This randomized, double-blind, placebo-controlled, phase II/III study was designed to evaluate safety, immunogenicity, and efficacy of an active anti-interferon-α (anti-IFN-α) vaccine in asymptomatic HIV-1-infected patients. The active immunization was aimed at inducing anti-IFN-α antibodies to counteract IFN-α overproduction. In all, 242 patients, recruited between December 1995 and July 1996 in eight centers in Europe and Israel, with CD4+ counts from 100 to 634 cells/mm3 who were receiving or not receiving antiretroviral therapy (including protease inhibitors) were randomized to receive either anti-IFN-α vaccine or placebo. The anti-IFN-α immunization regimen consisted of three priming injections delivered intramuscularly at 1-month intervals in a water-in-oil emulsion of inactivated recombinant IFN-α-2b (i-IFN-α) followed by intramuscular booster injections of i-IFN-α adsorbed onto calcium phosphate every 3 months. Immunogenicity to vaccine was defined as an increase of anti-IFN-α antibody level of more than twofold the preimmunization value. Clinical progression, changes in antiretroviral treatment, and decrease of CD4+ counts to <200 cells/mm3 were considered endpoints for efficacy evaluation. Contrary to our previous experience, in which six to seven oil priming injections induced a >90% response rate, the three oil-adjuvanted injections in this trial were suboptimal because only 40 of 122 vaccinees (33%) had raised anti-IFN-α antibody following immunization. In vaccinees, both antibody responders (AbRV) and nonresponders (AbNRV), the tolerance to the vaccine was good and was without evidence of significant safety concerns. During the course of the trial, 62% of vaccine responders, 64% of nonresponders, and 63% of placebo patients elected to add protease inhibitor-containing regimens as new treatment guidelines were established, resulting in a marked decrease in clinical and laboratory progression such that the expected endpoints of the study could not be achieved and further follow-up was halted. Despite the unexpectedly low immunogenicity and fewer than expected endpoints, anti-IFN-α vaccine recipients, in comparison with placebo recipients, showed a lower rate of disease progression, nonelective treatment changes, and/or CD4+ count decrease to <200 cells/mm3, but the difference was not statistically significant. Nevertheless, the subgroup of patients immunized to IFN-α who experienced a rise in anti-IFN-α antibodies had a significantly lower rate of occurrence of HIV-1-related events and of any combination of the endpoints compared with those of either placebo patients or vaccinees who failed to develop anti-IFN-α antibodies, the latter two groups behaving similarly. Further studies of this approach are warranted because these data suggest a beneficial effect of this adjuvant approach.