This chapter should be cited as follows:
Schleiss MR, Glob. libr. women's med.,
ISSN: 1756-2228; DOI 10.3843/GLOWM.419483

The Continuous Textbook of Women’s Medicine Series – Obstetrics Module

Volume 17

Maternal immunization

Volume Editors: Professor Asma Khalil, The Royal College of Obstetricians and Gynaecologists, London, UK; Fetal Medicine Unit, Department of Obstetrics and Gynaecology, St George’s University Hospitals NHS Foundation Trust, London, UK
Professor Flor M Munoz, Baylor College of Medicine, TX, USA
Professor Ajoke Sobanjo-ter Meulen, University of Washington, Seattle, WA, USA

Chapter

Cytomegalovirus

First published: May 2023

Study Assessment Option

By completing 4 multiple-choice questions (randomly selected) after studying this chapter readers can qualify for Continuing Professional Development awards from FIGO plus a Study Completion Certificate from GLOWM
See end of chapter for details

INTRODUCTION

Among agents that cause congenital infection, cytomegalovirus (CMV) is the most commonly encountered viral pathogen, and is globally the most important infectious disease responsible for adverse neurological development in infants. With an estimated birth prevalence in the developed world of 0.64%, it is currently estimated that 60,000 infants with congenital CMV (cCMV) are born in the United States and Europe annually. Numbers are likely higher in the developing world, but there are limited surveillance data available. Between 10–15% of infected infants will have long-term sequelae, most commonly sensorineural hearing loss (SNHL). Given the morbidity associated with cCMV infection, development of a safe and effective preconception vaccine is a major public health priority. Strikingly, maternal immunity provides, at best, incomplete protection against recurrent CMV infections during pregnancy. These recurrent infections can lead to vertical transmission and subsequent sequelae in infants. These observations complicate the design of vaccines targeted at preventing cCMV transmission. This chapter summarizes the current state of CMV vaccine development, with an emphasis on the areas of knowledge deficit regarding the protective correlates of maternal immunity.

CYTOMEGALOVIRUS VIROLOGY

As a member of the Herpesviridae family, CMV is a highly complex virus, with the largest genome of any human viral pathogen. In humans, there are eight distinct members of this family of viruses; CMV is designated as human herpes virus type 5, and is a member of the betaherpesvirus subfamily.¹ The betaherpesviruses, which include CMV and human herpesviruses type 6 and 7, share the propensity to infect leukocytes. CMV, like all members of the Herpesviridae, has a large, double-stranded DNA genome. The genome is divided into unique long (U_L) and unique short (U_S) segments bracketed by terminal repeats. As a result, the genome can form four different isomeric configurations during replication, a strategy which contributes to generation of strain variants and (possibly) escape from host immune control.² Escape from host immune clearance is also mediated by CMV genes that interfere with both innate and adaptive immunity. These include homologs of G-protein coupled receptors; chemokines; gene products that modify MHC class I and II processing; proteins that interfere with NK cell function; and proteins and microRNAs that both promote latency and inhibit interferon responses.^3,4

An understanding of the morphology of the virus particle helps to inform the current state-of-the-art with respect to vaccine design. As a prototypical enveloped virus, there are three distinct regions of the CMV virus particle. These include an inner icosahedral capsid, which contains a nucleocapsid consisting of the viral genome and accompanying viral proteins; the tegument, a protein-rich layer containing several proteins that are targets of the host T-lymphocyte response to infection (and hence relevant to vaccine design); and the envelope (derived from the host nuclear envelope during the virus replication cycle), which contains several virally encoded glycoproteins that are targets of host humoral immune responses and, as such, represent potential candidates for subunit vaccine development.⁵ CMV proteins that have emerged as potential vaccine candidates are summarized in Table 1, and reviewed in detail below.

