PAVEing the Way to a Smaller Trial

A smaller, more focused trial is now under consideration at NIAID to replace the proposed PAVE 100 study

By Andreas von Bubnoff

After lengthy deliberations, the National Institute of Allergy and Infectious Diseases (NIAID) announced on July 17 that the Phase IIb test-of-concept trial known as PAVE 100 will not take place. Instead, NIAID is now considering a smaller, more focused study to evaluate the efficacy of a prime-boost regimen of the DNA and adenovirus serotype 5 (Ad5) vaccine candidates developed by the Vaccine Research Center (VRC) at NIAID.

The PAVE 100 trial, originally slated to begin last year at multiple study sites, was to involve 8,500 HIV-uninfected men and women from the Americas and southern and eastern Africa. This trial would have evaluated whether this prime-boost regimen could protect against HIV infection, or lower the viral load in individuals who become infected despite vaccination. It would have been funded by NIAID and conducted by the HIV Vaccine Trials Network (HVTN), IAVI, the US Military HIV Research Program, and the US Centers for Disease Control and Prevention (CDC). However, it was one of several trials delayed when Merck’s candidate vaccine, known as MRKAd5, failed to show any efficacy in the STEP trial.

Further fallout from the Merck trial sent the PAVE protocol team back to the drawing board. Based on analyses of the STEP trial, which indicated that male volunteers who received the vaccine candidate had a higher risk of acquiring HIV if they were uncircumcised and had pre-existing immunity to the Ad5 vector, two smaller PAVE trials were proposed. PAVE 100A was to include only 2,400 circumcised men who have sex with men in the US without pre-existing Ad5 immunity. The second part of the trial, called PAVE 100B was to be conducted in Africa. This arm of the trial was later deferred.

On May 30, AIDS vaccine experts met in Bethesda, Maryland, as part of the AIDS Vaccine Research Subcommittee (AVRS) to discuss whether to proceed with PAVE 100A, the US arm of the trial. At the conclusion of that meeting, a clear majority of the AVRS members recommended that NIAID should conduct the PAVE 100A trial. NIAID director Anthony Fauci recalls that only a few were not in favor of moving ahead with PAVE 100A.

But the final decision was his and in July Fauci announced that he had decided not to proceed with either part of the PAVE 100 trial. “A and B are out,” says Fauci. The main reason for his decision was the size of the proposed trial. Fauci concluded that an even smaller study than the 2,400-person trial would be sufficient to primarily study an effect on viral load, which is what most researchers consider to be the best hope for this regimen. “There are very few people who believe that a T-cell vaccine of this sort is going to actually block acquisition of [HIV] infection,” Fauci says. The original plan for PAVE 100A would also have studied in detail the immune responses to the vaccine. But Fauci argues that immune correlates should be studied later, only after the regimen’s ability to lower viral load is established. “I say do a trial that’s big enough to give you the answer if it works,” he says. “If it does work then you have the option of going back and amplifying the study. Why invest in a trial that’s so large that you’ll be able to [study] immunological correlates right away whether [the vaccine candidate] works or not?”

Even though Fauci’s final decision seemed to go against the recommendation of the AVRS, he says there was a “lack of firmness” in the opinions of some committee members. Some who voted yes publicly, later contacted him and said they were not sure. Fauci also took into consideration those who had a vested interest in the trial. Fauci says, after “taking all of this into consideration, there was still a majority that voted yes, but the margin was closer than the count at the meeting on May 30.”

Fauci also consulted another advisory committee, the Strategic Working Group (SWG) of NIAID’s Division of AIDS (DAIDS). “The vote of the strategic working group was unanimously not to go ahead with the trial,” says Fauci, adding that this group consists of “outside, unbiased investigators who have no stake in any of the [AIDS vaccine] clinical trials.”

Defining a smaller trial

Fauci has asked Scott Hammer, chair of the PAVE 100 protocol team and a professor of medicine at Columbia University, and Larry Corey, the principal investigator of the HVTN, to come up with an alternate trial design that focuses on whether the VRC vaccine candidates could lower viral load in vaccinated individuals who subsequently become HIV infected. The question now is how small of a trial will provide that answer. Fauci says the new trial would likely be “significantly” smaller than the PAVE 100A design, perhaps involving only 1,200 to 1,400 volunteers. “This would fall under the category of a STOC trial,” Fauci says, referring to a screening-test-of-concept trial. This design allows investigators to conduct smaller, less expensive trials than Phase IIb test-of-concept studies, and still collect information about a candidate’s ability to reduce viral load in infected vaccinees (AIDS 21, 2259, 2007).

