Correspondence
Race and Genomics
N Engl J Med 2003; 348:2581-2582June 19, 2003
- Article
To the Editor:
The article on race and genomics by Cooper et al. (March 20 issue)1 claims that multilocus interactions are “both mathematically and biologically implausible” as risk factors for complex disease. This statement stands in strong opposition to considerable empirical research suggesting that complex phenotypes often exhibit epistasis in diseases such as hypertension, other cardiovascular diseases, Alzheimer's disease, and obesity.2,3 The data strongly support the presence of complex interactions either in the presence of detectable marginal effects or in their absence.4 The many reports cited in recent reviews3,4 and an understanding of the complexity of biologic processes contradict the conclusions of Cooper et al.
Similarly, there is a long-standing theoretical literature suggesting the importance of the genetic context (epistasis) in the development of phenotypes, dating back at least to Wright's 1932 article on mutation and evolution.5 Wright's article and subsequent literature clearly demonstrate the mathematical plausibility of multilocus effects on complex phenotypes such as common diseases. The data and theory contradict the conclusions of Cooper et al. and support the need to pay greater attention to the role of gene–gene interaction in the etiology of complex disease.
5 ReferencesScott M. Williams, Ph.D.
Vanderbilt University Medical Center, Nashville, TN 37232
smwilliams@phg.mc. vanderbilt. edu Alan R. Templeton, Ph.D.
Washington University, St. Louis, MO 631301
Cooper RS ,Kaufman JS ,Ward R . Race and genomics. N Engl J Med 2003;348:1166-1170
Full Text | Web of Science | Medline2
Williams SM ,Addy JH ,Phillips JA III , et al. Combinations of variations in multiple genes are associated with hypertension. Hypertension 2000;36:2-6
Web of Science | Medline3
Templeton AR . Uses of evolutionary theory in the Human Genome Project. Annu Rev Ecol Syst 1999;30:23-49
CrossRef4
Templeton AR. Epistasis and complex traits. In: Wolf JB, Brodie ED III, Wade MJ, eds. Epistasis and the evolutionary process. Oxford, England: Oxford University Press, 2000:41-57.
5
Wright S. The roles of mutation, inbreeding, crossbreeding, and selection in evolution. In: Jones DF, ed. Proceedings of the Sixth International Congress on Genetics. Vol. 1. Brooklyn, N.Y.: Brooklyn Botanic Garden, 1932:356-66.
To the Editor:
It is odd to talk about race as a genetic or biologic concept when racial categories are, by definition, “social-political constructs and should not be interpreted as being scientific or anthropological in nature.”1 Race is not a constant — a person who is “black” in the United States may be “white” in Brazil and “colored” in South Africa. In the United States, we will continue to collect racial data in health studies for many reasons, including the legal mandate to do so.2
Even if race and ethnic background are not biologic concepts, surely that does not imply that biomedical researchers are not interested in them. Social epidemiology3 — including the study of important nongenetic predictors of health, such as race or socioeconomic status — has been growing in parallel with genetic epidemiology. Like González Burchard et al. (March 20 issue),4 biomedical researchers should acknowledge the sociopolitical context of race and continue to study racial differences in populations. The fact that a variable is not genetic does not mean that it is not important.
4 ReferencesKaren C. Swallen, M.P.H., Ph.D.
University of Wisconsin, Madison, WI 53706
kswallen@ssc.wisc. edu 1
Revisions to the standards for the classification of federal data on race and ethnicity. Washington, D.C.: Office of Management and Budget, 1997. (Accessed May 30, 2003, at http://www.whitehouse.gov/omb/fedreg/ombdir15.html.)
2
NIH policy on reporting race and ethnicity data: subjects in clinical research. Notice no. NOT-OD-01-053. Bethesda, Md.: National Institutes of Health, 2001. (Accessed May 30, 2003, at http://grants1.nih.gov/grants/guide/notice-files/NOT-OD-01-053.html.)
3
Kawachi I . Social epidemiology. Soc Sci Med 2002;54:1739-1741
CrossRef | Web of Science | Medline4
Gonzalez Burchard E ,Ziv E ,Coyle N , et al. The importance of race and ethnic background in biomedical research and clinical practice. N Engl J Med 2003;348:1170-1175
Full Text | Web of Science | Medline
Author/Editor Response
Dr. Swallen offers an unexpected response to our article, since we appear to be in substantial agreement on the issues she raises. Our purpose in the article was to argue that continental race is not a useful genetic category. We pointed out, however, that as a surveillance tool, “race,” as defined by the U.S. government, does identify population groups with very different rates of illness and is therefore important in the practice of public health. We also asserted that variation in these rates “can be understood as the result of differential exposure to environmental causes.” The study of those environmental causes has always been the core function of epidemiology, and the surveillance of disease patterns and causal exposure according to demographic groupings is thus a useful endeavor, as we have argued in more detail elsewhere.1-3 Our opposition to the view of González Burchard et al., for example, relates to the acceptance of race as a useful proxy for genetic factors in the prediction of susceptibility to disease. We believe, instead, that racial differentials in the rates of common diseases in the United States reflect the effects of discrimination and exploitation that medicine and public health officials are loath to acknowledge.
Williams and Templeton focus on our treatment of the issue of gene–gene interactions. We raised that issue in a discussion of the strength of potential genetic explanations for racial differences in the rates of common disease. Two potential alternative explanations were put forward. Under the “common disease, common variant” hypothesis, marked differences in the frequencies of a few genes might explain racial differentials. However, common variants are old and therefore tend to be shared by all human populations. Under the “common disease, many rare variants” hypothesis, one would need to invoke interactions among multiple loci. It is this logic — that racial differentials could be explained by a specific set of interactions that varied according to race — that we contend is “mathematically and biologically implausible.” We do not question the existence of gene–gene interactions in general. As Williams and Templeton point out, these represent a well-established biologic phenomenon. Rather, we question whether this phenomenon is a plausible explanation for differences between racial groups in the rates of disease.
3 ReferencesRichard S. Cooper, M.D.
Loyola Medical School, Chicago, IL 60153
rcooper@lumc.edu Jay S. Kaufman, Ph.D.
University of North Carolina School of Public Health, Chapel Hill, NC 27599-74351
Cooper RS . Health and the social status of blacks in the United States. Ann Epidemiol 1993;3:137-144
CrossRef | Medline2
Cooper RS . A case study in the use of race and ethnicity in public health surveillance. Public Health Rep 1994;109:46-52
Web of Science | Medline3
Kaufman JS ,Cooper RS . Considerations for use of racial/ethnic classification in etiologic research. Am J Epidemiol 2001;154:291-298
CrossRef | Web of Science | Medline
- Citing Articles (2)
Citing Articles
1
Hossein Bahrami, David A. Bluemke, Richard Kronmal, Alain G. Bertoni, Donald M. Lloyd-Jones, Eyal Shahar, Moyses Szklo, João A.C. Lima. (2008) Novel Metabolic Risk Factors for Incident Heart Failure and Their Relationship With Obesity. Journal of the American College of Cardiology 51:18, 1775-1783
CrossRef2
A. R. Miserez, H. Scharnagl, P. Y. Muller, R. Mirsaidi, H. B. Stahelin, A. Monsch, W. Marz, M. M. Hoffmann. (2003) Apolipoprotein E3Basel: new insights into a highly conserved protein region. European Journal of Clinical Investigation 33:8, 677-685
CrossRef
