Medicine, Genetics and Race: The Case of Cardiovascular Diseases

National concern over health disparities in the USA has been expressed for several decades through several Government reports [1]. In response to those reports, the Office of Research on Minority Health was created in 1990 by the Acting Director of the NIH. Since then, a series of initiatives to include minorities in clinical research and increase training of minority health professionals has increasingly focused attention on racial and ethnic disparities in health in the USA. In 1998, the President‘s Office set the goal of eliminating disparities by the year 2010 and, 2 years later, Congress enacted The Minority Health and Health Disparities Research and Education Act of 2000. This legislation resulted in the establishment of the National Center of Minority Health and Health Disparities (NCMHD). As a result of these initiatives, reports documenting racial and ethnic differences in incidence, mortality and severity of disease appear on a regular basis in the biomedical literature [1,2]

During this period, other NIH parallel developments were taking place regarding another milestone project: the sequencing of the human genome. In 1989, The National Human Genome Research Institute (NHGRI) was originally established as the National Center for Human Genome Research (NCHGR). Its role was to lead the NIH efforts to the Human Genome Project. This mission was completed in 2004 with the publication of the full and complete sequence of the human genome [3]. This accomplishment facilitated the task of other related programs, including the Genetic Variation Program and the International HapMap Project that analyzed the DNA from 269 people from four large populations (Yoruba tribesmen in Ibadan, Nigeria; Japanese in Tokyo; Han Chinese in Beijing; and Utah residents with ancestry from northern and western Europe) [4]. These four populations were selected to include people with ancestry from widely separate geographic regions. The findings from the HapMap project, as well as other parallel efforts, revealed that the human populations represented by those samples share the common haplotype patterns and that the overall organization of genetic variation was very similar [5]. Therefore, the current information shows a greater variation within populations than between populations [6].

This knowledge has brought the tasks of the NCMHD and the NCHGR to a common ground. Many symposiums and papers have attempted to properly define race, ethnicity and ancestry and describe how they relate to the new knowledge extracted from the ‘genomes‘ [7–13]. Moreover, ample discussion exists regarding the relative relevance of social and environmental factors versus biological/genetic factors in determining those health disparities that should be eliminated in the coming years [2]. These discussions bring us again to the long-existing dilemma of nature versus nurture as determinants of health and disease [14,15]. The goal of this editorial is not to recreate again the contents of the excellent and comprehensive discussions that have been published in recent years, but to focus on a few specific examples to underscore the importance in the current race/ethnicity/ancestry/genetic polemic of allying rather than antagonizing nature and nurture. In this regard, cardiovascular diseases (CVDs) can be used as the paradigm for the study of ancestry-related differences in CVD incidence and mortality as well as the aforementioned interlocution between nature and nurture. Two different but complementary aspects will be discussed. First, how ethnicity may affect disease risk and, second, how ethnicity may affect response to drug therapy.

CVD risk & ethnicity

Ethnic differences in atherosclerosis, CVD and lipid metabolism have been thoroughly reviewed by Kuller [16] and Forouhi and Sattar [17]. Whereas our current understanding of CVD is derived primarily from white subjects, the reality is that subjects from certain other ethnic groups experience a disproportionately greater burden of CVD.

The speed at which the changes in disease incidence have taken place supports the notion that environmental factors may outweigh genetic factors as drivers of this epidemic. This concept has been supported by the interpretation of the findings from a number of natural experiments involving geographical migration from one region to another with different lifestyles and rates of disease, but also by studies examining the fast in situ behavioral ‘migration‘ that is taking place in many parts of the world from a traditional to a more ‘Westernized‘ lifestyle. Probably the most cited study in support of the importance of environment over ethnicity (genetic factors) is the Ni-Hon-San Study, which compared CVD rates and risk factors in Japanese men living in Japan, Hawaii (Honolulu HeartProgram [HHP]) and California. This study showed that coronary heart disease (CHD) and stroke mortality rates in Hawaii were intermediate between rates in Japan and California. Gradients in CVD risk factors were similar to the gradients in disease rates [18]. This and other observations have been used by some to emphasize the importance of the environment as a disease risk factor but, in fact, what these studies reveal is the presence of a very complex network of genetic and lifestyle interactions. Moreover, it is important to keep in mind that subjects migrating in space (from one region or another) or in time (moving through the westernization process within their own geographical boundaries) may not be representative of the whole population. Depending on the specific drivers determining the transition, migrants may be better educated, wealthier and healthier than native or non-migrant populations. Alternatively, the opposite situation may exist by which people migrate to another place in search of better medical care that is not available in their points of origin. Either way, their rates of disease and risk factor distribution may be different to the non-migrants [19]. This point had already been expressed by the HHP investigators who compared trends in incidence rates for CHD, stroke and cause-specific mortality between 1966 and 1984 in studies of Japanese men in the Pacific region (8006 participants and 3130 nonparticipants) [20]. CHD and stroke rates declined by approximately 40% for the total HHP cohort; however, the decline was much larger (over 60%) in the nonparticipants. There was also a much greater decline in total mortality and cancer mortality rates in the nonparticipants. The finding that nonparticipants in epidemiological studies can show differences in incidence trends to participants suggests that caution should be used in interpreting trends limited only to participants.