CMV transmission in pregnancy: maternal screening and therapy

CMV infection is ubiquitous, and usually asymptomatic. Transmission of CMV occurs chiefly via contact with infected secretions, and the virus is often transmitted to susceptible children early in life via breast milk, saliva, and urine. Conditions such as crowding, lower socioeconomic status, and utilization of group day-care all contribute to the spread of infection in young children. Infections acquired by toddlers in group day-care are of particular concern, since such infections are commonly transmitted back (in the family's home) to a child's susceptible parent, particularly the mother. If infection is transmitted to a pregnant woman, this can increase the risk of cCMV in a subsequent pregnancy. Hence, infections in toddlers can be a significant risk factor for young women that are pregnant or contemplating pregnancy. CMV is also commonly acquired in young adults through sexual activity. In older children and adolescents, a “heterophile-negative” mononucleosis syndrome can be observed with such primary CMV infections.

With respect to CMV infection in pregnant patients, most infections are asymptomatic in nature. It is important to consider whether CMV infections in pregnancy are primary infections or non-primary infections, since primary infections during pregnancy carry a much greater risk of fetal transmission (30–40%) than do non-primary infections (3%); moreover, some studies suggest that fetal infections that occur in the context of a primary maternal infection are more likely to be associated with long-term sequelae.^6,7 It is controversial whether serological screening for CMV antibodies should be offered to all women during pregnancy. The Society for Maternal-Fetal Medicine in the USA does not recommend routine CMV serological screening during pregnancy.⁸ On the other hand, the Canadian counterpart, the Society of Obstetricians and Gynaecologists of Canada (SOGC), does recommend that CMV sero-screening during pregnancy “can be offered”, particularly for patients at high risk, including young women of child-bearing age who have a child under 3 years of age at home.⁹ Table 2 provides a summary of criteria that can be considered by the clinician in making the decision to offer serological screening. The assessment of sero-status optimally includes serological examination for anti-CMV IgM and IgG antibodies, as well as an assessment of the IgG avidity index. Primary infection can be defined, through a combination of positive IgM antibodies; demonstration of seroconversion from IgG negative to IgG positive status; demonstration of CMV, usually by PCR testing for viral nucleic acids, in blood, saliva, urine and other body fluids (including amniotic fluid); and evolution from low-avidity to high-avidity anti-CMV antibody titer.¹⁰ Unfortunately, there are no clinically useful diagnostic studies available that can identify non-primary infections. These infections can be caused by reactivation of a latent strain of CMV or, more commonly, re-infection with a novel, heterologous viral strain.^11,12

Maternal infection during gestation, whether primary or non-primary, rarely causes symptoms. If symptoms are noted, they may include malaise, fever, myalgia, cervical lymphadenopathy, and, less commonly, pneumonia and hepatitis.¹³ Laboratory studies may reveal atypical lymphocytosis or elevated transaminases. Ultrasound is generally insensitive with respect to diagnosis of maternal–fetal CMV transmission, but some findings, in particular echogenic fetal bowel, are highly suggestive, and should prompt evaluation for fetal CMV transmission.¹⁴ Fetal MRI may be warranted, particularly when a screening ultrasound identifies fetal brain abnormalities. Fetal CMV infection can be confirmed by PCR testing of amniotic fluid. Amniocentesis, if performed, is recommended >6 weeks after a presumed onset of maternal infection, and after 17 weeks of gestation, since earlier amniocentesis may yield false-negative results if the fetal CMV infection has not progressed to the stage of virus being shed in fetal urine.¹⁵ Valaciclovir (8 g/d), given from the diagnosis of primary maternal infection up to amniocentesis, decreased vertical transmission rates from 29.8% to 11.1% in a randomized controlled trial of 90 pregnant women; the benefit was most notable for first-trimester, and not peri-conceptual, maternal infections.¹⁶ In a phase II open-label trial, valaciclovir given to pregnant women with a mildly symptomatic fetus was associated with a two-fold enhancement in the likelihood of delivering an asymptomatic neonate with cCMV, compared to historical controls.¹⁷

