A paper published online by the Journal of Experimental Medicine (JEM) on Monday describes a mechanism by which the Merck HIV vaccine may have enhanced the risk of HIV infection among some participants in the recent STEP trial. The Merck vaccine used a crippled form of a cold virus called adenovirus serotype 5 (Ad5) as a carrier (called a vector) for the HIV components it aimed to induce immune responses against, and the STEP results showed that people with pre-existing antibody responses against Ad5 (because of a natural exposure in the past) faced an elevated risk of acquiring HIV infection if they received the vaccine. In contrast, vaccine recipients in STEP with low or absent antibody responses against Ad5 were unaffected.
The new research paper, from Guiseppe Pantaleo’s laboratory at the University of Lausanne, indicates that the presence of anti-Ad5 antibodies causes the Ad5 vaccine to trigger the development of "immune complexes" consisting of the vaccine, anti-Ad5 antibodies and dendritic cells. The researchers show that the type of immune response induced by the vaccine vector in the presence of anti-Ad5 antibodies is somewhat different due to the formation of these immune complexes. In particular, the type of cytokines produced differ (in the presence of the complexes, there is a more inflammatory profile) and the CD8 T cell responses triggered are less polyfunctional than those induced when anti-Ad5 antibodies are not present. Although the paper only looks at Ad5-specific CD8 T cell responses, a study presented at the recent HIV vaccine conference (abstract appended below) offers evidence that the HIV-specific CD8 T cell responses induced by the vaccine may have been similarly affected. If confirmed, these findings might provide an explanation as to why a statistically significant inverse correlation was seen between the magnitude of the vaccine-induced HIV Gag-specific CD8 T cell response and post-infection viral load setpoint, but only among study participants who lacked anti-Ad5 antibodies (i.e. the HIV-specific CD8 T cell responses among Ad5-negative individuals may have been more functional).
The JEM paper also reports that in vitro HIV replication is enhanced roughly threefold in the presence of the immune complexes. However, the only virus used in this experiment is the LAV isolate which enters cells via the CXCR4 co-receptor. Since almost all HIV transmission involves isolates that use the CCR5 co-receptor, it is somewhat mystifying that more HIV variants were not tested (particularly given the publicity that this paper was likely to generate). Additional research will be needed to ascertain if this enhancement of HIV replication holds true for other isolates. Another potential outstanding issue is that immunization with Ad5 vectors induces Ad5-specific antibodies, so once the Ad5-negative negative participants in STEP received the first of their three immunizations they would also have had circulating antibodies, yet their risk of acquiring HIV infection did not appear to be raised. Perhaps it is possible that antibodies induced by natural exposure to Ad5 are different from those induced by Ad5 vectors, but this needs to be specifically evaluated.
Published online November 3, 2008
doi:10.1084/jem.20081786
The Journal of Experimental Medicine
BRIEF DEFINITIVE REPORT
Matthieu Perreau1, Giuseppe Pantaleo1,2, and Eric J. Kremer3,4
1 Service of Immunology and Allergy, Department of Medicine, Centre Hospitalier, Universitaire Vaudois, Lausanne, University of Lausanne, CH-1011 Lausanne, Switzerland
2 Swiss Vaccine Research Institute, 1011 Lausanne, Switzerland
3 Institut de Génétique Moléculaire de Montpellier, Centre National de la Recherche Scientifique 5535, 34293 Montpellier, France
4 Université Montpellier I and II, 34967 Montpellier, France
The STEP HIV vaccine trial, which evaluated a replication-defective adenovirus type 5 (Ad5) vector vaccine, was recently stopped. The reasons for this included lack of efficacy of the vaccine and a twofold increase in the incidence of HIV acquisition among vaccinated recipients with increased Ad5-neutralizing antibody titers compared with placebo recipients. To model the events that might be occurring in vivo, the effect on dendritic cells (DCs) of Ad5 vector alone or treated with neutralizing antiserum (Ad5 immune complexes [IC]) was compared. Ad5 IC induced more notable DC maturation, as indicated by increased CD86 expression, decreased endocytosis, and production of tumor necrosis factor and type I interferons. We found that DC stimulation by Ad5 IC was mediated by the Fcreceptor IIa and Toll-like receptor 9 interactions. DCs treated with Ad5 IC also induced significantly higher stimulation of Ad5-specific CD8 T cells equipped with cytolytic machinery. In contrast to Ad5 vectors alone, Ad5 IC caused significantly enhanced HIV infection in DC–T cell cocultures. The present results indicate that Ad5 IC activates a DC–T cell axis that, together with the possible persistence of the Ad5 vaccine in seropositive individuals, may set up a permissive environment for HIV-1 infection, which could account for the increased acquisition of HIV-1 infection among Ad5 seropositive vaccine recipients.
AIDS Vaccine 2008, Cape Town, South Africa, October 13-16, 2008
Abstract# OA05-04
Effects of Pre-existing Vector Immunity on the Quality of Vaccine-induced T-Cell Responses in an Ad5-HIV-1 Vaccine Trial
SO Pine1, J Kublin2, D Casimiro3, S Hammer4, S De Rosa5, and MJ McElrath6 1 University of Washington and Fred Hutchinson Cancer Research Center, Seattle, WA, USA; 2Fred Hutchinson Cancer Research
Center and the HVTN, USA; 3HIV Vaccine Trials Network and Merck Research Laboratories, West Point, PA, USA; 4HIV Vaccine Trials Network and Columbia University, New York, NY, USA; 5Vaccine and Infectious Disease Institute, FHCRC, Seattle, WA, USA; 6Department of Medicine, University of Washington, HVTN and FHCRC, Seattle, WA, USA
Background: The challenges facing HIV-1 vaccine development were recently underscored by the lack of efficacy demonstrated in the STEP Trial with testing of a replication-deficient adenovirus serotype 5 (Ad5) construct. Although the vaccine effectively stimulated HIV-specific T-cell responses, natural immunity to the vector itself appeared to have an impact on the study outcome. Little is known about how pre-existing vector immunity affects the quality of T-cell response to a vaccine. To address this, we examined peak responses in a previous phase I trial (HVTN050/ Merck V520 Protocol 018) that tested a MRKAd5 HIV-1 Gag vaccine in persons with and without previous Ad5 immunity. Methods: Pre-existing Ad5 immunity was quantified by baseline neutralizing antibody (nAb) titers, and vaccinees were stratified as either Ad5 naive (n = 30, nAb 18) or Ad5-immune (n = 30, nAb > 200). An ex vivo T-cell assay to measure 30 cytokines simultaneously was combined with intracellular cytokine staining to allow a thorough evaluation of the quality of antigen-specific T-cell responses in vaccinees. Results: Preliminary analyses showed significantly higher amounts of IL-15, IL-2, IP-10 and TNF-a (p-values = 0.02), but not IFN-g, were secreted in response to Gag peptides in the Ad5-naive compared to the Ad5-immune group. The Ad5-naive group also had higher levels of poly-functional Gag-specific CD8 T-cells secreting IFN-g, IL-2 and TNF-a (0.32% vs. 0.04% CD8 T-cells, p = 0.002). Conclusion: Higher levels of T-cells secreting IL-2 and IL-15 in the Ad5-naive vaccinees suggests that these T-cells may have a higher proliferative capacity. Moreover, elevation of IP-10 and TNF-a without a concomitant increase in IFN-g may indicate that antiviral pathways are qualitatively different in the absence of preexisting vector immunity. We conclude that the quality of a T-cell response to the Ad5/HIV vaccine is altered with pre-existing vector immunity, although the net effect of these responses in HIV protection remains to be determined.
Comments