Sex-Based Differences in Mumps Immunity After a Third Dose of Measles-Mumps-Rubella Vaccine

Abstract

Background:

A third dose of measles-mumps-rubella (MMR3) vaccine is recommended for mumps outbreak control, but the magnitude and durability of MMR3-induced immunity and the effect of sex on those outcomes, remain incompletely characterized.

Methods:

A total of 214 healthy individuals with two prior MMR doses received MMR3. Blood samples were collected at baseline (D0), day 28 (D28), and 18 months (M18) post-vaccination. Mumps-specific IgG, IgG avidity, and neutralizing antibodies (nAb) were measured by IgG ELISA and plaque reduction microneutralization assay. Cytokine and chemokine secretions from peripheral blood mononuclear cells (PBMCs) were profiled using multiplex assay.

Results:

Median IgG sample index values increased from 2.12 at D0 to 2.64 at D28 (p = 0.0019), and 2.85 at M18 (p <0.001). IgG avidity rose slightly from 79.54% at D0 to 81.88% at D28 and 86.87% at M18. nAb titers (ND₅₀) increased from 50.48 at D0 to 67.00 at D28 (p = 0.0014), and 60.18 at M18 (p = 0.044). Despite these statistically significant changes, fold increases across time points were small (IgG: 1.15 – 1.13; avidity: 1.03 – 1.08; and nAb: 1.2 – 1.18). Of the 21 cytokines and chemokines analyzed, only interferon-γ induced protein 10 (IP-10/CXCL10) showed significant increase and was negatively correlated with D28/D0 changes in IgG titers (r = −0.23; p = 0.014), but not with M18 changes, avidity, or nAb responses. Biological sex significantly influenced both IgG and IP-10 responses.

Conclusions:

MMR3 conferred limited additional humoral and cellular immune benefits, with notable sex-dependent effects. These findings support MMR3 use primarily in outbreak settings and identify IP-10 as a potential biomarker associated with MMR3 immunogenicity.

1. Introduction

A third dose of the measles-mumps-rubella (MMR) vaccine, or MMR3, has been recommended during mumps outbreaks as a strategy to boost mumps-specific immunity [1–3], with some studies reporting significant reductions in mumps attack rates following MMR administration [4–6]. Although transient increases in mumps-specific IgG titers and their kinetics following the MMR3 have been well documented in young adults [7–10], the breadth and quality of the immune response, particularly antibody avidity and neutralizing capacity, remain incompletely characterized. In addition, the cellular immune response, including cytokine and chemokine profiles, elicited by mumps vaccination, is not well defined. Evidence from our group and others indicates that biological sex is an important determinant of mumps-specific antibody responses following the second MMR dose [11, 12]; however, whether similar sex-based differences influence immune responses to a third dose remains unknown. Given the continued resurgence of mumps in populations previously vaccinated with two MMR doses [13–19], a comprehensive evaluation of the mumps-specific immune responses elicited by MMR3, including potential sex-dependent effects, is essential to guide its optimal use.

In this observational study, we comprehensively characterized both humoral and cellular immune responses to the mumps component of MMR3 in 214 healthy adults with prior two MMR doses. We quantified changes in mumps-specific IgG titers, antibody avidity, and neutralizing antibody titers at baseline (D0), 28 days (D28), and 18 months (M18) post-vaccination, along with a panel of 21 cytokines and chemokines measured at D0 and D28. We also examined the influence of demographic and clinical variables, including sex, age, body mass index (BMI), and time since the second MMR dose on the magnitude and quality of immune responses.

2. Methods

Methods used in this study were similar or identical to the ones reported in our previous studies [11, 20–22].

2.1. Ethics statement

The study was approved by the Mayo Clinic Institutional Review Board (IRB# 17–008166). Written informed consent was obtained from each participant before any study procedures were initiated, in accordance with the Declaration of Helsinki.