Conversely, the relevance of genetic factors, and more specifically the interaction between genetic and environmental factors, can be seen by another interesting natural experiment occurring in Asia, and more specifically, Singapore, which has experienced a dramatic modernization in lifestyle during the last few decades. It has been suggested that Singapore represents a population laboratory within which to examine the effect of urbanization on the health of various ethnic groups and the data show that ethnic differences in CHD risk are over and above the effects of westernization and/or urbanization [21]. For this reason, the following discussion will focus on research from Singapore [22].

Singapore is a small, fully urbanized island nation situated in Southeast Asia. According to the 2000 census, the population stood at 4.02 million, with a largely third- and fourth-generation immigrant population comprising 76.8% Chinese, 13.9% Malays, 7.9% Asian–Indians and 1.4% other ethnic groups [101]. All three major ethnic groups living in Singapore experienced urbanization simultaneously and, although some minor ethnic disparities in socioeconomic status exist, they are negligible when compared with most other countries and the standard of living is relatively high and comparable between ethnic groups. Similarly to other developed countries, CVD accounts for the largest proportion of deaths. CHD mortality, although initially low, rose to levels that were similar to those experienced in the USA more than a decade ago [23]. The small size of the country, coupled with complete urbanization, high standards of living for all ethnic groups and readily accessible healthcare, reduce some of the confounding factors seen in many previous studies.

Despite the similarities in living conditions, the impact of rapid urbanization appears to have had a differential effect on CHD risk in each ethnic group. Several studies have now shown that Asian–Indians in Singapore have an approximately threefold increased risk of CHD compared with Chinese and Malays exhibit an intermediate level of risk >[24–26]. The ethnic differences in risk for CHD are accompanied by ethnic differences in serum lipid concentrations. Total cholesterol was found to be highest in Malays and may contribute to the higher CHD risk in Malays compared with Chinese [27]. However, despite rates of acute myocardial infarction that were almost twice that seen in the Malays, the low-density lipoprotein cholesterol (LDL-C) levels were similar between Asian–Indians and Malays. Instead, several features of the metabolic syndrome are more frequently seen in Asian–Indians [28,29]. These include greater obesity and the lowest high-density lipoprotein cholesterol (HDL-C) concentration. These ethnic differences in HDL-C, both alone and in combination with hypertriglyceridemia, have been demonstrated in several separate studies [27–29]. They have also been shown to be independent of obesity and the presence of glucose intolerance [28,29]. HDL-C concentration is the one lipid parameter that differentiated Asian–Indians from Malays and may contribute to the higher risk of CHD seen in Asian–Indians.

However, the most striking clinical differences among these three ethnic groups were that of diabetes mellitus. A recent study aimed to investigate whether the risk of ischemic heart disease (IHD) associated with diabetes mellitus differs between ethnic groups [30]. The investigators used registry linkage to identify IHD events in 5707 Chinese, Malay and Asian–Indian participants from three cross-sectional studies conducted in Singapore between 1984 and 1995. These investigators assessed the interaction between diabetes mellitus and ethnicity in relation to the risk of IHD events using Cox proportional hazards regression. As previously known, diabetes mellitus was more common in Asian–Indians. Furthermore, diabetes mellitus was associated with a greater risk of IHD in Asian–Indians. The hazard ratio when comparing diabetes mellitus with nondiabetes mellitus was 6.41 in Asian–Indians and 3.07 in Chinese with a very significant disease by ethnicity interaction. Differences in the levels of established IHD risk factors among diabetics from the three ethnic groups did not appear to explain the differences in IHD risk. These findings demonstrate that Asian–Indians are more susceptible to the development of diabetes mellitus than Chinese and Malays. When Asian–Indians do develop diabetes mellitus, the risk of IHD is higher than for Chinese and Malays. Consequently, the prevention of diabetes mellitus among this ethnic group is particularly important for the prevention of IHD in Asia, especially given the size of the population at risk. Elucidation of the reasons for these ethnic differences may help us understand the pathogenesis of IHD in those with diabetes mellitus and to palliate the problem that Asian–Indians and other populations have following their migration to other sites with more Westernized lifestyles [31].

