One of the promises of the human genome project is individualised pharmacological therapy. An individual’s genetic type would be determined, and the resulting diseases which the person was susceptible to would be determined. In addition, the drugs to which an individual would respond well would be enumerated, and the drugs which were more likely to be toxic could be established. At present we are just beginning to determine those associations which would help us tailor an individual’s pharmacological therapy so as to maximise efficacy and minimise toxicity. In this report we review the considerations involved in determining the pharmacogenetic profile of any given drug and to review the current level of understanding of the association between genetic polymorphisms and drug toxicity/efficacy in rheumatology.

Establishing an association between genetic types and toxicity is perhaps an easier task to accomplish than understanding the association between a given genetic type and drug efficacy. Drug toxicities are generally discrete events and a specific toxicity can often be traced to the agent in question. Thus, it is fairly easy to study a given population and define those with a specific drug toxicity. Nonetheless, a sufficiently large population must be studied to establish an association. Although this restates the obvious, most drugs for chronic conditions such as rheumatoid arthritis (RA) are toxic to only a modest proportion of the patients taking them otherwise they would have long ago been pulled off the market. To establish a relative risk association for a toxicity that occurs in only 5–10% of patients taking a given drug might require hundreds or even thousands of patients if the observed increase in risk of toxicity is modest.

There are many reasons why any given drug may fail in any given individual. Clearly, genetically associated resistance may account for failure of a given …