2.2. Study cohort

A total of 239 healthy adult participants were enrolled in this study between April 2019 and March 2021 [22]. All participants had previously received two doses of the MMR vaccine and were administered a third dose (MMR-II^®, Merck & Co., Inc.). Peripheral blood samples were collected prior to immunization (D0), at 28 days post-vaccination (D28), representing the peak antibody response, and at 18 months post-vaccination (M18), representing a longer-term timepoint to determine if immune responses returned to pre-vaccination levels or if homeostasis with increased immunologic memory was achieved. Demographic and clinical data, including age, sex, race/ethnicity, body mass index (BMI), and vaccination history, were recorded for all participants.

2.3. Preparation of serum and peripheral blood mononuclear cells (PBMCs)

A blood sample was collected at each timepoint as follows: (i) 90 mL in BD Vacutainer^® tubes containing sodium heparin for peripheral blood mononuclear cell (PBMC) isolation, and (ii) 10 mL in clot-activator BD Vacutainer^® tubes for serum preparation. PBMCs were isolated by gradient centrifugation using Ficoll Paque Plus tube (GE Healthcare Life Sciences, Uppsala, Sweden). Purified PBMCs were resuspended in freezing medium (RPMI 1640 supplemented with 20% heat-inactivated fetal calf serum [FCS] and 10% dimethyl sulfoxide [DMSO]), at a final concentration of 1×10⁷ cells/mL and cryopreserved in liquid nitrogen. For serum preparation, blood tubes were centrifuged at 2,000×g for 15 min at room temperature, and the serum was collected and stored at −80 °C until analysis.

2.4. Detection of mumps IgG antibodies

Mumps virus (MuV)-specific IgG were quantified using the Zeus ELISA Mumps IgG Test System (catalog no. 9Z9281G; Zeus Scientific, Inc., Branchburg, NJ) according to the manufacturer’s instructions. Serum was diluted 1:21, tested in duplicate on plates precoated with inactivated MuV antigen (Enders strain). IgG titers were expressed as a sample index, calculated as the ratio of the optical density (OD) of each sample to the OD of the kit-supplied calibrators. The sample index was then interpreted as either negative (≤0.9), equivocal (0.91 – 1.09), or positive (≥1.1).

Antibody avidity was also assessed using the same ELISA platform, as described previously [22]. Briefly, each serum sample was tested in parallel under two conditions: one following the manufacturer’s standard protocol, and the other including an additional wash step with 35 mM diethylamine (DEA, pH=10) after antigen-serum incubation. Avidity was expressed as a percentage, calculated as the ratio of the OD in the DEA-treated wells to the OD in untreated wells. According to the manufacturer, the assay has an intra-assay coefficient of variation (CV) of 5.7%, and an inter-assay CV of 7.2%, a sensitivity of 96.6%, and a specificity of 90.4%.

2.5. Detection of MuV-specific neutralizing antibodies (nAb)

To quantify MuV-specific nAb titers, we adapted our previously developed high-throughput fluorescence-based plaque reduction microneutralization (PRMN) assay for measles-specific nAbs [20, 23]. The modified assay used a recombinant MuV Iowa G strain expressing green fluorescent protein (MuV-GFP), generously provided by Dr. Biao He (University of Georgia). This strain was derived from a wild-type MuV isolated during the early phase of the 2006 mumps outbreak in Iowa, USA. Briefly, heat-inactivated sera (56 °C for 30 min) were serially diluted (4-fold, 4 replicates per dilution) in Opti-MEM medium (Gibco; Invitrogen, Carlsbad, CA) and mixed 1:1 with MuV-GFP inoculum, yielding final dilutions from 1:8 to 1:8,192). After a 1-h incubation at 37 °C, 50 μL of each mixture was added to Vero cell monolayers (2×10⁴ cells/well in DMEM with 5% FCS and Penicillin/Streptomycin) in 96-well plates and incubated for 43 h at 37 ºC, 5% CO₂. Fluorescent plaques were imaged and quantified using the ImageXpress^® Nano platform with MetaXpress^® software (Molecular Devices). The 50% neutralizing dose (ND₅₀) was calculated using the Karber formula [20, 23]. Each assay batch included a reference standard (anti-Mumps IgG Human Serum 93/582, NIBSC, UK). The CV of the MuV-specific assay was 5.23%, comparable to that of the measles-specific PRMN assay (5.75%) [20].

