Author:

Keywords:

Mumps, long-term immunity

Abstract:

Summary Measles, mumps and rubella are viral infections which cause high morbidi ty in different age groups and populations. Even though measles, mumps a nd rubella vaccines have been available for decades, and their systemati c use has significantly reduced the incidence and associated complicatio ns due to these infections, measles and mumps are still a health concern and elimination appears hard to achieve. The main objective of this PhD thesis is to obtain a better insight in t he mechanisms underlying long-term protection after measles-mumps-rubell a vaccination, with a special focus on the mumps component of the vaccin e. Research questions related to different aspects of long-term protecti on after immunisation have been formulated and answered. First, the underlying reasons for a large outbreak of mumps in a high ly vaccinated school population are explored. To what extent are outbrea ks of mumps attributable to primary and/or secondary vaccine failure? In a retrospective study we investigated a mumps outbreak that occurred in 7 primary schools in the city of Bruges (school year 1995-1996) in wh ich the mean coverage rate for one MMR dose was 91.8% (range 81.8 - 94.4 %). The attack rate for mumps was 5.7% (range: 0.3 - 13.6%). The odds ra tio for developing mumps in the vaccinated versus the unvaccinated child ren was 0.33 (95% confidence interval: 0.19 – 0.55), and the overall vac cine efficacy in this particular outbreak was 64.0% (CI: 43.2 – 77.2%). The proportion of vaccinated children with mumps increased progressively by increasing time lapse since the last dose of mumps vaccine was admin istered. The estimated odds to develop mumps within one year after vacci nation were very low, i.e. 0.01 (0.01 – 0.03), but increased thereafter with 27% for each extra year since the last vaccination (OR: 1.27 (1.16 – 1.38)). Different hypotheses to explain the occurrence of this outbreak were con sidered: (i) suboptimal vaccination coverage, (ii) primary and/or second ary vaccine failure, and/or (iii) an inadequate vaccination schedule. Firstly, the vaccination coverage for one dose of MMR-vaccine in the aff ected school population was high (mean rate 91.8%). However, a few years after the outbreak in Bruges (1999), the estimated MMR-vaccination cove rage in a representative sample in the surrounding area of the outbreak (province of Western Flanders) showed to be only 79.8%. This may not be high enough to stop circulation of the mumps virus. Therefore, the actua l force of infection may have been higher than expected from the vaccina tion coverage of the school population in Bruges. Secondly, the primary vaccine failure is estimated to be 5-10% based on pre-licensure clinical trials. Following different outbreak studies a mu ch lower effectiveness was noted. Our study clearly showed that the risk to develop mumps disease increased with increasing time interval betwee n vaccination and possible exposure to mumps virus, which reflects secon dary failure of the vaccine related to waning immunity. Lastly, taking into consideration the risk of secondary vaccine failure, the Belgian vaccination schedule, with an interval of about 10 year bet ween doses (MMR-vaccinations at the ages of 12 months and 10-12 years), may not be adequate to compensate for the risk of outbreaks. A shift of the second vaccine dose to younger age (e.g. 4-6 years, as has been appl ied in several other European countries) would avoid the build-up of sus ceptible individuals over time and warrant higher vaccination coverage t han can be achieved in young adolescents. This would result in a better control of the circulation of the mumps virus and preclude further outbr eaks. We have shown that waning immunity is a risk factor in the development o f outbreaks of mumps in countries with a lower than optimal vaccination coverage and with a two-dose schedule interval of more than 4-5 years. To better understand the mechanism of secondary vaccine failure as a dri ving force to recurrent outbreaks, the following research questions focu s on important aspects of long-term immunity after MMR-vaccination. The persistence of specific circulating antibodies (humoral immunity), of ly mphoproliferative response (cellular immunity), and of humoral B-memory are studied here. Secondly, the long-term humoral immunity after MMR-vaccination was ex amined. Are circulating anti-mumps antibodies still present years after immunisation and what is their role in long-term protection against mump s? Do these observations also apply to measles and rubella? Mass vaccination with MMR-vaccine remarkably reduced circulation of meas les, mumps and rubella viruses. In the absence of repeated natural boost ing by wild-type virus both