Serological survey of mumps antibodies in adults in the Czech Republic and the need for changes to the vaccination strategy

ABSTRACT

Mumps outbreaks, especially in adolescents and young adults, have been reported in the Czech Republic. The aim of the presented study was to determine the seroprevalence of specific IgG antibodies against mumps in the adult population of the Czech Republic. The study was designed as a multicenter serological survey of adults aged 18 years and over. Specific IgG antibodies against mumps were detected in blood samples using an enzyme-linked immunosorbent assay (ELISA). A total of 1,911 serum samples were examined. The overall seropositivity reached 55.3%. In individual age groups, the highest seropositivity 63% (63.5–65.2%) was recorded in adults aged 40 years and over; the lowest seropositivity was found in adults aged 18–29 years (27.4%). The difference in seropositivity rate between the 18–29 years age group and the 40 years and over age groups was statistically significant (p < 0.001). Only the 18–29 years age group included both vaccinated and unvaccinated (born in the pre-vaccine era) individuals. In vaccinated individuals, seropositivity was reported in only 19.1% of persons; in unvaccinated individuals, seropositivity reached 48.2%. Our results demonstrate the long-term persistence of antibodies following natural infection and the decrease in seropositivity that occurs after vaccination over time. This immunity waning may account for the higher susceptibility of adolescents and young adults to mumps. Therefore, the current vaccination program in the Czech Republic could be considered as less effective. It will be modified with the shifting of the second dose of vaccine from two years of age to the preschool age.

Introduction

Vaccination against mumps has been broadly introduced into National Immunization Programs (NIP) in many countries, beginning in the 1970s. By the end of 2013, the vaccine became the part of the NIP in 120 countries around the world.^1-4 In the Czech Republic (CR), immunization against mumps was incorporated into the NIP in 1987 as a mandatory vaccination. Initially, starting in 1987, a live-attenuated bivalent vaccine against measles and mumps, or a monovalent vaccine against mumps, was used. In 1995, these vaccines were replaced by a trivalent measles, mumps and rubella vaccine. The Jeryl-Lynn mumps vaccine strain, which is used in the CR, is genotype A.⁵ The vaccination schedule consists of two doses. Currently, the first dose is administered to children from the 15^th month of life and the second dose is administered 6–10 months after the first dose. The vaccination coverage has been 98%.⁶

Before the introduction of mumps vaccination in the CR, mumps was a common disease that mainly affected school-age children. Similar to other countries, the introduction of vaccination resulted in a significant reduction in the mumps incidence. However, new outbreaks have been reported, and the age group with the highest incidence of mumps has shifted to adolescents and young adults (aged 15–19 and 10–14 years). Many cases of mumps were reported in vaccinated persons.^7-11 Similar to the CR, other countries with a high vaccination coverage and a long-term history of vaccination have reported outbreaks of mumps in both vaccinated and unvaccinated individuals, with the highest age-specific incidence occurring in adolescents and young adults.^12-16

Mumps virus is easily transmissible – the basic reproductive number (R₀) for mumps in a fully susceptible population is 4–7. However, the effective reproductive number should be determined in partially immunized populations. This number is influenced by vaccination coverage, vaccination effectiveness, and previous infection. These factors push the effective reproductive number below R₀. If vaccine effectiveness is assumed to be 85% after two doses of vaccine, a vaccine coverage of 88% and 98% will be needed to interrupt community transmission of mumps at reproductive values of four and six, respectively.¹⁷ Primary or secondary vaccine failure should be considered when assessing the cause of a mumps outbreaks. Primary vaccine failure is defined as the lack of an immunologic response to vaccination, whereas secondary vaccine failure is characterized by the loss of immune protection (the waning of immunity) over time after initial effectiveness.^18-19 At the same time, indirect (or herd) protection plays an important role in the protection of a population. A common implication of the term is that the risk of infection among susceptible individuals is reduced by the presence of immune individuals.²⁰ The herd immunity threshold for mumps is between 75% and 86%.²¹

The immune response to mumps virus infection likely results from of a complex interplay between both humoral and cellular immunity, and no definitive correlates of protection have yet been identified. However, humoral immunity is important in protection against mumps, and antibody measurements are often used as a surrogate measure of immunity to different viral infections.^22-23 Therefore, we chose to use a serological survey in our study design.

The active mumps infection provides long-term, probably lifelong immunity. In contrast, post-vaccination immunity is probably not lifelong, allowing outbreaks to occur even in immunized populations. Because of the recent mumps outbreaks, especially in adolescent and young adult populations, questions have arisen about the persistence of protection in adults who were vaccinated during childhood, and about the level of herd protection. The primary goal of this study was to determine the seroprevalence of specific IgG antibodies against mumps in the adult population in the CR. The secondary goal was to assess mumps susceptibility in adults and to evaluate the effectiveness of the mumps vaccination program in the CR.

Results

Demographic characteristics

In total, 1,911 serum samples were obtained from individuals aged 18–87 years, including 917 men (48%) and 994 women (52%). The demographic characteristics of the study participants are described in Table 1, including the number of subjects in each individual age group. Except for the 70 years and over age group, the distribution of study participants was balanced across each age group. A questionnaire that focused on medical history was filled out by 1,633 individuals (85.5%).

Table 1.

Demographic characteristics of study participants.

	Overall		Men		Women
Age group (years)	n	%	n	%	n	%
18 – 29	387	20.2	173	18.9	214	21.6
30 – 39	374	19.6	181	19.7	193	19.4
40 – 49	356	18.6	144	15.7	212	21.3
50 – 59	307	16.1	153	16.7	154	15.5
60 – 69	350	18.3	178	19.4	172	17.3
≥70	137	7.2	88	9.6	49	4.9
Total	1911	100	917	100	994	100
Average age	45.5		46.7		44.4

Serology

The overall seropositivity of IgG antibodies against mumps reached 55.3% (Table 2). Overall seronegativity was 44.7%, and 18.0% of seronegative samples repeatedly contained borderline levels of mumps-specific IgG antibodies. Some differences in seropositivity were identified between men and women, both overall or within individual age groups. Seropositivity was always lower in men than in women, except in the 70 years and over age group. Overall, the difference in seropositivity between men (52.3%) and women (58.0%) was statistically significant (p = 0.014) regardless of age. Within individual age groups, statistically significant differences in seropositivity were found between men and women aged 30–39 years (p = 0.005) and 50–59 years (p = 0.011). There was no statistically significant difference in seropositivity between men and women in other age groups.

Table 2.

Seroprevalence of mumps-specific IgG antibodies overall, by age group and by sex.

IgG		Positive		Negative		Total		p
Age group (years)	Sex	n	%	n	%	n	%	chi-square test
Overall (18–87)	Men	480	52.3	437	47.7	917	100	0.014
	Women	576	58.0	418	42.0	994	100
	Total	1056	55.3	855	44.7	1911	100
18–29	Men	47	27.2	126	72.8	173	100	0.930
	Women	59	27.6	155	72.4	214	100
	Total	106	27.4	281	72.6	387	100
30–39	Men	89	49.2	92	50.8	181	100	0.005
	Women	123	63.7	70	36.3	193	100
	Total	212	56.7	162	43.3	374	100
40–49	Men	88	61.1	56	38.9	144	100	0.444
	Women	138	65.1	74	34.9	212	100
	Total	226	63.5	130	36.5	356	100
50–59	Men	89	58.2	64	41.8	153	100	0.011
	Women	111	72.1	43	27.9	154	100
	Total	200	65.2	107	34.8	307	100
60–69	Men	107	60.1	71	39.9	178	100	0.154
	Women	116	67.4	56	32.6	172	100
	Total	223	63.7	127	36.3	350	100
≥70	Men	60	68.2	28	31.8	88	100	0.290
	Women	29	59.2	20	40.8	49	100
	Total	89	64.0	48	35.0	137	100

Comparison of achieved results in individual age groups showed the highest seropositivity 63% (63.5–65.2%) in individuals aged 40 years and over. The lowest seropositivity was recorded in adults aged 18–29 years (27.4%), followed by adults aged 30–39 years (56.7%). The differences in seropositivity between adults aged 18–29 years and all groups of adults aged 40 years and over were statistically significant (p<0.001) (Table 2, Fig. 1). The highest seronegativity was recorded in adults aged 18–29 years (72.6%), followed by adults aged 30–39 years (43.3%). Borderline results were frequently recorded (18% in the total study population, ranging from 15.5 to 21.9% across individual age groups). Importantly, this increased total seronegativity both overall and in individual age groups (Table 2).

Figure 1.

Prevalence of mumps-specific IgG antibodies by age group. Number of examined sera samples in individual age groups: 18–29 years of age (yoa) … 387, 30–39 yoa … 374, 40–49 yoa … 356, 50–59 yoa … 307, 60–69 yoa … 350 and ≥70 yoa … 137. Seropositivity in individual age groups: 18–29 yoa … 27.4%, 30–39 yoa … 56.7%, 40–49 yoa … 63.5%, 50–59 yoa … 65.2%, 60–69 yoa … 63.7% and ≥70 yoa … 64.0%.

The youngest age group (18–29 years) included both individuals vaccinated during their childhood (277 individuals; 71.6%) and unvaccinated individuals (110 individuals; 28.4%). In vaccinated individuals, seropositivity was reported in only 19.1% of persons, seronegativity was reported in 60.3% of persons, and borderline results were detected in 20.6% of persons. In unvaccinated individuals, seropositivity reached 48.2%, seronegativity was 37.3%, and 14.6% of results were borderline. In adults aged 18–29 years, the difference in the seropositivity rate of vaccinated individuals (19.1%) and unvaccinated individuals (people born before the introduction of vaccination; 48.2%) was statistically significant (p<0.001) (Fig. 2). This statistically significant difference in seropositivity rate persisted between vaccinated (19.1%) and all unvaccinated individuals (61.4%) regardless of age (p<0.001).

Figure 2.

Prevalence of mumps-specific IgG antibodies in adults aged 18–29 years. Total number of examined sera samples was 387 (277 samples from vaccinated individuals / 110 samples from non-vaccinated individuals). Seropositivity in vaccinated individuals was 19.1% and in non-vaccinated individuals 48.2%.

This study also evaluated measured concentrations of mumps-specific IgG antibodies in vaccinated and unvaccinated seropositive individuals. The median and geometric mean of antibody concentrations were calculated. Based on not normal distribution of data in individual groups, medians are presented. Overall, the median of antibody concentrations was 39.6 U/ml. In individual age groups, comparable medians were found (40.3 U/ml at 18–29 years, 38.7 U/ml at 30–39 years, 37.9 U/ml at 40–49 years, 39.2 U/ml at 50–59 years, 40.2 U/ml at 60–69 years, and 41.7 U/ml at 70 years and over). Similarly, the median of antibody concentration in vaccinated adults aged 18–29 years (39.4 U/ml) was not significantly different than that in unvaccinated individuals of the same age (41.8 U/ml), or in unvaccinated individuals across all age groups (39.6 U/ml).

Comparison of overall seropositivity between genders did not reveal any statistically significant difference in median of antibody concentrations between seropositive men (38.7 U/ml) and women (40.3 U/ml). However, a significantly higher median antibody concentrations was detected in women compared with men in the 50–59 years (36.0 U/ml in men, 42.8 U/ml in women) and 70 years and over age groups (39.2 U/ml in men, 50.8 U/ml in women) (p = 0.014; p = 0.019). In the other age groups, medians of antibody concentrations were similar in both seropositive men and women. Statistical significance was maintained whether medians or geometric means were used in the statistical analyses.

