Rubella Vaccines

Clinical Description

Acquired Rubella

Rubella is spread by large particle aerosols; primary implantation and replication occurs in the nasopharynx. The incubation period of rubella is 14–23 days, with most patients developing a rash 14–17 days after exposure. During the first week after exposure, there are no symptoms, but in the second week, lymphadenopathy may be noted, particularly occipital and postauricular, and virus cultures reveal rubella virus in the nasopharynx. Later in the second week, virus appears in the blood. About this time, there may be a prodromal illness consisting of low-grade fever (<39.0°C), malaise, and mild conjunctivitis. If not already present, lymphadenopathy is likely to develop.

At the end of the incubation period, a maculopapular erythematous rash appears on the face and neck in about two thirds of cases, but the clinical to subclinical ratio is variable. The rash may be difficult to detect, particularly on pigmented skin, and is more prominent after hot showers or baths. During a course of 1–3 days, the rash spreads downward and begins to fade. Pharyngeal virus excretion is present in high titer up to 4 days after rash onset, and virus excretion in the pharynx and urine may continue for another 1–2 weeks, but viremia ends with the appearance of antibodies, which is more or less simultaneous with the onset of the rash. Fig. 54.1 illustrates the evolution of acquired rubella.

Although acquired rubella is thought of as a benign disease, arthralgia and arthritis commonly are observed in adults, particularly women, associated with virus replication or latency in the synovia. ^, Chronic arthritis has been reported after rubella infection, although the evidence is weak. ^, Other less-common complications are thrombocytopenia and encephalitis, which may be fatal. ^, Encephalitis has been reported to occur in approximately one in 6,000 cases and is of the postinfectious type, although the limited available pathologic data show little evidence of demyelination. However, the incidence of encephalitis was one in 1,600 cases of rubella in a Japanese outbreak ; and in the South Pacific islands of Tonga and Samoa, the incidence was estimated as between one in 500 and one in 1,000 cases. In 2011–12 in Tunisia, 280 cases of rubella were confirmed, including 39 cases of encephalitis that occurred mainly in adolescents and adults. Another report from the 2011–2012 rubella outbreak in Tunisia described nine cases of encephalitis, most of which were in children. A study done in China between 2010 and 2012 using a multiplex reverse transcription–polymerase chain reaction (PCR) showed that 3.3% of viral encephalitis was caused by rubella. In addition, there is a rare late syndrome of progressive rubella panencephalitis that is inexorable and may also be seen in congenital infection. ^, By analogy to subacute sclerosing panencephalitis caused by measles, progressive rubella panencephalitis is thought to be related to persistence of viral antigens. Guillain-Barré syndrome after rubella has also been reported, although the causal relationship is uncertain. Antibodies to rubella and measles structural proteins are elevated in autoimmune chronic active hepatitis, but there is no evidence for the presence of rubella virus in this disorder.

Rubella virus is one of the agents that causes Fuchs heterochromatic anterior uveitis. This condition is a chronic inflammatory syndrome that may lead to cataracts and glaucoma. Rubella virus RNA and high-titer antibodies have been demonstrated in intraocular fluid of patients with this condition. Intraocular rubella virus was detected in a 28-year-old survivor of congenital rubella syndrome, indicating the ability of the virus to persist and cause Fuchs uveitis. The prevalence of the syndrome has decreased in the United States since the introduction of vaccination.

Congenital Rubella Syndrome

Rubella is the archetypical fetal infectious pathogen (see “Pathogenesis as It Relates to Prevention”). Because all organs of the fetus are affected, it is not surprising that CRS comprises a lengthy list of abnormalities, including anatomic changes resulting from interference with organogenesis and inflammatory effects involving organs such as the liver and spleen ( Table 54.1 ). Histopathological study of several congenital rubella syndrome cases showed widespread presence of rubella antigen in fibroblasts lining the heart and blood vessels, alveolar macrophages, progenitor cells of the outer granular layer of the brain, as well as capillary endothelium in the placenta. Rubella virus replicates in the nasopharynx of affected infants and presents a risk to others. In a Japanese study, 17% of infected infants were still excreting rubella virus 1 year after birth.

The time of infection during gestation is important in relation to the fetal outcome. Infection in the first 2 months often leads to spontaneous abortion; if the pregnancy survives, serious ocular or cardiac disease is likely to result. In contrast, infection late in the first half of pregnancy is more likely to result in isolated hearing impairment. However, the relationship between gestational age and abnormality can be overemphasized because fetal infection, once established, spreads to all organs, and damage may be progressive. Rubella virus probably reaches the placenta through infected monocytes, where it replicates in many different types of cells, including cytotrophoblasts, endothelial cells, amniotic epithelium and cells in the decidua.

