The efficacy of standard clinical trials in medicine has recently become the subject of contentious debate between those practitioners who maintain that such trials, despite admitted shortcomings, still represent the best means we have for learning about the effects of pharmaceutical drugs on the human body, and those who maintain that the current system of collecting knowledge of such effects is but one possibility and most likely not the most efficacious one. Gimley and Lebsmith, in their recent work, fall into the latter camp and indeed go further by challenging the idea that the standard medical trials can yield meaningful information on pharmacogenetics, or how a drug interacts with the human body.

Gimley and Lebsmith's foremost criticism is that the effect of a drug differs depending on a person's physiology. To be sure, there are cohorts, or groups, that react to a drug in a specific manner, but clinical trials are unequipped to identify such groupings. The main reason is that clinical trials are allied to the notion that the larger the number of subjects in a study, the greater the validity of a drug, should it show any promise. Therefore, even if a drug can exercise a marked effect on a subset of subjects within a trial, this information will be lost in the statistical noise.

Another criticism of Gimley and Lebsmith concerns the very idea of validity. Pharmaceutical companies will run hundreds of trials on hundreds of different medications. Given the sheer number of trials a few are likely to yield positive results, even if there is no demonstrable effect. Gimley and Lebsmith cite the fact that most pharmaceuticals that have exerted a positive effect in the first round of testing are likely to fail in the second round of testing.

Gimley and Lebsmith argue that a more effective approach is to identify groups who exhibit similar genetic subtypes. This very approach is currently in use in groups possessing a particular molecular subtype of breast cancer. Furthermore, these groups are not only trying one specific drug, but also a combination of such drugs , subbing them in and out to measure the effects on a subject, a procedure Gimley and Lebsmith endorse. Nonetheless, such an approach is often both time- consuming and costly. However, given the constraints of current medical trials, trials that target subtypes-even if they do not yield any significant advances-will encourage a culture of experimentation on how clinical trials are conducted in the first place.