Abstract
In this observational study, mumps-specific in vitro lymphoproliferation was measured in 24 subjects with low antibody titers and 24 subjects with high antibody titers who received their last vaccine dose up to 16 years previously. Overall, a significant lymphoproliferative response was found in 32 subjects (66.7%)—namely, in 13 (54.2%) of those with low antibody titers and 19 (79.2%) of those with high antibody titers. The mean stimulation index for subjects with low antibody titers was 4.47, whereas that for subjects with high antibody titers was 8.31 (P=.032). Mumps vaccine–induced cell-mediated immunity appears to be more persistent than the antibody response
In countries with long-standing, national, 2-dose measles, mumps, and rubella (MMR) vaccination programs, the incidence of mumps and the rate of associated complications have been significantly reduced [1]. In 2006, however, a large outbreak of mumps occurred in the United States, in which 67% of all affected individuals had received 2 doses of MMR vaccine [2]
In Belgium, it is recommended that children receive 2 MMR doses, the first of which is suggested to be given at 15 months of age (since 1985) and the second at 10–12 years of age (since 1995). From 1985 until 1993, the Urabe strain was used, and, since that time, the Jeryl Lynn strain has been administered. After the initiation of the MMR vaccine program in Belgium, the incidence of mumps decreased rapidly, and the current prevalence of mumps has been low, with <1 case reported per 100,000 inhabitants annually over the past few years (V. Van Casteren, written communication, 2001, and T. Lernout, written communication, 2006)
Prelicensure studies of the Jeryl Lynn and Urabe mumps vaccines in seronegative children showed high seroconversion rates (∼95%) after vaccination [1]. Despite the initial success observed, several studies have shown that antibodies induced with 1 dose of vaccine tend to wane over time, and this decrease coincides with mumps outbreaks among children in primary and secondary schools [3]. Two recent studies also demonstrated a waning of antibodies in subjects who received 2 doses, the second of which was given 15 years earlier [4, 5]
Like measles and rubella vaccines, mumps vaccines also induce cellular immunity in addition to the humoral immune response. Until recently, this aspect of mumps vaccination has been poorly examined. In 2001, Gans et al. [6] showed that the mumps vaccine is capable of inducing a lymphoproliferative response in infants vaccinated at 6, 9, or 12 months of age. Recent studies of the long-term persistence of cell-mediated immunity showed that mumps-specific lymphoproliferative responses are still present in 70%–98% of adults given 2 doses of vaccine with mumps component, irrespective of the presence or absence of circulating anti-mumps antibodies [7–9]. The recent outbreak in the United States (in which most affected individuals had received 2 vaccine doses) raises the question of whether memory T cells are able or sufficient to prevent mumps outbreaks in the absence of antibodies
To extend our insight regarding the persistence of mumps-specific T cell responses induced by 1 or 2 vaccine doses given up to 16 years previously, we compared the mumps-specific lymphoproliferative responses in groups of vaccine recipients with low and high levels of anti-mumps antibody
Materials and methodsFrom a group of university students in whom the humoral immune response had been assessed previously [10], 48 students were selected. All subjects had proof of receiving at least 1 dose of vaccine containing mumps virus, were free of acute or chronic ailments, and did not have immunodeficiency induced by disease or medication. Vaccination dates were double-checked using school health records
Students donated 50 mL of venous blood. Humoral immunity against mumps was determined using a commercial EIA. The test was performed using an automated system, and absorbance values were interpreted according to the manufacturer’s instructions. Mumps-specific antibodies were expressed in titers. To examine whether cellular immunity persisted in the context of waning anti-mumps antibodies, 24 subjects with undetectable levels of mumps antibody (negative or low antibody group [antibody titer, <230]) were selected to participate in this study, as were an equal number of subjects with high antibody titers (high antibody group [antibody titer, >2200 in 20 of these 24 subjects]). According to the manufacturer’s instructions, protection was assumed if subjects had a titer of >500
