Immune Activation with HIV Vaccines

See allHide authors and affiliations

Science  04 Apr 2014:
Vol. 344, Issue 6179, pp. 49-51
DOI: 10.1126/science.1250672

The development of a safe and effective HIV vaccine is perhaps the most important and challenging goal remaining in HIV-AIDS research. Recent progress using a poxvirus vector prime and envelope protein boost strategy demonstrated a modest but statistically significant level of efficacy and established the concept that a vaccine could prevent HIV infection (1), and approaches to boost durability and efficacy are currently in the planning stages (2). But the results of two vaccine concepts based on recombinant adenovirus serotype-5 (rAd5) (35) pointed to a potential major problem—that such vaccines might increase susceptibility to HIV infection. This also raised the question of whether the problem extends to some or all of the other recombinant adenovirus vectors currently in development or to other vector-based vaccines.

Last year, the U.S. National Institute of Allergy and Infectious Diseases (NIAID) convened the Mini-Summit on Adenovirus Platforms for HIV Vaccines (6) to investigate this issue. Furthermore, the question was raised whether increased susceptibility might be seen with any HIV vaccine that activates the immune system rendering activated CD4+ T cells more susceptible to HIV infection, while at the same time inducing little or no protective effect against HIV acquisition (6).

Between 2005 and 2013, two rAd5 vaccines for HIV were assessed in three efficacy studies (35). The first study (Step), using three doses of the Merck rAd5 vaccine containing genes encoding three HIV-1 proteins (gag, pol, and nef), was stopped for futility. In addition, a statistically significant trend toward increased HIV infections in vaccine recipients was observed (3, 7). The group at highest risk was uncircumcised men who both had sex with men (MSM) and had high titers of preexisting antibodies against Ad5. The following Phambili trial of the same Merck rAd5 vaccine, conducted in South Africa, was closed and unblinded early during the enrollment period. Few participants received the planned three doses of vaccine. Analysis of the data showed no increased risk of HIV infection (5). However, data from the long-term unblinded follow-up of Phambili participants suggested an increased risk of infection in vaccinated men relative to unvaccinated controls (8).

In 2009, a different rAd5 vaccine was tested in the HIV Vaccine Trial Network (HVTN) 505 trial. It contained three doses of a DNA prime (a plasmid of non-HIV DNA and certain HIV genes) followed by a single boost of rAd5 expressing HIV envelope and viral structural antigens. The study restricted enrollment to circumcised MSM who lacked preexisting antibodies to Ad5, because no level of increased risk had been seen in this group in the Step trial (7). The HVTN 505 trial was halted prematurely because it met futility criteria; however, there was no evidence of increased risk of HIV infection in the vaccinated subjects (9).

Results of a Meta-Analysis

A meta-analysis of the Step, Phambili, and HVTN 505 trials was performed by statisticians from the HVTN Statistics and Data Management Center and from NIAID. Combining data from the three studies, there was an overall hazard ratio of 1.33 (P < 0.01) associated with vaccination. However, almost all of the increased risk of HIV acquisition was driven by the Merck vaccine (Step and Phambili: hazard ratio = 1.41, P = 0.005) with the Step trial contributing most infection endpoints. HVTN 505 considered alone did not show any trend toward infection risk (6, 10). It could not be determined whether the lack of increased susceptibility in the latter trial was due to population or regimen (inclusion of env, DNA prime, and single rAd5 boost with differences in vector backbone).

Other Ad-Based Vaccines

Alternative Ad vectors for use in tuberculosis or malaria vaccines have not been evaluated in trials in sufficient numbers of adults at risk for HIV infection to provide useful information about possible interactions with HIV. However, active-duty U.S. Army recruits have been vaccinated with Ad4/7 to prevent highly contagious respiratory illnesses in close quarters during training (11). From 1999 to 2011, the Army interrupted and then resumed vaccination with Ad4/7, creating three separate cohorts for comparison. The serologic data were reviewed retrospectively for these three cohorts, and no changes in HIV incidence were detected (6).

Potential Mechanisms for rAd5 Effects

In 2008, NIAID held an HIV Vaccine Summit, which fostered studies to understand the possible mechanisms of the effects seen in the Step study (12). Results revealed at the 2013 NIAID Mini-Summit included those from nonhuman primate studies of the Merck rAd5-gag/pol/nef vaccine constructs. The data showed an increase in activated CD4+ T cells in the gastrointestinal mucosa after rAd vaccination and increased simian immunodeficiency virus (SIV) acquisition after rAd-SIV vaccination; the empty rAd vector did not increase SIV acquisition ( 13). Human gut biopsies of individuals vaccinated with the DNA prime and rAd5 boost as part of HVTN 204 (low risk) and HVTN 505 (high risk) revealed high numbers of activated Adspecific CD4+ T cells, with increased expression of C-C chemokine receptor type 5 (the co-receptor for HIV), but no HIV-specific cells in multiple samples from the rectum and colon (6). Furthermore, the activated T cells were concentrated in unevenly distributed foci, making extensive sampling a necessity (14). Although there is no current evidence of increased risk of HIV acquisition for other Ad vectors, the presence of rAd5-activated T cells in tissue and the degree of shared epitopes between Ad serotypes that are recognized by CD4+ T cells should be considered as a possible area of concern.

