Conclusions

Discussions on strategies of how to develop HIV vaccines are often hampered by the confrontation of two truisms: One states that the more information we obtain from basic research, the better off are we to develop more effective HIV vaccines. The other argues that laboratory research alone will never be a substitute for large-scale clinical trials to obtain definitive information on vaccine efficacy. A passionate and uncompromising defense of either position will not help those people who have to take the practical decisions, nor will this effectively promote HIV vaccine development in general. Thus, a sensible strategy is to accept the uncertainties of proceeding with efficacy trials of available products which have met previously defined minimal requirements [134], and at the same time continue basic research to obtain additional information on the nature of protective immune responses in humans, some of which would likely be derived from the efficacy trials themselves.

From a practical point of view, it would be important to address three questions: (1) What type of additional information is necessary to proceed to efficacy trials with the greatest likelihood of success? (2) How realistic are the expectations that relevant information will be obtained from additional laboratory, animal protection or natural history studies in the absence of efficacy trials? (3) From the candidate vaccines which have entered Phase I/II trials, are there products which meet minimal conditions to proceed to Phase III trials?

Answering the above questions is not easy. Since natural immune responses to HIV are complex (including both humoral and cellular responses) and obviously not very efficient, focusing laboratory research on limited aspects of the human immune response to HIV infection and disease may lead in false directions. Likewise, great uncertainties remain concerning the relevance of animal models as predictors of vaccine efficacy in humans.

Several candidate vaccines, based on different concepts, are at different stages in the HIV vaccine development ‘pipeline’. Candidate vaccines based on the subunit recombinant envelope concept and produced in mammalian cells, have been shown to protect chimpanzees from HIV-1 infection, and to be safe and reasonably immunogenic in humans, inducing neutralizing antibodies. A second generation of candidate vaccines, which are based on live vectors expressing the envelope and other HIV-1 genes, and which are capable of inducing CTLs, are beginning to be evaluated in human trials. Newer generations of candidate vaccines now being mostly explored in animal experiments are using combinations of subunit recombinant proteins or live-vectored vaccines with other immunogens (such as synthetic peptides or pseudovirions), or are based on more novel approaches, including nucleic acid immunization and perhaps whole-inactivated or live-attenuated vaccines.

With our present state of knowledge, it is not possible for laboratory assays to accurately predict which vaccine concept, or concepts, will induce protection against HIV infection in humans. Unless major advances are made in our understanding of the nature of protective immune responses to HIV-1 in humans, that information will only be obtained through the conduct of Phase III field efficacy trials. However, in view of the rate of progression of the HIV pandemic, especially in developing countries, it would not be ethical to wait in the hope that such advances will occur soon, thus postponing trials with candidate vaccines. In fact these trials, conducted in parallel or sequentially, may represent our best chance to enhance our basic knowledge of the nature of protective immune responses to HIV infection.

Thus, in order to avoid the unacceptable alternative of perpetual uncertainty, or to delay the development of a much needed vaccine, there is no other choice but to effectively integrate further basic research with the initiation of large-scale field efficacy trials in the process of HIV vaccine development. These Phase III trials will present unique opportunities to: (1) establish if different vaccine concepts can induce protection in humans; (2) validate the primate models presently being used in HIV vaccine research; (3) obtain information on immune correlates of vaccine-induced protection; (4) explore the significance of viral genetic variability in relation to vaccine-induced protection; (5) evaluate different end-points for vaccine efficacy (prevention of infection, establishment of chronic infection, or disease); and (6) generate additional data on vaccine safety.

The development of an HIV vaccine will be a long and difficult process. Multiple efficacy trials and case-control studies will ultimately be required before a safe, effective and affordable vaccine is available for widespread public health use. With more than 6000 new infections occurring every day worldwide, there is urgency to proceed.