2.6. Detection of MuV-specific cytokines and chemokines

Cytokine and chemokine secretion from PBMCs stimulated in vitro with inactivated MuV (Enders strain) was quantified using a multiplex electrochemiluminescence kit (Meso Scale Diagnostics, Rockville, MD), as previously described [22]. The cytokine and chemokine panel was customized to include mediators released from innate and T cell activations as well as to reflect inflammation stages. Prior publications reporting on the innate response to infection with any of the three viruses (measles, mumps, rubella) were also used to select analytes. Our panel included Eotaxin, Eotaxin-3, IFN-α2a, IL8-HA, IP-10, MCP-1, MCP-4, MDC, MIP-1α, MIP-1β, TARC, IL-1β, IL-2, IL-4, IL-6, IL-8, IL-10, IL-12p70, IL-13, IFN-γ, and TNF-α [24–26]. Since cytokine and chemokine responses occurs shortly after vaccination, we chose to profile cytokine and chemokine responses at baseline and 28 days following vaccination. Briefly, PBMCs (200,000 cells/well) were cultured in triplicate under three conditions: (i) RPMI medium (unstimulated control), (ii) MuV Enders strain (MOI = 0.5), or (iii) phytohemagglutinin (PHA, 10 µg/mL; positive control) for 48 h at 37 °C, 5% CO₂. Supernatants were collected, stored at −80 °C, diluted 1:2, and analyzed in triplicate. Plates were read, and results were analyzed at the Mayo Clinic Immunochemical Core Laboratory. The coefficient of variation for each analyte, as specified by the manufacturer, was <20%. MCP-1 and IL-6, which exceeded the assay’s detection limit, were measured separately by ELISA (BD OptEIA^™ kits) using serum dilutions of 1:256 for MCP-1 and 1:16 for IL-6, per manufacturer’s instructions.

2.7. Statistical analysis

Cytokine and chemokine concentrations were normalized by subtracting values in unstimulated wells from those in MuV stimulated wells. The Wilcoxon rank-sum test was used to compare paired IgG, nAb, cytokine, and chemokine levels across three timepoints (D0 vs D28; D0 vs M18; and D28 vs M18), and between males and females. The associations between immune response outcomes and variables such as age, time since the second MMR dose, and BMI were assessed by Spearman’s rank correlation. A p-value <0.05 was considered statistically significant. All analyses and figure generation were performed in RStudio.

3. Results

3.1. Demographic characteristics of study cohort

Demographic characteristics of the study cohort were partially described in our previous publication [22, 27]. Briefly, 239 healthy individuals with documented receipt of two prior MMR vaccine doses were enrolled. For the current analysis, 25 individuals were excluded due to incomplete immunologic assessments, primarily resulting from inadequate blood sample volume. The final analytic cohort comprised 214 individuals, including 130 (60.7%) females and 84 (39.3%) males, with a median age of 35.9 years (range: 21.5 – 45.8). Most participants self-identified as White (n = 207), with smaller numbers identifying as Asian (n = 4), Alaskan Native (n = 1), 1 African American (n = 1), and multiracial (n = 1). The first MMR dose was administered at a median age of 15.6 months (IQR: 15.0 – 17.7) and the second dose at a median age of 12.5 years (IQR: 11.3 – 17.2). The median BMI was 27.5 kg/m² (IQR: 24.3 – 32.3).

