Dear Editor,

Suarez-Kurtz focuses on Brazil and clearly demonstrates the extensive genetic admixture there. He quotes two areas in our paper where we do address admixture, but nevertheless contends that we paid little attention to it. A more careful reading of our paper1 would have revealed several other examples of germane content. In various sections we highlight the need for careful definition of genetic differences; acknowledge genetic variation within subpopulations; specifically distinguish countries or groups that are likely to have, like Brazil, large degrees of allelic variation (African-Americans, Caribbeans, Panamanians) and those that are less genetically diverse (sub-Saharan Africans, Inuits, Finns); call for caution in interpreting and using the results of studies; and underline the need to understand each group chosen for study because “the races that comprise the human species are far more heterogeneous than previously thought”. In general we call for better scientific studies, data and interpretation, and for avoiding conflating scientific data and social constructs.

Interestingly, Suarez-Kurtz agrees with us that some markers occur with greater frequency in some communities of common geographical ancestry than others. He mentions the L3d haplotype that is “characteristic of West African populations”, and several others.

Importantly, Suarez-Kurtz predicates his arguments mainly on SNPs. In our paper we highlight other sources of variation, such as gene duplications and deletions, mutations in regulatory genes and large-scale copy-number variations. Stefansson et al., for example, have recently described a polymorphic chromosomal inversion that is much more common in some populations than others, and that interestingly seems to be under active selection pressure2.

We should also note that admixture can be a valuable tool in the study of complex diseases. The potential was recognized more than 50 years ago3. In our paper we highlighted the work of the International Multiple Sclerosis Genetics Consortium, which is attempting to understand the genetic contribution to MS susceptibility through admixture mapping. A pioneering study based on admixture mapping, published at about the same time as ours, has revealed two chromosomal regions in African-Americans that affect susceptibility to hypertension4.

On 16 June 2005 the USFDA's (US Food and Drug Administration's) Cardiovascular and Renal Drugs Advisory Committee meets to review BiDil5, and there is much speculation that it will be approved for use specifically in African-Americans. This is undoubtedly controversial, but the published evidence seems compelling that its use results in significantly better outcomes in African-Americans that have heart failure than when a placebo is used6. In our paper we described a situation in which harmful or ineffective drugs that were tested on mixed populations in the West might be resuscitated for use in developing countries. A recent example is Iressa (gefitinib), which was found to be ineffective against lung cancer and the USFDA is now considering pulling it out of the US market. However, it seems to be more effective in Asian patients and AstraZeneca is altering its marketing strategy to focus on selling it in Asia7. More data is needed to confirm these findings.

The day might come when boutique personalized medicine (pharmacogenetics based on the testing of individuals) will be available for all, but that might take a very long time, especially for people in developing countries. We argue that we should be exploring what we can do in the meantime to make pharmacogenetics relevant to the developing world, and thinking about resuscitating drugs such as BiDil, which would have been discarded had it not been for the way of thinking that we are advocating. BiDil might even be useful for other populations in the future — a hypothesis that could not be tested if it wasn't around to study.

Finally, a word of explanation about our use of “they” (developing countries pharma companies) and “our” (knowledge) — which Suarez-Kurtz has misunderstood. In a paper advocating the consideration of genomics as a global public good, we have argued, as have others, that knowledge is the ultimate global public good8. So we used “our” in the paper to highlight humankind's knowledge, not that of two professors at the University of Toronto.

Yours sincerely,