Understanding the Hunt for Immune Correlates of Protection from RV144

How can the recently conducted AIDS vaccine trial in Thailand help inform researchers about the types of immune responses that can provide protection against HIV?

A recently completed efficacy trial in Thailand, known as RV144, showed that two vaccine candidates administered sequentially in what is referred to as a prime-boost regimen could reduce the risk of HIV infection by 31.2% (see VAXOctober 2009 Spotlight article, Vaccine Research Gains Momentum). This trial provided the first evidence of efficacy for any AIDS vaccine candidate.

Vaccines work because they train the immune system to produce various types of immune cells and proteins, referred to as immune responses, against a specific virus or bacteria. A vaccine can induce many different types of immune responses, including antibodies (Y-shaped proteins that bind to viruses and prevent them from infecting cells), cellular immune responses (CD4+ and CD8+ T cells that orchestrate the killing of virus-infected cells), as well as natural or innate immune responses. For HIV, researchers do not yet know which types of immune responses are necessary for protection. While the vaccine candidates tested in RV144 only provided a modest efficacy, this trial provides scientists with the first opportunity in humans to try to figure out which immune responses induced by these candidates actually protected some of the volunteers against HIV infection. The specific immune responses that are responsible for protection are referred to as the immune correlates of protection (see VAX November 2006 and December 2006 Primers on Understanding Immune Correlates of Protection, Part I and II).

If the immune correlates of protection can be identified from RV144, it would be a significant boost for AIDS vaccine research. Scientists could then design AIDS vaccine candidates that induce these specific responses at higher levels, and thereby improve upon the modest efficacy seen in RV144. But identifying the correlates from this trial will not be an easy task. Overall, relatively few volunteers became HIV infected during RV144, so researchers have a limited pool of individuals in which to search for immune correlates.

The hunt

Researchers will begin the hunt for correlates by screening the thousands of blood and cell samples collected from volunteers during RV144 for different types of immune responses. Specifically, researchers will look closely at the samples taken from the 51 participants who received the vaccine candidates yet still became HIV infected through natural exposure to the virus. Samples from these volunteers will be compared to those from the 8,146 participants who received the vaccine candidates but did not get infected.

Because there are many more uninfected vaccinees than infected vaccinees, scientists will try to match each HIV-infected vaccinee with four to five who are uninfected and who share a similar demographic and genetic profile. Matching the samples as closely as possible in terms of gender, race, age, level of HIV risk, and genetic background will enable researchers to rule out the role these other factors may have played in the immune responses to the vaccine candidates.

Researchers will use multiple laboratory tests, or assays, to try to tease out if a specific immune response—including antibodies or cellular immune responses—occurred more frequently or at a higher level in vaccinated volunteers who did not acquire HIV compared to those who did. Although RV144 showed some efficacy in preventing acquisition of HIV, the prime-boost regimen did not appear to have any impact on the amount of virus circulating in the blood of individuals who became HIV infected despite vaccination. This result has led many researchers to assume that the modest level of protection was more likely due to antibody responses than cellular immune responses. Antibodies are the key to protection for most, if not all, existing vaccines. Still, researchers will carefully analyze all categories of immune responses in the hunt for correlates from RV144.

Limited samples

Scientists will have to carefully choose which assays to conduct because a limited number of specimens were collected during RV144. This trial was not designed to determine the correlates of protection. No samples were collected until after the six-month vaccination period, and researchers did not collect cell or tissue samples from mucosal sites at which sexual transmission of HIV occurs.

The search for correlates is also hampered by the fact that researchers do not know exactly when people were exposed to HIV and whether the HIV-uninfected vaccinees were even exposed to HIV at all.

Additional studies

There are also additional studies that can help identify the correlates of protection. One of these studies, known as RV152, is already ongoing. It involves the 51 individuals who became infected in RV144 despite vaccination. Information collected from these volunteers may shed light on the characteristics of the virus that infected these individuals. Investigators are also considering whether to conduct a smaller trial with the same prime-boost regimen that is designed specifically to try to determine the immune correlates.

Researchers may also be able to collect valuable clues from studies in nonhuman primates. If they can replicate the protection seen in RV144 in nonhuman primates, they could then use this model to try to identify the immune responses that are responsible for protection.