MATERNAL CORRELATES OF PROTECTIVE IMMUNITY AGAINST CCMV INFECTION

Historically, our understanding of the mechanisms of immune responses capable of preventing or controlling infection have centered around antibody responses. Evidence for this comes from adoptive transfer studies using therapeutic IgG that demonstrate its ability to prevent or ameliorate CMV disease, particularly in transplant patients that are at high risk following the iatrogenic immunosuppression that is required to prevent organ rejection.¹⁸ The presumed protective mechanism has been IgG-mediated virus neutralization. Virally encoded envelope glycoproteins are the target of such protective humoral immunity. The capacity for anti-glycoprotein antibodies to neutralize virus effectively appears to relate to their ability to block interaction with cell-surface receptor complexes. Optimal humoral immunity appears to depend on generation of antibody responses to more than one envelope glycoprotein complex, since pathways of CMV entry into the host cell are diverse and cell-type-specific. For endothelial and epithelial cells, cell entry is mediated by endocytosis in a pH-dependent fashion, whereas entry into fibroblasts is non-endocytic, and pH-independent.¹⁹ Antibodies to the glycoprotein B complex are of particular interest for blocking both pathways of virus entry, whereas antibodies to the viral pentameric complex (consisting of glycoproteins H, L, UL128, UL130 and UL131) are critical for preventing the endothelial/epithelial cell entry pathway.^20,21

Unfortunately, the enthusiasm for a potential protective role for antibody in preventing vertical transmission of CMV has been dampened by data from recent placebo-controlled trials that have failed to demonstrate any impact of CMV hyperimmune globulin (CMV-Ig) on the transplacental spread of virus in pregnant patients with primary CMV infection. A phase IIB, randomized, double-blind study of CMV-Ig conducted in Italy, when compared to saline placebo, did not demonstrate a statistically significant impact of CMV-Ig on the rate of CMV transmission in pregnant women with primary CMV infection.²² The National Institutes of Health (NIH) recently sponsored a multicenter placebo-controlled trial to examine the impact of passive CMV-Ig therapy on congenital CMV transmission in pregnant women with primary CMV infections. This study also failed to demonstrate any benefit of CMV-Ig in reduction of CMV transmission to the fetus.²³ Subsequent to the publication of these studies, it was suggested that CMV-Ig therapy during pregnancy may require both a more frequent dose interval than what has heretofore been employed, and that administration should be performed at earlier time points in pregnancy than that which has been utilized in previous studies of women with documented primary CMV infections; more studies, accordingly, may be warranted.²⁴

Recent studies have also raised questions about whether the beneficial effects of CMV IgG responses are indeed mediated by neutralization of the virus, or whether other, non-neutralizing functions might be important in the control of infection.²⁵ So-called “elite neutralizers” of CMV demonstrate broad specificities against multiple viral glycoproteins, and possess potent effector functions, including antibody-dependent cellular cytotoxicity (ADCC) and antibody-dependent cellular phagocytosis (ADCP), that mediate an antiviral effect through the Fc component of IgG.²⁶ In an analysis of immune responses in a study of glycoprotein B subunit vaccine in CMV seronegative women – a study which demonstrated ~50% vaccine efficacy against acquisition of CMV infection²⁷ – it was noted that ADCP, and not virus-neutralizing antibody, represented the major mechanism of protection.²⁸ In another study of effectors of protection against cCMV transmission, CMV-specific IgG binding and antiviral functions in paired maternal and cord blood sera from transmitting and non-transmitting mother–infant dyads were compared. The investigators found that high-avidity IgG binding to CMV and ADCP were associated with reduced risk of cCMV.²⁹ In light of these observations, past assumptions about protective mechanisms of maternal immunity to CMV, as well as vaccine design strategies, will have to be re-considered.