Ethnicity & drug response

Studies indicate that, overall, African–Americans are less likely to achieve control of hyperlipidemia compared with white people and respond slightly less to statins than white subjects in clinical studies [32,33]. Additional support for this notion comes from a population-based study that has examined the effect of race on achieving target LDL-C goals among treated individuals [34]. This study involved 16,052 new statin users (32.5% African–American) and found that African–American patients initiating statin therapy are less likely to achieve ideal LDL-C levels, even after controlling for adherence differences and other factors, suggesting that African–Americans may require different pharmacological management. It is enticing to hypothesize that the small (3–6%) but consistent ethnic differences in LDL-C response have some genetic basis. In fact, it is possible that apolipoprotein E (APOE) could be a player in these differences. We and others have shown that, on average, carriers of the APOE4 allele respond to statins with less of a reduction in LDL-C levels compared with subjects who did not carry the APOE4 allele [35,36]. It should be noted that the APOE4 allele is more common in black (∼22%) than in white people (∼12%) [37]. Therefore, if this hypothesis is shown to be true, perceived race would be irrelevant for determining differential response to statins and the effect could be due to an allele that is shared among ethnic groups but more frequent in black people.

Many of the studies showing ethnic-related differences are based on ecological observations or subjected to the classical confounders (i.e., socio-economic status and education level among others). We could control the effects of some of those confounders by returning to examine the information that in this regard is emerging from Singapore. In this country, evidence-based consensus guidelines have recommended consideration of increasingly stringent cholesterol-lowering goals, yet most patients do not meet these targets. Specifically, those with higher baseline LDL-C levels or Malaysian ethnicity were less likely to achieve LDL-C goals, which may suggest some ethnic-related differences in response [38]. However, the observational design of the study cannot rule out potential cultural confounders. A more rigorous and mechanistic approach was taken by Lee and colleagues who focused on the ethnic and genetic differences of rosuvastatin response [39]. The background for the study is based on the observation that systemic exposure to rosuvastatin appeared to be approximately twofold higher in Japanese subjects living in Japan compared with white subjects in Western Europe or the USA [40,41]. The organic anion transporting polypeptide 1B1 contributes to the hepatic uptake of rosuvastatin. Polymorphisms (i.e., T521>C) in the solute carrier organic anion transporter 1B1 (SLCO1B1) gene can lead to reduced transport function in vitro. Lee and colleagues prospectively examined the pharmacokinetics of rosuvastatin in white and Asian individuals living in Singapore to determine whether the pharmacokinetic differences between Japanese and white subjects extended to other Asian ethnic groups and to determine whether polymorphisms in the SLCO1B1 gene contribute to any pharmacokinetic differences observed [39]. Whereas SLCO1B1 genotypes did not account for the observed pharmacokinetic differences between Asians and white subjects, it became evident that plasma exposure to rosuvastatin and its metabolites was significantly higher in Chinese, Malay and Asian–Indian subjects compared with white subjects living in the same environment. Collectively, these results strongly suggest a genetic basis for interindividual variability in rosuvastatin pharmacokinetics for which relevant mechanisms have not been determined. Moreover, it is unclear whether the ethnic differences in rosuvastatin disposition also have an impact on lipid-lowering effects or susceptibility to known adverse effects of statins. We can expect that, through targeted clinical and basic research, combining the information from race and ethnicity with genetic variants, a better prediction of interindividual variability in drug response will improve clinical outcomes and optimize drug development.

There is evidence supporting the theory that ancestry is a significant determinant of both disease risk and therapeutic responses. This may be due to the different frequencies of alleles associated with risk, therapeutic response or both. However, most of these alleles are found across ethnic groups and the effects associated may not be exclusive to any given group. As previously stated by the Race, Ethnicity and Genetics Working Group: “Efforts to move past the use of racial and ethnic categories in genetics research often will require consideration of a very broad range of additional variables including racism and discrimination, socioeconomic status, social class, personal or family wealth, environmental exposures, insurance status, age, diet and nutrition, health beliefs and practices, educational level, language spoken, religion, tribal affiliation, country of birth, parents‘ country of birth, length of time in the country of residence, and place of residence along with genetic variation” [42]. Such studies will be able to provide more definitive information regarding personalized medicine and the possibility of eliminating health disparities well beyond the current approaches based on social definitions and classifications.

Acknowledgements

Supported by the National Institutes of Health, National Institute on Aging, grant number 5P01AG023394–02 and NIH/NHLBI grant number HL54776 and contracts 53-K06–5–10 and 58–1950–9–001 from the US Department of Agriculture Research Service.