The impact of a natural mumps infection on antibody seroprevalence was also evaluated. Overall 762 individuals (46.4%) reported natural infection, 595 individuals (36.3%) did not suffer by mumps and 190 individuals (17.3%) didn´t know about mumps infection in their medical history. Most of participants (75.3%) suffered by mumps in childhood up to 5 years of age. No correlation was found between seropositivity and the age of participants. Individuals with a history of mumps infection had a 66.3% seropositivity rate, whereas in individuals without history of the mumps infection, the seropositivity rate was only 55.0%. However, this difference was not statistically significant. These findings may reflect a relatively high incidence of an asymptomatic form of the mumps, in which many patients present with a negative medical history even though their seropositivity suggests that they were exposed to the infection.

Discussion

Currently, despite the significant benefit of vaccination, mumps outbreaks have been reported in the CR, and the highest incidence of mumps has shifted to older age groups. One major underlying cause of these outbreaks could be a low level of herd immunity. This hypothesis is supported by the results of our study. The overall seropositivity of mumps antibodies reached only 55.3%, despite a high overall vaccination coverage in the CR. The highest seropositivity (over 63%) was reported in individuals aged 40 years and over, while the lowest seropositivity was reported in individuals aged 18–29 years (27.4%) and 30–39 years (56.7%). This study also frequently detected borderline IgG levels (18% overall). Similar large proportion of borderline results has also been noted in other studies of mumps serology.^24-26 In this study, results that were repeatedly borderline were considered negative, thus increasing the proportion of individuals classified as seronegative.

The results of our study are similar to those described in a national serological survey of the CR conducted in 2013. The highest seropositivity rate was observed in children aged 2–4 years (84.0–76.5%). This rate declined to 59.4% in children 6 years of age and to 51.0% in children 9 years of age. In adolescents aged 10–20 years, the seropositivity rate ranged from 28.6 to 42.3%, whereas in young adult age 21–27 years, it ranged from 33.7 to 46.2%. These results suggest that the increased susceptibility of adolescents to mumps infection may probably result from waning immunity rather than a primary failure of vaccination in childhood. In the 2013 survey, a similar trend was observed in adults; the highest seropositivity, similar to our results, was observed in adults aged 30–64 years, where the seropositivity rate was 74.5–85.6%. Insufficient herd immunity was also detected in the CR in a previous serological survey from 2001. This survey reported mumps seropositivity in 70–86% of vaccinated children under the age of 15 years and in 75–90% of individuals between the ages of 15 and 40 years. These findings were inconsistent with the level of vaccination coverage, which ranged from 97–100%.^27-28 The decline in the mumps seropositivity over time correlates, to a certain extent, with the age-specific incidence of the disease. Peak mumps incidence shifted from preschool age children in the pre-vaccination era to the 15–19 years age group during outbreaks that have occurred in the post-vaccination era.^7-8

Only the youngest age group in this study (adults aged 18–29 years) included both individuals vaccinated in childhood and unvaccinated whose seropositivity developed from a natural mumps infection. Other subjects enrolled in this study were not vaccinated. A statistically significant difference in seropositivity was found between vaccinated and unvaccinated persons. The lower seropositivity in unvaccinated individuals aged 18–29 years (48.2%) compared with unvaccinated individuals in the older age groups (56.7–65.2%) probably reflects a less intense exposure to the mumps virus, which has limited circulation within the population in the post-vaccination era. In contrast, the very low seropositivity in vaccinated individuals might reflect the waning of the immune response. Our results demonstrate that there is a long-term persistence of antibodies following natural infection, and that seropositivity decreases over time after vaccination in country with a long history of mandatory mumps vaccination and long-term high vaccination coverage. The significant difference in the seropositivity rates between vaccinated and unvaccinated individuals suggests that post-infection immunity lasts longer than post-vaccination immunity. The waning of immunity after vaccination may lead to a higher mumps susceptibility in adolescents and young adults. This supports the view that post-vaccination immunity is more short-lived than previously assumed. According to some authors, the odds of mumps increases for twice-vaccinated individuals by 10% for every year that has passed since the second dose.²⁹ Based on the serological survey in the CR in 2013, it is obvious that there is low seropositivity, even in children aged 4–9 years. However, this low seropositivity rate is not associated with a higher mumps incidence in this age group. Therefore, it is necessary to also consider the involvement of cellular immunity in the protection against mumps.²¹ Similar low mumps seropositivity rates, as well as the decrease in antibody concentrations and the related waning of protection, have also been observed in other countries.^12,14,30-32

