Cytomegalovirus Vaccines: At Last, A Major Step Forward

In 1992, Drs. Martha Yow and Gail Demmler published an editorial entitled, “Congenital Cytomegalovirus Disease-20 Years Is Long Enough”¹. This editorial pointed out that, in spite of the long-standing recognition of the major public health importance of congenital cytomegalovirus (CMV) infection, that few solutions had been forthcoming. In no area has the lack of progress been more frustrating than in the realm of CMV vaccine development. Although a number of vaccine strategies have been considered and developed, few have advanced in the clinic to the stage of efficacy testing. One vaccine candidate that reached the stage of efficacy testing was a live, attenuated CMV vaccine, the Towne vaccine. As reviewed by Dr. Stanley Plotkin², three randomized, controlled, double blind studies in renal transplant patients demonstrated that vaccination with Towne did not prevent these patients from being infected with CMV, but that vaccination did considerably modify the severity of disease in this setting. In contrast, the results of a study in adult women with young children attending day-care centers, who were given either Towne vaccine or placebo, were disappointing. The infection rate did not differ between vaccinated and control mothers, although the resistance of naturally infected women to CMV infection was striking. The conclusion was that, although the Towne strain vaccine was immunogenic, the immunity engendered does not appear to be as robust as that which occurs after natural infection, possibly due to the titer of neutralizing antibodies, which was lower after Towne vaccination than after natural infections.

With these results in perspective, preclinical development of many new CMV vaccine approaches in the past 15 years has focused on strategies designed to either optimize the neutralizing antibody response, or to improve the immunogenicity of live, attenuated vaccines³. The CMV envelope glycoprotein B (gB) has been focus of vaccine approaches aimed at inducing neutralizing antibody responses. A variety of expression technologies have been developed to express gB [Table 1], and these vaccines are undergoing evaluation in several clinical studies. Results in animal models have been encouraging with respect of the ability of a gB vaccine to prevent congenital CMV infection and disease⁴. Disappointingly, however, to date no clinical trial of any CMV vaccine has been shown to have an impact on acquisition of CMV infection. To the great excitement of the CMV vaccine community, this situation has now changed. The recent Congenital Cytomegalovirus Workshop at the Centers for Disease Control in Atlanta, Georgia (www.cmvconference2008.com), organized under the outstanding leadership of Drs. Michael Cannon (CDC) and Lenore Pereira (University of California, San Francisco), featured a session on new developments in CMV vaccines. At this meeting, Dr. Robert Pass (University of Alabama-Birmingham) reported the results of a phase II, randomized, double-blinded, placebo control of a recombinant gB vaccine conducted in postpartum women. The vaccine, manufactured by Sanofi Pasteur, consisted of 20 μg of recombinant gB mixed with the adjuvant MF59, administered at a 0, 1 and 6 month schedule to healthy women within 12 months of birth of a newborn. The gB vaccine was found to be safe and well-tolerated, and in an intent-to-treat analysis, vaccination had a significant impact on the probability of a study participant remaining CMV seronegative through the 42 month follow-up period. These data, currently in press in the New England Journal of Medicine, are the first that demonstrate significant efficacy of a CMV vaccine for prevention of infection.

Table 1.