3.2. Significant increase of mumps-specific IgG in females, but not males after MMR3

Mumps-specific IgG was quantified by ELISA and reported as a sample index. Since serum samples were unavailable for ELISA in 64 participants at the M18 timepoint, analyses were restricted to 150 participants (106 females, 44 males) with samples available at all three timepoints (D0, D28, and M18). At baseline (D0), the median sample index was 2.12 (IQR: 1.58 – 2.94); 21 participants (14%) were seronegative, 5 (3.3%) equivocal, 124 (82.7%) seropositive. No associations were observed with sex, age, time since the second MMR dose, or BMI (Supplementary Figure S1). By D28, the median sample index increased significantly to 2.64 (IQR: 1.93 – 3.35; p = 0.0019; Figure 1A), with a reduction in seronegativity to 11 individuals (7.3%). At M18, the median sample index was 2.85 (IQR: 1.93 – 3.57), significantly higher than D0 (p <0.001; Figure 1A), but not D28 (p = 0.13; Figure 1A). Seronegativity declined to 4 participants (2.7%) at M18. As with baseline, no demographic associations were identified at D28 and M18 (Supplementary Figure S1).

Figure 1. Impact of sex on antibody responses following a third dose of MMR vaccination (MMR3).

Mumps-specific antibody responses were assessed in serum collected at baseline (D0), 28 days (D28), and 18 months (M18) after MMR3. Mumps IgG antibodies, measured by ELISA and expressed as sample index in (A) the entire cohort, (B) females, and (C) males. IgG avidity is presented in (D) the entire cohort, (E) females, and (F) males. nAb titers are presented in (G) the entire cohort, (H) females, and (I) males. In panel A-C, blue and red dashed lines indicate positive (1.1) and negative (0.9) sample index cutoffs; values between 0.9 – 1.1 are considered borderline. In panel D-F, green dashed lines represent a 30% cutoff for dichotomizing low and high antibody avidity. In panel G-I, orange dashed lines indicate a proposed correlate of protection (16 ND₅₀) for nAb.

Compared with baseline, the median fold-increase in sample index was 1.15 (IQR: 0.99 – 1.56) at D28. Although the sample index remained elevated at M18 relative to D0, no further significant change (increase or decrease) occurred between from D28 to M18 with median M18/D28 fold-change of 0.99 (IQR: 0.86 – 1.25). Notably, females demonstrated a significant rise in sample index, from 2.18 (IQR: 1.49 – 2.94) at D0 to 2.82 (IQR: 2.00 – 3.36) at D28 (p = 0.0022; Figure 1B), and to 2.93 (IQR: 1.99 – 3.62) at M18 (p <0.001; Figure 1B). In contrast, no significant changes were detected among males at either D28 or M18 (p >0.05 for both timepoints; Figure 1C). Changes in sample index were not associated with other demographic measures (Supplementary Figure S2).

3.3. Slight increase in antibody avidity in females, but not in males

Mumps-specific IgG avidity was assessed in the same 150 participants with serum samples available at all three study timepoints. At D0, the median IgG avidity was 79.54% (IQR: 65.72 – 88.35%) and showed no association to sex, age, time since receipt of the second MMR dose, or BMI (Supplementary Figure S3). Using a 30% threshold [28], 6 participants had low-avidity IgG and 144 had high-avidity at D0. Median avidity did not increase significantly at D28 (81.88%; IQR: 70.33 – 89.22%; p = 0.079; Figure 1D), corresponding to a median fold-change of 1.03 (IQR: 0.95 – 1.18). In contrast, by M18, median avidity increased significantly to 86.87% (IQR: 79.44 – 94.28%; p <0.001; Figure 1D), with a median M18/D0 fold-change of 1.08 (IQR: 1.00 – 1.24). All six participants with low-avidity IgG at D0 remained low-avidity at D28 but converted to high-avidity IgG by M18. No associations were observed between avidity at D28 or M18 and demographic variables (Supplementary Figure S3). Consistent with IgG sample index findings, increases in IgG avidity were significant among females (Figure 1E), but not males (Figure 1F), and were unaffected by other demographic characteristics (Supplementary Figure S4).

