Email Updates

Search form

You are here

Clinical Trials

Candidate strategies that meet certain criteria in laboratory and animal studies move into clinical trials in humans. Clinical trials enroll human participants to evaluate scientific or medical interventions like drugs and vaccines. There are many kinds of clinical trials. This section specifically discusses clinical trials to test biomedical HIV prevention strategies. Clinical trials generally happen in a sequence:

Phase I trials are the first stage of testing an experimental product in human participants. These trials carefully evaluate the candidate's safety in humans. These trials typically involve several dozen healthy participants. They can also gather other types of information. A Phase I vaccine trial might measure the immune responses generated by the vaccine. A Phase I microbicide trial might look at changes in vaginal tissue after application of the product. A Phase I PrEP study might look at the levels of the drug in the blood over time.

Phase II trials are usually larger than Phase I studies and may involve a more diverse group of participants. Phase II studies can enroll anywhere from a few dozen to several hundred participants. These studies gather more information about safety and may ask many of the same types of questions asked in Phase I studies, perhaps with more details or different measures. Phase I and Phase II studies may both look at the effects of different doses or dosing schedules for an intervention and gather information about which strategy seems best for testing in larger trials. Some of these "dose-ranging" studies are described as Phase IIa studies.

Phase IIb studies are also known as "proof-of-concept" or "test-of-concept" trials. As the name suggests, the goal of these studies is to get an early indication of whether a given strategy might be effective at reducing the risk of HIV infection or transmission. The size and design of Phase IIb studies depend on the questions being asked and the intervention being tested. In general, they're designed in a way that gives an idea about whether an intervention has a benefit for HIV prevention. If there is an indication of benefit, then additional studies are done to learn more and try to confirm the finding.

Phase III trials are large-scale studies that examine whether an experimental strategy has a benefit for HIV prevention. They often involve thousands of participants and also gather additional data about safety. For a microbicide, the HIV prevention benefit could be reducing the risk of HIV infection in the person using the microbicide. For a vaccine trial, the benefit could be reduced risk of HIV infection or it could be reduced viral load in the person who gets the vaccine and later acquires HIV. (The vaccine itself does not cause HIV.) Phase III trials are designed to give a clear result about whether the candidate has any benefit. Even with a clear answer from a single Phase III trial, there might still be additional Phase III trials to look at the same strategy in other populations or to confirm the result.

How do prevention trials test whether a product has an HIV prevention benefit?

All biomedical prevention strategies are evaluated for safety in humans before moving into efficacy or effectiveness trials, which may be called Phase III, Phase IIb, test-of-concept, or proof-of-concept trials. There are technical reasons why some trial designs are called efficacy and others are called effectiveness studies. Both terms refer to trials that look at whether a candidate has an HIV prevention benefit. These benefits can include reducing an HIV-negative individual's risk of getting HIV; reducing an HIV-positive individual's risk of transmitting HIV; and reducing HIV viral load in a person who received the intervention when he or she was HIV-negative, and later became HIV-positive through sexual or other exposure.

The design of an efficacy trial depends on what intervention is being studied. Many biomedical prevention trials enroll HIV-negative participants and evaluate new strategies to see if they reduce participants' risk of becoming HIV-infected. This type of trial is described below, followed by details about other types of designs. An efficacy trial of a strategy to be tested in HIV-negative individuals starts where all trials start: by providing community education that allows people to learn about the goals of the trial and decide whether they might want to participate. Different trials use different strategies to provide this information. Potential participants visit the trial site to learn more and decide whether to participate. Those who are eligible and pass the screening tests (each trial has different eligibility criteria, depending on the product and population in which the intervention is being tested) may decide to enroll by providing informed consent. This is an important process that should continue throughout the trial. Visit our Research Ethics in Action page to learn more about informed consent.

All participants in biomedical HIV prevention trials receive an HIV prevention package including condoms, counseling, and STI treatment. Needle exchange and drug substitution therapy should be provided in any efficacy trials involving injection drug users. One group of participants is randomly assigned to receive the experimental intervention. This could be a candidate vaccine or microbicide, or it could be another strategy, like male circumcision. The other group of participants is assigned to receive a placebo. A placebo is something that looks exactly like the experimental candidate, but contains no active ingredients. That means a placebo has no effect on the body and cannot prevent HIV transmission. In the case of male circumcision, participants were asked to delay male circumcision for the duration of the trial.

The majority of biomedical HIV prevention trials are double-blind, which means that neither the participant nor the trial staff knows who is receiving the experimental candidate or placebo—and all trial participants are counseled that there's no way of knowing whether they received the placebo or experimental candidate. They are also reminded at each visit that it is not known whether the experimental candidate provides any benefit and that they should keep on using proven prevention strategies. Over the course of the trial period, some participants get infected even though they are being counseled and receiving prevention services. This is consistent with what we know about the AIDS epidemic: even with information and services, not everyone can protect himself or herself all the time.

At the end of a trial, researchers compare the rates of new infections in the participants who received the experimental strategy and in those who received placebo. They may also look at the viral load (amount of virus circulating in the body) in participants who got infected, comparing the experimental and placebo groups. If there are significantly fewer new infections or lower viral load in the experimental group, that is, if the difference is greater than can be reasonably attributed to chance, this suggests that the experimental strategy is beneficial.

A trial looking at whether an experimental strategy reduces an HIV-positive person's risk of transmitting HIV works in a similar way, with some important differences. These trials generally enroll serodiscordant couples: an HIV-positive person and a primary sexual partner who is HIV-negative. Both the HIV-positive person and the HIV-negative person undergo informed consent, and both receive prevention services, although the two partners may have different schedules for study visits. One group of HIV-positive participants is assigned to receive the experimental strategy plus the standard prevention package. The other group receives the standard prevention package and (with a few exceptions, like male circumcision) a placebo. Both the HIV-positive and HIV-negative partners are followed over time. At the end of the trial, researchers analyze the data to see whether there are significantly fewer new infections in the HIV-negative partners of participants who received the experimental strategy versus those who did not.