Table 54.2 demonstrates that organ specificity is only generally related to the stage of gestational infection with rubella virus. The most common congenital defects are sensorineural deafness, cataracts, pigmentary retinopathy, and patent ductus arteriosus; however, malformations of the pulmonary artery or its branches can be demonstrated in 78% of CRS patients in whom cardiac studies are done, and patent ductus arteriosus is found in 62%. In addition, myriad other abnormalities occur, including glaucoma, endocrinopathies such as diabetes, hyperimmunoglobulinemia M, microcephaly, and mental disability. Ocular defects are particularly varied, including abnormalities of the cornea, the lens, the retina, and uveitis. All infants with congenital cataracts should be screened for CRS, which will turn up many cases. An analysis of prospective studies for distribution of clinical manifestations of CRS is summarized in Table 54.3 .

A comprehensive review of the retrospective and prospective studies estimated that hearing, cardiac, ocular, and intellectual manifestations were found, respectively, in 70%, 31%, 16%, and 6% of CRS infants. However, if mixed-cohort and infant-case series are included, the percentages of cardiac, ocular, and intellectual manifestations increase to 46%, 35%, and 40%, respectively. Various other estimates have been made of the incidence of fetal abnormalities after maternal infection. ^, Many prospective studies of the incidence of fetal rubella have accepted clinical diagnosis of rubella in the mother for case inclusion, but if only virologically confirmed maternal rubella is considered, the rate of transmission to the fetus during the first 10 weeks of pregnancy is as high as 90%. ^, Table 54.4 tabulates data from the two most reliable studies, one from United States and the other from United Kingdom. The first 12 weeks of pregnancy are clearly the most dangerous time for rubella infection in the mother. The incidence of fetal disease declines during the next 4 weeks, and in the 16th to the 20th weeks, only deafness has been reported as a complication. Preconceptional rubella rarely results in fetal infection, but rashes that occur within 12 days of the last menstrual period carry proven risk. ^, In an Irish epidemic, 71% of fetuses infected early in pregnancy were affected.

Damage caused by congenital rubella infection is not always evident at birth: glaucoma may become apparent and late cataracts are possible. Retinal detachments and esophageal problems are common later in life. Of interest, autism is a feature of some late-onset neurologic disease in CRS patients. In addition, a variety of syndromes thought to be autoimmune are seen, including diabetes mellitus and thyroiditis. ^{, ,}

Diabetes developed in 12% to 20% of American, European, and Australian CRS survivors. ^, However, a 40-year follow-up of Japanese CRS patients showed that only 1.1% were diabetics. Although autoimmunity is involved in ordinary diabetes mellitus, a thorough study of CRS patients showed no increase in autoantibodies to pancreatic proteins.

Virology

The agent of rubella is a cubical, medium-sized (50–85 nm), lipid-enveloped virus with an RNA genome belonging to the Matonaviridae family and the genus Rubivirus . Although the other togaviruses are arthropod-borne, there is no evidence for such transmission of rubella. Frey has reviewed the virology of rubella. Apart from the complex lipid envelope derived from the host cell, rubella virus is composed of three proteins: two embedded in the envelope in the form of spikes (E1 and E2) and one (C) composing the capsid. E1 is a glycoprotein with neutralizing and hemagglutinating epitopes, while the function of the E2 glycoprotein is unclear. The three proteins, which have molecular masses of 60,000 kilodalton (kDa) (E1), 42,000–47,000 kDa (E2), and 30,000 kDa (C), are derived from a polypeptide of 110 kDa that is translated from a 245-kDa messenger RNA.

The positive-sense RNA genome of rubella virus contains approximately 9,800 nucleotides and is infectious. ^, The replication strategy of rubella virus is similar to that of the alphaviruses in that both full-length and subgenomic RNAs are produced, and it is from the subgenomic RNA that viral structural proteins are translated. Three other proteins are produced by the virus in infected cells but are not incorporated into the virion. The structural capsid protein interferes with cellular mitochondrial function, inhibiting transport of proteins into the mitochondria and causing morphologic alterations of those organelles. A receptor for the virus has not been identified definitively, but myelin oligodendrocyte glycoprotein may be involved in spread to the placenta and fetal brain.

