Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Cytomegalovirus Vaccines
Cytomegalovirus (CMV) is a ubiquitous β-herpesvirus that only replicates in human cells and hence is only transmitted by person-to-person spread. CMV infections are generally asymptomatic in immunocompetent individuals but may produce a heterophile-negative mononucleosis syndrome in approximately 10% of primary infections in older children and adults. Like other herpesviruses, it becomes latent after primary infection but may reactivate from latency, particularly in the setting of immune suppression, leading to disease. Infection typically requires intimate contact with body fluids (blood, urine, saliva, breast milk). There is no seasonality to infection. Patient populations with an increased incidence of primary infection include breastfeeding infants, toddlers and care providers in group childcare settings, and sexually active adolescents. Seroprevalence is higher among nonwhites and among individuals of lower socioeconomic status.
From a public health perspective, the most important medical impact of CMV is the damage caused to a fetus when infection occurs in utero . A meta-analysis of published studies concluded that the overall birth prevalence of congenital CMV infection was 0.64%, although this study also noted that rates varied considerably among different study populations. Current estimates suggest that there are approximately 60,000 congenital infections annually in the United States and Europe ( Table 18.1 ). Nonwhite race, low socioeconomic status, premature birth, and neonatal intensive care unit admittance are risk factors for congenital CMV infection. In the United States, congenital CMV is increasingly recognized as a disease associated with health disparities, disproportionately affecting black infants. Congenital infections correlate directly with maternal CMV seroprevalence in the subpopulation assessed. Recent efforts to develop and implement newborn screening programs for congenital CMV should provide more rigorous estimates regarding the overall prevalence of this infection in future studies. , Notably, the prevalence of congenital CMV infection in developing countries is even higher, ranging from 0.6% to 6.1%. The risk of intrauterine transmission is highest when primary infection occurs during pregnancy, with a significantly increased risk of adverse fetal effects if infection occurs during the first half of pregnancy (and in particular, the first trimester). , CMV may also cause disabilities following second and third trimester infection, although the magnitude and severity are reduced. a CMV infection of the placenta interferes with the maintenance and differentiation of trophoblast progenitor cells, possibly resulting in decreased transport of oxygen and substrate to the fetus and leading, through indirect mechanisms, to intrauterine growth retardation. , Among congenitally infected infants, approximately 10% have signs and symptoms of disease at birth, and these symptomatic infants are reported to have a 40– 90% risk of subsequent neurologic sequelae, including mental retardation, microcephaly, developmental delay, seizure disorders, and cerebral palsy. Although the remaining 90% of infants are asymptomatic at birth, between 7% and 20% of these children are reported to subsequently develop permanent sequelae, particularly sensorineural hearing loss. Overall, it has been estimated that in a given cohort of 1,000 infants with congenital CMV, 170–190 will have permanent sequelae, of whom one-third are from the symptomatic group and two-thirds are from the asymptomatic group ( Fig. 18.1 ).
TABLE 18.1
Annual Impact of Estimated Congenital Cytomegalovirus Infection and Disease in Europe and the United States
| Category of Infants | Total |
|---|---|
| Total number of live births per year | 8,600,000 |
| Average rate of congenital CMV infection (%) | 0.7 |
| Total number of newborns with CMV infection | 60,200 |
| Number with symptomatic infection (12.7%) | 7,645 |
| Number with fatal disease (∼4%) | 306 |
| Number of survivors with sequelae (40–58%) | 3,058–4,434 |
| Number with asymptomatic infection (87.3%) | 52,555 |
| Number with sequelae (13.5%) | 7,095 |
| Total number with sequelae or death | 10,459–11,835 |
CMV, cytomegalovirus.
From Dollard SC, Grosse SD, Ross DS. New estimates of the prevalence of neurological and sensory sequelae and mortality associated with congenital cytomegalovirus infection. Rev Med Virol . 2007;17:355–363 .
Current estimates of magnitude of long-term sequelae in infants with congenital cytomegalovirus (CMV) infection in the United States. In a given cohort of 1,000 infants with congenital CMV infection, 170–190 will have permanent sequelae, with one-third from the symptomatic group and two-thirds from the asymptomatic group. HCMV, human cytomegalovirus.
(Data from Dollard SC, Grosse SD, Ross DS. New estimates of the prevalence of neurological and sensory sequelae and mortality associated with congenital cytomegalovirus infection. Rev Med Virol. 2007;17:355–363).
