Could a vaccine against immune-evading cytomegalovirus become a reality?

Keywords::

• basic reproductive number
• glycoproteins
• primary infection
• transplant patients

Human cytomegalovirus (HCMV) remains an important human pathogen in the neonate and the immunocompromised host, and there is evidence linking HCMV infection to an immune-risk phenotype in the very elderly that predisposes to early death [1,2]. The virus has a large number of genes in its armamentarium to manipulate the host environment to enable its survival. These include proteins targeting the innate and adaptive immune response, the cell cycle, apoptotic pathways and cytokine/chemokine networks [1]. The human immune system devotes a substantial proportion of its activity against HCMV through the use of CD4⁺ and CD8⁺ T cells that recognize almost the entire HCMV proteome, with responses against proteins such as the pp65 lower matrix protein reaching upwards of 5% of immune activity [3]. In addition, antibody responses against the virus are equally broad, although the majority of neutralizing antibodies are directed against the surface glycoprotein complexes [1]: glycoprotein (g)B, gM/gM, gH/gL and the pentameric complex involving gH/gL,UL128, UL130 and UL131, which is gaining importance in the field [4].

The reasons why HCMV needs so many immune manipulation genes has been partly revealed in recent studies using a rhesus CMV model where deletion of the class I HLA downregulatory genes led to the inability of the virus to re-infect a previously CMV-infected animal [5]. Re-infection of immune hosts leading to infection with multiple strains of HCMV has been reported in HIV patients and transplant recipients; this observation poses a further hurdle in that development of a vaccine is required that protects not only against primary infection but also re-infection [1].

Coupled with the fact that HCMV establishes a latent infection, there are a number of perceived and real challenges in producing a vaccine against HCMV; however, are they insurmountable? Two studies have estimated the basic reproductive number (Ro: the number of new infections arising a single infected individual) in North Europe and the UK showing that Ro is between 1.7 and 2.5, although it can be higher in lower socioeconomic groups (Ro = 3.5) and in non-Hispanic black Americans (Ro = 5.7) [6,7]. Overall, this Ro value is surprisingly low and means that the critical vaccination proportion (Pc) needed to reduce Ro to <1, and therefore eradicate HCMV, is between 41 and 62%. This value is comparable to that needed to eradicate smallpox, but much lower than the vaccine coverage needed to eradicate measles and polio from the human population (where Pc is >90%). Within the host, Ro has been estimated in liver transplant recipients in the high-risk (D⁺R^-) setting and also the intermediate-risk (R⁺) groups as approximately 15 and 2.5, respectively, although these estimates will be influenced by the net state of immunosuppression [8]. While primary vaccination and eradication must remain the a high priority, as identified by the Institute of Medicine report some 11 years ago [101], in the setting of transplantation, the key role that viral load plays in pathogenesis means that if a vaccine enhanced immunity sufficiently so that virus replication was substantially diminished, but not necessarily prevented, it may have a significant impact on disease (and the need for antiviral chemotherapy) but not necessarily infection per se[9].