vaccine-induced and naturally acquired antib odies may wane over time. As a consequence, in accordance with the inter val since the last vaccination, the number of susceptible subjects in th ose populations increases progressively. Accumulation of susceptible sub jects may lead to outbreaks, most likely in young adults, if the virus i s re-introduced. Long-term immunity after MMR-vaccination is therefore a prerequisite for sustained protection against measles, mumps and rubell a. To examine the potential booster effect of the second MMR-dose on lon g-term protection, we compared the persistence of circulating antibodies induced by one or two doses of MMR-vaccine in Belgian students who had received the first vaccine dose at least 16 years earlier. In this cross-sectional study a blood sample from university students im munised with either one or two doses of measles-, mumps and/or rubella-c ontaining vaccines during childhood was analysed with commercial Enzyme Immuno Assays (EIA) to determine the humoral immunity against measles (E nzygnost, Siemens), mumps (Enzygnost, Siemens), and rubella (MEIA Axsym Abbott). The mean age at which the first dose was administered did not d iffer significantly between both groups for each component. The mean int ervals between the blood sampling and the last vaccine dose were ~19 yea rs and ~8 years for the one-dose and two-dose groups, respectively, and did not differ significantly between the vaccine components in each grou p. In the entire study population (n = 160) the percentage of students with positive antibody titres was lowest for mumps (64.4%), followed by meas les (72.5%) and rubella (98.1%). About one in four students (n = 39, 24. 4%) had no positive titre against at least 1 of the vaccine components, and 49.4% (n = 79) had antibodies against all three viruses. The percent age of protected subjects was significantly higher in those who received two vaccine doses for measles (77.1% versus 58.7%, p = 0.05) and mumps (67.5% versus 55.6%, p = 0.04). Seroprotection for rubella showed the sa me trend, but did not reach significance (99.2% versus 71.4%, p = 0.08). Comparable trends were seen for GMT’s for mumps (878.9 titers versus 50 8.3 titers, p = 0.013) and measles (796.7 IU/L versus 444.1 IU/L, p = 0. 0154), but not for rubella (48.7 IU/L versus 39.5 IU/L, p = 0.55). We found that 2 doses of vaccine provided a significantly better humoral protection against measles and mumps in terms of seropositive rates and GMT. The difference in GMT for rubella did not reach statistical signif icance (p>0.05). However the statistical power was very low, since only 7 subjects were assigned to the one dose group. Although this second dose of MMR is traditionally seen as a catch-up dos e for measles and mumps, our findings suggest that the second dose has a booster effect and is more than just a catch-up to compensate for prima ry vaccine failure or missed vaccine administration. This booster effect may be important to confer long-term protection in the absence of natur al boosting by circulating wild type viruses, and to compensate for the waning of vaccine-induced antibodies over time. Furthermore, this study confirmed the published data from Finland (high coverage rate of a two-dose MMR-schedule since 1982) and showed that pro tection against rubella remains very high (99.2%) up to 8 years after a second dose of a rubella-containing vaccine. This finding is reassuring in view of the increasing age of first pregnancy in Western female popul ations. Antibodies are considered as the first line of defence against pathogens and important in acquiring immediate protection. However, vaccine-induc ed antibodies wane over time in the absence of natural boosting. For thi s reason it is important to evaluate the presence and role of other immu nologic processes, such as cellular immunity (research question three) a nd humoral immune memory (research question four) and to examine possibl e correlations between the humoral and cellular arms of the immune syste m. Thirdly, we investigated the presence of cellular immune responses ag ainst mumps in a group of mumps-vaccinees to find an answer to the follo wing questions: 1) Does T-cell-mediated immunity to mumps persist after one or two doses of MMR-vaccine? and 2) What is the role of T-cell immun ity in the long-term protection against mumps after immunisation? Like measles and rubella vaccines, mumps vaccines also induce cellular i mmunity in addition to humoral immune responses. Until today this aspect of mumps vaccination has been poorly examined. One study showed that mu mps vaccine is capable of inducing a lymphoproliferative response in inf ants vaccinated at 6, 9 or 12 months of age. Recent studies on long-term persistence of cell-mediated immunity (CMI) showed that mumps-specific lymphoproliferative responses are still present in 70 to 98% of adults i mmunised with 2 doses of mumps-containing vaccine during childhood. In t hese studies antibody titres were not measured or not correlated with CM I responses. In the light of the recent outbreak in the US where most ca ses had received two vaccine doses, the question rises whether memory T- cells are capable to prevent mumps outbreaks in the absence of antibodie s. To study the potential impact of persistent mumps-specific T cell respon ses induced by one or two vaccine doses on long-term protection against mumps, we compared the mumps-specific lymphoproliferative responses in g roups of subjects with low (n = 24) and high anti-mumps antibody levels (n = 24) who received the last vaccine dose up to 16 years before. Overall, 32 of the 48 subjects (66.7%) had a positive stimulation index (SI ≥ 3), and this proportion was higher in the high antibody group (79.2%) than in the low antibody group (54.2%), but the differenc e did not reach statistical significance (p = 0.12). The mean stimulatio n index (SI) in subjects with low and high antibody titres was 4.47 and 8.31, respectively (p = 0.032). In total, 61.5% of subjects who received only 1 MMR-vaccine dose with a mean interval to serum sample of 16 year s still had a positive response to the lymphoproliferation assay (LPA), compared to 72.7% of the two-dose vaccinees (p = 0.96), who received the ir last dose with a mean interval of 8 years since blood sampling. The o utcome of the LPA was neither influenced by the interval since last immu nisation (p = 0.47). Our study showed that mumps-specific lymphoproliferative responses can b e elicited in vitro up to 16 years after the administration of one or tw o mumps vaccine doses, even in subjects with no or low anti-mumps antibo dy levels at that time. These observations, along with the recent US outbreak, raise the questio n to what extent this persistent cellular immunity contributes to the lo ng-term protection against mumps infection and disease. Sustained antibo dy-mediated protection relies largely on long-lived IgG producing plasma cells and memory B-cells. When antibodies wane or even disappear due to a lack of recurrent exposure to wild-type virus or revaccination, prote ction against infection will depend on the speed with which the immune s ystem is able to mount a protective antibody response and can prevent th e virus from invading the human host. Since we and others demonstrate th at cellular immune responses to the mumps virus can be revealed in the m ajority of vaccinated individuals, irrespective of their anti-mumps anti body status, outbreaks such as the one experienced in the US in 2006 sug gest that circulating antibodies are the most important defence mechanis m against mumps. Moreover, in our study we show that 61.5% of the one-do se vaccinees still had positive lymphoproliferative responses despite th e fact that they were immunised up to 16 years before. Several outbreak studies have shown that the risk to develop mumps after only 1 dose of m umps vaccine increases with time. If cellular immunity in the absence of circulating wild type mumps virus would protect against disease, the li kelihood of outbreaks after one dose would be low. Our findings suggest that immune protection against mumps is more depend ent on high antibody titres than on cellular immune responses and emphas ize the importance of circulating antibodies in the protection against m umps disease. Fourth, we examined whether one to two decades after MMR-vaccination memory of humoral immunity against measles and mumps is yet present in t he blood, when circulating antibodies are not measurable any longer. To this end, we have transferred peripheral mononuclear blood cells (PBM C’s) into conditioned NOD-SCID mice in order to reveal the presence of m umps- and measles-specific memory B-cells in the circulation of vaccine recipients. Via the analysis of plasma of chimeric mice, this model allo ws for the detection of antibody production by human memory B-cells that are present in the circulation of subjects without detectable measles- and mumps-antibodies in the serum. PBMC’s of 6 donors (3 subjects with low and 3 with high antibody titres against measles and mumps) were injected in the spleen of conditioned NO D-SCID-mice (3 per subject). In vivo production of human antibodies agai nst measles and mumps was evaluated in mouse plasma on days 7, 10 and 13 with a commercial ELISA (Enzygnost measles and mumps, Siemens). For mumps, 3 donors had undetectable antibody titres and 3 had very high antibody titres, whereas for measles 2 donors had undetectable antibodi es, 1 donor had a very low antibody titre and 3 donors had high antibody