In contrast to humoral immune responses that may be important in protection of the maternal–fetal dyad against cCMV transmission, less is known about cellular responses and their role in protection. CD4+ and CD8+ cells both play roles in protection against CMV dissemination and disease in patients undergoing hematopoietic stem cell and solid organ transplantation,³⁰ and in the normal development of anti-CMV immunity following infection.^31,32 However, although adoptive transfer of T cells has been shown to be beneficial in primate models of congenital CMV,^33,34 T-cell mediated protection has been more challenging to demonstrate in human pregnancy.^35,36

IMMUNIZATION TO PREVENT CCMV TRANSMISSION DURING PREGNANCY

Given the impact of cCMV on newborn health (in particular, the lifelong disabilities associated with symptomatic infection) development of a vaccine to prevent maternal–fetal transmission is considered to be a major public health priority. One strategy to prevent transplacental transmission of CMV involves passive immunization, via the administration of high-titer CMV immune globulin during pregnancy. As noted above, the use of human CMV-Ig has been commonplace in solid organ transplant recipients, for which good evidence for protection against CMV disease has been shown.¹⁸ Unfortunately, clinical trials examining this strategy in an attempt to prevent cCMV transmission in pregnancy have yielded disappointing results.^22,23 Recent data suggest that a more frequent dosing interval of CMV-Ig may prove beneficial in preventing transmission and/or fetal disease;³⁷ thus, additional trials are probably warranted to continue to explore this approach. At the present time, however, administration of CMV-Ig during pregnancy – either to treat a fetus with documented infection in utero or to prevent cCMV transmission in a high-risk pregnancy – cannot be recommended.

Another strategy is to develop a pre-conception vaccine against CMV that could be given to women of child-bearing age. The intrinsic challenge for this approach is of course the fact that “natural immunity” to CMV is imperfect, and re-infections occur commonly, resulting in some cases in vertical transmission of virus. Vaccines evaluated in clinical trials to date have generally focused on eliciting responses against humoral and cellular targets encoded by the virus (Table 1). In broad terms, this has been approached through two general approaches: live, attenuated (or replication-disabled) vaccines; or subunit vaccines (based on combinations of specific CMV recombinant proteins and/or peptides, or vectored vaccines expressing these immunogens of interest).

Live, attenuated vaccination has included trials of a highly tissue culture-passaged clinical CMV isolate known as the “Towne” strain. Although this strain of CMV was observed to be very safe in clinical trials (the Towne virus was never shown to be shed in vaccinees, nor establish latent infection), as a vaccine it was only weakly immunogenic, and appeared incapable of conferring protection against acquisition of primary CMV infection in young women.³⁸ Efforts to “re-engineer” Towne to enhance its immunogenicity by expressing regions of the genome of another minimally laboratory-passaged strain of virus, the “Toledo” strain, met with some success.^39,40 However, these so-called “Towne-Toledo” chimeras are no longer being evaluated in any clinical trials, for two major reasons. First, neither Towne nor the re-engineered Towne-Toledo chimeric vaccines express the functional PC,⁴¹ the protein complex responsible for CMV infection of epithelial and endothelial cells and, as such a key vaccine target (Table 1). Secondly, since these are replication-competent viruses, they pose at least a theoretical risk of transplacental transmission if inadvertently administered to a pregnant individual – their outstanding safety record notwithstanding.

A novel approach to vaccine design that could induce responses to multiple viral proteins, while demonstrating enhanced safety compared to live-virus vaccines, is the use of a disabled, infectious, single-cycle (or DISC) vaccine. A replication-incompetent DISC vaccine, referred to as V160, is currently in clinical trials. The vaccine includes a laboratory-adapted isolate of CMV, the “AD169” strain, which expresses an intact PC protein. The virus is further modified by linking two essential CMV gene products, the UL51 and IE1/2 proteins, as in-frame fusions with the destabilization domain of the FK506 binding protein, ddFKBP.^42,43,44 This modification renders generation of virus in culture completely dependent on the presence of a small molecule, Shield-1, supplied in trans. Virus can only form in cell culture containing this molecule, since Shield-1 is a purely synthetic ligand that does not exist in nature. Hence the vaccine is safe in the human host, since it can only undergo one round of replication.

