develops more accurate tools to identify the molecular underpinnings of drug response, the need for classification by race will be replaced by more accurate and specific identification of each individual person's likelihood of responding to a particular drug therapy.
Such instrumental characterizations construct race as a useful fiction--a means to an end that is to be discarded once its temporary utility has faded. Once again, in this grand march forward to pharmacogenomics, race, like Friedrich Engles' state, is supposed to “wither away.”
Some biomedical professionals take issue with this approach. One recent study by Yen-Revollo, Aumen, and McLeod specifically considered the legitimacy of using race as an interim surrogate for genetic variation:
Currently, genotyping of all patients before starting a drug regimen is impractical. Since many polymorphisms occur at varying rates in different racial groups, we investigated whether a patient's race could predict presence of drug-relevant genetic variants well enough to be used as a substitute for individual genotyping.
The study found that,
Our results clearly demonstrate that racial generalizations for treatment recommendations are not valid and are consistent with other recent publications regarding the suitability of using patient race to determine medical treatment.
The majority of genomic heterogeneity in pharmacologic genes was determined to be due to individual variation rather than racial grouping. Thus, race should not be used as a predictive substitute for individual patient genotyping.
Significantly, even as it debunked the legitimacy of using race as a surrogate, the study also considered its appeal:
Although progress has been made in identifying gene polymorphisms that influence response to several drugs, the promise of personalized medicine has not been realized as of yet because personal genotyping is cost prohibitive and most drug-genotype interactions remain unknown.Since individual causative alleles usually have distinct frequencies across the ‘Old World’ populations, there is potential utility in using race labels as a surrogate for genetic information, as a means to the ultimate goal of individualized therapy.
In this scheme, cost and the unknown are the operative factors leading toward the use of race “as a means to the ultimate goal of individualized therapy.”Theoretically, as costs come down and knowledge grows, the need for race will diminish. Yet means have a curious way of devolving into ends, and stepping stones may somehow remain underfoot long after a particular destination has been reached. So it appears to be with the unfolding case of warfarin in pharmacogenomics.
Warfarin: The Poster Child
In 2009, the FDA declared warfarin to be a “pharmacogenomic opportunity.” As one leading warfarin researcher had recently noted, “[W]arfarin is the ideal drug to test the hypothesis that pharmacogenetics can reduce drug toxicity: it is commonly prescribed, has a narrow therapeutic/toxic ratio, and is affected by common genetic polymorphisms.” Similarly, in January, 2009, after he stepped down from his post as director of the National Human Genome Research Institute under President George W. Bush and before his later elevation to director of the National Institutes of Health by President Barack Obama, Francis Collins noted that warfarin “has become a poster child for the future of pharmacogenomics.”
Warfarin, an anticoagulant, is among the most widely prescribed drugs in modern medicine. In 2004, more than 30 million prescriptions were written for the drug in the United States alone. Sales of warfarin in the U.S. were approximately $500 million in 2002. There was a 1.5-fold increase in warfarin prescriptions between 1999 and 2005, perhaps reflecting the demographic shift toward an aging population, which is typically a primary target of warfarin therapy. It is commonly prescribed to patients who are at risk of developing blood clots, such as persons with atrial fibrillation, recurrent strokes, deep venous thrombosis, pulmonary embolism, or those who have received heart valve replacements. It is difficult to calibrate the right dose for an individual patient because warfarin has a narrow therapeutic window of efficacy and a wide-range of inter-individual variability in response. Finding a correct dosage can be a delicate matter, involving the gradual upward titration of an initially low dose with regular monitoring of the coagulation rate using the “international normalized ratio” or INR (INR compares the blood's clotting ability at a given moment to a standardized measure) and adjusting the dosage until the appropriate rate of coagulation is obtained. Too much warfarin places a patient at risk of a potentially fatal hemorrhage while too little may increase a risk of blood clots and stroke. The complexity of warfarin dosing is indicated by the fact that warfarin is the second most common drug (after insulin) implicated in emergency room visits-- causing more than 43,000 emergency cases per year.