3.4. Significant increase in neutralizing antibodies among males, but not females

Mumps neutralizing antibodies (nAbs) against Iowa G wild-type strain were quantified in 214 participants (136 females, 78 males) and reported as ND₅₀ values. At D0, the median ND₅₀ was 50.48 (IQR: 29.44 – 92.35), with no associations with sex, age, time since the second dose, or BMI (Supplementary Figure S5). Using an ND₅₀ titer ≥16 as a proposed correlate of protection against wild-type mumps virus [29], 202 participants (94.4%) were classified as seroprotective and 12 (5.6%) as sub-protective. Median ND₅₀ increased significantly at D28 to 67.00 (IQR: 37.31 – 123.21; p = 0.0014; Figure 1G) and remained elevated (but slightly lower) at M18 (60.18; IQR: 35.18 – 112.93). All participants who were sub-protective at D0 remained sub-protective after MMR3. ND₅₀ values at both D28 and M18 showed no associations with demographic variables (Supplementary Figure S5).

Consistent with the changes observed in sample index and IgG avidity, increases in nAb titers at D28 were limited but statistically significant. The median fold-change was 1.2 (IQR: 0.98 – 1.61) from D0 to D28 and 1.18 (IQR: 0.96 – 1.50) from D0 to M18. A decline in nAb titers was observed between D28 and M18, with a median fold-change of 0.84 (IQR: 0.75 – 1.18), indicating early antibody waning. Notably, the initial increase in nAb titers was sex-dependent: significant increases were observed in males from D0 to D28 (p = 0.0038; Figure 1I), whereas no significant change occurred in females (p >0.05; Figure 1H). Age and BMI were not associated with nAb responses (Supplementary Figure S6).

3.5. Significant increases in IP-10 secretion in males following MMR3

Cytokine and chemokine secretion by PBMCs following in vitro MuV stimulation was assessed in culture supernatants at D0 and D28 among the 214 participants. Of the 21 analytes measured, 17 were detectable in response to MuV stimulation; eotaxin-3, IL-4, IL-8HA, and IL-12p70 were not produced and were excluded from further analysis (Supplementary Figure S7 and S8). Weak to moderate correlations were observed among cytokines and chemokines at both baseline and D28 (Supplementary Figure S9 and S10).

Among the 17 responsive analytes, only interferon-γ induced protein 10 (IP-10/CXCL10) increased significantly following vaccination (Figure 2A; Supplementary Figure S11), rising from a median of 1,526.98 pg/mL (IQR: 628.71 – 1,857.93) at D0 to 1,687.82 pg/mL (IQR: 924.16 – 1,906.78) at D28 (p = 0.045). This increase was sex-dependent, with a greater fold change observed in males (median, 1.06; IQR: 0.92 – 1.69) than females (median, 1.03, IQR: 0.74 – 1.42) (p = 0.044, Figure 2B). No significant sex-related differences were observed for the remaining analytes (Supplementary Figure S12).

Figure 2. Effect of sex on modest increase of IP-10 following MMR3.

(A) Of the 21 cytokines and chemokines, only IP-10 significantly increased 28 days (D28) post-MMR3 compared to baseline (D0). (B) The increases in IP-10 were higher in males than in females.

3.6. Correlation between the change of IP-10 level and mumps antibodies

Because IP-10 was the only analyte to show a significant increase post-vaccination and a sex-dependent effect (Figure 2), mirroring patterns observed in antibody responses (Figure 1), we further assessed its relationship with mumps-specific immunity. Changes in IP-10 were negatively correlated with D28/D0 changes in IgG sample index (r = −0.23; p = 0.014; Figure 3A). However, IP-10 changes were not significantly associated with any other humoral immune response outcomes measured between D0 to M18 (Figure 3B-3F).

Figure 3. Correlations between changes in IP-10 and antibody responses following MMR3.

Changes in IP-10 from D0 to D28 (D28/D0) were analyzed in relation to antibody changes (A-C) from D0 to D28 (D28/D0) and (D-F) from D0 to M18 (M18/M0), including sample index (A, D), IgG avidity (B, E), and nAb (C, F). The blue lines represent linear trends, and gray shadings indicate the 95% confidence interval.

