VaxGen vaccine trial fails the test but may offer insights

Presentation of data draws criticism as premature’

AIDSVAX, the first AIDS vaccine to complete human trials, showed disappointing results, with only a 3% protection rate. However, researchers were encouraged by preliminary data showing that the antibody response to the vaccine offered protection, particularly in a subgroup of African-Americans and Asians.

"When all volunteers were analyzed, the vaccine did not appear to be effective," says Michael Para, MD, a principal investigator in the study. "On the other hand, the black/Asian subgroup did appear effective."

AIDSVAX reduced the rate of infection by 3.8% in people who received it compared to a control group that received placebo injections — far from the 70% minimum efficacy experts estimate is needed to gain approval for widespread use.

VaxGen officials had said an efficacy of 30% might be enough to make the product useful in some populations.

The vaccine appeared to be effective, however, in a subgroup of subjects, notably African-Americans. Among them, 2% who received the vaccine became HIV-infected, compared to 8.1% of the placebo group — a statistically significant difference.

When Asians and mixed-race volunteers were added to the group of blacks — a total of about 500 of 5,000 volunteers — the protective effect was nearly as strong.

While the findings in minorities could have worldwide implications, vaccine experts cautioned it was dangerous to jump to conclusions so early.

"Until the data is crunched and the analyses are finished, it is completely premature to determine whether or not AIDSVAX is truly effective for particular populations," says Pat Christen, executive director of the San Francisco AIDS Foundation.

The Gay Men’s Health Crisis (GMHC) in New York City, the nation’s oldest AIDS organization, criticized the company for putting a positive spin on the results and offering false hope, particularly to minority populations.

"Subset analyses are problematic in the best of cases. With small numbers of African-Americans and Asians in the trial and wide confidence intervals associated with the results, making any statements about efficacy in this subpopulation is grossly premature," says Gregg Gonsalves, director of treatment and prevention advocacy at GMHC. "VaxGen’s assertions of its vaccine’s efficacy among blacks are based on 13 infections in this population in a trial of more than 5,000 participants. The assertions about efficacy among Asians are based on only four HIV infections in the study."

Some AIDS researchers went beyond expressing disappointment in the results to call for new strategies in vaccine development. The results "require that we reassess our strategy of vaccine development," says AIDS Health Foundation president Michael Weinstein. "Clearly, work should continue where the trial showed promise, but the general results suggest that 20 years into this pandemic, we really need to stop and take a long, hard look at this issue."

The foundation’s chief of medicine, Charles Farthing, MD, who doubts that a preventive vaccine can be developed any time soon, suggested more vaccine resources be shifted to providing antiretroviral therapy in poor countries.

In coming months, review of the data will focus on the robustness of the findings, particularly the efficacy reported in the subgroups. The enrollment in those groups was so small that a few more or less infections could have made the results insignificant. For example, there were only 13 infections among blacks.

"This data could have been all over the map," says one investment analyst. "Two patients in the black group could have completely changed the statistical significance of this study."

"Yes, you have to be skeptical," agrees Phillip Berman, PhD, VaxGen’s senior vice president of research and development and inventor of the vaccine. "When you look at just the blacks, you are starting to talk about a few infections here and there. On the other hand, the antibody response in the people who were protected was clearly higher. Also, it looks like the vaccinated persons got infected with a different [strain of] virus than the placebo ones. And in the few people who were infected, you see better CD4 response and a little better viral suppression. So, yes, the numbers are small, but you start to get a consistent story in the analysis, so each of those things makes me increasingly comfortable with the small numbers. We are on a track."

Whatever the vaccine’s future, AIDS organizations called the completion of the trial, which had to overcome substantial and unique ethical and regulatory hurdles, a feat in itself.

"We applaud VaxGen for showing that a Phase III trial is possible for an AIDS vaccine and for proving successful enrollment and retention of trial participants is possible," says Christen. "It is a remarkable and historic accomplishment."

Participation was high

A total of seven inoculations of the vaccine were given over three years to high-risk gay and bisexual men and female partners of injection drug users. At least 5,000 received one vaccination, and more than 4,500 completed the trial. The high level of participation and completion was "truly remarkable," Para added.

"We were disappointed the vaccine did not induce a reduction in all populations," says Para. "Certainly the vaccinated subjects didn’t show a reduction in HIV infection. However, the differences in infection rates in the black/Asian subgroup was quite remarkable."

The trial analysis showed no real difference in efficacy by age, geographic location, education, and most importantly, sexual behaviors, in any groups, particularly in minorities, he points out.