The public health impact of congenital CMV infection is significant and underrecognized, and carries with it substantial costs to society. In the early 1990s, the disease burden associated with congenital CMV infection was estimated to cost the U.S. healthcare system approximately $1.9 billion annually, with an estimated cost of more than $300,000 per child. Although partially effective antiviral therapy is available, development of a CMV vaccine is the most promising strategy for addressing the problem of congenital CMV. An effective vaccine could, by preventing neurological sequelae and other disabilities, provide a lifetime of benefit for newborns. A study modeling the financial impact of routine vaccination of adolescent females demonstrated that such a strategy would be highly cost-effective. A report from the Institute of Medicine of the National Academy of Sciences categorized development of a CMV vaccine for adolescents as a “Level 1” (highest level) priority. The optimal timing for administration of a CMV vaccine aimed at preventing congenital infections may depend on the age-dependent seroprevalence in the population being vaccinated. Some models suggest that immunization of toddlers and young children may be preferable to vaccination of adolescents, as this strategy could protect young women of child-bearing age from exposure to CMV by decreasing or eliminating infectious virus in the urine and secretions of their own children that acquire the infection via their attendance at group child-care centers.
In addition to young women of childbearing age, patient populations that could benefit from development of a CMV vaccine include immunocompromised patients, in particular solid organ transplant (SOT) patients and hematopoietic stem cell transplant (HSCT) recipients, who are at high risk for CMV disease (manifested variably as interstitial pneumonia, hepatitis, nephritis, encephalitis, bone marrow depression, and potentiation of bacterial and fungal infections). CMV-seropositive HSCT recipients and seronegative recipients with a seropositive donor have an enhanced and persistent risk when compared with seronegative recipients of a seronegative donor. SOT recipients are also at risk for CMV disease and graft rejection, particularly seronegative recipients of organs from seropositive donors. , , Although the advent of highly active antiretroviral therapy has dramatically decreased the risk of CMV disease in HIV-infected patients in the developed world, HIV-CMV interactions continue to be of substantial importance in the developing world, where HIV therapies may be less available. This may have particular importance in the context of vertical transmission, since maternal HIV infection confers an increased risk of congenital CMV infection, even if the infant remains HIV uninfected. CMV vaccines may ultimately also prove to be useful in this high-risk setting. In addition, an emerging body of evidence suggests that CMV infection may play a role in the pathogenesis of atherosclerosis, autoimmune diseases, and malignancies, in particular glioblastoma multiforme. CMV serostatus may also impact the clinical course of burns, trauma, and sepsis. CMV seropositivity has also been shown to affect B-, T-, and natural killer (NK)-cell responses to influenza vaccination. Finally, studies have noted an intriguing association between CMV infection and immunosenescence. In light of these observations, universal vaccination against CMV might confer benefits to all individuals over the course of a lifetime, and not just women of reproductive age and the special populations described above.
CORRELATES OF CYTOMEGALOVIRUS IMMUNITY: IMPLICATIONS FOR VACCINE DESIGN
Congenital CMV infection can occur as the result of a primary infection, reactivation of latent infection, or reinfection with a new strain, although the relative importance of reactivation versus reinfection has not been fully elucidated. The degree of effectiveness of natural immunity in preventing congenital CMV infection and its attendant sequelae is not established with certainty. A number of studies have described fetal CMV transmission in women with preconception immunity, generally believed to be a result of reinfection with new strains of CMV or, less likely, reactivation of latent infection. Such infections can produce congenital infection with signs and symptoms of disease in the newborn indistinguishable from those associated with primary maternal infection during pregnancy. Moreover, sequelae in the setting of nonprimary maternal infection identical to those observed in congenitally infected infants born to women with primary CMV infection during pregnancy can ensue. These observations complicate CMV vaccine design, and suggest that (a) for full protection, a CMV vaccine may need to stimulate responses superior to those conferred by natural immunity, and (b) there may be a strong rationale for vaccinating seropositive individuals, with the goal of increasing the level of baseline immunity conferred by prior infection, toward the goal of preventing reinfection and, in the context of pregnancy, subsequent vertical transmission.
