Towards an AIDS vaccine: the role of nonhuman primates

Over the last 10 years, about 20 human immunodeficiency virus (HIV) vaccine candidates have been tried in humans, with disappointing results as gauged by limited immune responses or protection against infection. These difficulties suggest that a new strategy is needed to test systematically new vaccine candidates. That opportunity is now afforded by nonhuman primate models with SIV, which have been shown to provide an excellent mirror of HIV infection in humans. The recent introduction of SHIVs, chimeric viruses that carry the HIV envelope and are able to infect and cause AIDS in monkeys, also has added an important additional research tool. These models can be used to address a series of questions, including the following: (1) Can protection be provided by partial immunity or is sterilizing immunity required? (2) What are the immune parameters that best predict protection against a potentially pathogenic challenge? (3) What role does mucosal immunity play and can it be induced by practical modes of immunization? (4) Can an attenuated virus be selected that is both protective and safe? An orderly strategy for the evaluation of vaccine candidates could be adopted that would involve several phases: (a) the selection of a limited set of challenge models, ranging from very severe to mild and requiring consideration of primate species, age, route of infection, and challenge viruses; (b) the assessment of candidate vaccines using comparable virus challenges; and (c) accelerated testing in humans of any candidate vaccines that have met a 'proof of efficacy' in primates.