1. Home>
2. Specialties>
3. HIV/AIDS Clinical Care>
4. News in Context

HIV-1 Vaccine Might Increase Infection Risk in Certain Subgroups

The HIV-1 vaccine used in the STEP trial has already been shown to be ineffective, but might it actually cause harm?

Researchers are now concerned that the experimental HIV-1 vaccine used in the STEP trial might increase susceptibility to HIV-1 infection in certain populations. Previous analyses from the trial focused exclusively on the 1500 study participants who had low levels of preexisting immunity to the vector used in the vaccine (adenovirus serotype 5 [Ad5]). When the vaccine failed to prevent HIV-1 infection in these individuals, the Data and Safety Monitoring Board recommended that immunizations be halted (ACC Oct 1 2007), and researchers began to analyze data from the larger study population, which also included people with higher levels of preexisting Ad5 immunity.

Overall, 49 cases of HIV-1 infection occurred among the 914 men who received at least one dose of vaccine, compared with 33 cases among the 922 men who received placebo. Among the 1058 men who had low baseline levels of Ad5 immunity (titer 200 units), the number of new HIV infections did not differ substantially between the vaccine and placebo groups (28 vs. 24). However, a more pronounced difference (21 vs. 9) was seen among the 778 men who had high baseline levels of Ad5 immunity (titer >200 units). Unfortunately, the two populations are not identical; for example, the population with high Ad5 titers was younger, less likely to live in the U.S., and less likely to be circumcised. Thus, confounding factors could not be ruled out. Nevertheless, these results suggest that the vaccine might have increased susceptibility to HIV-1 infection in individuals with high levels of preexisting immunity to Ad5.

The mechanisms underlying these puzzling results are not clear, and further studies are needed. In the meantime, the study participants are being told whether they received vaccine or placebo and what their adenovirus titers were. They will also receive additional risk-reduction counseling and will continue to be followed by study investigators.

Comment: Several points are worth noting. First, the vaccine itself did not cause HIV infection, although it may have made some people more vulnerable. Second, the individuals who became infected during this study (whether in the vaccine group or the placebo arm) continued to engage in very high-risk sexual behavior. In contrast, those who reduced such behavior did not become infected, thereby reinforcing the importance of risk-reduction measures. Finally, I think that the study investigators did the right thing by unblinding the participants. Doing otherwise would have offered some statistical and scientific advantages, but the safety of the volunteers comes first, and, to that end, unblinding was best.

Overall, these results have massive implications for future vaccine trials. Before the findings were released, the HIV Vaccine Trials Network (which cosponsored this study with Merck) was poised to start PAVE 100, a large trial using a DNA prime/Ad5 boost vaccine developed by the NIH. However, this trial is now on hold, and the AIDS Vaccine Advisory Committee will meet soon to decide how to proceed.

— Carlos del Rio, MD

Dr. del Rio is the Principal Investigator for the Atlanta site of the STEP trial and is an HIV Vaccine Trials Network co-investigator.

Published in Journal Watch HIV/AIDS Clinical Care November 19, 2007

Citation(s):

Data from STEP study presented at open scientific session confirm Merck’s investigational HIV vaccine was not effective [press release]. Seattle: Merck & Co., Inc., and HIV Vaccine Trials Network; Nov 7 , 2007. (http://www.hvtn.org/media/pr/step.html)

2011 Year in Review