Researchers Present Much-Anticipated Correlates Data from RV144

At the first plenary session of AIDS Vaccine 2011, which opened this morning in Bangkok, Thailand, researchers presented the results of a two-year-long effort to try to identify an immunological explanation for the modest 31% efficacy afforded by the prime-boost vaccine regimen tested in the RV144 trial, the first to show any efficacy in protecting against HIV infection. Barton Haynes, who led the scientific steering committee that oversaw the correlates search, reported that two immune responses were found to be significantly correlated with the HIV infection rate in RV144 vaccine recipients, a finding that generated several hypotheses about protective immune responses following vaccination and helped dispel doubts about whether the modest efficacy seen in the trial was real. "The findings lend credence to the vaccine efficacy seen in the RV144 trial," said Haynes, calling the results "intriguing clues on how the RV144 vaccine regimen may have protected."

The first immune response researchers identified that correlated with HIV infection risk was immunoglobulin (Ig)G antibodies that bind to the V1/V2 variable loops in HIV Envelope. The presence of IgG binding antibodies to V1/V2 correlated with a 43% reduction in HIV infection rate. For volunteers with high V1/V2 responses as compared to those with medium- or low-level responses, there was a 71% reduction in HIV infection rate.

The second immune response identified as a correlate of risk was plasma IgA antibodies that bind HIV Envelope. These IgA responses were directly correlated with HIV infection rate; presence of these antibodies was correlated with a 54% increase in the HIV infection rate. This suggests that high IgA responses interfere with protection against HIV infection. However, when researchers compared HIV-infected and vaccinees with placebo recipients, they found that the HIV infection rate was the same, suggesting that although high levels of IgA antibodies decreased vaccine efficacy, the presence of these antibodies did not enhance an individual's risk of HIV infection. Haynes said one hypothesis for the correlation between IgA antibodies and increased risk of infection may be that IgA blocks antibody-dependent cellular cytotoxicity (ADCC; a mechanism by which anitbodies help faciliate the destruction of HIV-infected cells). Blocking of ADCC by IgA has been seen in cancer, according to Haynes. Because no mucosal samples were collected in RV144, researchers were only able to analyze IgA in plasma, which is a different form of IgA than what is found at the mucosa. "It's an open question about what implications this finding may have for mucosal immunity," said Haynes, adding that collection of mucosal secretions and tissues will be a key part of the RV144 follow-up studies.

Starting two years ago when the RV144 results were first presented at the AIDS Vaccine conference in Paris, researchers began preparing plans to conduct a correlates analysis. After issuing a broad call for suggestions about how to proceed, a broad swath of researchers collaborated on a series of pilot studies to identify the best assays to use for the case-controlled correlates analysis. Careful selection of assays was necessary because there was a paucity of samples to work with. When RV144 was in the planning stages, the plans for sample collection were scaled back to cut costs. Given this and other limitations, the correlates search was comparable to looking for a needle in a haystack, Haynes said he was surprised that any correlates were identified.

After much consultation, researchers settled on six primary and approximately 30 secondary assays for the correlates analysis of 41 HIV-infected vaccinees, 205 uninfected vaccinees, and 40 placebo recipients. The six primary assays that were selected measured the following immune responses: binding IgA antibodies in plasma; IgG avidity to A244 gp120 (the antigen used in the vaccine candidates); ADCC; neutralizing antibodies; binding IgG antibodies to the V1/V2 loops of HIV Envelope scaffolded onto a gp70 from murine leukemia virus; and CD4+ T-cell responses as measured by secretion of the cytokines interferon-gamma, interleukin-2, tumor necrosis factor-alpha, and expression of the CD154 surface marker.

The statistical analysis plan for the correlates determination was developed by Peter Gilbert and colleagues at the Statistical Center for HIV/AIDS Research and Prevention (SCHARP). The analysis was 80% powered to detect an approximately 50% reduction in HIV infection rate and controlled for variables, including gender and baseline behavioral risk. The final statistical analysis developed at SCHARP was then validated by an independent team of statisticians.

These results give direction to future studies, including clinical trials that are being planned as follow-up studies to RV144. Haynes said that the immune responses identified as correlates in RV144 would also be evaluated in passive protection studies in non-human primates. These results are also leading to the design of new immunogens. "They [the correlates] give us an important lead on improving on these responses," said Haynes. "Now we have informed hypotheses that come from a trial."

But Jerome Kim, deputy director (Science) at the US Military HIV Research Program, cautioned that these correlates may be unique to this vaccine regimen. "We have to bear that in mind as we look to the next steps in HIV vaccine development."