This past Wednesday May 11, the journal Nature published a vaccine experiment in rhesus macaques that has drawn widespread media coverage. Led by Louis Picker at the Oregon Health and Science University’s Vaccine and Gene Therapy Institute (VGTI), researchers used cytomegalovirus (CMV) as a vaccine vector to immunize against the highly virulent and pathogenic SIVmac239 challenge virus. In 12 out of 24 macaques that received the vaccine, an unprecedented degree of long-term control of SIV replication was observed; after transient post-infection peaks ranging from 60 to 10,000,000 copies, there were only occasional blips above the limit of detection that diminished over time. In four of these animals that were euthanized after 52 weeks of follow up and had multiple tissues analyzed, replication-competent virus could not be found and viral RNA and DNA levels were extremely low, leading to the suggestion that SIV was being progressively cleared. Another group of nine macaques immunized with a DNA plus adenovirus serotype 5 (Ad5) regimen did not exhibit comparable control of the challenge virus, rather their average viral loads were slightly lower than unimmunized controls out to around 100 days of follow up but were not significantly different thereafter.
The researchers note that these results echo and extend those obtained in a prior study in which 4/12 macaques appeared almost completely protected from SIVmac239 after vaccination with the CMV vector. Evidence suggests that the mechanism of protection in both experiments was a sustained and active “effector memory” CD8 T cell response against SIV. Effector memory or Tem CD8 cells cannot proliferate very well upon encounter with their target, but are very efficient killers of virus-infected cells.
The only other approach that has shown better results in this animal model is a live attenuated SIV vaccine, which is considered too dangerous to try and adapt for use in humans. The researchers argue that this is not the case for CMV, a claim that is likely to be extremely controversial. In the media stories about the study, Louis Picker states that CMV can be dangerous in certain groups of people such as individuals with immune deficiencies and pregnant women, but does not cite the growing body of literature indicating that infection with the virus is associated with accelerated senescence of the immune system and increased rates of morbidity and mortality as people age. These harmful effects appear to be related to the wear-and-tear caused by CMV’s potency as a chronic immune system stimulus, and this same potency may also underlie its efficacy as a vaccine vector. If these two aspects of CMV are inextricably intertwined, it seems unlikely that the approach can be rendered safe for human use. Even attempting to establish safety would present profound challenges given that it takes decades for the pernicious immune senescence-related aspects of CMV infection to manifest.
Another argument in favor of human studies mentioned in some of the press articles is the ubiquity of CMV infection in humans; Picker states in the BBC coverage that ~99% of people in sub-Saharan Africa are already CMV positive. Unlike some other vaccine vectors, this pre-existing immunity is not likely to affect how a CMV vector would work because the virus has evolved immune evasion mechanisms that endow it with the ability to re-infect. But that does not necessarily mean that adding a persistent CMV vector on top of natural infection would be neutral in terms of safety, as studies have found that the intensity of the immune response to the virus is linked to poor health outcomes.
Although the issue of CMV and immune senescence has not cropped up in the coverage of the Nature paper, Picker and colleagues are certainly aware of it (according to IAVI Report’s coverage of the AIDS Vaccine 2010 conference, Picker posed the salient question there: “is this protection without practicality?”). It is possible that evidence will yet emerge that CMV’s dangerous properties can be disentangled from its salutary ability to sustain Tem CD8 T cells and, even if not, the new study provides a strong impetus to seek safe ways to induce similar immune responses. Perhaps most encouragingly, Picker’s work has demonstrated that the right type of vaccine-induced immune response can exert an extraordinary degree of control over a virus notorious for overwhelming everything the immune system throws at it.
Nature. 2011 May 11. [Epub ahead of print]
Hansen SG, Ford JC, Lewis MS, Ventura AB, Hughes CM, Coyne-Johnson L, Whizin N, Oswald K, Shoemaker R, Swanson T, Legasse AW, Chiuchiolo MJ, Parks CL, Axthelm MK, Nelson JA, Jarvis MA, Piatak M, Lifson JD, Picker LJ.
Vaccine and Gene Therapy Institute, Departments of Molecular Microbiology and Immunology and Pathology, and the Oregon National Primate Research Center, Oregon Health & Science University, Beaverton, Oregon 97006, USA.
The acquired immunodeficiency syndrome (AIDS)-causing lentiviruses human immunodeficiency virus (HIV) and simian immunodeficiency virus (SIV) effectively evade host immunity and, once established, infections with these viruses are only rarely controlled by immunological mechanisms. However, the initial establishment of infection in the first few days after mucosal exposure, before viral dissemination and massive replication, may be more vulnerable to immune control. Here we report that SIV vaccines that include rhesus cytomegalovirus (RhCMV) vectors establish indefinitely persistent, high-frequency, SIV-specific effector memory T-cell (T(EM)) responses at potential sites of SIV replication in rhesus macaques and stringently control highly pathogenic SIV(MAC239) infection early after mucosal challenge. Thirteen of twenty-four rhesus macaques receiving either RhCMV vectors alone or RhCMV vectors followed by adenovirus 5 (Ad5) vectors (versus 0 of 9 DNA/Ad5-vaccinated rhesus macaques) manifested early complete control of SIV (undetectable plasma virus), and in twelve of these thirteen animals we observed long-term (≥1 year) protection. This was characterized by: occasional blips of plasma viraemia that ultimately waned; predominantly undetectable cell-associated viral load in blood and lymph node mononuclear cells; no depletion of effector-site CD4(+) memory T cells; no induction or boosting of SIV Env-specific antibodies; and induction and then loss of T-cell responses to an SIV protein (Vif) not included in the RhCMV vectors. Protection correlated with the magnitude of the peak SIV-specific CD8(+) T-cell responses in the vaccine phase, and occurred without anamnestic T-cell responses. Remarkably, long-term RhCMV vector-associated SIV control was insensitive to either CD8(+) or CD4(+) lymphocyte depletion and, at necropsy, cell-associated SIV was only occasionally measurable at the limit of detection with ultrasensitive assays, observations that indicate the possibility of eventual viral clearance. Thus, persistent vectors such as CMV and their associated T(EM) responses might significantly contribute to an efficacious HIV/AIDS vaccine.