HIV immunology: what lessons can we learn from recent vaccine trials?Results from two large HIV vaccine studies, the STEP and RV144 trials, gave surprising and divergent results that have raised more questions than answers in efforts to understand the immunology of HIV-1 infection and to develop successful vaccines against this condition.
The STEP trial was a double-blind Phase 2 study, in which 3000 high-risk healthy uninfected volunteers were randomly allocated to receive three injections of an adenovirus serotype 5 viral vector expressing three HIV-1 genes (gag, pol, nef) or placebo . The aim of the study was to test the capacity of the vaccine to reduce acquisition of HIV-1 infection or to decrease viral load set-point in vaccinated individuals who subsequently had breakthrough infection. A major issue for this vaccine was the high prevalence of adenovirus-specific antibodies as a result of prior exposure to the virus, particularly in sub-Saharan Africa, which would have been expected to lead to reduced immunogenicity.
The STEP trial was stopped early when a planned interim analysis showed vaccinated trial participants with high adenoviral antibody titres had increased rates of HIV-1 acquisition compared to subjects with low or absent adenoviral antibody titres . The adenoviral 5 HIV-1 gag/pol/nef vaccine induced strong, durable HIV-1-specific CD8 and CD4 immune responses in a large proportion of vaccine recipients; however, there was no difference in magnitude and quality of CD8 T cell immunity between vaccinated persons who developed HIV-1 infection and those who did not . Finally, viral load in vaccine recipients with HIV-1 infection was similar to that observed in subjects who were allocated to the placebo arm .
The results of the STEP trial have raised a number of critical issues for the field of HIV-1 immunology, namely the interaction between immune responses to viral vector and HIV-1 immunogens, the utility of currently used immune monitoring assays to measure potential correlates of protective HIV-1 immunity and the suitability of nonhuman primate models of infection to guide the development of HIV-1 vaccines .
The RV144 trial enrolled 16,402 low-risk subjects in Thailand into a multicentre double-blind placebo-controlled efficacy study. Vaccine recipients received, over a 6-month period, four doses of a canarypox viral vector. It encoded the HIV-1 subtype B gag and protease proteins and a chimeric envelope protein consisting of subtype E gp120 linked to subtype B gp41 domain (ALVAC). It was designed to prime HIV-1-specific cell-mediated immunity and was then followed by two injections of a recombinant glycoprotein gp120 subunit vaccine containing concentrations of subtype B and E antigens designed to boost HIV-1 cell-mediated immunity and stimulate neutralising HIV-1 antibodies (AIDSVAX) . The HIV-1 immunogens were chosen to match strains circulating in North America and Thailand.
The aim of the study was to determine if this prime boost vaccine combination reduced HIV-1 infection and early HIV-1 viraemia. The decision to proceed with the study was highly controversial as a previous version of the AIDSVAX failed to show any benefit in two previous studies and the ALVAC was considered be insufficiently immunogenic to proceed to an efficacy trial .
The results of RV144 showed that the prime boost HIV-1 vaccine combination led to a modest reduction (26-31%) in acquisition of HIV-1 infection and when subjects with HIV-1 infection at the time of randomisation were excluded, the reduction in rate of new HIV-1 infections in vaccine recipients was statistically significant (P=0.04). The vaccine did not have any effect on early HIV viral load or CD4 T cell count in vaccinated subjects who eventually became infected. This suggests that vaccination may have induced good mucosal immune responses that do not influence HIV-1 replication in peripheral blood.
As with the STEP study, the results of RV144 have significant implications for HIV-1 immunology and vaccine research which include an opportunity to identify markers of protective immunity, the need to study HIV-1-specific immune responses in mucosal specimens, a re-evaluation of the role of innate immunity in control of HIV-1 infection and agreement on laboratory tests that could be used to look for correlates of protective immunity. In addition, follow-up studies of the RV144 will need to answer a number of questions, such as:
* Are the needs for protection against HIV-1 transmission different in high-risk as opposed to low-risk populations?
* What was the duration of the vaccine effect?
* Which component of the vaccine (ALVAC, AIDSVAX or both) was responsible for the observed reduction in HIV-1 infection?
* Could immune responses to the canarypox viral vector have influenced the outcome of this trial?
* Will the results will be applicable to other regions of the world with different circulating strains to those in the RV144?
* What were the effects of confounding factors that can influence HIV-1 transmission (e.g. male circumcision and rates of intercurrent sexually transmitted infections [7,8]) on the rate of HIV-1 infection in vaccine recipients?
The most important contribution of these HIV vaccine trials, especially the RV144 study, will be the opportunity to investigate potential correlates of protective immunity against HIV-1 infection. The establishment of such biomarkers is the central issue in HIV vaccine development and will have a significant influence on the design of vaccine and clinical trials to assess their efficacy.