titres. None of the mice injected with PBMC from subjects with undetect able antibody titres for measles and mumps developed detectable human an tibody titres against either virus. All mice injected with PBMC from sub jects with high(er) antibody titres against mumps and/or measles, also s howed detectable antibody titres for measles and/or mumps in the animal plasma. We showed here for the first time that vaccine-induced memory B-cells ca pable of secreting measles virus and mumps virus–specific IgG can be rev ealed by intrasplenic transfer of human PBMC into the SCID-NOD mice. In fact, a strong correlation was noted between the presence or absence of virus-specific IgG in the sera of cell donors and the capacity or inabil ity of the respective PBMC to secrete detectable levels of virus-specifi c IgG in the chimeric mouse plasma. These data suggest that in vaccinees without detectable serum antibodies against mumps and/or measles viruses, no or extremely low numbers of me mory B-cells against these viruses are circulating in the venous blood. In vaccine recipients that have retained high-titred anti-measles or ant i-mumps IgG in circulation, memory B-cells able to produce these antibod ies seem to be circulating at frequencies that allow for the production of detectable virus-specific IgG responses in vivo in the transplanted m ice. Individuals without circulating antibodies and memory B-cells may theref ore be at increased risk of infection and clinical disease upon encounte r with the wild-type virus. The findings of this study along with the ob servation of waning mumps immunity in twice-vaccinated individuals could explain why an outbreak of mumps occurred in the US in 2006. The data on measles in this study are encouraging since one subject with a measles antibody titre just above the detection limit, developed a de tectable IgG response in all three recipient mice. Moreover, the issue o f waning immunity is less prominent for measles than for mumps and the p roportion of subjects with little or no antibodies left up to 10 to 15 y ears after the second vaccine dose is limited. Nevertheless, individuals without circulating antibodies may also be at risk of developing a symp tomatic measles infection upon encounter with the wild-type virus, since no circulating memory B-cells could be demonstrated by intrasplenic tra nsfer of PBMC in NOD-SCID mice. Finally, we looked at the MMR-vaccination coverage in three different age groups (toddlers, 7-year-olds and adolescents) in Flande rs as a proxy of the level of herd immunity in the population. The WHO-Europe’s goal is to eliminate measles and rubella by 2010 which requires a coverage rate of 95% for both MMR-vaccine doses. To achieve e limination of mumps a similar vaccination coverage should be reached. In Belgium, two doses of MMR-vaccine are currently recommended at 12 month s and 10-12 years of age. In Flanders (Northern region of Belgium), reco mmended vaccines are offered free of charge to infants mainly through we ll-baby clinics, and to children and adolescents in a school-based immun isation program. General practitioners and pediatricians play a compleme ntary role in this free of charge vaccination system. In accordance with the Flemish school health legislation, the School Health Services have to monitor the vaccination status of all children under supervision, inf orm parents and children about recommended vaccinations, and offer recom mended vaccinations, including catch-up vaccinations. However, the schoo l health legislation does not imply mandatory immunisation for school en try. Parents are free to accept the offer of free immunisation by well-b aby clinics and school health services, or to have their child vaccinate d by a GP or pediatrician. In a survey conducted according to the WHO Ex panded Programme on Immunisation (EPI) method in 1999, the coverage of t he first MMR-dose was estimated to be 83.4% in 18-24 month old toddlers in Flanders. In 2005, MMR-vaccination coverage was measured in a new study in which t he same EPI two-stage random cluster sample technique was used to survey in 1500 toddlers (18-24 months of age), 900 primary school children (bo rn in 1997) and 1500 adolescents (born in 1991) living in Flanders (Belg ium). Documented MMR-vaccination was recorded and a questionnaire on soc iodemographic factors was completed at home by trained interviewers. Mis sing data were retrieved from well-baby clinics and school health servic e documents. The overall response rate was 89.5%, leaving 3490 subjects fit for analysis. The mean age was 1.67 ± 0.10 years (range: 1.48 - 2.15) in toddlers, 7.8 ± 0.26 years (range: 7.4 – 8.4) in primary school children and 13.9 ±0.27 years (range: 13.4 - 14.4) in adoles cents. MMR coverage (first dose) was 94.0% in