4. Discussions

A third dose of MMR vaccine (MMR3) has been recommended as an outbreak-control measure [1]. Although estimates of vaccine effectiveness remain uncertain, available evidence suggests that MMR3 can reduce attack rates [4–6], and that early administration during mumps outbreaks may help limit transmission [30]. Nevertheless, studies evaluating the immunological impact of MMR3 have reported heterogeneous findings. In contrast to reports describing robust immune boosting after MMR3 [7–9], our data demonstrated only modest increases in antibody responses at 28 days and no additional improvement at 18 months post-vaccination (Figure 1), consistent with other studies reporting limited or transient immunologic effects [10, 31]. Several reports have also demonstrated declines in antibody titers after MMR3, with a return to baseline levels within one to two years [10, 32]. In our study, mumps-specific IgG levels were largely maintained over time, whereas nAb titers declined. The basis for this discrepancy remains unclear and warrants further investigation.

Although a definitive correlate of protection for mumps has not been established and may vary depending on the antigenic match between vaccine and contemporary strains [29, 33, 34], existing evidence suggest that vaccine-induced antibodies can still effectively neutralize circulating strains [35, 36]. In this study, neutralization assay using the Iowa G wild-type strain demonstrated that all participants with sub-protective nAb titers at D0 remained sub-protective after vaccination, reinforcing the limited immunological benefits of MMR3. Our observations are consistent with a previous report showing that antibody titers against genotype G remained largely unchanged following MMR3 administration [32]. Similarly, prior studies have reported that MMR3 fails to elicit protective nAb titers in approximately 10% of seronegative adults [31]. The mechanism underlying this insufficient boosting remains unclear and warrants further investigation. A deeper understanding of contributing factors will be essential for developing strategies to enhance mumps-specific immunity in at-risk populations. Enhancement of cellular immune responses was also minimal in our study, with IP-10 emerging as the only analyte that increased following vaccination. Collectively, these findings suggest that MMR3 confers modest immunological benefit in our study cohort.

We detected significant cytokine and chemokine secretion in PBMCs in response to mumps stimulation at baseline, including key mediators of T cell response such as IL-2, IL-4, IL-10, IFN-γ, TNF-α, MIP-1α, and MIP-1β [37]. These findings indicate that mumps-specific T cell responses were retained for a prolonged period following the second MMR dose (Supplementary Figure S7). However, only minimal differences were observed between the baseline and post-vaccination (Day 28), suggesting that MMR3 provided little enhancement of mumps-specific cellular immunity (Supplementary Figure S8). Given the central role of T cells in supporting antibody production [38], the limited cellular responses observed here may help explain the modest antibody responses. Moreover, because T cell-mediated immunity contributes directly to protection against mumps infection [39, 40], these findings further underscore the limited immunological benefits of MMR3 and highlight the need to explore alternative strategies for achieving more robust and durable immune responses to mumps.

Sex is a well-recognized biological variable influencing vaccine immunogenicity, with females generally mounting stronger immune responses than males [41, 42]. We and others have previously demonstrated higher mumps-specific IgG antibody titers in females compared with males following vaccination [11, 43–45]. In this study, however, sex-associated effects were heterogeneous and varied across immunological outcomes. Specifically, IgG titers and avidity increased significantly in females but not males (Figure 1A-F), whereas increases in nAb titers and IP-10 were observed in males (Figure 1G-I, Figure 2B). This discordance suggests that quality and functional characteristics of mumps-specific immunity induced by MMR3 may differ by sexes. Although the mechanisms underlying these differences remain undefined, they may reflect the influence of sex hormone signaling, X-linked immune gene regulation, or sex-specific immune response kinetics [46–48]. The broader implications of these findings are notable. If MMR3 preferentially enhances antibody quantity and avidity in females and antibody functionality and associated cytokine responses in males, the net protective benefit of the vaccine may also differ by sex [48]. Such heterogenicity highlights the value of incorporating sex-stratified analyses into vaccine evaluation. More broadly, our findings align with emerging evidence supporting precision vaccinology approaches that account for biological variables such as sex, age, and genetic background to optimize vaccine-induced protection [49, 50].

