The Myth of Equipoise in Phase 1 Clinical Trials

Adil E. Shamoo, PhD


Medscape J Med. 2008;10(11):254 

In This Article


A seminal paper by Miller and Brody[1]clearly laid the general framework on why we should abandon the term "clinical equipoise" in research. Their paper argued that the concept of equipoise (also known as risk-benefit)[2,3,4,5,6,7,8,9,10,11] in research with human subjects is conceptually "flawed" and unnecessary. Clinical equipoise (ie, the idea that there should be genuine disagreement about which arm of a clinical trial is superior before initiation of the study) is conceptually flawed because the ethics of clinical research differs from that of clinical medicine. Physicians who are treating patients have a moral obligation to promote the patient's best interests, whereas clinical researchers who are conducting experiments involving human subjects are primarily obligated to "science" but secondarily have an obligation to prevent harm to the experimental subject who is still primarily a patient.

The requirement that the arms of a clinical research protocol be in clinical equipoise makes sense only under the assumption that clinical researchers have an obligation to promote the best interests of research subjects, since the subjects' best interests could be promoted by enrollment in any arm of the study. Clinical research and the clinical researcher are promoting the best interest of the public and not the individual subject. Many researchers hoped that this paper would put an end to the concept of equipoise in research; unfortunately, it did not.

There are 2 probable explanations for the lack of acceptance of such a clear and logical presentation and refutation of the concept of equipoise in human subject research. The first explanation is that we live in an individualistic society, which deemphasizes utilitarian (or collectivist) reasoning. We often justify policies and decisions by appealing to human rights or autonomy, even if we go through intellectual acrobatics to force-fit our reasoning into this framework. Or, we simply use a utilitarian calculus in our reasoning unwittingly and without disclosing it. As if to hide their true intentions from those who condemn utilitarian reasoning, Miller and Brody used the term "utilitarian" only once in their lengthy article, when they described the aim of clinical research as "a frankly utilitarian purpose," with which I agree that research is a utilitarian pursuit.


This discussion has 2 separate and important objectives: 1) Is there equipoise in phase 1 clinical research trials? and 2) Is there a need for equipoise in phase 1 trials? Phase 1 clinical trials clearly illustrate the moral conflicts between the individual interests and the public good. However, I am aware that the analysis for phase 1 used here applies to all research with human subjects. This is because nearly all ethical analyses include phase 1 studies in terms of the issue of equipoise or risk-benefit.[2,3,4,5,6,7,8,9,10,11]Some may claim that consideration of the equipoise concept is for randomized, clinical trials (RCTs) only. Fries and Krishnan[12]studied 45 phase 1 RCTs from a meeting on arthritis and found that they all violated equipoise. Also, the prestigious report of the National Bioethics Advisory Commission illustrates equipoise analysis using phase 1 trials.[4] Many phase 1 clinical trials, especially vaccine trials, are RCTs.[13,14,15,16] I will attempt to move away from subjective and anecdotal evidence to justify clinical equipoise in research to the application of accepted scientific standards.

Phase 1 Trials

There is little ethical discourse and few publications of phase 1 trials involving healthy volunteers.[17,18] Phase 1 clinical trials are the gatekeepers for all clinical research. The advancement of drug-based therapeutic treatments rests in the conduct of these trials. Medical science would come to a halt if we stopped using phase 1 trials, or the like, in clinical trials. Phase 1 trials use as few healthy subjects as possible, typically 20 to 80; however, not all phase 1 trials use healthy subjects -- for example, cancer patients are used in phase 1 cancer trials. While no data are available for the number of cancer patients used in phase 1 trials, we can safely presume that the number is small compared with the total number of healthy human subjects in all phase 1 trials.[9,19] In other words, during phase 1 trials we have literally no idea (except from what has been learned from animal experiments) what would be the outcome of the introduction of a drug or a chemical -- and at what dosage -- into a human being.

The healthy subjects in phase 1 trials are given an escalating dose of the "chemical-potential drug" to determine the maximum tolerable dose (MTD). In other words, MTD is the maximum dose of the chemical to induce the first symptoms of toxicity, with the hope that the toxicity is reversible. The tolerable dose is then reduced to a less toxic dose for subsequent phases of clinical trials. As Freedman[3]expressed while discussing phase 1 trials: "[T]he reason for conducting the trial is to discover the point at which a compound is too poisonous to administer."


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.