STOC trials would only require about 1,000 study volunteers, depending on the HIV incidence in the trial cohort, and would provide answers about which candidates are worth pursuing in larger trials.

At the May 30 AVRS meeting, John Moore, a professor of microbiology and immunology at Weill Cornell Medical College, suggested doing a STOC trial, saying that he didn’t believe the VRC vaccine could be expected to have an effect on HIV acquisition, and therefore the initial trial didn’t have to be large enough to show that. Fauci agrees. “Moore actually recommended what we ultimately decided to do,” he says.

IAVI president and CEO Seth Berkley said he welcomed the fact that NIAID is considering a smaller trial. “These kinds of smaller studies, such as the screening-test-of-concept trial proposed by IAVI, should become the norm, to test for a sign of promise before proceeding to large efficacy trials,” Berkley says.

But Eric Hunter, chair of the AVRS who expressed support for conducting the PAVE 100A trial at the May 30 meeting, says while a smaller trial is cheaper, it may delay the overall process. “There is a trade-off,” Hunter says. “Certainly such a [small] trial would be much more cost effective in the short term, but [it] has the downside that if it is successful then you have to go back and restart the whole thing again to get an explanation as to why it might have been successful.”

In a statement responding to NIAID’s announcement, the AIDS Vaccine Advocacy Coalition (AVAC) called for the Institute to “act swiftly to clarify the path ahead.” Fauci says he expects a recommendation from Hammer and Corey on a smaller trial design within weeks. “We are in the process of developing a protocol that we feel will respond to the challenges outlined by Dr. Fauci,” says Corey.

So far the response to his decision has been “overwhelmingly positive,” Fauci says, adding that when “you get 100 emails and 99 of them say you have made the right decision, that’s what I would consider [positive].” Stanley Plotkin, an advisor to Sanofi Pasteur and AVRS member, agrees with Fauci’s decision to consider a smaller trial. “I think that the decision is a reasonable compromise,” says Plotkin, who recommended conducting the PAVE 100A trial at the May 30 meeting. “I would say that the PAVE 100 trial was not cancelled, but rather scaled down to determine if the DNA/Adeno approach can do what it is supposed to do, reduce viral load after infection.”

Indeed Fauci does still think it is worth testing the VRC vaccine candidates. “I didn’t completely scrap it because I felt that [it was] different enough from the STEP product that it was interesting to see if it worked,” he says.

Sorting out the differences

The VRC regimen is somewhat similar to MRKAd5 in that both use Ad5 to introduce HIV genes into the vaccinees. But while the STEP regimen involves three vaccinations with the same Ad5 candidate, the VRC prime-boost vaccine regimen involves three vaccinations with DNA encoding clade B Gag, Pol, Nef, and Env from HIV clades A, B, and C, followed by a vaccination with the Ad5 vector encoding clade B Gag and Pol, and Env from clades A, B and C. MRKAd5 did not contain HIV env. “I felt that the idea of a DNA containing an [HIV] envelope [gene] was an interesting enough concept,” Fauci says.

At the AVRS meeting, Julie McElrath, director of laboratories at the HVTN, presented a preliminary comparison of immune responses in humans collected from the STEP trial to data from a Phase II trial (HVTN 204) of the VRC vaccine regimen. This data showed some differences. The VRC regimen showed stronger CD4+ T-cell responses, but CD8+ T-cell responses and the breadth of responses—the number of HIV epitopes recognized—were similar for MRKAd5 and the VRC’s DNA/Ad5. However, the predominant responses were to different HIV proteins for the two regimens—DNA/Ad5 recipients responded predominantly to Env and Gag epitopes, whereas MRKAd5 recipients predominantly responded to Pol epitopes.

Hunter says antibody or CTL responses directed toward HIV Env epitopes might play a role in reducing viral load. “There is some preclinical data that would suggest addition of env might facilitate control of viral load even in the absence of a neutralizing antibody response,” he says.