There is only one serotype of rubella virus, and analyses of sequence variation among occidental isolates show high conservation of amino acid structure (0% to 3.3% differences), with less conservation of isolates from Asia (up to 7%). ^, However, there are two clades and 13 genotypes. ^, For rubella isolates collected before 2000, isolates from North America, Europe, and Japan are closely related to each other and form clade I, whereas clade II comprises some strains from China, Korea, and India. However, in China, between 2001 and 2007, there was a shift of genotypes toward predominance of Clade I genotypes 1E and 1F. Similarly, in France, genotype 1E has become predominant. Isolation of virus from cataracts in Indian CRS cases revealed genotype 2B. Of rubella isolates collected in recent years, four genotypes (1E, 1G, 1J and 2B) had wide geographic distribution, whereas others occurred sporadically or were geographically restricted. ^, Other studies identified the circulation of multiple genotypes, including in Africa, with 1B, 1E, 1G, and 2B being prominent. ^, Genotype 1G seems to be particularly prevalent in Africa. Genotype 2B predominates in India. The genetic differences between clades do not appear to translate into antigenic differences, despite amino acid changes of 3% to 6% in viral proteins. Moreover, isolates from CRS cases are not genetically distinct from isolates from acquired rubella. No significant antigenic drift has occurred in recent years, and isolates from vaccinees showed little variation from vaccine strains. However, recent data show many intercontinental transfers of clades I and II. Because clade II was clearly of Asian origin, and clade I isolates from Asia are highly variable, an Asian origin of rubella virus has been suggested. Genetic analysis allowed the confirmation of the epidemiologic data that implicated spread of endemic rubella strains from Greece to the United Kingdom in 1999.

Two zoonotic viruses genetically related to rubella virus have been identified. One, called Rustrela, was found in various animals in a German zoo. Another, called Ruhugu, was isolated from bats in Uganda, and showed clear genetic relationship to rubella. How Ruhugu evolved to rubella is uncertain, but this finding suggests that, as for other human viruses, pathogens for humans evolved from viruses of animals. ^,

Rubella virus grows in many different primary, semicontinuous, and continuous cells of mammalian origin. In human amnion cells, it produces a subtle cytopathic effect. More pronounced cytopathic effects, sufficient to allow plaque formation, are produced in continuous cell lines, such as rabbit kidney (RK13) and baby hamster kidney (BHK-21). Even in those cell lines, fresh isolates frequently are not cytopathogenic and require adaptation by serial passage. High passage generates defective interfering RNA and particles. Fig. 54.2 describes the replication of rubella virus in cell culture.

Recently, rubella virus has been shown to adapt to growth in granulomas present in immunocompromised vaccinees, as described in the section on vaccine reactions. The chronic replication in the cells of the granuloma results in genetic drift.

Virus isolation is generally performed in primary African green monkey kidney (AGMK) cell culture in which virus growth is detected by “challenging” the cultures with a cytopathogenic agent such as echovirus 11. If rubella virus has infected the cultures, replication of echovirus 11 is blocked, probably by interferon induction; the presence of rubella virus is inferred by this interference. Confirmation is performed by another technique, such as neutralization or fluorescence with specific antirubella serum.

Pathogenesis as It Relates to Prevention

The pathogenesis of acquired rubella infection provides two points at which immune intervention could have an effect. The first point is in the nasopharynx, where the virus first replicates and from which it spreads to local lymph nodes. Secretory immunoglobulin (Ig) A antibody in the nasopharynx, induced by prior disease or vaccination, can block mucosal replication. The second point begins about a week into the incubation period, at which time the viremia can be blocked by the presence of antibody, either passively or actively acquired.

During viremia in a pregnant woman, the virus often infects the placenta. Placental replication appears to precede fetal infections, leading to entrance of virus into the fetal circulation, from which it infects fetal organs. In vitro experiments show that human embryonic cells of many different lineages are susceptible to viral replication and develop chronic noncytopathic infection. Human stem cells differentiate less well after rubella virus infection. The same phenomenon occurs in vivo , except that only a few cells are infected at any one time. If the infected cells are stimulated to divide, either artificially in vitro or in the course of embryologic development in vivo , there is an inhibition of mitosis ^, that may be mediated in part by a soluble protein inhibitor or by the induction of apoptosis. Although low-passage strains produce more pronounced apoptosis, the effect is only moderate in human fibroblasts in culture. In a few organs, including the lens, cochlea, and brain, the damage caused by the virus is more cytopathic. Viral infection of the ciliary body may contribute to cataractogenesis and wild virus destruction of endothelial cells may lead to vasculitis and ischemia. Cytologic studies show that the cell skeleton and mitochondria both are damaged by rubella virus replication. Atreya and coworkers proposed that rubella virus replicase interacts with retinoblastoma tumor suppressor protein and the enzyme citron-K kinase, leading to cell-cycle arrest in S phase and activation of apoptosis. Although studies have shown that rubella infection inhibits cell division and apoptosis in infected cells, Adamo and coworkers emphasized the importance of the innate immune response to rubella infection, leading to interferon induction that causes apoptosis.

