Hopes Dashed Again for HIV Preventive Vaccine

SAN FRANCISCO, Sept. 24 -- In another setback to the effort to create a preventive vaccine against HIV, a key trial has been halted because of a lack of efficacy, according to Merck and the HIV Vaccine Trials Network.

With 3,000 volunteers enrolled, a data monitoring committee called a halt to the so-called STEP trial because the vaccine was not blocking HIV infections.

Also, the vaccine did not reduce the level of HIV in the bloodstream of those who became infected - a second major goal of the trial.

Susan Buchbinder, M.D., of the San Francisco Department of Public Health, a study co-chair, said she was speechless when the data monitoring committee delivered its report. "I think we were all shocked."

But Dr. Buchbinder said there's a silver lining. The trial was designed to get a quick and clear answer to the efficacy question and it did so.

On one hand, she said, the researchers are "tremendously disappointed." On the other hand, "we got the answer as quickly as possible."

The study enrolled its first patient in December 2004, but enrollment didn't really get under way until 2005 - meaning that the study delivered "a clean result" in just over 30 months.

She said the study should be a model for future clinical trials in the HIV vaccine field, even though the result "took us all by surprise."

Her view was echoed by Wayne Koff, Ph.D., the research vice-president of the New York-based International AIDS Vaccine Initiative, which was not involved in the STEP trial.

"This was a candidate and the candidate has failed and we don't know why," Dr. Koff said. "We're going to learn a lot from it when we look at the data - and that's why we do clinical trials."

He added that while the trial delivered results quickly, "we still have to do it faster and better."

One possible lesson from the trial, Dr. Koff said, is that viruses used in animal models of HIV disease are not the same.

In this case, he noted the virus did very well protecting macaques against a laboratory virus called SHIV, which combines genes from HIV and from Simian Immunodeficiency Virus (SIV). But against the naturally occurring SIV, it had very little effect.

This is the first piece of data suggesting the SIV is a better predictor of vaccine efficacy, Dr. Koff said, although he cautioned that the vaccine used in the animal studies was not exactly the same as that used in the STEP trial.

The phase II international trial was regarded as "test of concept" of the vaccine - which included three recombinant HIV genes delivered by a weakened adenovirus vector - in uninfected volunteers at high risk for acquiring HIV infection.

The volunteers were mainly homosexual men, but included some women in North and South America, the Caribbean, and Australia.

A planned efficacy analysis looked at about 1,500 volunteers regarded as being most likely to do well because they began the study with low level immunity to the adenovirus vector.

But among the 741 volunteers who got at least one of the three planned doses of vaccine, the data monitoring committee saw 24 cases of HIV, compared with 21 among the 762 participants who got placebo.

In a subgroup of participants who had received at least two vaccinations and remained HIV-negative for at least the first 12 weeks of the trial, the researchers saw 19 cases of HIV in the 672 volunteers who received vaccine and 11 in the 691 who received placebo.

The vaccine also had no effect on HIV levels among those who became infected.

Two or three months after infection, the average level of HIV RNA in the bloodstreams of those who had been given vaccine was about 40,000 copies per milliliter of plasma, compared with 37,000 copies among those who got placebo.

Meanwhile, a parallel study in South Africa, which began earlier this year using the same vaccine, has been put on hold, according to the National Institute of Allergy and Infectious Diseases, one of its sponsors. The institute is also the sponsor of the vaccine trials network.

That study, known as the Phambili trial, is examining the effect of the vaccine in a region where HIV is mainly causing a heterosexual epidemic.

The vaccine was intended to generate a cell-mediated immune response. The hope was that the three HIV genes - gag, pol, and nef -- would stimulate the body to generate an HIV-specific immune response through the body's own CD8 T-cells.

In fact, Dr. Buchbinder said, early data seem to show that the vaccine did stimulate an immune response. "It just wasn't a protective response," she said, and researchers will now have to go back to animal studies and earlier human trials to see why not.

Dr. Koff noted that some 30 HIV vaccine candidates remain in play, all of them aimed at generating cell-mediated immunity "because we don't know how to generate an antibody response."

But he said it's extremely likely that an effective HIV vaccine will have to do both.