the toddler group, 88.0% i n the 7-year-olds, and 80.6% in adolescents. The 10-12 year-dose was doc umented in 83.6% of the adolescents, but only 74.6% had proof of both MM R-vaccines. In 2008, the study was repeated in toddlers and adolescents. The MMR-vac cination coverage for toddlers was 96.6 %. In adolescents the first MMR-vaccination was documented in 88.1%, the second dose in 90.6%, neve rtheless, only 83.5% had documentation of both MMR-doses. The vaccination coverage measured in both studies (2005 and 2008) reflec ts rates based on reliable documented vaccination either at home or in m edical files. However, in adolescents and primary school children the co llection of data was seriously hampered by the lack of vaccination docum ents at home. Recently, a centralised web-based vaccination database has been introduced in Flanders, where all vaccinators are required to regi ster the vaccines they administer. Registration of MMR-immunisation into this database will be a valuable tool to enhance documentation of immun isation in primary school children and adolescents in the future. However, to reach the WHO goal - elimination of measles and rubella by 2 010 - continued efforts are required to administer the first and second MMR dose to at least 95% of all children between 12 months and 12 years of age in Flanders. Our study showed that Flanders is not ready to elimi nate measles, mumps and rubella yet, since in all age groups the goal fo r elimination is not reached. Nevertheless, there is a tendency towards elimination since in the youngest age group a coverage rate of 94% is ac hieved in 2005, which further increased to 96.6% in 2008. In order to re ach and sustain elimination for measles and rubella Flanders must engage in additional actions to enhance MMR immunisation coverage for both dos es. Stable but suboptimal coverage causes continuation of disease transm ission within the population, and a shift in the age of measles acquisit ion. Cohorts with increased susceptibility for measles and mumps will gr adually replace the actual adults who have high levels of naturally acqu ired immunity. Moreover, complications of measles, mumps and rubella are more frequent in adults, with acute encephalitis being the major cause of death for measles, and rubella contracted by pregnant women being the cause of the congenital rubella syndrome. A lower MMR-coverage was noted in children of single or divorced parents (toddlers, adolescents), families with more than four children (toddler s, adolescents), children of non-Belgian origin (children, adolescents), lower education or unemployment of parents (toddlers, children, adolesc ents), low family income (children, adolescents), vaccination by the GP (toddlers, children), and education-related factors (children, adolescen ts). These lower immunisation rates can be the result of either a true n on-immunisation situation or a loss of documents, but it is more likely related to a combination of factors: (i) the lack of knowledge about imm unisations, or (ii) language skills of parents may not be sufficient to fully understand the information on vaccinations, or (iii) the time need ed to review this information or to store it in a safe way is simply mis sing. Through the school-based immunisation program, family- and school- related indicators for non-immunisation can be easily picked up to ident ify pupils who did not receive two doses of MMR-vaccine. Additional supp ort for school health services to provide a strict follow-up for familie s and children at risk for non- or partial MMR-vaccination, may improve immunisation of these children. The recommended WHO coverage rate of 95% for MMR was achieved in toddler s in 2008. Documentation of vaccination is a major concern in older age groups, and may explain lower coverage estimates. Children growing up in a less privileged environment deserve special attention. Overall, in this doctoral thesis epidemiological as well as immunolog ical aspects of mumps vaccination, and where possible measles and rubell a immunisation, were studied and linked to each other. The most importan t conclusion from this thesis is that the protection following two doses of mumps-containing vaccines is not life-long. More specifically, circu lating antibodies wane over time which is associated with the absence of mumps-specific memory B-cells. In addition, long-term protection may no t be guaranteed by the cellular immune system after MMR-vaccination. Mor eover, in Flanders the vaccination coverage for two dos es of MMR-vaccine is not high enough to eliminate measles, mumps and rub ella. There is a risk that a population of susceptible individuals is ac cumulating over time, which could increase the risk of outbreaks of mump s in older age groups in Flanders.