The outbreaks of mumps may also have other causes, including suboptimal vaccine efficacy, differences in vaccine strains and circulating strains of the mumps virus, or the absence of an effective natural booster because of the limited circulation of causative agent in the general population. However, primary vaccine failure resulting from genotype mismatch between the vaccine virus strain and wildtype virus strains is not likely to play a key role.^33,34

Despite the occurrence of mumps outbreaks in the CR in recent years, the two-dose vaccination program should be considered effective and successful. A significant decrease in mumps incidence was reported after vaccination implementation, and the significant preventive effect of the two-dose vaccination scheme against mumps complications has been demonstrated.³⁵ However, the occurrence of mumps outbreaks and the findings of serological surveys has emphasized the need for changes to the vaccination program. One of the possible changes to the vaccination program could be to add a third dose of vaccine at the adolescent age.^36-38 Unfortunately, there is no general consensus around the administration of a third vaccine dose in routine vaccination. Data on the immunogenicity of the third dose are limited. In a study assessing mumps antibody titers pre- and post-third dose of vaccine, an initial temporary increase in response to the third dose was observed, but antibody titers declined to near-baseline levels 1 year later.³⁸ These findings support the use of a vaccine dose to help control outbreaks.²⁹ On the other hand, a third vaccine dose may have limited value for routine use in a vaccinated population. The implementation of the third dose of vaccine into the NIP in the CR is not expected.

Different intervals in mumps vaccination schedules are used around the world and even within European Union. It is not clear which two-dose vaccination schedule is most effective. Outbreaks have been reported in countries with different intervals between vaccine doses. It is necessary analyze each epidemic individually because outbreaks could be observed in countries with high vaccination coverage but with certain population groups with very low coverage. Those clusters of individuals can undermine the benefits of herd immunity in a highly vaccinated society.^36,39 However, according to studies that compare epidemiological data with vaccination coverage in different European countries, a longer interval between the first and second dose of mumps vaccine is associated with a reduced risk of mumps epidemic.⁴⁰

The authors are aware of some weaknesses and limitations of this study. First, there is no definitive correlate of protection for mumps; both humoral and cellular immunity play a role. The loss of antibody concentrations over time does not necessarily imply a loss of clinical protection. In addition, the retesting procedure described in method could lead to ambiguous results at the edge of borderline and positive results. Furthermore, it should be noted that the vaccination status of each person enrolled in this study was not collected from personal vaccination certificates. Instead, it was based on the age of the participants and the term of implementation of mumps vaccination into the NIP. However, the CR is a country with long-term high mumps vaccination coverage.

Currently, there is short interval between mumps vaccine doses in the CR. Based on the results of serological surveys, which indicate that there is a decrease in the immune response several years after vaccination and that there is a low mumps seropositivity rate in adolescents and young adults despite high vaccination coverage, the interval between vaccine doses will change starting in 2018. Administration of the second dose of vaccine will shift to the age 5–6 years, shortly before children begin to attend school. This may extend the interval of protection and thus extend the duration of protection of vaccinated individuals against mumps. Finally, ongoing surveillance is needed to assess of future impact of planned changes.