In summary, the action of rubella virus on organogenesis is mediated by varying combinations of intracellular pathology, inhibition of cellular replication, and apoptosis. ^,

Diagnosis

Methods for the diagnosis of acquired and congenital rubella have been reviewed and are summarized in Table 54.5 . Clinical diagnosis of acquired infection is so inaccurate as to be useless without laboratory support. When suspected measles cases are subjected to laboratory tests, rubella infection is often identified. Isolation of virus can be accomplished from the blood and nasopharynx during the prodromal period and from the nasopharynx for as long as 2 weeks after eruption, although the likelihood of virus recovery is sharply reduced by 4 days after the rash. The Vero cell line, primary AGMK cells, or the RK13 cell line are generally used for virus isolation; but because of the slow growth of the virus in tissue culture, virus isolation is often bypassed in favor of serologic diagnosis. Isolation of virus is more common from throat swabs than from serum or plasma.

The PCR has been adapted to the detection of rubella RNA by reverse transcription and amplification. The method appears to be 100% sensitive and approximately 90% specific and is particularly useful for prenatal detection of rubella infection of the fetus (see below). Reverse transcription–PCR was shown to permit diagnosis of rubella during the rash before appearance of IgM antibodies. The Taq Man real-time PCR assay has been applied to rubella virus detection.

Initially, serologic diagnosis became feasible because rubella virus hemagglutinates red blood cells, particularly of avian origin, permitting measurement of antibodies by hemagglutination-inhibition (HI) testing. Because HI and neutralization tests are labor intensive, other serologic examinations that are amenable to mass testing have come into use, such as latex agglutination, indirect hemagglutination, enzyme-linked immunosorbent assay (ELISA), single radial hemolysis, and fluorescence inhibition. ^, Any of these methods may be used to measure antibodies to rubella. Detection of antibodies in saliva or urine could simplify testing in developing countries, but commercial kits are not available. Nevertheless, identification of IgM antibody in oral or crevicular fluid permitted efficient diagnosis of rubella in the United Kingdom over a 10-year period.

Serologic diagnosis depends on the demonstration of a fourfold rise in IgG titer between acute and convalescent specimens or a demonstration of IgM antibody in the acute specimen. For IgM testing, ELISA, in a capture format to avoid false positivity because of IgG antibody, is the predominant assay in use; results may be positive for up to 6 weeks after the acute infection.

Susceptibility to rubella infection has been defined as an HI antibody titer of less than 1:8, a hemolysis-in-gel result of less than 10 IU, or an optical density by ELISA at less than the limit set by the manufacturer. However, women may lose detectable antibody and T cell proliferation in the presence of rubella antigens leading to suspicion of susceptibility but yet have protective recall responses when exposed. However, negative immunoblots identify women who are truly susceptible. Waning antibody may be restimulated on exposure, such that protection results, and epidemics have not occurred in well-vaccinated countries even when significant numbers of vaccinees no longer have antibodies at the 10 IU level. Titers at the borderline level are difficult to evaluate, but the consensus is that, although they reflect immunity in most cases, the safest response is to vaccinate the individual with equivocal levels of antibody.

Assays for antibody avidity and for responses to specific proteins or peptides have come into use for diagnosis of recent infection. ^, Low-avidity antibodies indicate recent infection, with maturation to high avidity by about 2 months post exanthem. A careful study of antibody responses to primary rubella infection found an initial low-avidity IgM response, followed by low-avidity IgG ₃ and IgA responses, and then, finally, IgG ₁ responses maturing from low to high avidity. Liebert and coworkers studied various serologic modalities and found that low levels of antibody to the major neutralizing epitope on the E1 protein, poor response to the E2 protein, and low IgG avidity for rubella virus were highly predictive of recent infection in both mother and infant. ^, Antibodies to E2 develop months after infection or vaccination. Rules for diagnosis of rubella in pregnancy have been well summarized.

In contrast to the situation with acquired infection, the recognition of the combination of cataracts, heart disease, and deafness provides a clinical means of diagnosis of CRS that is reasonably accurate. However, laboratory confirmation is always desirable, both because isolated abnormalities may occur and for public health reasons.

Congenital rubella infection in the infant is associated with the detection of virus, viral genome, IgM antibodies by capture techniques, low-avidity IgG antibodies, lack of antibodies to the E1 protein, or antibodies persisting beyond the predicted decay of passively transmitted maternal antibodies. Virus often can be isolated from tissues obtained at biopsy or autopsy or during surgical procedures such as cataract extraction, but more often nasopharyngeal swabs, urine specimens, or cerebrospinal fluid serve as the sources. Almost always, one or more of these sources are positive for virus at birth, gradually becoming negative during the first year of life. In severe cases, virus excretion may persist for several years. A negative result from PCR performed on fetal samples or amniotic fluid during pregnancy is good evidence against intrauterine rubella.

IgM antibodies are present in the congenitally infected infant for as long as a year after birth; however, the IgM titers decline over 6 months, whereas low-avidity IgG antibodies may persist for longer periods. Persistence of IgG antibodies beyond 6 months of age can be detected in 95% of infants with CRS. Salivary specimens are also useful in CRS for detection of antibodies. An infant with seropositive results for rubella after 6 months of age and who has not received rubella vaccine is likely to have been congenitally infected. However, CRS patients may lose their antibodies over time: 60 years after fetal infection, 40% were seronegative. Moreover, sensitization to proliferation of lymphocytes after stimulation with rubella antigens is often negative in CRS infants, even when they are seropositive, and this test may be used for diagnosis in children younger than 3 years of age. Congenital rubella can also be diagnosed by PCR tests of lens aspirates in cases of congenital cataracts.

Diagnosis of CRS has been focused on the first year of life when IgM antibody and viral shedding are likely to be present. However, a recent study of school-age children with CRS identified several serologic markers of fetal infection, including higher rubella-specific IgG and higher relative concentrations of antibodies to the core C and E2 proteins of the virus, but lower relative concentration of antibodies to the E1 protein compared with normal children. Using those markers, CRS could be identified retrospectively with 65% sensitivity and 100% specificity. Reinfection with rubella virus has been documented to rarely cause CRS, in which case IgG3 antibody is usually absent, in contrast to CRS after primary infection.

Significance as a Public Health Problem

The seriousness of congenital rubella as a public health problem can be gauged by the results of the last major American epidemic in 1964–1965 (see Table 54.6 ). An estimated 12.5 million rubella cases occurred, including approximately 2,000 cases of encephalitis. More than 30,000 pregnancies were affected by this epidemic. Of these pregnancies, approximately 5,000 pregnant women chose to have surgical abortions, surely an underestimate, and approximately 6,250 lost fetuses to spontaneous abortions. Another 2,100 infants were stillborn or died soon after birth. CRS occurred in 20,000 infants who survived pregnancy. Of these, 11,600 were deaf, 3,580 blind, and approximately 1,800 mentally disabled. The human misery imposed by such severe damage can well be imagined. The economic burden was estimated to be US$221,660 per child with CRS, and the total cost of the epidemic may have been US$1.5 billion in 1965 dollars.

A more recent economic analysis of the impact of CRS calculated the disability-adjusted life-years and cost of care. For a CRS child in a low-income country, 29 disability-adjusted life-years and US$11,266 in cost of care would be incurred, whereas for a high-income country the figures are 19 disability-adjusted life-years and US$934,000 in cost of care. Before the use of vaccine, rubella epidemics involved approximately 5% of the population, although only approximately 10% of these cases were reported to public health authorities. In the years between epidemics, rates were about one-tenth of the epidemic peaks, but CRS continued at a low endemic rate even in those years.

Analysis of the age distribution of acquired rubella in the prevaccine era showed 60% of cases in children younger than 10 years of age and 23% in persons older than 15 years of age. As discussed subsequently, application of vaccine in children did not reduce the incidence in adolescents and adults by much, and CRS continued to occur for some years after the introduction of vaccination.

Lest it be thought that rubella infection in pregnancy has lost its danger, in 1991, a rubella outbreak among unvaccinated Amish persons living in Pennsylvania serves as a corrective. Young Amish women were 20% seronegative at the beginning of the outbreak, and infections in pregnancy were common. More than 8% of Amish infants born after the outbreak had laboratory evidence of infection, and more than 2% had CRS. Of infants born after first-trimester infection, 90% had confirmed or possible CRS.

In the United Kingdom, CRS estimates before the use of vaccine were only about 200–300 cases per year, but that figure was from a passive ascertainment system. In Australia, about one baby with CRS per 2,000 births was recorded before vaccine. During an outbreak of rubella in Israel, there were 1,441 confirmed infections in pregnant women, most of whom had abortions. Projection of the experience to the United States would be equivalent to 75,000 pregnancies complicated by rubella. In France, 16% of congenital cataract cases were attributable to CRS. A follow-up study of an epidemic in Poland showed that 15% of pregnant women had been infected. The rate of CRS after first, second, and third trimester infection was 78%, 33%, and 0%, respectively. The picture of rubella in the developing world is complex. ^, Although a frequent seroepidemiologic finding has been a high rate of infection early in life, there are many exceptions, including island populations, several West African countries, the city of Calcutta, and Morocco. Thus, the pattern is that of a disease that comes in epidemic waves, but that may be absent from an area for some time.