IP-10 levels increased after MMR3, with a greater rise in males (Figure 2). Interestingly, IP-10 was negatively correlated with MMR3-induced changes in mumps-specific IgG (sample index) (Figure 3A), and males did not mount significant IgG increases (Figure 1C), raising a possibility that heightened IP-10 may contribute to their suboptimal antibody responses. IP-10, a CXC chemokine secreted in an IFNγ-dependent manner from multiple cell types [51], exerts diverse effects, including chemotaxis, proliferation, and apoptosis, via CXCR3 receptor on T cells, NK cells, dendritic cells, and subset of B cells [52]. Elevated IP-10 has been reported in several infections, including mumps, respiratory syncytial virus, and measles [51, 53, 54], and may play a role in mumps orchitis-associated infertility [55, 56]. In measles, higher serum IP-10 levels were linked to lower neutralizing antibody titers and antibody avidity [53], consistent with the negative correlation observed in this study. In contrast, vaccine studies of rubella [57], SARS-CoV-2 [58], and Ebola [59] have reported positive correlations between IP-10 and antibody responses, and mechanistic studies suggest that IP-10 can promote B-cell differentiation and antibody production [60]. Collectively, these findings suggest that the influence of IP-10 on humoral immunity may be sex-dependent and shaped by immunological context and pathogen-specific factors, underscoring the need for further investigation.

In our cohort of 214 healthy participants previously vaccinated with two doses of MMR vaccine, MMR3 conferred limited additional humoral and cellular immunologic benefits. The findings further support the use of MMR3 primarily as a targeted intervention during mumps outbreaks rather as a routine booster in well-vaccinated populations. From a public health perspective, the modest immunologic enhancement observed suggests that the primary value of MMR3 may lie in transiently increasing population immunity to interrupt transmission in high-risk settings, such as universities, as previously reported [5]. The limited immunologic gains underscore the importance of continued research into correlates of protection, which remain poorly defined for mumps, mechanisms underlying waning immunity, and potential strategies to improve long-term mumps vaccine effectiveness. Additionally, we observed a discordant influence of biological sex on both the magnitude and functional quality of mumps-specific antibody responses (Figure 1) as well as on IP-10 responses (Figure 2). Although the mechanisms underlying this observation remains unclear and warrant further investigations, they may offer potential insight into pathways contributing to sex-biased vaccine responses.

This study has several limitations. First, the study cohort reflected the local demographic structure and had limited ethnic diversity, which may restrict the generalizability of the findings. Second, the limited immunologic benefits observed were based on immunological endpoints rather than clinical protection, as no breakthrough infections or substantial exposure risks occurred within the study population. Third, cytokines and chemokines were measured from PBMC culture supernatants, leaving the cellular source of these analytes unresolved; moreover, the in vitro PBMC culture system may not fully reflect the in vivo kinetics of cytokine responses. Fourth, all participants lacking protective nAb titers at baseline remained unprotected after MMR3, and the mechanisms underlying this non-responsiveness remain unclear, warranting further mechanistic studies. Without a better understanding of these mechanisms, the potential for additional MMR booster doses to meaningfully enhance mumps immunity in this subgroup remains uncertain.

In conclusion, this study characterized both humoral and cellular immune responses to mumps following a third dose of MMR vaccine. Overall, MMR3 elicited modest enhancement of mumps-specific immunity. We identified a pronounced influence of biological sex on post-vaccination immune responses and found IP-10 to be associated with reduced vaccine responsiveness. Together, these findings provide additional insights into the heterogeneous effects of MMR3 and underscore the need for alternative strategies to achieve more robust and durable protection against mumps.