Peripheral blood mononuclear cells (PBMCs) were isolated from heparinized whole blood by means of ficoll-hypaque density gradient centrifugation performed in accordance with standard procedures. The cells were washed twice in Hanks’ balanced salt solution, and, after resuspension in fetal bovine serum (Biowithaker), they were supplemented with 10% DMSO (Sigma). They remained cryopreserved in liquid nitrogen until analysis was performed. The PBMCs were analyzed once all blood samples had been collected. After thawing, the PBMCs were washed 2 times with Hanks’ balanced salt solution and resuspended in complete RPMI 1640 (Gibco) containing 10% fetal bovine serum. The viability of the cultured PBMCs ranged from 86.9% to 98.4%, as determined by propidium iodide exclusion in flow cytometric analysis. Cell viability exceeded 95% in >87% of the samples. PBMCs were stimulated with active monovalent mumps vaccine (Mumpsvax; Merck) at a MOI of 25×105. The vaccine was reconstituted in 1 mL of complete RPMI 1640. Cultures containing media without antigen served as negative controls, and cultures restimulated with either tetanus toxoid (TT) or varicella zoster virus (VZV) lysate served as positive controls. Phytohemagglutinin (Sigma), 4 μg/mL, was used to demonstrate the capacity of the PBMCs in each specimen to proliferate. After 2 days (for phytohemagglutinin) and 5 days (for mumps virus, TT, and VZV), 0.5 μCi [3H]-thymidine (Amersham) was added to the cultures, and [3H]-thymidine incorporated during the final 16 h of the culture was quantified by liquid scintillation counting. The geometric mean of the number of counts per minute for unstimulated PBMCs and for PBMCs stimulated with mumps vaccine virus, TT, and VZV lysate were calculated. The stimulation index (SI) was the ratio of the geometric mean of the number of counts per minute in antigen-stimulated and unstimulated control wells. Responses were considered to be positive when the SI was ⩾3
Cellular immune responses were compared using Student’s t test, on the continuous scale, and dichotomized responses were compared using Fisher’s exact test. Differences in the SI, according to the number of doses and the interval between vaccination and collection of blood samples, were assessed using analysis of covariance. P<.05 was considered to be statistically significant. All statistical analyses were done using the R program (version 2.6.1; The R Foundation for Statistical Computing)
The present study was approved by the ethics committee of the University Hospitals of the Katholieke Universiteit Leuven
ResultsTable 1 summarizes the demographic and vaccination characteristics of the study group according to their anti-mumps antibody status (i.e., presence of a low or high titer). The mean age (±SD) at the time that blood samples were collected was 20.0 ± 1.0 years, and 87.5% of the subjects (n=42) were female. Both subgroups were comparable with respect to sex, age at blood sampling, age at receipt of the first or second vaccine dose, and the interval between the last mumps vaccination and the date of blood sampling. Twenty-two students (45.8%) had received 2 doses of mumps vaccine, and 8 of these students belonged to the low antibody group (8 [33.3%] of 24 students), whereas 14 belonged to the high antibody group (14 [58.3%] of 24)
Demographic and vaccination characteristics and lymphoproliferative responses noted in both study groups, according to anti-mumps antibody (Ab) status
Demographic and vaccination characteristics and lymphoproliferative responses noted in both study groups, according to anti-mumps antibody (Ab) status
Table 1 summarizes the number and proportion of subjects with a positive lymphoproliferative response (SI, ⩾3) according to their anti-mumps antibody titers. Overall, 32 (66.7%) of the 48 subjects had an SI ⩾3, and this proportion was higher in the high antibody group than in the low antibody group, although the difference did not reach statistical significance (P=.12). The mean SI for the group with low antibody titers was significantly lower than that for the group with high antibody titers (4.47 vs. 8.31; P=.032) (figure 1). When the lymphoproliferation assay (LPA) response was compared according to the number of doses of vaccine containing mumps virus, 16 (61.5%) of 26 subjects who received only 1 vaccine dose and for whom the mean interval to serum sample collection was 16 years still had a positive LPA response, whereas 16 (72.7%) of 22 subjects who received 2 doses of vaccine still had a SI ⩾3. However, the outcome of the LPA was not influenced either by the number of doses of mumps vaccine virus received (P=.96) or by the interval since the last vaccination (P=.47)
Comparison of the magnitude of the mumps antigen–specific lymphoproliferative response between subjects with low antibody titers (n=24) and those with high antibody titers (n=24). Differences were tested for significance by use of Student’s t test. Data are shown as box plots denoting the median value and 25th and 75th percentiles, as well as the individual results on a log scale. Whiskers denote the data range, except for outliers. The stimulation index (SI) was the ratio of the geometric mean of the number of counts per minute in antigen-stimulated and unstimulated control wells. Responses were considered to be positive when the stimulation index was ⩾3. + symbols indicate individual SI values
Comparison of the magnitude of the mumps antigen–specific lymphoproliferative response between subjects with low antibody titers (n=24) and those with high antibody titers (n=24). Differences were tested for significance by use of Student’s t test. Data are shown as box plots denoting the median value and 25th and 75th percentiles, as well as the individual results on a log scale. Whiskers denote the data range, except for outliers. The stimulation index (SI) was the ratio of the geometric mean of the number of counts per minute in antigen-stimulated and unstimulated control wells. Responses were considered to be positive when the stimulation index was ⩾3. + symbols indicate individual SI values
