The Ramazzini Institute
for Occupational and Environmental Health Research   

Who Are We?

HELEN PEARSON ©Nature News Service

Race should not influence drug prescriptions, warn geneticists. Genetic differences between individuals give a better indication of who will respond well to a medicine, a new study shows.
     Clinical trials for new drugs often note the participants' ethnic origin in an effort to identify groups who may react badly to a drug or be immune to its effects.
     Classifying people according to small differences in their DNA sequence creates groups that are better associated with drug response, David Goldstein of University College London and his team propose. "Ethnicity is not a reliable guide," says Goldstein.
      Goldstein's team compared 40 variable genetic regions, unrelated to drug reaction, in nearly 400 people from all over the world. They divided the people into four groups according to differences between these regions. These groupings predicted drug response better than did ethnic labels such as black, Caucasian and Asian in four out of six cases.
     Geneticists have known this for a while. "It's no surprise that skin pigment is a lousy predictor of physiology," says Howard McLeod of Washington University School of Medicine in St Louis, Missouri. This study is the first to prove it.
     Small variations in genes that control, say, the proteins that break down a drug influence how active the drug will be in the body and hence an individual's reaction to it.
     "Understanding molecular mechanisms is more important than crude skin colour," explains McLeod.
     In some cases, these differences do correlate with ethnic background. Between 5% and 10% of Europeans carry a form of the gene CYP2D6 that affects the actions of more than 40 drugs. For example, it renders the painkiller codeine useless, as such people cannot process it into its active component, morphine. Less than 1% of Japanese people carry the same form of the gene. But this ethnicity relationship cannot be extrapolated to all drugs.
     Moreover, any ethnic group will still contain individuals who respond to a drug and those who do not. "There is so much variation within a population, you cannot use [ethnicity] alone," says Wendell Weber, who studies population genetics at the University of Michigan in Ann Arbor.
     Ultimately, many hope that the future will see personalized prescriptions based on an individual's DNA; pharmacogenetic medicine. At present, however, our knowledge of which genes are important in disease and drug response is scant. "We're trying to get to the relevant genes," says Claibourne Stephens of Genaissance Pharmaceuticals in New Haven, Connecticut.
     One way to start finding the genes involved -a route being pursued by companies including Genaissance -uses common single-letter DNA changes. These 'single nucleotide polymorphisms' (SNPs) are scattered throughout the genome. When new drugs enter clinical trials, companies plan to analyse their participants' SNPs.
     Sets of SNPs (called haplotypes) that are commonly associated with a certain drug response will then be identified. "Clinicians can begin to use this for predicting whether you can administer a drug safely," says Weber. Recent results suggest that particular haplotypes do correlate with drug response, independently of ethnic origin. http://www.nature.com/nsu/ Wilson, et al. Population genetic structure of variable drug response. Nature Genetics advance online publication (2001).

Future Profiles

     Genetic Profiles is an attempt to create a catalog of "ecological" profiles. Besides genomic information, they will include social, economic, demographic, psychological, biomedical, and environmental data. Each of the affected populations face special problems, but share common research, legislative, employment, insurance, and health care needs. There may also be unexplored environmental questions that need to be highlighted, not only in terms of causation, but also in terms of protection as a population especially vulnerable to contaminated environments. Some members of each population are likely to have “membership” in other populations, creating an overlap of biomedical and environmental research and protection needs.