A variety of factors can influence individual response to warfarin, including dietary intake of green leafy vegetables, alcohol consumption, age, weight, and liver function. In addition, the current label lists approximately 130 specific drugs reported to interact with warfarin. For many years, researchers have also observed population-based variation in response to warfarin associated with different races or ethnicities, particularly “Asian” or “east Asian.” That is, studies have observed that on average, subjects identified as Asian may tend to have a different response to warfarin than subjects identified as belonging to different races, particularly “Caucasian.” Based on such observations, the FDA-approved label for Bristol Meyers' brand name warfarin, called Coumadin, states, “Asian patients may require lower initiation and maintenance doses of warfarin.” Race then has long been a part of the clinical conceptualization of warfarin response. It was widely assumed that a significant proportion of such variation was likely due to differing frequencies of certain alleles that affected drug response. In the absence specific information relating to such alleles, race was presumed to be a reasonable surrogate to finding the right dose of warfarin. As one 2005 study concluded, “Warfarin dose requirements vary across ethnic groups even when adjusted for confounding factors, suggesting that genetic variation contributes to interpatient variability.” As in the Albain et al. article, “other factors” here again created a space for geneticizing racial difference.
In the past decade great strides have been made toward identifying specific genetic variations that have a significant impact on individual response to warfarin. In particular, specific polymorphisms in the CYP2C9 gene and VKORC1 gene have been identified as accounting for 30%-50% of variation in individual response to warfarin. CYP2C9 affects pharmacokinetics, or what a body does to a drug. People with certain CYP2C9 alleles metabolize, or break down, warfarin more slowly than average; thus, they would need a lower dose of the medication. VKORC1, in contrast, involves pharmacodynamics, or what a drug does to a body. It affects the production of vitamin K, which is vital to blood clotting. Warfarin works, in part, by suppressing the production of vitamin K. Individuals with certain VKORC1 alleles might also need a lower dose of warfarin. Each person has two copies of each gene. Carriers of two CYP2C9 *1 alleles, known as the “wild type” or standard type, are extensive metabolizers of warfarin. The two most common relevant CYP2C9 variants are referred to as CYP2C9*2 and *3. The most common relevant VKORC1 variant is referred to as VKORC1 3673(-1639G> A). These variants have become particular targets for genetic testing.
With the proliferation of genetic data, one might think that race would cease to play a significant role in studies of warfarin response. Yet, as genetic studies have expanded, so has the use of race and related categories to assess variable frequencies of particular polymorphisms in specific population groups. Numerous studies have observed that some relevant CYP2C9 and VKORC1 alleles vary in frequency across certain ethnic or racial groups. Usually these studies employ such broad categories as “Asian,” “Caucasian” “Hispanic” or ““African-American,” but some studies are more nation specific, identifying allele frequencies and response, for example, in Swedes, Koreans, Iranians, Japanese, and Israelis.
Ironically, there seems to have been an increase in such racial, ethnic, or nation-specific studies of allele frequencies in recent years, as the significance of specific genetic variations has been more fully elaborated and characterized. In recent work on warfarin, racial categories have persisted, and even increased in use, alongside the production of specific genetic information. In studies of the impact of genetics on warfarin response, it seems to have become an unstated norm to characterize gene frequency with reference to whatever racial, ethnic, or national group on which the study happened to be performed.
For example, a report on warfarin and genetic testing issued in 2008 by the American Medical Association, the Critical Path Institute, and the Arizona Center for Education and Research on Therapeutics highlighted the significance CYP2C1 and VKORC1 but also emphasized for both that the “prevalence of gene variation differs depending on racial background,” stating, for example, that “[a]pproximately 37% of Caucasians, 14% of African-Americans and 89% of Asians carry at least one variant copy of VKORC1.” The Pharmacogneomics Knowledge Base (PharmGKB) lists studies of this VKORC1 variant finding a range of frequencies in European populations from 39% in “Swedish” to 54% in ““Spanish” with frequencies of 91% and 93% in “Chinese” and “Japanese” respectively. This begs the question of why one would care about frequencies across racial or ethnic groups when it is possible to test directly for the gene itself, regardless of race? Yet genes, it seems, somehow take on an added interest when connected with race, or perhaps vice versa. Far from withering away, race is persisting and even proliferating as genetic information increases.
