Seronegative Vaccinees May Not Benefit From Multiple Booster Doses of MMR Vaccine in Restoring Immunity Authors: Gabriele Anichini, Chiara Terrosi, Giulia Alessandri, Giovanni B. Miceli, Emma Borrelli, Simonetta Fabrizi, Gianni G. Savellini, Maria G. Cusi Published: 2024 ABSTRACT ABSTRACTDespite the availability of a highly efficacious vaccine, a global resurgence of measles infections has occurred, largely due to decreased vaccination coverage and waning immunity following the two‐dose vaccination schedule. This study aims to assess the cellular immune response in individuals who did not respond to the two‐dose MMR vaccine and evaluate the efficacy and durability of immune responses after booster doses. An observational study was conducted involving 24 individuals who were seronegative for measles years after completing the two‐dose MMR vaccine schedule. Of these, 11 seroconverted after a single booster dose, while 13 required an additional booster. Antibody levels were monitored 1 and 3 years post‐booster. Cellular immune responses were evaluated in 10 nonresponders by in vitro stimulation of PBMCs with UV‐inactivated measles virus. In nonresponders, measles virus stimulation did not induce significant alterations in T, B, or NK cell populations, indicating a lack of measles‐specific immune memory. However, stimulation with rubella induced a significant immune response. Booster doses in subjects with waning antibodies resulted in low IgG titers, which significantly declined over 3 years. Neutralizing antibody titers were also low for both double and single booster recipients. This study highlights the limited long‐term efficacy of booster doses in individuals with waning immunity or nonresponsiveness to the measles vaccine. Current vaccination strategies relying on booster doses may not be sufficient for sustained protection in these individuals. Improving vaccine formulations to elicit stronger and longer‐lasting immunity is essential for effective measles control and elimination. Introduction Despite the availability of a safe and highly efficacious live‐attenuated vaccine, there has been a recent surge in measles infection in various countries [1, 2, 3, 4, 5]. This surge can be largely attributed to a global reduction in vaccination coverage, a trend partly exacerbated by the SARS‐CoV‐2 pandemic [6, 7, 8, 9]. The phenomenon of primary vaccine failure or waning immunity following two‐dose measles vaccination [10, 11, 12, 13, 14, 15, 16, 17, 18] presents a significant challenge in maintaining efficient population‐level immunity against this highly contagious virus. Addressing this issue requires a deeper understanding of the underlying immunological mechanisms and the possible need for booster doses to restore and sustain immunity. Notably, previous studies provided evidence supporting the utility of booster doses in bolstering immune responses and prolonging immunity in individuals with waning antibody titers [19, 20, 21]. These findings underscore the importance of booster vaccination strategies in mitigating the risk of measles resurgence and achieving sustained population immunity. However, it remains unclear whether those individuals who become seronegative years after getting vaccinated remain susceptible to the infection or retain an immunological memory that is promptly awakened in the event of re‐exposure. Similarly, it is necessary to verify the response in vaccinated subjects who have received a further booster dose to confirm whether the conferred protection is long‐lasting. This study aims to assess the cellular immune response against measles virus in subjects who exhibited nonresponsiveness, despite completing a two‐dose MMR vaccine schedule. In addition, it aims to monitor levels of protection over time in subjects who have responded to multiple booster doses of the MMR vaccine. Materials and Methods Study Population Participants in this observational study were students, postgraduates, medical doctors, and health care workers subjected to routine analysis for the biological risk assessment at the Centre of Preventive Medicine and Health Surveillance of the University Hospital “Santa Maria alle Scotte” in Siena between December 2023 and January 2024. All participants provided us with their vaccination records against measles, consisting of two doses of M‐M‐RII (Merck & Co. Inc. Rahway, NJ, USA). Moreover, at the times of blood collection and the administration of the booster dose(s), all participants were declared healthy and were not taking any medication. We selected 24 subjects (9 males, 15 females; median age 29.1 years, 95% CI: 27.4–30.8) who tested negative for circulating measles IgG antibodies years after being administered two doses of MMR vaccine (average elapsed time from second dose administration 15.3 years; 95% CI: 14.3–16.4), although responding against rubella and mumps. Among them, 11 subjects (four males, seven females; mean age 26.9 years, 95% CI: 24.2–29.5) seroconverted after the administration of a further booster dose, while the remaining 13 (five males, nine females; mean age 31.0 years, 95% CI: 29.3–32.6) needed an additional dose, which was administered 1 month after the previous one. Subsequently, anti‐measles circulating antibodies were evaluated about 1 year later (1.14 years for single‐booster receivers, 1.25 years for double receivers) to assess anti‐measles response. Subjects were monitored once again 3 years after the booster administration. Both circulating IgG and neutralizing antibody titers were evaluated. Among them, 10 subjects (five males, five females, average age 32.4 years) who did not respond against measles virus even after the administration of the additional second booster dose (thus named nonresponders), were selected. Blood was collected from each nonresponder subject to analyze cell‐mediated immune response after in vitro stimulation of isolated PBMC with UV‐inactivated measles virus by cytofluorimetric analysis. Despite being seronegative, these subjects were checked to understand whether they still maintained an immunological memory after viral re‐exposure. Finally, five of them were also stimulated with inactivated rubella virus as a positive control to validate our stimulation protocol. This research was carried out according to the principles of the Declaration of Helsinki. Ethical approval was obtained from the local Ethical Committee for clinical trials (approval no. 25626, 20/11/2023) (Comitato Etico Regionale per la Sperimentazione Clinica della Toscana—sezione AREA VASTA SUD EST) in terms of General Data Protection and Regulation (GDPR) upon written informed consent signed by all subjects before participating in this study. Cells and Viruses Vero cells (ATCC CCL‐81) were grown as a monolayer in Dulbecco's modified Eagle medium (DMEM) (Euroclone, Milan, Italy), supplemented with 100 U/mL penicillin and 100 µg/mL streptomycin (Euroclone, Milan, Itay) and 5% heat‐inactivated fetal calf serum (FCS) (Euroclone, Milan, Italy) in a humidified 5% CO2 atmosphere at 37°C. Then, PBMCs were isolated from selected donors’ blood upon written informed consent by using Ficoll‐Hypaque gradient separation. PBMCs were then cultured in RPMI (Euroclone, Milan, Italy), supplemented with 100 U/mL penicillin/streptomycin (Euroclone, Milan, Italy) and 10% heat‐inactivated FCS (Euroclone, Milan, Italy) in a humidified 5% CO2 atmosphere at 37°C. