Decades of research have failed to produce a slam-dunk AIDS vaccine, but the fruitless efforts have proven a boon for basic immunologists, adding new tools and knowledge to their repertoire.

Most of these gains derive from something that initially stymied HIV vaccine researchers: animal models don't accurately predict which AIDS vaccine will work in people. Rather than test their candidates in mice and nonhuman primates, researchers were forced to assess them in people.

Because of this, “there's been an explosion in the field of human immunology,” says Rick Koup, chief of immunology at the US National Institutes of Health's Vaccine Research Center. “Before, the vast majority of immunology research was done in mice, and there was almost no strong basic immunology work going on in either humans or nonhuman primates,” Koup says.

Before, ...there was almost no strong basic immunology work going on in either humans or nonhuman primates. Rick Koup, US National Institutes of Health's Vaccine Research Center.

When scientists began studying human immunology, they realized the picture was much more complicated than previously thought. This first became apparent when HIV vaccine researchers tried to mimic traditional vaccines by injecting proteins from the coat of the HIV virus. It didn't work.

The first HIV vaccines stimulated antibodies, just like more traditional vaccines. But the antibodies generated by the first AIDS vaccines did not prevent the virus from infecting people, or from progressing after the initial infection.

Scientists then began working on an HIV vaccine that could stimulate robust cellular response in addition to the strong antibody response. But even vaccines that invoke a strong cellular response did not protect people from HIV infection. This stumbling block has led to another breakthrough in understanding of the immune system: the cellular immune response to vaccines is itself quite nuanced.

Scientists have since realized that chronic infections, such as HIV, interact with the immune system in a completely different way than do acute infections. For instance, Giuseppe Pantaleo and coworkers have found that the human CD8 T-cell response to HIV infection is skewed so that less protective cells dominate (Nature 410, 106; 2001). His group has also shown that HIV-positive people who stay relatively healthy over a long time have a more diverse helper T-cell repertoire than those who rapidly become ill (Blood 103, 966; 2003).

Eliciting the right balance of T-cell subpopulations is now a main goal in HIV vaccine research. Scientists are aided by the development of new technologies, such as flow cytometry, that allow the in-depth study of the immune system, and that were developed in part by the HIV vaccine field.

By examining why HIV vaccines have failed, researchers have also learned more about the immune factors—dubbed 'correlates of protection'—that make any vaccine work. That has led to the idea of rational vaccine design, in which researchers try to engineer a vaccine to elicit a particular protective response. If it works, this idea of protection by design—rather than by a lucky guess—may be the HIV vaccine field's most lasting legacy for immunology.