The discordance between partial protection from HIV-1 transmission and the lack of effect of the vaccine on early HIV-1 viral load and CD4 in recipients with breakthrough retroviral infection may lead to new insights into host resistance to HIV-1 infection. Models of non-human primate retroviral infection suggest that initial HIV replication is confined to small foci within cervical submucosa within the first couple of days of transmission . As most episodes of HIV-1 transmission occur across urogenital and gastrointestinal mucosal surfaces, evaluation of the anatomical distribution of vaccine-induced responses may prove very relevant in future efforts to develop more effective HIV-1 vaccines.
A critical issue arising from the HIV vaccine trials will be the need to determine the nature of both innate and adaptive mucosal immune response to HIV-1 infection. The innate immune response serves as a first-line defence against viruses. What the contribution of soluble innate molecules in the genital or gastrointestinal tract (lysozyme, lactoferrin, secretory leukocyte protease inhibitor, defensins) and of innate cellular immune responses (dendritic cell, natural killer cell, natural killer T cell and [gamma][delta] T cell) is to HIV control, and how to exploit innate immunity to induce immunological memory, should be an important priority for on-going research.
It is not known if the RV144 vaccine combination directly elicited other front-line immune defences within the genital tract such as either resident mucosal effector memory T cells or marginal zone effector memory B cell immune responses that may have contributed to partial protection against HIV-1 transmission without influencing HIV-1 replication in peripheral blood. Hansen and colleagues recently showed that continuous expression of SIV proteins using a primate CMV viral vector designed to directly stimulate mucosal CD8 effector T cells successfully protected some macaques from mucosal SIV infection . This vaccine was not designed to elicit systemic CD8 T cell immune responses and as such provides indirect evidence that successful immune responses to HIV may occur in distinct anatomical compartments.
Marginal zone B cells can produce IgA antibodies within hours ofexposure to glycosylated proteins without cognate T cell help. Efforts to determine whether the RV144 trial vaccines can elicit IgA T cell-independent neutralising antibodies need to be undertaken. Future work in this field should try to determine whether mucosal antibodies exhibit other functional features, for example, complement fixation, inhibition of epithelial transcytosis, blockade of cell-cell transmission across infectious synapses or antibody-dependent cell cytotoxicity, which may inhibit HIV-1 transmission. Although there is considerable overlap in specificity of mucosal and peripheral blood CD8 T cell responses to HIV-1 proteins, more sophisticated analysis of CD8 T cell function using polychromatic flow cytometry shows mucosal and peripheral blood immunity are not the same in a significant proportion of patients. Measurement of HIV-specific T cell responses in mucosal tissue may provide a better index of cervical status than those observed in blood [11,12]. Hence, it may be necessary to analyse immune responses to HIV-1 in tissues rather than blood to define protective correlates immunity.
It should be appreciated, however, that the immune response to HIV-1 may be a double-edged sword. Evidence from primate models of SIV transmission indicates that some inflammatory immune responses following retroviral challenge may facilitate viral infection by increasing numbers of activated CD4+CCR5+ target cells . Although initial studies following the STEP trial showed that there was no association between T cell proliferative and cytokine responses to adenovirus 5 in those with high levels of pre-existing adenoviral 5 antibodies [14,15], recent work raises the possibility that the increased rate of HIV-1 transmission observed in vaccine recipients with a history of significant exposure to adenovirus 5 may have resulted from vaccine-induced migration of activated memory CD4+CCR5+ target T cells to the gastrointestinal tract [16,17]. One potentially important lesson for future HIV vaccine studies is the need to circumvent CD4+CCR5+ activation at mucosal sites that are important for HIV transmission.
Another major lesson for HIV immunology is what measures of possible immune correlates of protective T cell immunity should be employed in future vaccine trials. Data from the STEP trial show that strength of vaccine-induced cytotoxic T lymphocyte activity as measured by the IFN-[gamma] secretion in blood does not provide a robust indication of functional antiviral activity . Current efforts to assess potential correlates of T cell immunity to HIV infection emphasise that the quality rather than the magnitude of CD8 antiviral activity may be more important in host control of HIV infection. Most authorities on HIV-1 immunology believe that polyfunctional T cell immune response, defined as the secretion of at least two cytokines towards a HIV-1 antigen, should be the aim of current or future HIV-1 vaccines . A major advance in studies of HIV-1 immunology will be to agree on what constitutes a polyfunctional immune response to a HIV-1 antigen, what type of immune cells are involved and where this immune response occurs (peripheral blood versus mucosa). Over 50% of the HIV-specific CD8 T cells elicited after vaccination in the STEP trial produced IFN- alone and did not exhibit IL-2 or TNF- responses, suggesting that this HIV vaccine failed to stimulate a polyfunctional T cell immune response [2,3]. Studies in animal models of Leishmania infection , and evidence from HIV-1 elite controllers , indicate that the capacity of the immune system to secrete IL-2 in combination with other cytokines is associated with increased resistance to infection or delayed disease progression. Future definition of a polyfunctional immune response may need to encompass both the number and identity, particularly IL-2, and the concentration of secreted protein by different T cell subsets in peripheral blood and mucosal samples.