Although imperfect, several studies suggest that preconception immunity does confer reduction in both the risk of congenital CMV transmission and the risk of sequelae if transmission occurs. Reported rates of transmission of CMV to the fetus in the setting of proven primary maternal infection are quite high, with ranges spanning 24–75%; an overall risk of transmission of approximately 30% was estimated in a 2007 metaanalysis. Transmission rates are much lower (in the range of 1–2%) in women with preconception immunity who are reinfected during pregnancy, compared to women with primary infection, although the relative contributions of reinfections due to secondary infection with novel strains, versus reinfection due to reactivation of latent strains, remain incompletely defined. Several studies have compared transmission of CMV to the fetus from seronegative women and seropositive women, in an attempt to gauge the protective benefit of preconception immunity. Attributable protection in seropositive women was 60% in one study and 91% in another. , In a study of seropositive women with serologic and/or virologic evidence of recent reinfection, a fetal transmission rate of 3.4% was reported in the setting of nonprimary maternal infection, considerably lower than the 30–40% rate of cCMV transmission in the setting of primary CMV infection. Another recent retrospective study of saliva screening in newborns demonstrated that the risk of fetal transmission is fourfold higher in primary maternal infection than in nonprimary maternal infection. A recent analysis of existing studies of the impact of maternal CMV serostatus on placental transmission concluded that the calculated placental transmission rate was lower in nonprimary infection than that observed during primary infection.
In addition to decreasing the risk of infection, there is some evidence that preconception immunity also reduces the risk of severe sequelae in those infants that do acquire infection in utero , compared to infected infants born to women with primary infections during pregnancy. It has been reported that 25% of congenitally infected infants born to mothers with a primary CMV infection during pregnancy had at least one sequela, compared with only 8% in infants born to women with recurrent infection. In infants with congenital CMV, sensorineural hearing loss is of reduced severity in those born to women with preconception immunity, compared to children delivered to women with primary CMV infections acquired during pregnancy. , In a retrospective cohort study, neuroimaging abnormalities in CMV-infected infants were less frequently observed in the setting of nonprimary maternal infections (8.3%) compared to those born to women with primary infection (76.8%). In a study of 237 mother-infant pairs in Texas identified through a combination of maternal CMV IgG and IgM antibody serologic screening and newborn screening using urine CMV culture followed longitudinally with serial hearing evaluations up to 18 years of age, sensorineural hearing loss was more common after maternal primary infection. On the other hand, two recent Italian studies that compared infant outcomes of congenital cytomegalovirus infection following maternal primary and nonprimary infections reported no differences in neurodevelopmental and hearing sequelae attributable to the type of maternal CMV infection. , A meta-analysis examining studies in which neonatal symptoms and/or long-term sequelae in infants and children with cCMV were described in the context of whether maternal infections were primary or nonprimary similarly concluded that maternal infection status did not influence neonatal symptomatic disease, risk of developing hearing loss, or other neurologic outcomes. Clearly, more studies are needed to resolve this question, which has major implications for CMV vaccination. An increased emphasis on maternal CMV serological screening in pregnancy, coupled with universal congenital CMV screening programs linked to long-term audiologic and neurodevelopment follow-up, could help provide new knowledge that could help to resolve this question.
Another major hurdle encumbering CMV vaccine development is the lack of clearly defined correlates of protective immunity for the pregnant woman, the maternal-placental interface, and the developing fetus. Considering that human CMV is a large and complex virus encoding at least 165 gene products as well as 20 viral microRNAs, it is not surprising that immune control involves more than one arm of the immune system. Several lines of evidence suggest a key role for antibody in protection against congenital infection and disease. In an animal model of congenital infection, antibodies to virally encoded envelope glycoproteins have been shown to be protective against congenital guinea pig CMV infection. In a study in infants performed in Ugandan mother-infant pairs, the IgG titer to the immunodominant envelope glycoprotein B (gB) was found to be higher in infants without CMV infection and was moderately associated with delayed CMV acquisition at 6 months of age. Much of the presumed benefit of anti-gB antibody response has been attributed to the virus-neutralizing capabilities of these antibodies. On the other hand, other studies have failed to demonstrate any relationship between anti-gB and/or virus-neutralizing titers with respect to the risk of development of sequelae following congenital CMV infection. Other parameters of the IgG response, in particular IgG avidity index, may be more important in mitigating maternal-fetal transmission. Mechanistic considerations of the CMV antibody response that inform and direct current vaccine design are considered below.
