The interplay between antibodies and T cells in HIV risk

From the HVTN and Vaccine and Infectious Disease Division

            HIV is a chronic infection that affects people worldwide. Due to its ability to mutate rapidly and ever-changing nature, the virus is very difficult to protect against. There is currently no approved vaccine for HIV acquisition. Therapy consists of a lifelong drug regiment, which reduces viral load but does not cure disease. Development of new vaccines is hindered by researchers’ lack of knowledge as to robust correlates of protection. With little protection seen in past vaccine trials, there are little data available to determine correlates of protection; however, this lends itself to identify possible biomarkers that could be used to identify correlates of risk. In a recent study published in the Journal of Infectious Diseases, researchers from the Vaccine and Infectious Disease Division at Fred Hutch revisited the HVTN 505 study, a trial in which a DNA HIV-1 Env, Gag, Nef, Pol prime was followed by a recombinant adenovirus type 5 (rAd5) HIV ENV, Gag-Pol fusion protein boost. Samples from this study were taken 1 month after the final vaccination and were assessed for correlates of risk for HIV infection.

            Vaccine-induced cellular and humoral immune responses were assessed. Env-antibodies were tested for binding, neutralizing, and antibody-dependent cell-mediated cytotoxicity (ADCC) and epitopes were mapped using a peptide array. Both IgG and IgA antibody responses were found along with low level tier 1 neutralizing responses. The ADCC response was not statistically significant. In addition, the group looked at specific antibody types and epitopes and whether they could account for differences in protection. Previous studies suggested that both IgA and  gp41 binding antibodies may increase risk. However, in the new study, this does not seem to be the case. HIV-specific CD8 T cell responses were also documented. In response to the effect of the two arms of the immune system on risk (defined as the probability of infection from month 7 to 24, based on data), first author Dr. Fong stated, “We found that vaccine-induced T cell and antibody responses were synergistic in their association with HIV infection risk in the HVTN 505 efficacy trial.”

Figure: Env- CD8 T cell polyfunctionality (colors) combined with effects of varying amounts of antibody (x axis) on HIV risk (lines).

Their analysis found that when CD8 Env-specific responses are low, higher IgG Env-specific responses are needed to reduce risk (see figure, black line). When HIV-specific CD8 responses are low to medium, the antibody response did not significantly reduce risk (see figure, red and blue line). If, however, the polyfunctionality of CD8 cells are taken into account, the higher it is the lower the HIV risk, regardless of antibody status (compare all CD8 lines).

            These findings led the authors to two possible explanations. “The big question now is whether the observed risk gradient means that the vaccine was protective for some but harmful for others depending on their immune response levels or that the immune responses do not actually cause the change in risk, but merely coincide with another, truly causal factor that potentially has nothing to do with the vaccine. One way we could study this question is to examine how HIV variant-specific immune responses associate with variant-specific infection risk” stated Dr. Fong. Further research is needed in order to tease apart these two hypothesizes. However, it is apparent from this work, that both humoral and cellular responses are needed after vaccination to reduce the risk of HIV acquisition.

Fong Y, Shen X, Ashley VC, Deal A, Seaton KE, Yu C, Grant SP, Ferrari G, deCamp AC, Bailer RT, Koup RA, Montefiori D, Haynes BF, Sarzotti-Kelsoe M, Graham BS, Carpp LN, Hammer SM, Sobieszczyk M, Karuna S, Swann E, DeJesus E, Mulligan M, Frank I, Buchbinder S, Novak RM, McElrath MJ, Kalams S, Keefer M, Frahm NA, Janes HE, Gilbert PB, Tomaras GD. 2018. Vaccine-induced antibody responses modify the association between T-cell immune responses and HIV-1 infection risk in HVTN 505. J Infect Dis.

Funding was supplied by the National Institutes of Health, US Public Health Service grants, and NIH/NIAD Duke Center for AIDS Research.

Fred Hutch/UW Cancer Consortium faculty members Youyi Fong, Nicole Frahm, Peter B Gilbert, Holly E Janes, and Julie McElrath contributed to this work.