A major rationale for the STEP trial was the hypothesis that the induction of conserved regions of the HIV gag/pol and nef gene would correspond to strains circulating in the population, leading to either to reduced acquisition of HIV-1 or slower disease progression. Data from Phase studies I preceding STEP showed vaccination elicited a limited number of CD8 T cell immune responses to a limited number of HIV-1 immunogens (median number 3) compared to the much greater numbers (median number 14) observed in natural infection . This may have been an important factor in the lack of vaccine efficacy in the STEP trial. Current studies to address this issue are under way and will compare viral sequence obtained from infected vaccine recipients and those present in the HIV vaccine. If the limited breadth of CD8 T cell immune response to HIV-1 immunogens is shown to be an important factor for the disappointing results seen in the STEP trial, efforts to design vaccines to stimulate multiple conserved regions of HIV-1 virus and to elicit immune responses against genetically diverse strains of circulating isolates of HIV-1 should be a major focus for future research. Finally, the specificity of cytotoxic lymphocyte responses may matter; the breadth of gag-specific responses but not other responses is associated with lowered viral loads in natural infection .
Results from the STEP trial raise the issue that immune responses to viral vectors might have a deleterious effect on HIV-1-specific immune responses. Since it is likely that viral vectors will continue to be used in the near future in HIV vaccine studies, it will be important to assess the impact of vector-specific immunity on the quality and magnitude of HIV-1 immune responses. Since most individuals born after 1974 have not been immunised against smallpox and would therefore have little pre-existing immunity to this virus, use of pox-related viral vectors may offer more advantages that adenoviral-based vectors. In addition, advances in nanotechnology and polymer chemistry may allow immunologists to use these materials as novel vaccine adjuvants and to target vaccines more effectively to mucosal surfaces . The large number of immunisations used in the RV144 trial would make it very difficult to administer more successful versions of this vaccine in routine clinical settings. If the DNA prime boost strategies using pox viral vectors were found to bypass the need for multiple injection courses, then this would have significant implications for the use of successful HIV vaccines in the clinic .
The large number of potential immune correlates of protection raises the need to perform focused studies to determine the biomarkers of protective HIV-1-specific immune responses given the limited access to mucosal samples that are available from the RV144 study, and the limited number ofmucosal mononuclear cells and smaller volumes of mucosal secretions compared to those that can be obtained from peripheral blood samples. Biomarkers of correlates of HIV-1-specific immune protection may be obtained from those infected persons termed 'elite controllers' who maintain a viral load below the limit of detection . Mechanisms underlying spontaneous immune control of HIV-1 replication are likely to be heterogeneous but in some cases will be likely to be associated with immune control of the virus.
Efforts to study persistently exposed seronegative persons may also offer new insights into host protection against HIV-1 acquisition and inform strategies in HIV-1 vaccine development . Distinguishing between cause and effect will be difficult in these individuals; however, insights into immune correlates of protection may be gained by systematic study of immune responses following successful vaccination against other human viral pathogens such as measles, hepatitis A and yellow fever [4,21].
Standardisation of non-human primate models used to guide HIV vaccine development are needed as differences in the composition of challenge--virus, dose, route of challenge, immune readouts--make it very difficult to compare relative efficacy of various vector and immunisation strategies. One lesson from the STEP trial is that the SIVmac239 model, which predicted the failure of the adenoviral-based vaccine strategy, is probably the best model that is available at this point in time. However, to be a fully validated model it would need to predict vaccine success.
Mechanistic studies looking at mucosal immune responses elicited by vectors used in the RV144 trial using well-characterised, non-human primate models may offer clues to HIV vaccine development. Natural SIV infection of non-human primates has highlighted important aspects of retroviral immunology, such as immune activation, paucity of CD4+CCR5+ at mucosal surfaces and the combination of antiviral innate and adaptive immune responses, which may all be pertinent to the aims of HIV vaccine trials . Designing a vaccine that focuses solely on increasing adaptive immune responses may not be enough to either inhibit transmission or attenuate progression to AIDS.
Finally, the use of systemic biology, including genome-wide association studies, proteomics, and bioinformatic tools, in large HIV-1 patient cohorts and in future HIV vaccine trials may also provide important leads into host mechanism of immune HIV-1 controls. Such studies are likely to play an important role in future HIV-1 research.