The concept that anti-CMV immunoglobulin has the potential to prevent CMV infection and disease has been demonstrated in SOT recipients, in whom good evidence for a magnitude of approximately 50% with respect to prevention of CMV disease in the post-transplant setting has been noted. Although this is not a pregnancy population, the benefits of antibody are nonetheless clear. In addition to its ability to modify the risk of infection in transplant recipients, the protective value of antibodies has also been demonstrated in studies of premature infants in the context of postnatally acquired CMV infections, which are typically acquired by consumption of CMV-positive breast milk from seropositive lactating women. ,
Another important setting in which the potential benefit of anti-CMV antibodies has been evaluated is in studies focused on preventing transplacental transfer of CMV to the developing fetus in women that acquire primary CMV infections during pregnancy. In addition to the potential for preventing transfer of CMV to the fetus, there has been hope that antibody administration may treat a CMV infection acquired by a fetus in utero if vertical transmission has occurred, thereby potentially ameliorating the severity and sequelae of CMV disease after birth. In an uncontrolled study conducted by Nigro and colleagues, administration of hyperimmune CMV immunoglobulin was associated with significantly lower risks of congenital CMV infection and disease. However, in a Phase IIB, randomized, double-blind study conducted by Revello and colleagues (CHIP study), CMV hyperimmunoglobulin, when compared to saline placebo, did not have a statistically significant impact on the rate of CMV transmission in pregnant women with primary CMV infection. , The National Institutes of Health (NIH) recently sponsored a multicenter trial to try to resolve the question of the impact of passive CMV immunoglobulin therapy on congenital CMV transmission in pregnant women with primary CMV infections, with primary infection diagnosed prior to 24 weeks’ gestational age. These results have recently been reported. Outcome data were available for 394 women in this multicenter trial. The results demonstrated that administration of CMV hyperimmune globulin did not result in a lower incidence of a composite of congenital CMV infection or perinatal death than placebo. , a The methodology employed in this NIH-sponsored trial has been questioned by some investigators, and other studies suggest that any benefit of passively administered CMV-Ig during pregnancy requires a more frequent dose interval, and administration at earlier time points in at-risk pregnancies, than has been utilized in previous studies. ,
Complicating the analysis of the contribution of anti-CMV IgG to protective immunity for the developing fetus is the uncertainty of the mechanism by which IgG confers protection. For decades, the protective mechanism of IgG was presumed to be mediated by virus neutralization, with the key targets being virally encoded envelope glycoproteins ( Fig. 18.2A ). CMV-neutralizing antibodies were typically measured in the laboratory based upon their ability to prevent CMV infection of human fibroblast cells. Most of this CMV-neutralizing antibody response was believed to target the envelope glycoprotein B (gB). Although inclusion of gB has been a cornerstone of CMV vaccine design, recent years have seen emergence of other glycoprotein targets in CMV vaccine discussions. In 2005, Wang and Shenk undertook a detailed analysis of CMV infection of epithelial cells, and it was soon demonstrated that infection of these cells was dependent on the presence of intact CMV genes capable of encoding proteins UL128, UL130, and UL131. The corresponding open reading frames for these gene products, however, were found to be routinely deleted or mutated in viral strains that had undergone a high number of serial tissue culture passages in fibroblast cells. These proteins were found to be essential components of a pentameric complex (PC) of proteins that included the gH and gL glycoproteins ( Fig. 18.2B ). The gH and gL proteins can also form a trimer with another CMV glycoprotein, gO, although the role of this complex in vaccine design has not been explored. The emergence of the PC in CMV vaccine design, however, has represented a major paradigm shift. This is driven by the observation that virus neutralization at the virus-epithelial cell interface can be mediated by antibodies to the PC (described below; see Table 18.2 ). Indeed, an analysis of the virus-neutralizing characteristics of CMV hyperimmunoglobulin performed by serial depletion of antibody to cell-surface expressed surface proteins indicated that the major neutralizing antibody response was directed to PC, suggesting that the contribution of anti-gB antibodies to the overall CMV neutralization profile may have been overemphasized in past studies. Studies by the group in Pavia, Italy, suggested that early appearance of antibodies against the PC correlated with protection against transplacental virus transmission in women with primary CMV infection, while development of gB antibodies was not protective. Other analyses have suggested that the quantity of anti-PC antibody produced during primary maternal infection correlates with the prevention of transmission. On the other hand, when antibody depletion studies using highly purified pentamer complexes were performed on sera in a Brazilian cohort, it was found that there was no difference in the titer of anti-PC antibody in groups of immune women who transmitted CMV to the fetus, and those that did not. Clearly, even as candidate CMV immunizations including the PC as a vaccine target move forward in clinical trials, more work is needed in elucidating the role of the PC in protective immunity against transplacental CMV transmission.
