Yesterday saw an avalanche of media coverage relating to the results of RV144, a large HIV vaccine trial that took place in Thailand involving 16,402 volunteers, most at a low risk of exposure to HIV infection. Few details were initially released (at 2am EST time in the US) beyond the bare bones of the primary efficacy analysis, which found that 74 out of 8,198 volunteers who received placebo immunizations became infected with HIV compared to 51 out of 8,197 volunteers who received a combination of two vaccines, ALVAC vCP1521 and AIDSVAX B/E (seven volunteers were excluded from the efficacy analyses after it was found that they were HIV-infected at the time of receiving their first vaccination). The result equates to a protective efficacy of a little over 31%, and is statistically significant with a p value of 0.039. The result has been widely hailed as historic in the press, and is certainly welcome given the many unpleasant surprises that have befallen the HIV vaccine field over the years.
However, the 95% confidence interval associated with the efficacy result spans from 1.1% to 52.1%, indicating a large degree of uncertainty and marginal significance (a lower bound of less than 1 means that significance has not been achieved). It seems that perhaps a difference of just two or three infections less in the placebo group (or more in the vaccine group) could have caused the result to be non-significant, which is somewhat concerning.
As background, the trial was a collaborative endeavor involving the Thai Ministry of Health, the US Military HIV Research Program and NIAID. At the time the idea for the trial originated, around the mid-to-late 90s, the two candidates used in the trial were considered to be the best available agents for inducing cellular and humoral immune responses against HIV, respectively. However, it was also clear both had significant shortcomings. ALVAC induces HIV-specific CD8 T cell responses in only around 20% of recipients, and although the majority of recipients of AIDSVAX develop antibody responses, the antibodies are completely incapable of neutralizing circulating HIV isolates. Furthermore, AIDSVAX failed to offer any protection in two large phase III efficacy trials (one involving Thai intravenous drugs users, the other primarily involving gay men). It is the poor immunogenicity of the candidates and the dismal track record of AIDSVAX that caused many people to be surprised by the efficacy results (the full text of a Phase 1/2 immunogenicity study is available free online).
However, the vaccine combination has been shown to be effective at inducing Env-specific CD4 T cell responses, as measured using lymphoproliferation assays (upwards of 60% of recipients develop these responses). It has also been reported to induce non-neutralizing but “binding” antibodies, which some scientists have suggested could play a role in protection. An additional consideration may be that, in a low-risk population, there is likely to be a longer lag time, on average, between receipt of a vaccine and exposure to HIV, potentially giving more time for vaccine-induced immune responses to mature. So while it is understandable that many people thought a positive outcome of the trial was implausible, there are some reasons – albeit highly speculative and unlikely – why it may not have been impossible.
The most important next step is release of detailed data, which is slated to happen at the annual AIDS Vaccine conference in Paris next month. It was also stated on yesterday’s conference call that a manuscript is complete and ready for submission to a scientific journal. Key questions include the extent of the availability of samples from the study and whether it will be possible to uncover any correlates of protection (assuming the significance of the result holds up under scrutiny). It is also unclear if the relative contribution of ALVAC and AIDSVAX to the outcome can be disentangled; it seems unlikely, and this is a concern that TAG raised about the trial when it was first launched (there is no data on the efficacy of ALVAC alone).
In terms of benefits to the trial participants, the efficacy threshold for licensure in Thailand that the investigators had set was 50%, and that has not been reached. There hasn't been much public discussion yet regarding whether placebo recipients in the trial should be offered the choice to receive the vaccine, but the trial investigators have stated that they're open to discussing the matter further. The overall sense being conveyed by trial sponsors is that if the results prove robust, further trials will be needed to try and understand them.
Below is a selection of the many news articles and blog postings discussing the RV144 announcement. AVAC also has a page of extensive resources, including a recording of yesterday’s conference call with the trial investigators.
Massive AIDS Vaccine Study a "Modest" Success
Jon Cohen, Science
HIV vax testers react to Thai trial
Bob Grant, The Scientist
Scientists Puzzle Over Minor Success Of AIDS Vaccine
David Brown, Washington Post
An unpopular vaccine study produces surprising result
Keith Alcorn, AIDSMap
MSF Welcomes HIV Vaccine Trial With Cautious Optimism
Médecins Sans Frontières
For First Time, AIDS Vaccine Shows Some Success
Donald G. McNeil Jr.
Comments