[1.] Buchbinder SP, Mehotra DV, Duerr A et al. Efficacy assessment of a cell mediated immunity HIV-1 vaccine (the STEP Study): a double blind randomised, placebo-controlled, test of concept trial. Lancet, 2008, 372, 1881-1893.
[2.] McElrath MJ, De Rosa SC, Moodie Z et al. HIV-1 vaccine-induced immunity in the test-of-concept Step Study: a case-cohort analysis. Lancet, 2008, 372, 1894-1905.
[3.] Corey L, McElrath MJ, Kublin JG. Post-Step modifications for research on HIV vaccines. AIDS, 2009, 23, 3-8.
[4.] Sekaly R-P. The failed Merck vaccine study: a step back or a launching point for future vaccine development. J Exp Med, 2008, 205, 7-12.
[5.] Rerks-Ngarm S, Pitisuttihum P, Nitayaphan S et al. Vaccination with ALVAC and AIDSVAX to prevent HIV-1 infection in Thailand. N Engl J Med, 2009, 361, 2209-2220.
[6.] Cohen J. Surprising AIDS vaccine success praised and pondered. Science, 2009, 326, 26-27.
[7.] Berkhout, B, Paxton WA. HIV vaccine: it may take two to tango, but no party time yet. Retrovirology, 2009, 6, 88-91.
[8.] Dolin R. HIV vaccine trial results-an opening for future research. N Engl J Med, 2009, 361, 2279-2280.
[9.] Broliden K, Haase AT, Ahuja SK et al. Introduction: Back to basics: mucosal immunity and novel HIV vaccine concepts. J Intern Med, 2009, 265, 5-17.
[10.] Hansen SG, Vieville C, Whizin N et al Effector memory T cell responses are associated with protection of rhesus monkeys from mucosal simian immunodeficiency virus challenge. Nat Med, 2009, 15, 293-299.
[11.] Shacklett BL, Critchfield JW, Ferre AL, Hayes TL. Mucosal T-cell responses to HIV: responding at the front lines. J Intern Med, 2009, 265, 58-66.
[12.] Ferre AL, Hunt PW, Critchfield DH et al. Mucosal immune responses to HIV in elite controllers: a potential correlate of immune control. Blood, 2009, 113, 3978-3989.
[13.] Li Q, Estes JD, Schlievert PM et al. Glycerol monolaurate prevents mucosal SIV transmission. Nature, 2009, 458, 1034-1040.
[14.] O Brien KL, Liu J, King SL et al. Adenovirus-specific immunity after immunization with an Ad5 HIV-1 vaccine candidate in humans. Nat Med, 2009, 15, 873-875.
[15.] Hutnick NA, Carnathan DG, Dubrey SA et al. Baseline Ad5 serostatus does not predict Ad5 HIV vaccine-induced expansion of aadenovirus-specific CD4+ T cells. Nat Med, 2009, 15, 876-878.
[16.] Benlahrech A, Harris J, Meiser A et al. Adenovirus vector vaccination induces expansion of memory CD4 T cells with mucosal homing phenotype that are susceptible to HIV-1. Proc Natl Acad Sci USA, 2009, 106, 19940-19445.
[17.] Chakupurakal G, Onion D, Cobold M et al. Adenoviral vector-specific T cells demonstrate a unique memory phenotype with high prolifertive capacity and coexpression of CCR5 and integrin alpha4beta7. AIDS, 2010, 24, 205-210.
[18.] Douek DC, Roederer M, Koup RA. Emerging concepts in the immunopathogenesis of AIDS. Ann Rev Med, 2009, 60, 471-478.
[19.] Darrah PA, Patel DT, De Luca PM et al. Multifunctional TH1 cells define a correlate of vaccine mediated protection against Leishmania major. Nat Med, 2007, 13, 8443-8850.
[20.] Betts MR, Nason MC, West SM et al. HIV nonprogressors preferentially maintain highly functional HIV-specific CD8 T cells. Blood, 2006, 107, 4781-4789.
[21.] Walker BD, Burton DR. Towards an AIDS vaccine. Science, 2008, 320, 760-764.
[22.] Hubbell JA, Thomas SN, Swartz MA. Materials engineering for immunomodulation. Nature, 2009, 462, 449-460.
[23.] Cohen J. HIV natural resistance field finally overcome resistance. Science, 2009, 326, 1476-1477.
[24.] Sodora DL, Allan JS, Apetrei C et al. Towards an AIDS vaccine: lessons from natural simian immunodeficiency virus infections of African non human primates. Nat Med, 2009, 15, 861-866.
Division of Infectious Diseases, Imperial College, London, UK
Correspondence to: Peter Kelleher
Immunology Section, Division of Infectious Diseases
Imperial College, Chelsea and Westminster Hospital
369 Fulham Road
London SW10 9NH, UK
Email: p. firstname.lastname@example.org