Although they establish the feasibility of combining zidovudine with cellular immune reconstitution in treating patients with HIV infection, our results show that any benefits from such combination therapy are at best transient. Future attempts at cellular immune reconstitution may need to use improved antiretroviral regimens as well as immunization of donors with HIV-specific antigens.

Development of an effective vaccine for prevention of infection with HIV would provide an important mechanism for controlling the AIDS epidemic. In the current study, the first clinical trial of a candidate HIV-1 vaccine initiated in the United States, the safety and immunogenicity of escalating doses (10-1,280 Aig) of recombinant gp160 (rgpl6O), were evaluated in 138 HIV-negative volunteers. Maximal antibody responses, as evaluated by ELISA, were seen after immunization with three doses of 1,280 ;&g rgpl6O. Responses to some specific epitopes of HIV gp160, including the second conserved domain and the CD4 binding site, were seen more frequently than after natural infection. Neutralizing antibodies to the homologous HIV strain, but not heterologous strains, were induced by this regimen. Blastogenic responses to rgpl60 were seen in most volunteers receiving at least two doses of 2 20 Mg. These envelopespecific T cell responses were also seen against heterologous strains of HIV. No major adverse reactions were seen after immunization. Thus, rgpl60 is a safe and immunogenic candidate HIV vaccine; further studies are needed to determine if it will provide any clinical benefit in preventing HIV infection. (J. Clin. Invest. 1993. 92:919-928.)

A Phase I study of recombinant interferon-gamma (rIFN-gamma) was conducted to determine the toxicity and pharmacokinetics of this lymphokine in acquired immunodeficiency syndrome (AIDS) patients with Kaposi's sarcoma (KS). Sixteen patients with AIDS/KS were entered into a fixed-dose trial at either 0.001, 0.01, 0.1, or 1.0 mg/m2 of rIFN-gamma. rIFN-gamma was initially administered either as a single 24-hr continuous iv infusion or as a single im injection, followed 4 days later by a 10-day course of daily therapy by the same route. Following a 1-week washout period, this sequence of administration was then repeated, with the drug given by the alternate route. Pharmacokinetic analysis of the 1.0-mg/m2 group revealed that peak serum levels of up to 153 U/ml occurred 2-4 hr after im injection and that steady-state levels of up to 40 U/ml were reached approximately 7-12 hr after beginning iv infusion. Dose-related toxicities in this trial included fever, headache, fatigue, nausea, and hepatitis, all of which were most severe at the two highest doses. Dose-dependent depression of the total white blood-cell (WBC) count, affecting both granulocytes and lymphocytes, was the most common laboratory abnormality. Natural killer (NK)-cell activity was slightly enhanced at a dose of 0.1 mg/m2 but suppressed at 1.0 mg/m2 of drug; monocyte-mediated cytotoxicity, in contrast, was significantly increased only at the highest dose. No dose-related changes were noted in KS lesions, HLA-DR expression by peripheral blood mononuclear cells, lymphocyte blastogenesis, or the ability to culture cytomegalovirus (CMV) from body fluids. We conclude that a maximally tolerated dose (MTD) for this drug is in the range of 0.1-1.0 mg/m2 and that at least modest evidence of systemic immunomodulation may be seen when rIFN-gamma is given at doses at or near this MTD.

Our objective was to map serial patterns of Western blot reactivity over time of a cohort of initially ELISA-negative, Western blot-indeterminate individuals from a high-risk group and to determine if these individuals were at increased risk of harboring occult HIV-1 infection. A 2-year prospective study used serial ELISA, two types of Western blot, immunologic profiles, HIV-1 culture, and analysis by polymerase chain reaction. Subjects were 20 ELISA-negative, Western blot indeterminate homosexual volunteers and 20 matched seronegative controls. Results showed that 19 of 20 study subjects completed a mean of 17.0 months of clinical and laboratory follow-up. Reactivities with p24 and/or with p55 were the two most commonly observed Western blot patterns, occurring in 70% of individuals. Specific Western blot reactivity was dependent upon the particular immunoblot preparation being used and varied considerably on a longitudinal basis. No individual pattern appeared predictive of an increased likelihood of subsequent seroconversion to HIV-1 relative to controls. By all other criteria including polymerase chain reaction analysis, samples from 17 of 19 individuals remained negative for HIV-1 at each time point. Two individuals evolved from an indeterminate to a positive Western blot and, simultaneously, from a negative to a positive polymerase chain reaction analysis, during follow-up. Our conclusions were as follows. ELISA-negative, Western blot-indeterminate individuals from a high-risk group show marked variability in immunoblot findings over time, and these patterns do not appear predictive of an increased likelihood of infection.(ABSTRACT TRUNCATED AT 250 WORDS)