I n your doctor’s office, sometime in the future, you spit a sample of mouthwash into a vial. The follow- ing day, the doctor advises you not to take the drug that he had considered prescribing to treat your condi- tion, because a genetic test for certain single nucleotide polymorphisms (SNPs) predicts that you could suffer a severe adverse reaction to it. By contrast, the same test indicates that you are expected to show an excellent response to a different medication with little chance of side effects, and you are given the appropriate prescrip- tion. This is the promise of pharmacogenetics – the optimization of drug therapy based on the individual patient’s genetic profile. How far into the future is such a day? How are recent and rapid technological devel- opments in this long-established field of research changing drug development paradigms in the phar- maceutical industry and the practice of clinical pharma- cology, and what has happened recently to accelerate the pace of progress in pharmacogenetics? A historical perspective When looking to the future, it is instructive to gain context by considering the past. The concept that inter-patient variation in drug response occurs, and that this can have important implications, has been appre- ciated for centuries. Sigmund Freud, in his 1885 monograph on the pharmacological effects of cocaine (Uber Coca), stated: ‘It is well known that… even ani- mals of the same species differ most markedly from one other in those chemical characteristics that determine the organism’s receptivity to foreign substances.’ In addition, contrary to popular belief in certain quarters of the biotechnology sector, the recognition that much of this variation is genetically based – and hence phar- macogenetics as a scientific discipline – is not new either. The concept was put forth by Motulsky in 1957 (Ref. 1), the term ‘pharmacogenetics’ was coined by Vogel in 1959 (Ref. 2), and by 1962, the first book on the subject by Werner Kalow 3 had already compre- hensively documented several confirmed examples of inherited traits that markedly affect drug response or toxicity in human populations. Indirectly, physicians have always attempted to pre- scribe drugs with the patient’s genetic make-up in mind – the patient’s family history is routinely considered when making decisions regarding drug dosage. Such anecdotal information is often inadequate, however, in assessing the importance of inheritance as a factor in accurate and rational drug prescribing. In clinical prac- tice, patients given medications often find either that they do not work or that they are accompanied by unacceptable side effects, and that they must return repeatedly to the doctor for a series of hit-or-miss pre- scriptions until, hopefully, the right drug is found. This process is not only inefficient but clearly suboptimal from the viewpoint of patient care and safety. Historically, those classical pharmacogenetic traits that were discovered and characterized based on the observation of marked toxicity or aberrant drug response in a subset of the population have largely involved monogenic defects in enzymes of drug biotransfor- mation. Examples include: the prolonged apnea observed following succinylcholine administration for muscle relaxation in general anesthesia (mutant forms of plasma cholinesterase); the increased occurrence of peripheral neuropathy during isoniazid therapy for tuberculosis (mutations in arylamine N-acetyltransferase NAT2); and orthostatic hypotension during antihypertensive therapy with debrisoquine or tetanic uterine contrac- tions with the oxytocic agent sparteine 4 (variant forms of the cytochrome P450 isoform CYP2D6). Such traits, however, represent only the most easily detectable tip of the ‘pharmacogenetic iceberg’. First of all, considering that roughly one in every three hun- dred nucleotide bases in the human genome is expected to be polymorphic, a vast amount of functionally important variation in targets of drug action remains to be discovered. Even more important, the majority of phenotypic variations in overall drug response can be expected to result from the complex interplay among multiple potentially polymorphic gene products, including those involved in drug absorption, transport, pharmacodynamics (drug targets), biotransformation and excretion. Until recently, however, the ability to genetically analyse these polygenic traits in a com- prehensive fashion has been hampered by technical limitations and by a lack of knowledge regarding the identities of each potential gene product involved in the complex network of events between the administration of a drug and the patient’s response to it. Pharmacogenetics and pharmacogenomics As alluded to above, the historical starting point for classical pharmacogenetic investigations has been the 334 0167-7799/00/$ – see front matter © 2000 Elsevier Science Ltd. All rights reserved. PII: S0167-7799(00)01463-3 TIBTECH AUGUST 2000 (Vol. 18) FEATURE A SNPshot: pharmacogenetics and the future of drug therapy Dale R. Pfost, Michael T. Boyce-Jacino and Denis M. Grant Pharmacogenetics holds great promise for the optimization of new drug development and the individualization of clinical therapeutics in the 21st century. In this brief review, we trace the historical roots of pharmacogenetics, discuss its rapidly evolving processes and paradigms, and look towards future applications of pharmacogenetics in enhancing the efficiency of the drug development pipeline and in improving patient care. D.R. Pfost (dpfost@orchid.com), M.T. Boyce-Jacino and D.M. Grant are at Orchid BioSciences, 303 College Road East, Princeton, NJ 08540, USA.