Methods

Study design

This study was a multicenter observational study aimed at assessing the seroprevalence of specific IgG antibodies against mumps in adults aged 18 years and over. To create a representative sample of the population, study participants were recruited in three study centers in geographically different regions of the CR during the period October 2011 – April 2012. The participants were required to provide a blood sample. The obtained sera were stored at a temperature of minus 20°C until serological measurements that were carried out at the same time under identical standardized conditions. Study participants were also required to fill out a questionnaire about their history of mumps vaccination, previous mumps infection and exposure to mumps infection within the last 5 years. Information from these questionnaires was used for interpret study results.

Study participants

Study participants from the general population were recruited on the “first-come, first-served” basis based on an advertising campaign in the media in the area of study centers. The balanced distribution of the study participants in each individual age group – 18–29 years (years of birth 1982–1993), 30–39 years (1972–1981), 40–49 years (1962–1971), 50–59 years (1952–1961), 60–69 years (1942–1951), and 70 years and over (1924–1941) – reflected the age distribution of the population in the CR. The only recruitment criterion was age. Because of number of subjects who did not filled questionnaires, mistakes or missing answers and uncertainty about vaccination history, the vaccination status of each person enrolled in study was not collected from questionnaires. The vaccination status of the study participants was determined by their year of birth and the NIP in place in the years following their birth. Vaccination against mumps was implemented into the NIP of the CR in 1987. Thus, it was assumed that any participant born in 1986 or later had been vaccinated against mumps. Therefore, only the 18–29 years age group included both vaccinated and unvaccinated individuals. The other age groups included only unvaccinated individuals who were born in pre-vaccine era.

Laboratory methods

The presence of specific IgG antibodies against mumps in collected sera samples was determined by an enzyme-linked immunosorbent assay (ELISA). We used the commercial RIDASCREEN Mumps Virus IgG kit (R-Biopharm, Germany). The assays were performed in an automated ELISA analyzer TrinLab D2/DS2 according to the manufacturer's user manual. Results were reported in units/per milliliter (U/ml). All measurements were performed in doublets, using one batch of the diagnostic set. All measurements included a commercially available negative control and two positive controls. All results were assessed as positive, negative or borderline according to limits given by the diagnostic set producer. Measurements that were over 24 U/ml were considered positive, measurements that were under 14 U/ml were negative and measurements that ranged from 14–24 U/ml were borderline. Serum samples with a borderline result were retested again in doublets. If the retested sample was positive or negative, then the third measurement of this sample was performed. The serum was considered positive or negative when two examinations were positive or negative. In the case of a second borderline result, the sample was considered negative according to the manufacturer's instructions. The concentration of IgG antibodies was calculated from the measured values of optical density, using a four-parameter equation where the constants were calculated by the manufacturer from the 7-point calibration range for each batch of the product and together with the calibration curve are listed in the set's certificate. If the measured optical density of the sample occurred above the upper limit of the calibration curve, the determination was repeated with diluted serum. The antibodies in vaccinated individuals were classified as post-vaccination, whereas those in unvaccinated individuals were classified as post-infection.

Statistical analysis

NCSS 9 software was used for statistical analysis. In addition to descriptive statistics, t-test (alternatively, nonparametric Mann-Whitney and Kolmogorov-Smirnov tests), the nonparametric Kruskal-Wallis one-way analysis of variance followed by Dunn's multiple comparison test with the Bonferroni correction, the nonparametric Spearman rank correlation coefficient, the chi-square test of independence in contingency tables and Fisher's exact test were used. Results were considered statistically significant at p < 0.05. Appropriate tests for statistical analyses were selected based on the distribution of values in each of the compared groups. The statistical tests that were used in our analyses are indicated with specific results.

The study protocol was approved by the Ethics Committee and investigation was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants before study entry.

Funding Statement

This study was supported by a Ministry of Defence of the Czech Republic – long-term organization development plan, Medical Aspects of Weapons of Mass Destruction of the Faculty of Military Health Sciences, University of Defence.

Disclosure of potential conflicts of interest

The authors report no conflict of interest.