TABLE 18.2
Cytomegalovirus-Encoded Proteins That Might Be Included in a Subunit Vaccine
| CMV Gene Product | Protein Function and Characteristics of Host Immune Response |
|---|---|
| E nvelope G lycoproteins | |
| gB | Major target of neutralizing antibodies; target of CTLs; fusogenic, involved in cell entry in all cell types; trimeric; prefusion conformation recently elucidated |
| gH/gL | Important target of neutralizing antibodies; required for host cell entry of virus; target of CTLs |
| gH/gL/gO | Trimer; with gB, required for entry of virus into fibroblasts |
| gH/UL116 | Glycoprotein complex; essential for gH trafficking and incorporation into virion envelope |
| gH, gL, UL128–131 pentameric complex (PC) | PC of gH/gL/UL128/UL130/UL131 on viral envelope; required for CMV entry, with gB, into endothelial/epithelial cells; target of neutralizing antibodies; antibodies neutralize CMV infection at epithelial and endothelial cell surfaces |
| gH/gL/gB | Trimer; key complex for cell fusion with virus |
| gM/gN | Targets of neutralizing antibody responses |
| S tructural P roteins | |
| pp65 | Major target of CTLs; target of non-neutralizing antibody responses |
| pp150, pp28 | Targets of CTLs and non-neutralizing antibody responses |
| pp50 | Target of CTLs |
| pp71, pp52 | Targets of non-neutralizing antibody responses |
| N onstructural P roteins | |
| IE1 | Major target of CTLs; target of non-neutralizing antibody responses |
| I mmunomodulatory P roteins | |
| US28 | Type 3 transmembrane protein; G-protein coupled receptor |
| RL11 | Type 1 transmembrane protein; NK cell evasin |
| UL5 | RL11 gene family; NK cell evasin |
| UL111a | Viral IL-10 |
| UL141 | Type I transmembrane glycoprotein; NK cell evasin |
| UL16 | Type I transmembrane glycoprotein; NK cell evasin |
CMV, cytomegalovirus; CTL, cytotoxic T lymphocyte; IE1, immediate-early antigen 1; PC, pentameric complex; NK, natural killer; IL-10, interleukin-10
Another area of active investigation that impacts on CMV vaccine design is the recent enhanced understanding of the importance of non-neutralizing antibody responses to viral proteins, both in the context of natural infection and in response to vaccination (discussed in detail below for subunit gB vaccines). Antibody-dependent cellular phagocytosis (ADCP) has recently been identified as a non-neutralizing component of anti-CMV immunity induced by gB vaccination. Other studies have demonstrated, in cell culture, that blocking membrane fusion mediated by gB can only be blocked by monoclonal antibodies targeting antigenic domain 5 (AD-5), underscoring the potential importance of targeting this domain in future vaccine studies. Protection against primary CMV infection in gB vaccine studies (summarized below) appears to be associated with serum IgG binding to gB molecules present on the cell surface but not to binding to soluble vaccine antigen, suggesting that IgG binding to the cell-associated conformation of gB, and not soluble gB, is the key immune correlate of vaccine efficacy. Vaccine-induced antibodies that target another specific antigenic domain of gB, antigenic domain 2 (AD-2; reviewed in more detail below), also appeared to be effectors of protection in SOT patients receiving gB vaccine, although the protective effect was mediated neither by neutralization nor antibody-dependent cellular cytotoxicity (ADCC). The AD-2 domain is also a target for IgA antibodies, elicited both in infected individuals and in those receiving gB/MF59 vaccine. Vaccination of seropositive individuals improved pre-existing gB-specific IgA and IgG levels and induced de novo gB-specific IgA and IgG responses in sero-negative recipients. When responses to the AD-2 domain were specifically examined, pre-existing IgG and IgA responses were boosted with vaccination in seropositives, but de novo AD-2 responses were not detected in seronegatives. The re-thinking of the immunological basis of protection against CMV conferred by IgG is further supported by mouse studies demonstrating protection against murine CMV disease conferred by non-neutralizing monoclonal antibodies in mouse challenge models.