All subjects, irrespective of their mumps antibody titer, displayed a positive lymphoproliferative response (SI, ⩾3) toward TT and VZV. This finding indicates that the PBMCs were of good quality and could mount a strong lymphoproliferative response to in vitro antigens and that the observed differences in the in vitro responses to mumps virus antigen were indeed associated with the differences in mumps antibody titers
DiscussionMumps-specific lymphoproliferative responses can be elicited in vitro up to 16 years after the administration of 1 or 2 doses of mumps vaccine virus, even in subjects who have either no or low anti-mumps antibody levels left at that time. Subjects with high levels of antibodies display positive lymphoproliferative responses (SI, ⩾3) more frequently than do subjects with no or low antibody levels (19 [79.2%] of 24 subjects vs. 13 [54.2%] of 24 subjects, respectively) (table 1), and these responses are of a higher magnitude (geometric mean SI, 8.31 vs. 4.47, respectively) (figure 1)
These observations, along with the recent mumps outbreak in the United States, raise a question about the extent to which persistent cellular immunity contributes to the long-term protection against mumps virus infection and disease. Is cellular immunity capable of effectively protecting individuals against mumps in the absence of anti-mumps antibodies?. In humans, protection against mumps is primarily based on the generation of mumps-specific antibodies. This, of course, requires the stimulation, expansion, and maturation of antigen-specific B cells, processes that need adequate support from CD4+ T helper lymphocytes. It has been shown that the T cell immune system plays an essential role in the establishment of high antibody titers, as well as the elimination of virus in instances of natural infection [11]
Sustained antibody-mediated protection largely relies on long-lived IgG-producing plasma cells and memory B cells [11, 12]. When antibodies wane or even disappear because of a lack of recurrent exposure to wild-type virus or revaccination, protection will depend on the speed with which the immune system is able to mount a protective antibody response and prevent the virus from invading the human host. For tetanus and diphtheria, it has been shown that circulating antibodies are crucial in the prevention of disease, whereas this seems to be less important for the hepatitis B virus, given its long incubation period [12]
Because we and other investigators have demonstrated that cellular immune responses to the mumps virus can be identified in the majority of vaccinated individuals, irrespective of their anti-mumps antibody status, outbreaks such as the one experienced in the United States in 2006 suggest that circulating antibodies are the most important defense mechanism against mumps. Experience with the Rubini mumps vaccine strain further supports this view. The Rubini mumps vaccine strain was capable of inducing strong lymphoproliferative responses in individuals who were vaccinated twice, whereas the antibody responses were rather weak. The numerous outbreaks that occurred in populations vaccinated with this strain suggest that immune protection against mumps is more dependent on high antibody titers than on cellular immune responses [13, 14]
Moreover, in the present study, we show that 61.5% of the recipients of 1 vaccine dose still had positive lymphoproliferative responses, despite the fact that they were vaccinated up to 16 years previously. Several studies of outbreaks have shown that the risk of developing mumps after receipt of only 1 dose of mumps vaccine virus increases with time [1, 3]. If cellular immunity in the absence of circulating wild-type mumps virus would protect against disease, the likelihood of outbreaks developing after 1 dose would be low
Overall, these findings emphasize the importance of circulating antibodies in the protection against mumps disease. Should we rethink the vaccination program for mumps?. In the latest US outbreak, the effectiveness of the 2-dose vaccine schedule was estimated to be 76%–88%, and no differences were observed between the attack rates noted for recipients of 1 and 2 vaccine doses. This vaccine effectiveness is similar to previous estimates of the efficacy of 1 vaccine dose [15]. Moreover, in a Finnish longitudinal study, 74% of the vaccine recipients who initially were seronegative still had detectable antibodies against mumps 15 years after receiving a second dose, but only 40% showed seroprotective titers in the ELISA that was also used in the present study. This study clearly shows a decrease in the proportion of individuals with high antibody titers, despite the administration of 2 doses. The pool of susceptible adolescents and adults will increase annually, and this may jeopardize the elimination of mumps that has been achieved in such countries as Finland [4]. If the persisting cellular immunity offers no or insufficient protection in the absence of circulating antibodies, than mumps vaccination policies should be reevaluated at the level of timing of and the number of doses to be administered
Acknowledgments
We thank Inna Ovsyannikova (Mayo Vaccine Research Group) and Sanofi Pasteur for providing the monovalent vaccines. We acknowledge the efforts of the students who willingly provided us with blood samples. We thank Sabrina Verlee and Thora Van Der Stock for their excellent technical assistance