Commercial Imperatives: Labels, Tests, and “Ethnic” Product Differentiation
By August, 2007, enough data on the genetics of warfarin response had been published to convince the FDA to authorize a labeling change to Coumadin to explain how people's genetics may affect their response to the drug. In a conference call announcing the change, the FDA's Lawrence Lesko noted that “this marks the first time that such pharmacogenomic information has been included in a widely used drug .... This means that personalized medicine is no longer an abstract concept, but has moved into the mainstream, where it is recognized as a factor in a product used by millions of Americans.” An article in the journal Medical Marketing & Media enthused, “The FDA rang in the era of personalized medicine with a labeling change on blood thinner warfarin cautioning that patients with either of two genetic variations might respond differently to the drug.”
News reports of the FDA mandated label change also noted some of its regulatory, legal, and commercial implications. First, Jane Woodcock, deputy commissioner and chief medical officer of the FDA, emphasized that the labeling update was “not a directive to doctors” to use genetic tests for warfarin therapy since current clinical studies do not definitively support such a recommendation. This caution was warranted given fact that no prospective clinical trials had yet been conducted comparing the outcomes of using genetic tests to guide warfarin dosing as compared to existing practices. It reflected well-established understandings of the FDA's role as regulating drugs, not medical practice, but indicated as well a second concern involving potential legal claims of malpractice liability. A report in the Wall Street Journal noted that prior to the labeling change, a medical group called the Anticoagulation Forum wrote a letter to Dr. Lesko warning that doctors might rely too heavily on the genetic tests and fail to monitor patients closely enough. Some doctors might even hold off in starting a patient on warfarin until they had the results of a test in hand. The group asked that any new label “reflect the uncertainty” so that doctors “wouldn't be held liable in court for failing to do the tests.” Third, even without a directive to test, large insurers like Aetna take account of such labeling changes when deciding whether to reimburse for a genetic test. The labeling change thus had significant implications for the growing industry of pharmacogdiagnostics. Following the labeling change, a number of companies petitioned the FDA for approval of diagnostic kits that tested for a variety of CYP2C9 and VKORC1 polymorphisms related to warfarin response.
Algorithms, Inertia, and Race as a Residual Category
FDA approval for a particular diagnostic test for warfarin response affirms the test's clinical validity. Such technical ability to test for a genetic variation is one thing, but establishing a genetic test's clinical utility (i.e., its effectiveness in real-life clinical settings) is another. For the tests to be worthwhile, a clinician needs to figure how to use the information they produce; this is where algorithms come in. A dosing algorithm for warfarin (or any drug) compiles diverse data relating to factors influencing drug response, assigns values to different factors, and applies them to compute an estimated optimal dosage for any given patient. Weight and age are common factors. For example, a dose of two 200mg tablets of ibuprofen might be recommended for an average adult, whereas the recommendation for a 2-3-year-old child weighing between 24 and 35 pounds would be one 100mg dose. In between, the recommendation for an 11-year-old weighing between 72 and 95 pounds would be 300mg.
Warfarin dosing is far more complex because of the high inter-individual variability of response and the severe consequences of over- or under-dosing. Since the FDA announced its labeling change, much attention has been devoted to developing dosing algorithms that incorporate new genetic information regarding the CYP2C9 and VKORC1 polymorphisms. Such algorithms are designed to tell doctors what they should do with the data produced by diagnostic genetic tests.
Prominent among recent genetic dosing algorithms is one constructed by a group called the International Warfarin Pharmacogenetics Consortium (IWPC), which was published in February 2009 in the New England Journal of Medicine. The IWPC is comprised of eminent biomedical researchers and institutions from around the world and is based in Stanford, California under the auspices of the PharmGKB. To develop a genetic dosing algorithm, it analyzed warfarin dose-response information from over 21 sites in nine countries for inclusion in the study. The study was not the first to show the advantage of incorporating genetic information into prescribing patterns, but it was by the largest and most inclusive to date with information on more than 5,000 patients.
The study was based on the perceived need to develop a dosing algorithm using both genetic and clinical data gathered from a “diverse and large population.” In this context, “diverse” included the concepts of race or ethnicity. From the outset, the study's use of these concepts was problematic. As stated in the “Methods” sections, “Information of race or ethnic group was reported by the patient or determined by the local investigator.” On the one hand, this reflects the simple reality of using data collected from numerous sites across the globe. On the other hand, both self-report and external ascription of race raise significant issues in a biomedical context. As bioethicists Mildred Cho and Pamala Sankar note, “Individual self-classification is not stable; for example, one US study found that one-third of people change their own self-identified race or ethnicity in two consecutive years.” Complicating matters still further, a study by Condit et al. found people often have very incomplete knowledge of the biological ancestry. Of a sample of 224 subjects interviewed for a study on attitudes toward race-based pharmacogenomics (the tailoring of drugs to genetic profiles), Condit et al. found that 39.6% did not know all four of their biological grandparents. In such situations, self-declared race may fail to capture significant variation in biological ancestry.
