Did patients in a failed HIV vaccine trial halted in 2007 become more susceptible to the virus due to the adenoviral vector used to deliver the experimental vaccine? Researchers have speculated this may have been the case, and a new study proposes a mechanism for how this could have occurred.

The in vitro study, published in the Proceedings of the National Academy of Sciences (PNAS) this week, reports that immune cells from people previously exposed to an adenovirus serotype 5 (Ad5) -- similar to the vector used in the Merck-funded STEP trial -- generate activated T cells primed to migrate to mucosal surfaces such as the gut, where they are more available to HIV infection.

"I'm not saying you should rule out adenovirus for all time, but you should be aware that there are possible problems with the use of it," Imperial College London vaccine researcher and main author Steven Patterson told The Scientist.

The researchers stimulated T cell production in vitro by bombarding immune cells from 20 healthy adults with Ad5. The T cells generated in cultures from Ad5-seropositive participants expressed more α4β7 integrins and CCR9, proteins that help T cells migrate to the intestinal lining. In those people, "between 70 and 80% of [T] cells that proliferated in response to Ad5 were α4β7 positive," Patterson noted.

Based on these results, STEP trial participants with no prior exposure to Ad5 would have produced less "mucosal homing" T cells -- that is, T cells that head for the gut -- when jabbed with the Ad5 vaccine, the researchers speculated. "If it's the first encounter with the virus and it's through the intramuscular route, it will tend to induce a response where most of the cells are not mucosal homing," Patterson said.

But research published in Nature Medicine in July contradicted this suggestion, concluding that Ad5 played little, if any, role in the boosted infection rates seen in the STEP trial. In those studies, Harvard immunologist Dan Barouch and University of Pennsylvania researcher Michael Betts and their colleagues analyzed blood samples from human trials preliminary to the phase IIb STEP trial. Barouch found that it was Ad5-seronegative -- not Ad5-seropositive -- STEP volunteers who experienced the largest increase in Ad5-specific T cells. Betts showed that the T cells produced after vaccination in Ad5-seropositive vs. Ad5-seronegative individuals didn't differ in the cytokines they expressed and would not have migrated differentially to mucosal linings.

"While the results found by Benlahrech et al. are provocative and likely accurate for their in vitro model," Betts wrote in an email to The Scientist, "no such upregulation of mucosal targeting receptors was found in two independent studies published in Nature Medicine performed on samples directly isolated from Ad5-vaccinated humans."

"No data from humans vaccinated with Ad5 vectors are described in this paper, and thus the implications of these findings for clinical trials remain unclear," concurred Barouch in an email to The Scientist. "The findings reported in the Benlahrech paper are derived from an artificial in vitro experimental system and do not appear to be consistent with the data obtained by two independent groups utilizing samples from the human clinical trials of the actual Merck Ad5-HIV vectors."

Patterson contends that Barouch and Betts may have simply missed the mucosal homing CD4 T cells in their analyses of blood samples from the STEP trial participants. "The reason they didn't see them is because they migrated out of the blood to the mucosal tissues," Patterson said. "That's the easiest explanation."

"We're still puzzling over the STEP trial," Pat Fast, chief medical officer of the International AIDS Vaccine Initiative, told The Scientist. "This [paper] raises a lot of really important questions and points a way towards much more detailed analysis in phase I clinical trials of immune responses."

Meanwhile, researchers in California are suggesting a different approach to HIV vaccine vectors. In a paper also published in this week's PNAS, California Institute of Technology biologist David Baltimore and colleagues show it may be possible to deliver a vaccine directly to the immune system's dendritic cells using a lentiviral vector wrapped in glycoproteins derived from a Sindbis virus. They successfully tested a vaccine in mice and are starting trials now in non-human primates. Lentiviruses and Sindbis viruses are far less common in the general population than are adenoviruses, and some suggest that this may make them safer vectors for HIV vaccines.

Though uncertainty continues to surround the role of the Ad5 vector's role in the STEP trial's premature failure, Baltimore said that using a lentivirus instead of an adenovirus as an HIV vaccine's vector simply seems to be more effective.

"Our argument is that it's better than adeno[vectors], regardless of whether there's a safety issue or not," he told The Scientist.

The NIH's PAVE 100 trial is currently testing an Ad5-vectored candidate.


Related stories:

Vector did not kill HIV trial
[20th July 2009]

AIDS vaccine trial set to start
[3rd June 2008]

HIV vaccine trials stopped
[25th September 2007]