The trial found no evidence that the vaccine caused any safety problems or enhanced susceptibility to infection, which were two main barriers to getting Food and Drug Administration (FDA) approval, Para notes.

Vaccine recipients had a slightly higher rate of pain, swelling, and tenderness at the injection site compared to placebo recipients, he says.

A separate Centers for Disease Control and Prevention study indicated that volunteers did not increase their risk behaviors, and VaxGen’s preliminary analysis of the trial data indicates that risk behavior was reduced in both the placebo and vaccine groups.

The history of the gp120 vaccine has been a rocky one, beginning with the National Institutes of Health’s decision in 1996 not to fund a Phase III trial of the vaccine after "breakthrough" infections were reported in smaller trials. Undaunted, Donald Francis, MD, VaxGen’s cofounder and one of the first public health experts to claim that AIDS was caused by a virus, reorganized and sought private backing to complete the trials.

"The results from this groundbreaking effort will provide new insights into HIV and hopefully pave the way to even more effective vaccines," Francis says.

Indeed, the vaccine may provide the first evidence in humans that antibodies from a surface protein vaccine correlate with the protection needed for an effective vaccine.

"The correlates of protection are going to be very important," says one vaccine researcher. "We are dying for that information."

The antibodies from the immune response to the vaccine were higher in the black and Asian volunteers. Likewise, they were higher in those who remained uninfected. Also, it appeared that those who were vaccinated and became infected had lower viral loads and higher CD4 counts than those infected in the placebo group — an indication that the vaccine may have a therapeutic effect as well, Para notes.

Curiously, white and Hispanic volunteers seemed to develop consistently lower levels of protective antibodies following vaccination. VaxGen intends to conduct additional analyses to confirm if there was a direct correlation between the level of antibodies and the prevention of infection.

"We’re not sure yet why certain groups have a better immune response, but these preliminary results indicate that a surface-protein vaccine that stimulates neutralizing antibodies correlates with prevention of infection," says Para.

Among the factors the company will examine more closely are behaviors, geographic location, age, and sex.

"We need to investigate these and other possibilities before we can say it’s a real difference," Berman says. "There also could be differences in virus infectivity and strain variation."

VaxGen is nearing completion of its Phase III trial in Thailand, which is testing a formulation of AIDSVAX designed to protect against HIV subtypes B and E. The company expects to announce results of that trial in the second half of 2003. Subtype E is prevalent in Southeast and East Asia and the Central African Republic. Unlike the AIDSVAX B/B trial, which tested the vaccine against sexual transmission of the virus, the trial in Thailand is testing the vaccine against infection acquired by injection drug use.

The results from that trial could provide additional answers and possibly support the findings from the U.S. trial. "It could provide confirmation of what we saw here," Berman says.

No timetable for FDA review

While it is almost certain the FDA would not approve the vaccine for use in the general population, it is possible that with further study and changes AIDSVAX could be licensed for the subgroups, making it the first time a vaccine was approved for a subpopulation only.

VaxGen officials would not predict how soon the FDA would complete analysis of the data, nor what it would make of the findings in the subgroup. One possibility is that further studies would be required in that group.

"It’s our current objective to follow up on current formulation and do whatever additional studies are necessary for licensure," Berman says. "In parallel, we will look at factors that underlie differences in response and protection, and that may be formulation changes or other biotechnology changes."

The changes could range from enhancing the vaccine by adding more immune modulators and adjuvants to simply changing dosing and administration schedules. "We will be following all leads, including looking at the genetic backgrounds," Berman says.

Because the vaccine is relatively simple compared to other candidates, teasing out what works and what doesn’t would be relatively easy, requiring 20 volunteers instead of 2,000.

"Having a single protein design gives us a simpler case to study than with a live organism or split vaccine," Berman notes. "We know exactly what is in the product. So the possibility to nail this is much higher than with older technologies."

Had AIDSVAX proved highly effective and headed for FDA approval, the company was ready to ramp up production of the vaccine at several new facilities and have it available by 2005.

Currently, VaxGen is also in the early stage of developing a vaccine against HIV subtype C, prevalent in sub-Saharan Africa, India, and China. The company is committed to developing increasingly effective formulations of AIDSVAX that target all HIV subtypes, says CEO Lance Gordon, PhD.

"We intend to continue development of this vaccine through licensure, including additional studies as necessary, for use in groups in which the vaccine demonstrated a significant reduction in infection," he says.