Skip directly to content

PAVE Gives Pause

Position Statement on the Proposed PAVE100A Vaccine Trial

Also available in Spanish on the website of Grupo de Trabajo sobre Tratamientos del VIH

New York, NY, May 30, 2008 - Treatment Action Group (TAG) Position Statement on the Proposed PAVE100A Vaccine Trial

The failure of Merck's adenovirus serotype 5 (Ad5)-based HIV vaccine to protect against infection or reduce post-infection viral load, combined with the evidence of potentially enhanced susceptibility to HIV acquisition among a subset of trial participants, has inevitably cast a cloud across the Ad5 candidate slated for use in the proposed PAVE100A Phase IIb efficacy trial. TAG recognizes and appreciates the work and commitment of researchers at NIH's Vaccine Research Center (VRC) who have developed the DNA and Ad5 HIV vaccine constructs, and their colleagues involved in PAVE100 who have collaborated in an effort to redesign the trial to ensure safety in the light of the data that emerged from the Merck vaccine trial (STEP/HVTN 502, hereafter referred to as STEP).

However, TAG has serious concerns regarding whether the PAVE100A trial should move ahead. It is now being described as a "discovery" trial, but it is important to lucidly articulate what the trial might discover in the context of the data from STEP.

What is the evidence that the VRC prime-boost approach might lead to an outcome better than STEP?

  • Based on the data published and presented to date, the VRC approach appears to induce T cell responses against HIV which are capable of more functions than those induced by the Merck vaccine ("polyfunctional" responses).
  • These polyfunctional T cells also produce a greater amount of potentially important immune system proteins called cytokines when compared to T cells that display fewer functions.
  • The VRC approach induces a more balanced combination of CD4 and CD8 T cell responses due to the inclusion of a series of priming immunizations with a DNA vaccine followed by just a single shot of the Ad5 vaccine. The Merck Ad5 vaccine biased the immune response toward CD8 T cells and induced CD4 T cell responses in fewer participants. Evidence suggests that CD4 T cell responses are an important component of HIV-specific immunity.
  • The VRC vaccines also include HIV proteins that were not in the Merck candidate: Envelope (Env) proteins from three different HIV clades (A, B & C).
  • The overall magnitude of the T cell responses induced by the VRC vaccines may be somewhat higher than those seen in STEP, but direct comparative data are lacking.
  • Although there is essentially no evidence to indicate that these differences might lead to protection against acquisition of HIV infection, there is some plausible evidence to suggest that they could lead to better control of post-infection viral load among immunized individuals.

What is the evidence that argues against conducting PAVE100A?

  • There is continued uncertainty whether the addition of Env proteins to an HIV vaccine will improve the chances of a successful outcome; there are conflicting data and opinions on this question.
  • Data indicates that the presence of T cells that can respond to multiple different fragments (epitopes) from each HIV protein—particularly the HIV protein Gag—is an important factor in control of viral replication. The Merck vaccine contained Gag, but recipients only developed a response to an average of just one epitope from the protein, although there are multiple epitopes that can potentially be targeted. It is currently unclear how many different epitopes from each protein are targeted by the T cell responses induced by VRC's vaccines (data presented to date suggests they induce broader CD8 T cell responses to Pol and Env compared to Gag), and this is a question that needs to be addressed in detail.
  • Because of the safety issue that emerged from STEP regarding the use of Ad5 vectors—particularly in uncircumcised individuals with antibodies to the naturally occurring form of Ad5 (the subset of participants that appeared to face increased risk of HIV acquisition due to receipt of the vaccine)—the PAVE100A trial is being restricted to circumcised gay men with no detectable antibody responses to Ad5. While this may reduce the possible risks associated with the Ad5 vector boost, other confounding issues may emerge from limiting a trial to this group. For example, it is likely that some proportion of individuals who lack antibody responses to Ad5 have in fact been exposed to adenovirus infection, but their immune systems failed to generate antibodies against it (similar to the way a small subset of individuals fail to mount antibody responses to licensed vaccines such as hepatitis B vaccine). Such individuals appear to have poor immune responses generally, and thus may be least likely to benefit from receipt of an HIV vaccine candidate, even one that had the potential to benefit individuals who mount better immune responses. In other words, the trial population may be skewed in a way that could lead to an underestimation of any vaccine effect.
  • Conducting the trial in such a limited and selected population also raises the issue of how broadly applicable any data from the study would be (including data on correlates of viral load control, which differed in STEP based on Ad5 antibody status). Certainly, the adenovirus vector used in the study could not be employed in a broader population due to the safety concerns raised by STEP.
  • Although it is regrettable to have to consider factors and forces outside of the realm of data, the political, public, and scientific fall-out from the STEP trial was disturbing and discouraging. It is grim to contemplate the consequences of adverse results emerging from PAVE100A. There is also the danger that even a simple failure to show efficacy could lead to a premature abandonment of efforts to create effective T cell-based vaccines for HIV infection.

TAG's Take on a Tough Call

As an organization that supports the scientific endeavor to cure and prevent HIV infection, we are reluctant to simply oppose PAVE100A without proposing more promising alternative approaches. But our view is that serious consideration should be given to abandoning plans for this particular clinical trial and Ad5 vaccine candidate in favor of supporting research to rapidly address two key issues:

  1. The use of an alternate vector or DNA approach for the boosting immunization that could be studied in a broader population.
  2. Inducing broader responses to proteins contained in the vaccine, particularly HIV-1 Gag.

Based on the information that is currently available to us, we feel that the uncertainties argue against spending human and fiscal resources on PAVE100A, and instead suggest focusing on improving T cell-based immunogens so they can be studied in a broader population with a greater chance of success. 


Submitted to the AIDS Vaccine Research Advisory Subcommittee (AVRS) of the National Institute of Allergy & Infectious Diseases (NIAID), May 30, 2008