Such problems are only compounded in situations where race is externally ascribed by a local investigator. Robert Hahn, for example, has conducted a series of studies in the field of public health showing how external ascriptions of race are often unreliable and also may change over time. Despite these issues, the one sentence quoted above states the entirety of the study's consideration of how it used race. This contrasts starkly with the detailed elaboration of “Genotype Quality Controls” and ““Statistical Analysis” in the remaining two pages of the same “Methods” section.
To compose its algorithm, the IWPC researchers calculated warfarin dosing three ways: (1) based on standard clinical data; (2) based on clinical data and genetic variations; and (3) using fixed daily doses. In developing dosing algorithms for each patient, researchers genotyped three VKORCl alleles and six CYP2C9 alleles. They also collected clinical data on such characteristics as age, height, weight, use of certain other drugs, and, of course, race. Then, they compared how closely their computational predictions matched the actual, clinically derived stable warfarin dosage for each patient. This was, therefore, a retrospective study and did not measure prospectively whether the use of a genetic dosing algorithm actually reduced adverse events. The study concluded,
The use of a pharmacogenetic algorithm for estimating the appropriate initial dose of warfarin produces recommendations that are significantly closer to the required stable therapeutic dose than those derived from a clinical algorithm or a fixed-dose approach. The greatest benefits were observed in the 46.2% of the population that required 21 mg or less of warfarin per week or 49 mg or more per week for therapeutic anticoagulation.
Thus, while successful as proof of concept showing the possible utility of the genetic dosing algorithm, the “greatest benefits” accrued only to the roughly half of the patient population who were outliers at the high and low ends of warfarin doing.
As for race, at one point the study presented a figure comparing “the predicted doses according to representative clinical or demographic characteristics, genotype combinations, race, and use or nonuse of amiodarone (an important interacting drug).”
It concluded that the data “suggest that most of the racial differences in dose requirements are explained by genotype.” The text accompanying the figure itself states that, “racial differences in the estimated does become insignificant when genetic information is added to the model.” On page 15 of the supplemental materials, the study calculated the percentage of variance in dose explained by race (R2). In the study's own terms, race accounted for 14.2% of variation when it was the only thing in the model. This would comport with the notion that race serves as a potentially useful surrogate when specific genetic variables are unknown. When pharmacogenetic data was added to the model, however, the contribution of race went from 14.2% down to 0.3%, i.e., almost nothing.
It would seem then that with this algorithm, warfarin dosing had reached the promised land of truly individualized pharmacogenomic practice. The study case specific genetic information was cast as rendering race “insignificant,” just as advocates of using race as an interim proxy said it would. Yet, when we turn to the supplemental material provided with the study and examine the actual dosing algorithm used (and recommended for further use), there we find race, still a prominent factor to be used by every doctor in every dosing calculating apparently regardless of the fact that it is “insignificant”: [Table 1 Ommitted]
* * * The output of this algorithm must be squared to compute weekly dose in mg.
All referencesto VKORC1 refer to genotypefor rs9923231. The accompanying “legend for the use of algorithm” further specifies: “Asian Race = 1 if self-reported race is Asian, otherwise zero; Black/African American = 1 if self-reported race is Black or African American, otherwise zero; Missing or Mixed race = 1 if self-reported race is unspecified or mixed, otherwise zero.”
Most immediately striking here is the straightforward use of whiteness as an unmarked norm in biomedical research. The IWPC algorithm is thus typical, and yet stunning in that it is appearing at the end of a decade of heated discussion concerning the proper use of race in biomedical contexts-- discussion that directly critiqued the use of “white” as the unmarked standard from which all other races are cast as deviating. The algorithm only makes a person present as a racialized subject if he or she is not white. Here white people do not explicitly possess race--that is, their race is tacit and does not formally come into play in calculating dosage. They are the norm against which the race of other subjects is made to matter. Thus, “black” and ““Asian” dosages are calculated as deviations from the unstated white norm.