In addition to the key role that antibody plays as a component of protection, it is evident that cellular responses are also important in congenital CMV infection. In the context of CMV and transplantation, the correlation between recovery of host cellular immunity and freedom from CMV disease has been apparent for many years. During primary CMV infection in healthy, immunocompetent adults, early development of CMV cell-mediated immune responses, particularly to the immunodominant CMV tegument phosphoprotein, pp65 (also known as the UL83 gene product), and the immediate-early antigen 1, IE1 is noted. The magnitude of these responses, as measured by ELISPOT assay detecting IFN-γ secreting CMV-specific T-cell responses after in vitro stimulation with viral antigens, has been shown to correlate with viral clearance. CD4+ cells sensitized to the virus appear to be necessary to resolve the persistent excretion of CMV following infection of young children, and delayed development of a CMV-specific CD4+ T-cell response has been observed in infected mothers who transmitted virus to the fetus. When transmitting and nontransmitting women were studied in the setting of primary CMV infection during pregnancy, the ELISPOT response of CD4+ cells to pp65 was significantly higher in nontransmitting mothers ; moreover, it was noted that there was significantly higher IL-2 production by CMV-specific CD4+ T-cells among nontransmitters, suggesting a protective role for this cytokine against transmission. CD8+ cytotoxic T cells are also important in immunity to CMV. This has been confirmed by the demonstration of the therapeutic efficacy of adoptive transfer of both pp65-specific and mixed populations of cytotoxic T lymphocyte (CTL) cells to bone marrow transplant recipients in preventing development of CMV disease in the post-transplantation setting. , , Interestingly, CMV-specific CD8+ cells are also found in congenitally infected children who have infections that do not disseminate after birth, , although T-cell responses in these infants are reduced in both frequency and functionality when compared to adult subjects. The targets of CD8+ T cells that appear to be of paramount importance in protection are IE1 and the pp65 tegument protein. T-cell epitopes encoded by additional CMV proteins may also be required for optimal T-cell-mediated protection, and may be employed in future subunit vaccines.
Table 18.2 lists several of the CMV proteins thought to be responsible for the induction of neutralizing and non-neutralizing antibodies as well as cellular immune responses (particularly CTL responses) in the normal seropositive host, with an emphasis on those that have been evaluated in the clinical trials of candidate CMV vaccines that are described in more detail below.
VACCINES IN CLINICAL TRIALS
Recent years have seen a notable increase in the number of candidate CMV vaccines that are undergoing human trials. The history of efforts to develop a CMV vaccine has been reviewed. The various vaccine candidates are listed in Table 18.3 and are discussed in detail below. They can be broadly grouped into three categories: live attenuated vaccines ; subunit vaccines based on recombinant proteins and peptides; and vectored vaccines comprising combinations of key CMV immunogens in various expression systems. These categories are individually reviewed in the following sections.
TABLE 18.3
Status of Cytomegalovirus Vaccines Evaluated in Clinical Trials
| Live Attenuated and Disabled Virus Vaccines | |
|---|---|
| AD169 vaccine |
|
| Towne vaccine (± rhIL12) |
|
| Towne/Toledo chimera vaccines |
|
| V160–001 replication-defective vaccine |
|
| Subunit and Peptide Vaccines | |
| Glycoprotein B (gB; CHO cell expression) adjuvated with MF59 (Sanofi)/AS01 (GSK) adjuvants studies have been completed; gB/PC subunit protein vaccine vaccine (utilizing proprietery GSK adjuvant system) is currently enrolling subjects (ClinicalTrials.gov Identifier: NCT05089630) |
|
| PADRE-pp65-CMV and Tet-pp65-CMV fusion peptide vaccines ± CpG DNA adjuvant |
|
| Nonamer peptide, NLVPMVATV from pp65; spans the HLA-A*02:01 pp65 binding epitope; administered with a water-in-oil adjuvant, ISA-51 (Montanide®), plus imiquimod |
|
| eVLP Vaccine | |
| eVLP gB vaccine (HEK cells) ± alum adjuvant eVLP pp65 vaccine formulated with granulocyte-macrophage colony-stimulating factor for intradermal administration; or formulated with AS01B adjuvant and administered intramuscularly |
|
| Vectored Vaccines | |
| Glycoprotein B/canarypox vector |
|
| pp65 (UL83)/canarypox vector |
|
| pp65, IE1-exon 4, IE1-exon 5/MVA vector (Triplex) |
|
| gB/pp65/IE1 trivalent DNA vaccine; gB/pp65 bivalent DNA vaccine |
|
| gB/pp65/IE1 alphavirus replicon trivalent vaccine |
|
| gB/pp65 LCMV bivalent vectored vaccine |
|
CMV, cytomegalovirus; CpG, cytosine phosphate guanine; CTL, cytotoxic T lymphocyte cell; IE1, immediate-early antigen 1; IL, interleukin; LCMV, lymphocytic choriomeningitis virus; PC, pentameric complex; rhIL, recombinant human interleukin.