To date, a number of approaches to the production of a HCMV vaccine have been attempted with mixed success. These include live-attenuated viruses, chimeric HCMV strains, peptides, use of surface glycoproteins with immune-enhancing adjuvants, DNA vaccines and single-cycle recombinant viral vectors to deliver B- and T-cell targets [10]. One of the earliest vaccine studies was carried out using the live-attenuated Towne strain of HCMV in renal transplant recipients [11]. At the time the trial was undertaken in the late 1970s, the genetic basis for the attenuation was unknown, although we now know that the Towne strain has a number of mutations and deletions affecting a range of pathogenicity genes. Compared with placebo recipients, the Towne vaccine had no effect on the incidence of virus replication following renal transplantation; however, there was consistent evidence for a reduction in the severity of HCMV disease in the high-risk D⁺R^- group [11,12]. This observation is consistent with the aforementioned thesis – namely that a vaccine that is able to reduce viral load by upwards of 0.5 log could have a significant impact on HCMV disease. Despite these encouraging observations, vaccine development for HCMV remained in the doldrums for many years. In the last decade, with increasing understanding of the B- and T-cell immunobiology of HCMV, attention has turned towards HCMV vaccine production and evaluation. A modified recombinant glycoprotein B vaccine given with the adjuvant MF59 (a squalene-based moiety), originally developed by Chiron and now being developed by Sanofi-Pasteur, has been subjected to a number of Phase I trials [13]. The vaccine is safe and highly immunogenic in both HCMV seronegative and seropositive individuals with the latter group showing further boosts in immunity after vaccination. This vaccine has been subjected to a Phase IIb study in seronegative women whose previous children were attending play groups (and therefore at high risk of acquiring HCMV infection which could then be passed to their mothers) [14]. The results showed a 50% reduction in infection in vaccinated women. A very recent study by Griffiths et al. has assessed the safety, immunogeneicity and efficacy of this vaccine in solid-organ transplant patients [15]. Vaccination of 140 liver and kidney recipients prior to transplant showed that the vaccine was safe and able to elicit high levels of antibodies and could boost levels in previously seropositive individuals. In addition, in patients who proceeded to transplant, vaccinees in the high-risk D⁺R^- group experienced a significantly shorter duration of DNAemia and required fewer days of antiviral therapy. These results are important for two reasons: first, they indicate that vaccines that induce a predominantly B-cell response against HCMV are of value in the T-cell immunocompromised host and, second, that using virologic end points in transplant patients provides a tractable system for evaluating candidate HCMV vaccines.

In the stem cell setting, where the greatest risk for HCMV infection and disease frequently occurs in seropositive patients who receive a transplant from a seronegative donor, there is potential for donor vaccination in addition to recipient vaccination post-stem cell therapy. Although previous studies have shown the feasibility of this approach [16], the logistics of performing such a study are not insignificant. Consequently, a study of a HCMV DNA vaccine encoding both glycoprotein B and a kinase-deficient form of the pp65 protein has focused on vaccination of recipients prior to ablative therapy and following stem cell transplantation. The vaccine was formulated with the poloxamer CRL1005 and benzalkonium chloride to aid delivery. Phase I studies of this vaccine indicated moderate immunogenicity [17], however the placebo-controlled Phase II study in 80 stem cell transplant (SCT) patients has demonstrated enhanced T-cell immunogenicity, specifically against pp65 and significant reductions in DNAemia and a delay in the occurrence of DNAemia [102]. This vaccine is likely to be entering Phase III studies in the near future and has just been licensed to Astellas Pharma Inc, [103]. It is noteworthy that HCMV vaccines that work well in one setting may not translate to other settings where HCMV is a problem. For example, it may be that generation of T-cell responses is crucial to protect the SCT recipient and therefore a vaccine that only generated B-cell immunity may be less efficacious in this population. Nevertheless, the gB-MF59 vaccine has shown efficacy in both healthy women and T-cell immunocompromised solid-organ patients, illustrating the challenges in predicting the success of HCMV vaccines across different patient populations. It is likely that combination vaccines will offer the best way forward in full protection and, to date, the most encouraging data has come from an alphavirus replicon expressing gB and a pp65–IE fusion protein (produced by AlphaVax and being developed further by Novartis Vaccines and Diagnostics). In Phase I studies this vaccine produced high levels of gB antibodies and CD4 and CD8 T-cells against pp65 and IE1 [18]. Further development of this vaccine is awaited with interest since it seems to offer both high immunogenicity and the generation of high-quality B- and T-cell responses.

In conclusion, some 27 years after Plotkin et al. evaluated the Towne vaccine in renal transplant recipients, it is likely that in the early part of the 21st Century, we may eventually witness the development of safe, efficacious and noninfectious vaccines for HCMV to protect against congenital disease and the direct and indirect effects after transplantation. Indeed, deployment of vaccination may reveal other human diseases and conditions to which HCMV contributes, although we should not forget the possibility that HCMV infection could also have a positive influence on the development of our immune system [19]. Even with a licensed vaccine, the challenges of global eradication should not be underestimated, especially in areas of the world such as Africa and South Asia where seroprevalance rates are very high and where infection occurs at a very young age. Nevertheless, I would suggest that the availability of a licensed HCMV vaccine will shortly be a reality.

Financial & competing interests disclosure

The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.