Tutorial: An Introduction to Clinical Trials

The First Installment of NextGen's "Tutorials in Medicine" Series

Development and testing of medical treatments demands both high standards of scientific rigor and safety monitoring because of the unique ethical roles of humans as research subjects and as patients who will eventually receive a treatment. While observational studies such as case reports, surveys, and cohort studies may be relatively easy to conduct, their results are subject to bias because these studies cannot control the administration of a treatment to demonstrate cause and effect decisively. Results from experimental studies can be more valid; these include laboratory studies (on cell cultures, pathological specimens, or animal models) and clinical trials involving human subjects. Pre-clinical laboratory studies can provide information about pharmacology and toxicology of a newly designed drug. However, clinical trials are the only process by which a company or researcher can demonstrate that a drug, device, or biologic (a vaccine or treatment synthesized from living organisms) is safe and effective, and thus approvable by the Food and Drug Administration (FDA).

A primary goal of a clinical trial is to establish cause and effect. In order to reach this goal, the study must isolate the effect of a treatment and rule out factors that could lead to misleading findings. This may include the elimination or at least minimization of bias, confounding (where the estimate of the treatment effect is distorted due the impact of another influential variable), and randomness. A clinical trial is a prospective study (a study that is planned in advance in which data are then collected in the future, as opposed to using data that has already been collected), and so it permits manipulation and measurement of a treatment in a controlled setting and thus provides the optimal strategy to avoid bias and confounding, minimize randomness, and quantify the uncertainty in the study outcome. If done properly, therefore, it is highly likely that the results will be considered valid. Disadvantages of clinical trials, beyond their infeasibility in some situations due to ethical concerns, include their costly and time-consuming nature. From the synthesis of a new drug to its appearance on the market can take approximately seventeen to twenty years, and cost from 0.8 to 1.7 billion dollars.

After an experimental drug has been shown to be promising in laboratory and animal studies, an Investigational New Drug (IND) application may be filed with the FDA, requesting to begin experimentation in human subjects. If the IND application is approved, then a four-phase series of clinical trials progressively examines the optimal dose, safety, short-term and long-term effects on the body, and ultimately the effectiveness of an experimental therapy in humans.

Phase I Trials

Phase I trials assess the tolerance of a drug. Often performed on a small sample group of healthy people, these trials aim to find the Maximum Tolerated Dose (MTD) of a drug and assess its toxicity. The therapeutic effectiveness of the drug is less of interest than the pharmacodynamics (the effect of the drug on the body, e.g. heart rate and respiration), as well as the pharmacokinetics (the effect of the body on the drug including how it is absorbed, distributed, metabolized, and eliminated). Only one in 10,000 pharmacological compounds reach this phase, and of those only eight percent will eventually make it to the drug market.

Phase II Trials

Phase II trials are performed on a small number of people with the disease in order to find the optimal dose of the drug. As in Phase I trials, testing the efficacy of the drug is a secondary concern, although the short-term therapeutic activity may be assessed. These trials frequently employ surrogate endpoints, substitute variables that are used when a clinical endpoint is not convenient or feasible. For example, in hypertension studies, mortality is the most important endpoint, but raised blood pressure is often the endpoint that is measured.

Phase III Trials

Phase III trials are performed on a large number of people who are randomly assigned to a treatment group or a control group (which receives either a placeboâ an inactive treatment, or an active control such as an existing standard of care (SOC) drug). Phase III trials often use the population for which the drug is intended, so the subjects may be more heterogeneous than in a Phase II trial. Finally, Phase III trials may be blinded so that neither the subject nor the treatment provider knows whether the subject has been assigned to the treatment group or the control group as an additional way to eliminate potential bias. In order to file a New Drug Application (NDA) for approval of the drug by the FDA, with a few exceptions, it is necessary to conduct two consecutive trials with results statistically significant at the p=0.05 level.

Phase IV Trials

Phase IV trials are conducted after the drug has been approved, and are intended to investigate the drugâ long-term effects by following up with the subjects over a long period of time. The importance of such studies is underscored by the recent Vioxx case in which safety issues were noted after FDA approval. These trials may observe the effects of the drug in the settings where it is used in practice, may act as surveillance of safety, and/or may be used as a marketing tool to expand the use of the drug to new populations.

Crucial to the success of a clinical trial are its appropriate design, monitoring, and statistical analyses. The basic elements of these components will be discussed in Part Two of this tutorial, appearing in the next issue.

Miya Bernson is an Associate Editor of the Next Generation and a member of the Harvard College Class of 2006. Scott Evans, Ph.D. is a Research Scientist at the Center for Biostatistics in AIDS Research and the Department of Biostatistics of the Harvard School of Public Health where he focuses on clinical trials research.