Live Virus Vaccines
First Generation Live, Attenuated CMV Vaccines
The first approach to CMV vaccine development employed live attenuated strains of virus passaged in tissue culture. Elek and Stern initially used the AD-169 laboratory strain to immunize normal adults, but the strain was not further developed. Plotkin and colleagues isolated a strain (Towne) from a congenitally infected infant and passaged it in human embryo fibroblasts until the 125th passage, in the process performing three clonings by plaque purification; pools were then prepared at the 128th passage for vaccine trials. Initial clinical trials were performed in healthy adult volunteers. , In these studies when the vaccine was given subcutaneously or intramuscularly, seroconversion was seen in nearly 100% of volunteers, and all clinical laboratory monitoring yielded normal results. In contrast, intranasal administration did not stimulate immunity. No virus was excreted in the pharynx or urine and virus was not recovered from the blood, which was unexpected at the time, but today is understood to be, at least in part, secondary to genetic modification in the PC coding genes (discussed below). Subsequent studies showed the persistence of antibodies for at least several years in adult female pediatric nurses who were vaccinated parenterally, by subcutaneous route, with the Towne strain.
In the Towne parentally administered vaccine studies, lymphocyte proliferation assays demonstrated sensitization to CMV antigens induced by vaccination. CD8+ cell-mediated human leukocyte antigen (HLA)-restricted cytotoxicity of CMV-infected cells also was elicited in the Towne vaccinees. , Towne virus thus induced both cellular and humoral immunity to CMV. , Studies of cellular immune responses in Towne vaccine recipients performed by Jacobson confirmed that all vaccinees developed both CD4+- and CD8+-mediated responses, but over the course of 12 months those responses declined, particularly to the pp65 protein. Sera of vaccinees were tested for antibodies to early antigens (EAs) and immediate early antigens (IEAs) of the virus, antigens not present in the virus particle (and thus only synthesized during viral replication). , Assessment of these serologic data, as well as biopsy and polymerase chain reaction (PCR) data, led to the conclusion that Towne vaccine produced an abortive infection at the site of inoculation. Immune responses, including delayed-type hypersensitivity, were lower than those observed after natural infection, perhaps because no replication occurred after vaccination.
Vaccine Efficacy in Transplantation Patients
To evaluate efficacy of Towne vaccine, advantage was taken of the high morbidity and mortality that CMV causes in seronegative renal transplant recipients who receive a kidney from a seropositive donor. After a pilot study to demonstrate tolerance of the vaccine, controlled, double-blind trials were conducted in prospective renal transplant recipients at academic hospitals at both the University of Pennsylvania and the University of Minnesota. The seronegative patients were randomized to receive vaccine or placebo subcutaneously prior to transplantation. Following transplantation, the patients were observed clinically, virologically, and serologically and scored for CMV-associated disease severity by individuals blind to the patients’ vaccine status. Despite the induction of relatively poor antibody and cellular immune responses, the vaccine appeared to provide protection against severe CMV disease. Two additional studies in renal transplantation patients, one of which was a multicenter study, reached the same conclusion. , Thus, Towne immunization rendered seronegative patients more resistant to the effects of CMV infection. Notably, no study has ever demonstrated evidence of shedding of Towne virus in any vaccine recipient; this has been confirmed by examination of viral strains in subjects in Towne vaccine studies that were later found to be shedding (non-Towne) virus. In subjects shedding virus who received kidneys from seropositive donors, viral strains were shown by DNA restriction-endonuclease assays to be identical to strains that were latent in the donor kidneys that had been reactivated after transplantation, and not identical to the Towne strain.