Prioritization and Optimization of CMV Vaccines

Live Virus Vaccines
Vaccine	Advantages	Disadvantages	Solutions/Priorities
Towne Vaccine	Safety, immunogenicity No evidence for viral shedding in vaccines or for latent CMV infection	Failure to prevent CMV infection in transplant patients Failure to prevent CMV infection in young women	Add rIL-12 as vaccine adjuvant Increase dose/potency of current Towne vaccine
Towne/Toledo “Chimera” Vaccines	Safety, immunogenicity in phase I study No evidence for viral shedding in vaccines or for latent CMV infection	Theoretical safety concerns Has not been studied in seronegatives Vaccination did not “boost” immunity in CMV seropositives	Requires further study in CMV seronegative individuals
Subunit Vaccines
Purified Recombinant Glycoprotein B (gB)	Safety, immunogenicity Efficacy in phase II studies (see this report)	Does not contain CMI targets Optimal adjuvant undefined	Future additional efficacy study warranted
DNA Vaccination	Safety, immunogenicity in phase I study Contains CMI target (pp65) Ease of manufacture	Strategies for optimizing immunogenicity require further study Efficacy studies not completed	Efficacy study active in bone marrow transplant patients
Alphavirus Vector System	High level of recombinant protein expression Contains CMI targets (pp65, ie1)	Phase I studies underway No immunogenicity data yet published	Awaiting safety, immunogenicity data from phase I studies
MVA Vector System	Highly attenuated poxvirus vector Excellent safety profile Expresses gB, pp65, ie1	No clinical trial data available	Phase 1 studies envisioned
CMV Polyepitope Vaccine	Polyepitope vaccine which encodes gB and multiple CTL epitopes	No clinical trial data available	Phase 1 studies envisioned

Many important questions about the protective role of the gB vaccine remain to be answered. Is there a critical threshold of neutralizing antibody response required for protective immunity? What is the anticipated duration of protection? If vaccine-induced immunity wanes, would there still be sufficient transplacental immunity to prevent CMV transmission to the fetus, even if primary maternal infection occurs? Although answers to these questions will come in the course of future studies, the importance of this study to the field cannot be overstated. The current version of the gB vaccine may not ultimately be the one licensed for clinical use, but these extremely encouraging findings should help drive more aggressive testing of other CMV vaccines. Hopefully, this will be the first of several efficacy studies that help define the optimal gB expression strategy and the optimal adjuvant(s) for clinical use. Other gB expression strategies in preclinical trials or phase I study [Table 1] include DNA vaccines, being studied under the direction of Dr. Ron Moss and colleagues at Vical (San Diego, California)⁵ and “vectored” vaccines based on attenuated poxvirus vectors such as modified vaccinia virus Ankara, developed by Dr. Don Diamond at City of Hope (Duarte, California),⁶ and alphavirus replicons, pioneered by Drs. Jeff Chulay and Jonathan Smith at AlphaVax (Research Triangle Park, North Carolina)⁷. In addition, novel adjuvants are being examined in Phase 1 studies of purified recombinant gB protein vaccines. The protective efficacy of a CMV vaccine may require inclusion of cell-mediated immune targets such as pp65 or ie1, and a polyepitope vaccine which encodes the extracellular domain of gB and multiple HLA class I & II-restricted CTL epitopes from CMV as a contiguous polypeptide is approaching clinical trial evaluation under the direction of Raj Khanna (Australian Center for Vaccine Development)⁸. At the same time that these subunit-based approaches move forward, continued optimization of the immunogenicity of live, attenuated CMV vaccines requires further study. This might be enabled by inclusion of recombinant cytokines, such as IL-12, in the Towne vaccine⁹, or by genetic engineering of more immunogenic, less attenuated live virus vaccines, an approach taken by MedImmune laboratories (Moutain View, California)¹⁰. Irrespective of the vaccine(s) that eventually become licensed for prevention of CMV infection and disease, the field owes a debt of gratitude to Dr. Pass for his perseverance in the completion of the gB vaccine study that represents – at last – a major step forward. We can only hope that the forthcoming publication of this landmark paper will fuel interest in efficacy testing of other CMV vaccines. More rapid progress can be achieved through the combination of: 1) increased advocacy for public education about the problem of congenital CMV by the CDC; 2) more aggressive product development and testing by vaccine manufacturers; and 3) increased funding for CMV vaccine research by the National Institutes of Health. Given the enormous medical, financial and emotional impact of congenital CMV on babies and families, we can all hope this study ensures that we won’t have to wait another 20 years for a solution to this urgent public health problem.