The epidemiology of CRS is known in detail for only a few countries of the world. Clusters of CRS have been reported in developing countries, ^{, ,} and there is little doubt that, wherever seronegative pregnant women are exposed to a rubella outbreak, CRS cases will follow. A study conducted in Chennai, India, confirmed the presence of rubella virus in 18% of congenital cataract cases, with serologic evidence of infection in 25% (see “Epidemiology” earlier).

The most complete analysis of CRS globally was performed by Vynnycky and Adams. They modeled the incidence of the disease in all WHO member states for the years 1996 and 2010 and derived the estimate of 103,068 annual cases of CRS globally, although the confidence limits were wide, ranging from 17,146 to 269,223 cases. Table 54.8 shows their estimates.

A review of rubella incidence between 2007 and 2018 concluded that the burden of rubella throughout the world has been decreasing, but, not surprisingly, the risk is greater in the countries that have yet to introduce rubella vaccine. An analysis of rubella in high and middle income countries concluded that aside from routine childhood vaccination, informing women of childbearing age about the risks of rubella in pregnancy is crucial to the induction of serological screening and demand for vaccination.

Immune Serum Globulin

Because most adults have had rubella, immune serum globulin (ISG) contains rubella antibody. By the HI test, rubella antibody titers of approximately 1:16 are found in immunoglobulins. Before the development of vaccine, ISG frequently was offered to pregnant women who had been exposed to rubella in the hope that it would prevent fetal infection. The results were equivocal, but if large doses (20–30 mL) of material were given, frank symptoms and viremia could be prevented. Experimental studies confirmed the efficacy of passive antibody in preventing both viremia and clinical rubella, ^, but there were numerous failures of γ-globulin to prevent congenital fetal abnormality in actual practice. ^, In one study of CRS patients, 6% of mothers gave histories of receipt of γ-globulin after exposure. Thus, whatever the real efficacy of ISG, it is unlikely to be complete.

The sole current indication for ISG is the exposure to rubella of a pregnant seronegative woman who will not accept abortion if infection is proved. If 1 week or less has elapsed since exposure and a serum specimen is taken first to confirm susceptibility, ISG in large amounts (20 mL) can be given intramuscularly. Convalescent specimens should be obtained 3 and 4 weeks later to search for IgM antibody and a rise in IgG titer. The absence of a rash in the exposed woman does not mean that viremia and fetal infection have been prevented. The newer intravenous γ-globulins do not have high concentrations of rubella antibodies, presumably because the antibodies are now the result of immunization rather than disease.

Hyperimmune Rubella Globulin

To overcome the deficiencies of ISG, a hyperimmunoglobulin was prepared by Cutter Laboratories (Elkhart, IN) from the sera of normal individuals who had high rubella antibody titers. The titer of rubella virus-specific antibodies in this preparation was 1:8000 by HI. In the late 1960s and early 1970s, controlled clinical trial of the preparation was performed with volunteers who were inoculated first with live unattenuated rubella virus and then given hyperimmune rubella globulin 24–96 hours later. With a challenge given intranasally, viremia was detected in two of five control subjects and in one of 10 subjects given γ-globulin. Pharyngeal excretion was similar in the two groups. Although a previous experimental study with high-titer immunoglobulin had given better results, the production of hyperimmunoglobulin was stopped, and it is not now commercially available.

Nonliving Vaccines

Advances in molecular biology allow consideration of a subunit-inactivated vaccine against rubella. The genome of the virus has been sequenced, in particular the genetic code for the E1 protein, ^, which carries multiple neutralizing epitopes located between amino acids 214 and 285 of the 481 amino acids contained in the polypeptide. T-cell epitopes also were defined on the E1 protein, although no single epitope was recognized by a majority of individuals. The E1 protein has been produced in quantity in baculovirus vectors and Chinese hamster ovary cells by truncating the C terminus to allow secretion. The immunogenicity of bioengineered E1 has been moderate, but strong adjuvants have produced good responses in animals. Synthetic peptides also have generated neutralizing antibodies, and virus-like particles containing the three main viral proteins have been produced from transfected cell lines, with retention of immunogenicity. ^, Other possible strategies for rubella vaccines include nucleic acid vaccine. Pougatcheva and coworkers raised neutralizing antibodies in mice by injecting complementary DNA coding for the envelope glycoproteins. Although it is doubtful that an inactivated vaccine could provide protection from infancy throughout the childbearing period, it might be useful for immunizing adult women.