Challenge Studies in Immunocompetent Towne Vaccinees
After this demonstration of safety and efficacy, volunteer Catholic priests who lived in a closed community and who were CMV-seronegative were vaccinated with Towne vaccine to determine whether they could be protected against a subsequent artificially administered challenge. One year later, in company with unvaccinated seronegative and naturally seropositive priests, they were challenged subcutaneously with varying doses of an unrelated and less attenuated wild-type CMV isolate called Toledo. CMV strain Toledo is a low passage clinical isolate that contains significantly fewer genetic differences compared to strains AD169 and Towne, which both contain genomic deletions in the “unique long” (UL) region of the CMV genome, designated as the ULb’ region, spanning CMV open reading frames (ORFs) UL128-151. These sequences are present in all low-passage primary clinical isolates but undergo extensive deletion, rearrangement and mutation after serial passage of virus in cell culture, particularly in fibroblast cells. Toledo contains the ULb’ region, although it is inverted compared to wild-type strains; this inversion causes truncation of UL128 before the third exon, which is displaced along with the downstream polyadenylation signal sequence, and this modification abrogates synthesis of an intact UL128 (and hence generation of the PC). A challenge dose of 1,000 plaque-forming units (PFU) in this study caused illness, even in those who were naturally seropositive; therefore, this dose was not given to other groups. With a reduction to a 100-PFU dose, the challenge virus caused a mild infectious mononucleosis syndrome with virus excretion in seronegative individuals, but it was asymptomatic in naturally seropositive individuals. The illness was accompanied by atypical lymphocytosis, raised hepatic enzymes, excretion of CMV, and generation of CMV-specific immune responses. Towne vaccinees also were protected against disease caused by 100 PFU, but four of 7 vaccinees transiently excreted virus asymptomatically. After injection of 10 PFU, CMV infection and symptoms were seen in the seronegative individuals, whereas both vaccinated and naturally seropositive individuals remained asymptomatic, and 0/5 shed virus with this lower challenge dose. Challenge of CMV-seronegative subjects with 100 pfu of Toledo was noted to cause CMV disease and virus was isolated from two challenged, seronegative subjects. Volunteer 1 had positive throat and urine cultures at 6, 12, and 16 weeks postinoculation, while volunteer 2 had a positive throat culture at week 7 only. Thus, vaccination of normal individuals rendered them resistant to an artificial parenteral challenge, but with protection of lower magnitude than that conferred by natural immunity, and the vaccine virus was capable of replication in seronegative subjects.
Despite the protection thus demonstrated against parenteral infection, in another study Towne vaccine failed to prevent infection of seronegative mothers in close contact with CMV-excreting children, although naturally seropositive women were protected. This controlled trial showed no reduction in the infection rate of Towne-vaccinated mothers compared with placebo-inoculated seronegative mothers. The authors speculated that the 20-fold lower neutralizing antibody levels induced by Towne vaccine in comparison with those observed in the setting of natural seropositivity accounted for the vaccine’s failure. Route of infection may have also played a role, since these young women presumably acquired CMV infection at mucosal surfaces following exposure to infectious body fluids (urine, saliva), in contrast to the parenteral route of inoculation carried out in the Toledo challenge studies.
Strategies to Improve Towne Immunogenicity
Several strategies to improve the immunogenicity of the Towne vaccine have been explored. One approach was to co-administer Towne with recombinant human interleukin (IL)-12 (rhIL-12). The immunogenicity and safety of this approach was evaluated in a Phase I, dose-escalation, randomized clinical trial in CMV-seronegative healthy volunteers. The adjuvant effect of rhIL-12 was associated with dose-related increases in peak anti-CMV antibody titers and improved CD4+ T-cell proliferation responses. The vaccine with adjuvant rhIL-12 at doses up to 2 µg was well tolerated and did not lead to Towne persistence, confirmed by whole blood CMV DNA PCR and urine CMV culture.
In another approach aimed at improving the immunogenicity of the Towne vaccine, a CMV DNA vaccine (described in greater detail in the following section), originally designated as VCL-CT02, was used to prime for memory immune responses to subsequent Towne vaccination. The priming efficacy of VCL-CT02, containing the three CMV genes pp65, IE1, and gB cloned from the AD169 strain, along with Towne vaccine challenge was evaluated in a series of Phase I clinical trials in healthy CMV-seronegative volunteers. The median time to first pp65 T-cell response and gB antibody response after Towne vaccine was 14 days for DNA-primed subjects, and 28 days for controls who were administered Towne vaccine only, suggesting a more rapid induction of antigen-specific responses with the DNA vaccine-primed group. Vaccination with VCL-CT02 was found to safely prime for an anamnestic response following subsequent administration of Towne vaccine. Urine CMV cultures obtained after the Towne booster were negative in all subjects. Preclinical studies of DNA vaccine priming followed by boosting with Towne vaccine in mice demonstrated that priming significantly enhanced T-cell responses against gB, pp65, and IE1 as measured by interferon (IFN)-γ secretion.
You're Reading a Preview
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here