It is also curious to note that the algorithm lumps together “mixed race” and ““missing race.” Having mixed race is here made the equivalent of absent race. In both cases a sort of statistical guess is being made to encompass the unknown. Through this association, the algorithm renders mixed race as mysterious and uncertain, a category without clear boundaries that is treated as the equivalent of absence in order to contain any challenge it might pose to the model. Yet, how can this characterization of mixed race as a distinct category deal with estimates that anywhere between 30-70% of self-identified African Americans have some white relatives in their ancestral history or that a significant proportion of white-identified people have some multi-racial background? Many of these people would likely register as either “black” or “white.” Thus, a self-identified black person and self-identified white person each with “mixed ancestry” would likely register respectively as black, or white, even though both could also register under the same “mixed race” category. Moreover, once a value is assigned, the algorithm multiplies it by different amounts depending again on racial identification. Thus, a person of mixed African and European ancestry could be counted three different ways depending on whether they self-identified as “White,” “Black,” or “Mixed.” This would be the case for everybody's mixed-race person of the moment, President Barack Obama. Alternatively, three siblings each with the same proportion of “mixed” ancestry could be counted differently by this algorithm, again depending on how they self-identified. Perhaps more significantly, this model assumes some sort of “pure” notion of races as biologically bounded and distinct, perhaps capable of “mixture” but with such mixtures ultimately separable into preexisting purportedly pure kinds, thus reinforcing genetically essentialist conceptions of races as biologically distinct entities.
A similar dynamic is evident in a separate dosing algorithm published a year earlier by a team or researchers led by Brian Gage, of Vanderbilt University, a prominent warfarin researcher and member of the IWPC (as were several of the other researchers on the study). Declaring warfarin to be the “ideal drug to test the hypothesis that pharmacogenetics can reduce drug toxicity,” this study contained a single race variable, specified as ““African-American race” which ultimately accounted for 0.4% of variation in response to warfarin--almost the exact same (negligible) amount as in the IWPC algorithm. With these dosing algorithms, we have a formal institutionalization of the imperative to render ambiguous racial identifiers as genetically bounded and fixed. In this particular promised land of genetically guided dosing, race did not wither away; to the contrary it persisted--prominently and structurally.
In both the Gage and the IWPC algorithm, race appears to be operating as a sort of residual category that has persisted through inertia. Race had long been long a part of considering how individuals respond to warfarin. It was generally understood to be a surrogate for underlying genetic variation and therefore included as a variable in numerous pre-genomic studies. The use of race thus became a standardized norm in warfarin research. Once instantiated as normative practice, race simply appears to have persisted, almost reflexively, even in the face of data showing its minimal impact on dose response variation. The inertial power of race was so strong that it persisted in the IWPC algorithm even alongside explicit assertions of its insignificance. To the extent that race captured any substantive information (that is, to the extent that the 0.3-0.4% variation could be deemed significant), race must be understood as a sort of catch-all category encompassing a residuum variation for which the researchers had no specific explanation. This brings us back to observation by Yen-Revello, Aumen, and McCleod that race remains an appealing surrogate where “drug-genotype interactions remain unknown.” Apparently, however, race remains appealing even after drug-genotype interactions are discovered. This is because genotype generally will never explain all variation in drug response. There will always be some aspects of inter-individual variation in drug response that remain “unknown.” As long as the inertial force of race maintains it as a normative category in biomedical research, it may be possible for researchers to associate race with whatever residuum of unexplained variation they find in their studies.
The case of warfarin is emblematic of a larger dynamic coming to characterize the use of race in biomedical research, development, and marketing. Not only is general use of racial and ethnic categories in biotechnology patents increasing, subsequent race-specific trials, marketing campaigns, and clinical education are also on the rise. The case of warfarin brings to light a common dynamic underlying driving persistence of racial profiling in biomedicine, even as specific genes are being identified. The experience of warfarin calls into question the entire rationale for using race in the “meantime” between current reality and the promise future of truly individual genomic medicine. It also illustrates the inertial power of race to remain prominent in a conceptual system of biomedical analysis once introduced, especially when buttressed by commercial imperatives. Finally, it shows how race is being constructed as a residual category to explain any “unknown” aspects of drug response, creating a new space for the persistence of race in biomedicine.
. Jonathan Kahn, Ph.D., J.D.,is a Professor of Law at Hamline University School of Law.