In the area of HIV clinical research, the woods are strewn with the whitened bones of AIDS vaccines that fell by the wayside. Why did they fail?

Immunologist/virologist John Shiver, who directs viral vaccine research at Merck Research Laboratories in West Point, Pa., said, “The earlier vaccines focused on antibody responses to the HIV gp120 envelope protein. Most researchers in the field had concluded by the early 1990s that those recombinant viral envelope-targeting vaccines couldn’t make the kind of antibody responses you would need to protect against the AIDS virus.

“Their problem,” he continued, “was that it is very difficult to make antibodies with a vaccine or even from infection by HIV that can broadly neutralize HIV. It turns out to be very type-specific, down to the individual virus. That makes such vaccines not applicable to the tremendous diversity of HIV in the real world. In the early 1990s that became clear. Here at Merck we started focusing on T-cell responses to the virus, and how best to characterize cytotoxic T-lymphocyte [CTL] responses in humans infected with HIV. We studied vaccine approaches that make those responses.”

Shiver is first author of an interim report in today’s issue of Nature, dated Jan. 17, 2002, titled, “Replication-incompetent adenoviral vaccine vector elicits effective anti-immunodeficiency-virus immunity.” Its senior author is Emilio Emini, senior vice president of vaccine research at Merck.

“The findings we report,” Shiver told BioWorld Today, “give us guidance about which vaccine candidates we should focus on in moving toward clinical trials. Doing some of the comparative rhesus monkey experiments we describe takes from a large list of possible variables what vaccines we want to test. So we selected three leading HIV vaccine categories: DNA vaccines in several formulations; a poxvirus vaccine called MVA modified vaccinia Ankara; and adenovirus the common cold virus.”

Three Vaccines Compete Head To Head

“Our work had suggested for a number of years,” Shiver said, “that adenovirus [AV] made for an exciting class of vaccine in terms of its ability to achieve the kind of immune responses we’re focusing on. That is, producing T-lymphocyte immunity, particularly CTLs that can kill viruses in infected cells.

“The way we did this preclinical research was to compare head to head how these three different vaccines produce CTL immune responses, and then challenge these monkeys with an AIDS virus, to see if our vaccines could protect them. These experiments confirmed our hunch that AV made the most potent vaccine we’d seen.”

Shiver and his consortium of 52 co-authors set up two successive monkey-immunizing and -challenging experiments Study A and Study B.

“These studies, which we exemplify in the Nature paper,” Shiver pointed out, “are strictly gag vaccine antigens. So antibodies that neutralize the virus are not an issue. This is the clearest proof of concept to date that purely cytotoxic T-cell responses against an AIDS virus provide significant protection.

“Studies A and B,” Shiver recounted, “used the same vaccine antigen against HIV the gag gene. Gag is a structural gene. No neutralizing antibody epitopes are directed against it. It is one of the most highly conserved genes within HIV, so it’s much less variable from one virus to another unlike the envelope. Study A compared in five different vaccinated monkey cohorts either this gag antigen, the MVA poxvirus vaccine or just adenovirus. Each of the animals received only one of those three vaccine types. Our idea was to see how each vaccine stacked up against the other two.

“Study B,” he went on, “was different in that we took combinations of our best vaccines from Study A. The most potent was adeno, followed by the DNA vaccine, with a polymer adjuvant. Those were the best two vaccines in Study A.

“In Study B,” Shiver said, “we primed the monkeys with one type of vaccine, DNA, then boosted them with the other type, MVA. This experiment showed that priming with DNA is probably the best way to prime its response for eventual boosting with an adenovirus-based vaccine.”

Monkeys Are Useful, But Not Human

“Based on such data,” Shiver observed, “Merck is now conducting Phase I trials in human patients, with the DNA vaccine, with or without adjuvants and with the adenovirus vaccine. They are taking place at clinical sites in 22 states plus D.C. The DNA vaccine studies started in December 1999; the adenoviral vaccine trials a little less than a year ago.

“We’re doing our Phase I studies in limited numbers of people,” Shiver said, “to look at these different vaccines and formulations, focused on safety data. When we have enough data, hopefully we’ll expand those studies into Phase II, and see what happens. That will take time. These are novel vaccines. The initial phases of testing are quite slow again because safety is our predominant concern. Also [there are] a lot of other variables whether or not to have an adjuvant with the DNA, and what adenovirus doses we would like to work with.”

Can these and impending monkey trials extrapolate to the human condition? Shiver’s answer is: “No, the only way we’re ever going to find out if our vaccine can produce immune responses that purely protect against HIV infection in humans [is that] we’re going to have to do Phase III efficacy studies. Monkeys are useful; generally, immune responses in a monkey would be more predictive of results in humans than, say, in a mouse.”

Back-to-back with this Nature paper is a cautionary companion article titled, “Eventual AIDS vaccine failure in a rhesus monkey by viral escape from cytotoxic T lymphocytes.” Its main authors are at Harvard Medical School. To this caveat, Shiver told BioWorld Today, “Their vaccine study is the continuation of a collaboration between our two groups. Their vaccines were actually made by our laboratory, so are fairly close to the DNA vaccine in our current Nature paper except they also have an envelope construct in their vaccine. This doesn’t say we would never have escape of virus with one of our best vaccines. But I think it certainly reduces the likelihood of that happening. And our goal,” Shiver concluded, “is to take our best vaccines forward into efficacy studies down the road.”