HIV vaccine–induced immune response interactions.

Immune activation associated with HIV vaccination theoretically can lead to increased infection due to the activation of CD4+ T cells during the immune response. The level of protection seen with a vaccine can be viewed as the balance between the responses to the vaccine that lead to susceptibility to infection and by the responses that favor protection.


Ad serotype cross-reactivity can be attributed, in part, to recognition by CD4+ T cells and CD8+ T cells of highly conserved regions in Ad hexon protein across most human and primate Ad species (1517). Therefore, this issue must at least be considered in the risk-benefit analysis of potential HIV vaccine trials using alternate Ad vectors. However, in a phase 1 study using a single rAd26-HIV env vaccination, a limited set of rectal biopsies showed no evidence for increased Ad-specific T cell activation in low-risk volunteers and a different host gene activation signature from that in Ad5-vaccinated individuals (6). Nonetheless, a description of the rAd5 experience should be included in informed consents associated with HIV vaccine trials using alternate Ad vectors.

It is conceivable that increased risk of HIV infection could be associated with any vaccination strategy that activates T cells, especially at mucosal surfaces. If the balance between the stimulation of protective anti-HIV responses versus Ad-specific responses that activate CD4+ T cells leans toward an anti-HIV response, then protection might be seen despite the presence of susceptible T cell targets. However, increased risk to infection may be seen if the anti-HIV response is weak and does not counterbalance the increased susceptibility of activated CD4+ T cells; if the anti-HIV response wanes faster than the CD4+ T cell activation; or if the Ad-induced response is maintained and boosted by reexposure to alternative adenoviruses. Furthermore, cellular immune responses against Ad5, regardless of Ad serostatus, diminished anti-HIV responses to rAd5-HIV vaccination in Step (15). Thus, the efficacy of an HIV vaccine may reflect a balance of two competing activities—anti-HIV responses and T cell activation.

Considerations for the Future

Given the increased risk and the lack of efficacy in trials using rAd5, further HIV vaccine studies testing rAd5 vectors are not appropriate. When considering HIV vaccines that are designed to elicit a component of T cell immunity, a risk-benefit analysis should consider the balance between anti-HIV responses and vector-directed responses that activate CD4+ T cells (see the figure). This is particularly important when assessing viral vectors, including alternative Ad vectors. Future clinical testing of Ad-based vaccines should evaluate the amounts and distribution of both vector and insert responses in target tissues where HIV acquisition is known to occur.

Other research activities could be pursued to help clarify the roles of antivector responses in overall HIV vaccine efficacy. For example, nonhuman primate studies using empty vectors or vectors with non-HIV inserts as placebo controls could define the levels of antivector immunity and evaluate the effect on virus acquisition of this vector-related activation of CD4+ T cells independent of an anti-HIV response. Also, the field could benefit from additional nonhuman primate studies. For example, the identification of biomarkers in primates that indicate increased risk of acquisition (18) could be valuable to monitor for risk in early-phase human studies.

A better understanding of mucosal immune responses to HIV vaccination is also needed. The timing, location, and number of mucosal biopsies that define the vaccine-induced gut immune responses need clarification. Understanding the influence of the mucosal microbiome on vaccination (19) and the impact specifically of the virome will be important. Particularly for Ad-based vectors, understanding components of risk related to the level of Ad exposure and persistence will be essential.

For non-HIV vaccine trials using vectors that induce strong T cell immunity that are conducted in regions with high HIV incidence, it may be important to monitor for HIV acquisition, depending on the target population. In such studies where the population may be at risk of HIV exposure, HIV incidence should be monitored at the end of the study and for an appropriate follow-up period.

The experience with rAd5-based HIV vaccines has shown that vaccine-induced protection likely reflects the balance between beneficial anti-HIV responses and deleterious effects of immune activation that increases the susceptibility of CD4+ T cells to infection (see the figure). Among the spectrum of existing or planned vaccines, this phenomenon is likely unique for an HIV vaccine because the activated CD4+ T cell is the very target for the virus. These observations should be taken into consideration in future HIV vaccine research endeavors and underscores the importance of maximizing the specific anti-HIV responses of such candidates.


View Abstract

Stay Connected to Science

Navigate This Article