Supplementary Material

1

Supplementary Figure S1. Nonsignificant impact of demographics on sample index. Sample indices at three timepoints were comparable between males and females (A, E, I) and showed no significant correlation with age (B, F, J), time since the second MMR dose (C, G, K), or body mass index (BMI) (D, H, L). The blue lines in (B-D, F-H, J-L) represent linear trends, and gray shadings indicate the 95% confidence interval.

2

Supplementary Figure S2. Nonsignificant correlations of sample index changes post-MMR3 with demographic factors. Changes in sample index (A-C) from D0 to D28 (D28/D0), from D0 to M18 (M18/D0), and from D28 to M18 (M18/D28) showed no significant correlations with age (A, D, G), time since the second MMR dose (B, E, H), or BMI (C, F, I). The blue lines represent linear trends, and gray shadings indicate the 95% confidence interval.

3

Supplementary Figure S3. Minimal impact of demographics on IgG avidity. IgG avidity at three timepoints was comparable between males and females (A, E, I) and showed non-significant correlation with age, except for D0 avidity (B, F, J), time since the second MMR dose (C, G, K), or BMI (D, H, L). The blue lines in (B-D, F-H, J-L) represent linear trends, and gray shadings indicate the 95% confidence interval.

4

Supplementary Figure S4. Nonsignificant correlations of IgG avidity changes post-MMR3 with demographic factors. Changes in IgG avidity (A-C) from D0 to D28 (D28/D0), from D0 to M18 (M18/D0), and from D28 to M18 (M18/D28) demonstrated no significant correlations with age (A, D, G), time since the second MMR dose (B, E, H), or BMI (C, F, I). The blue lines represent linear trends, and gray shadings indicate the 95% confidence interval.

5

Supplementary Figure S5. Nonsignificant impact of demographics on neutralizing antibodies (ND₅₀). Neutralizing antibody titers at three timepoints were comparable between females and males (A, E, I) and showed no significant correlation with age (B, F, J), time since the second MMR dose (C, G, K), or BMI (D, H, L). The blue lines in (B-D, F-H, and J-L) represent linear trends, and gray shadings indicate the 95% confidence interval.

6

Supplementary Figure S6. Nonsignificant correlations of ND₅₀ changes post-MMR3 with demographic factors. ND₅₀ changes (A-C) from D0 to D28 (D28/D0), from D0 to M18 (M18/D0), and from D28 to M18 (M18/D28) showed no significant correlations with age (A, D, G), time since the second MMR dose (B, E, H), or BMI (C, F, I). The blue lines represent linear trends, and gray shadings indicate the 95% confidence interval.

7

Supplementary Figure S7. Baseline (D0) cellular responses. PBMCs collected at D0 were cultured RPMI medium (unstimulated control) or stimulated in vitro with Enders strain mumps virus (MuV). Expression levels of 21 cytokines and chemokines were quantified in cell culture supernatants using a multiplex assay.

8

Supplementary Figure S8. Cellular responses at 28 days (D28) post-MMR3. PBMCs collected at D28 were cultured RPMI medium (unstimulated control) or stimulated in vitro with Enders strain mumps virus (MuV). Expression levels of 21 cytokines and chemokines were quantified in cell culture supernatants using a multiplex assay.

9

Supplementary Figure S9. Paired correlation among 17 responsive cytokines and chemokines at baseline (D0). Numbers indicate pairwise correlation coefficients, with shades of blue representing positive correlations and shades of red representing negative correlations.

10

Supplementary Figure S10. Paired correlation among 17 responsive cytokines and chemokines at 28 days (D28) post-MMR3. Numbers indicate pairwise correlation coefficients, with shades of blue representing positive correlations and shades of red representing negative correlations.

11

Supplementary Figure S11. Nonsignificant changes of 16 cytokines and chemokines post-MMR3.

12

Supplementary Figure S12. Nonsignificant impact of sex on the changes of cytokines and chemokines from D0 to D28.

Data sharing statement

All data included in this study are presented as figures and tables and available from the corresponding author upon reasonable request.