Subunit vaccines for CMV are also in a variety of clinical trials. These vaccines focus on cloned, recombinant expression strategies aimed primarily at eliciting antibody responses to the immunodominant envelope glycoproteins gB and the PC and, to a lesser extent, cell-mediated immune responses to the tegument phosphoprotein, pp65 and to the immediate early gene product, IE1 (Table 1). Several expression strategies for CMV vaccines are in various stages of clinical trial evaluation.^45,46 Driven by the modest success of an adjuvanted protein vaccine based on the CMV glycoprotein B^27,47 in preventing primary CMV infection, the “next generation” of protein vaccines under study will include the recombinant PC (in addition to gB) along with novel proprietary adjuvants. CMV gene products including various combinations of gB, PC, pp65 and IE1 can be expressed using non-CMV viral vectors, including modified vaccinia virus Ankara (MVA), vesicular stomatitis virus, and lymphocytic choriomeningitis virus, and these “vectored vaccine” approaches are also in clinical trials.²⁷ Nucleic acid-based approaches, in which the CMV gene product of interest is expressed as a cloned “gene cassette” capable of driving expression of the immunogen, also show promise. Although a DNA vaccine failed to demonstrate efficacy against development of CMV disease in hematopoietic and solid organ transplant patients,^48,49 the mRNA approach to CMV vaccination appears to be substantially more immunogenic.⁵⁰ Driven, in part, by the success of mRNA vaccines against COVID-19 disease, a phase III study of a CMV mRNA vaccine, based on the gB and PC gene products, is ongoing.⁴⁶ Finally, so-called “nanoparticle” vaccines, in which multiple copies of an antigen (or antigenic domain) of interest are expressed in repetitive arrays or as virus-like particles, appear to drive more robust humoral immune responses compared to a soluble form of the gene product of interest,⁵¹ prompting interest in phase I studies of this technology applied to CMV vaccines.

As more knowledge is gained about the critical components of protection for the developing fetus, it is hoped that a licensed vaccine to protect against cCMV will become available in the not-too-distant future. Once licensed, it is unlikely that a CMV vaccine would be used during pregnancy; rather, the goal would be to confer durable pre-pregnancy immunity to the virus in order to prevent primary infection, or, for a women with pre-conception immunity, to enhance her immunity to prevent her against re-infection with a new strain, and to prevent transmission of that strain to the fetus. The target population remains to be defined. Although most modeling of the cost–benefit considerations of a CMV vaccine have focused on the vaccination of 12-year-old boys and girls (as a vaccine targeting adolescents), a case can be made for targeting infants, particularly before they enter child-care settings, since many maternal CMV infections are acquired during second pregnancies from their toddlers enrolled in group day-care.⁵² Immunization should eventually be offered to young women of child-bearing age prior to establishment of pregnancy, irrespective of whether they are CMV-seropositive or CMV-seronegative (this strategy would embrace the goal of “boosting” immunity to reduce the risk of re-infection in the already-seropositive woman). It remains to be elucidated whether the CMV vaccine platforms that are being evaluated for prevention or amelioration of CMV disease in hematopoietic stem cell and solid organ transplant patients will be relevant to prevention of congenital CMV; indeed, we may eventually see licensure of different types of CMV vaccines that target different populations in need of protection. CMV exerts a life-long toll on the immune system⁵³ and is associated (albeit controversially) with other diseases such as atherosclerosis and cancer.⁵⁴ Thus, the benefits of universal CMV vaccination may extend beyond the protection of women of child-bearing age to include all individuals at risk of CMV-associated diseases occurring over the life-course.⁵⁵

PRACTICE RECOMMENDATIONS

ACKNOWLEDGMENTS

Support from the March of Dimes Birth Defects Foundation (FY17–849) and the National Institutes of Health (HD079918) is acknowledged.

Online Study Assessment Option
All readers who are qualified doctors or allied medical professionals can now automatically receive 2 Continuing Professional Development credits from FIGO plus a Study Completion Certificate from GLOWM for successfully answering 4 multiple choice questions (randomly selected) based on the study of this chapter.
Medical students can receive the Study Completion Certificate only.

 

(To find out more about FIGO’s Continuing Professional Development awards programme CLICK HERE)

 

I wish to proceed with Study Assessment for this chapter