Live Virus Vaccine

Results of Vaccination

Immune Responses

Vaccination induces antibodies of both IgM and IgG classes and cellular immune responses. Secretory IgA responses are also induced.

Most studies of immunogenicity have been done by measuring HI responses, although the neutralizing responses may be more important biologically. By the HI technique, 95–100% of RA27/3 vaccinees experience seroconversion by 21–28 days after vaccination, with geometric mean antibody titers ranging from 1:30 to 1:300, depending on the method of titration. ^{, , , ,} Some of the apparent failures, at least in young adults, may be explained by preexisting low levels of antibody that neutralize the vaccine virus but are detectable only by sensitive tests. An ELISA titer of 10 IU is considered to confer immunity, but even lower titers may also be protective. Indeed, a meeting in 2017 called by WHO concluded that titers lower than 10 IU could be protective. By searching for antibodies against the E1 protein of the virus using immunoblot, Picone et al. showed that women with low titers by ELISA whose sera gave a positive immunoblot were actually immune. Several direct comparative immunogenicity studies have been done with different strains. In early trials, RA27/3 compared favorably with Cendehill and HPV-77, which are no longer available. The Japanese TO-336 and the Chinese BRD-II strains compared favorably with RA27/3 in parallel studies. ^, A vaccination study conducted in India using MMRV showed 99–100% seroconversion to rubella after one dose. RA27/3 elicits complement-fixing and precipitating antibody titers in nearly all vaccinees. In a precipitin test system, RA27/3 was the only vaccine strain that evoked antibodies to the iota internal antigen of rubella virus. Vaccination induces antibodies predominantly binding the E1 protein as detected by immunoblot, but those antibodies mature in avidity less rapidly than after natural infection and do not reach the same level.

The induction of neutralizing antibody is particularly significant and is seen regularly and promptly after the administration of RA27/3. Fig. 54.3 and Table 54.10 present comparative data obtained in New Haven, Connecticut, and in Sweden. Immunoblot analysis confirmed the persistence of antibodies for at least 3 years to the E1 protein that bears neutralizing epitopes. Antibodies to the C protein also persisted, but antibodies to E2 were often absent. A study in vaccinated children showed persistence of antibodies for at least 10 years. A crucial property of RA27/3 is its ability to induce secretory IgA antibody in the nasopharynx, which, as discussed subsequently, may prevent reinfection with wild virus. This property makes vaccination with RA27/3 similar to natural infection, which also induces local immunity. Although secretory IgA responses are higher after intranasal vaccination, they also are induced by subcutaneous vaccination with RA27/3 because of replication of the virus in the nasopharynx. IgA antibodies were detected 5 years after vaccination with RA27/3, whereas they had disappeared by that time after vaccination with other vaccine strains.

TABLE 54.10

Neutralizing Antibody Response in 114 Rubella-Vaccinated Women Who Demonstrated Seroconversion With Hemagglutination-Inhibiting Antibodies

Vaccine	Before Vaccination a	After Vaccination
		8 Weeks	2 Years
Cendehill	1/45	24/43 (56%)	27/33 (82%)
HPV-77 duck	2/29	23/29 (79%)	16/17 (94%)
RA27/3	1/40	37/39 (95%)	20/20 (100%)

a Number positive/number tested. From Grillner L. Neutralizing antibodies after rubella vaccination of newly delivered women: comparison between three vaccines. Scand J Infect Dis . 1975;7:169–172 .

Not surprisingly, vaccination, like disease, is followed by the early production of IgM-class antibodies. These antibodies reach a peak at 1 month after vaccination and last approximately l month more but may persist longer at low levels.

Cellular immune responses to vaccination have been studied, although they have no known significance to protection. A proliferation of lymphoblasts in response to rubella antigen appears 2 weeks after vaccination, without suppression of tuberculin hypersensitivity. Relatively short-lived cellular responses also were seen in other studies. Human leukocyte antigen (HLA)-restricted T-cell cytotoxicity was shown to increase after immunization. Morag and coworkers showed that cytotoxic T-lymphocyte activity was high in tonsillar lymphocytes after intranasal vaccination (with RA27/3) but was low after subcutaneous vaccination (with HPV-77DE). Unfortunately, subcutaneous administration of RA27/3 was not evaluated. An association of rubella antibody response with class II but not class I HLA antigens has been demonstrated. ^, Elevation of tumor necrosis factor-α (TNF-α), interleukin (IL)-4 (IL-4), and IL-10 after vaccination has been reported. A study of second-dose MMR revealed effects on the distribution of lymphocyte classes, but no functional changes.

A group from the Mayo Clinic studied single nucleotide polymorphisms (SNPs) in relation to both antibody and cellular responses to rubella revaccination and have found numerous SNPs associated with higher or lower responses, although none that eliminated responses entirely. The SNPs were located in HLA genes and cytokine receptor genes. These extensive studies have recently been summarized. Males may have somewhat higher immune responses than females. Viremia has been documented between 7 and 11 days after RA27/3 inoculation. However, the viremia is low and inconstant. Pharyngeal excretion of virus is more frequent, occurring from approximately 7–21 days after vaccination, in low titer of usually less than 10 PFU per swab. Excretion peaks on approximately the 11th day after vaccination; if properly tested, essentially all vaccinees will be shown to excrete virus from the nasopharynx. ^, Several groups have studied third doses of rubella vaccine. Serologic responses were high and were maintained for at least 1 year. A comprehensive review of the properties of the RA27/3 vaccine strain use recently published which concluded that is generates high immunogenicity persistent protection and good safety.

Contact Spread

In view of the excretion of rubella virus by vaccinees, considerable effort has been made to detect the spread of vaccine virus to susceptible contacts. Initially, contact studies were focused on children in institutions and on families, and no evidence for spread of vaccine virus was found. ^{, ,} For example, none of 393 seronegative family members was infected by contact with RA27/3 vaccinees. Veronelli studied 347 familial contacts of HPV-77DE vaccinees and also found no evidence of spread.

Experience in studies of contact spread produced largely negative results, with the rare asymptomatic seroconversion that could not be explained fully. ^, In the early 1970s, many studies were conducted in contacts of HPV-77DE or Cendehill strain vaccinees, including pregnant women, marital partners, school teachers, and the general population, with little or no evidence of spread. The general lack of evidence for spread by vaccine virus may reflect the maintenance of markers of attenuation by excreted virus, as demonstrated for RA27/3.

Rubella in the Measles-Mumps-Rubella Combination

Responses to rubella as part of MMR combinations are equal to those seen after rubella vaccination as a single antigen. Table 54.11 is taken from a study by Weibel and colleagues that compared MMR formulated with RA27/3 to RA27/3 alone. The excellent responses in both groups of vaccinees have been confirmed by other investigators, ^, who also found RA27/3 to be superior to the earlier HPV-77DE component. Seroconversion to rubella vaccination with any of the other triple combinations (see “Immune Responses” earlier) is usually 97–98%. ^, Bivalent measles-rubella and mumps-rubella vaccines also produce rubella antibody levels that are equivalent to those of the monovalent vaccine. ^, Comparative studies of the Merck and GlaxoSmithKline triple combinations showed excellent seroconversion to rubella after both vaccines. ^,

TABLE 54.11

Antibody Responses in Initially Seronegative Children Who Received Combined Measles (Moraten)-Mumps (Jeryl Lynn)-Rubella (RA27/3) or Monovalent RA27/3 Rubella Vaccine

Vaccine	Antibody Responses Versus
	Measles (HI) Conversion			(Mumps Neutralizing) Conversion			Rubella (HI) Conversion
	N/Total	%	Geometric Mean	N/Total	%	Geometric Mean	N/Total	%	Geometric Mean
Combined	64/68	94	57	65/68	96	8	68/68	100	136
Monovalent	–	–	–	–	–	–	67/67	100	159

HI, hemagglutination inhibition; N, number.

From Weibel RE, Carlson AJ, Villarejos VM, et al. Clinical and laboratory studies of combined live measles, mumps, and rubella vaccines using the RA27/3 rubella virus. Proc Soc Exp Biol Med . 1980;165:323–326 .

In summary, rubella vaccines induce immune responses that are similar in quality but lesser in quantity than those after natural disease. ^, The live virus vaccine produces viremia and pharyngeal excretion, but both are of low magnitude and are noncommunicable. IgM and IgG antibody responses follow vaccination. Natural infection elicits nasal secretory antibody that may be useful in the prevention of reinfection, and RA27/3 vaccine also has the same property.

Effect of Maternal Antibody

On average, maternal rubella antibody levels are detected only for approximately 2 months after birth in the infant, and thus rubella vaccine does not suffer the same problem as measles vaccination in early life, where maternal antibodies interfere with infant vaccination. Studies have documented that the seroconversion rate in infants at 9 months of age is similar to that of children older than 12 months of age. ^{, ,}

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here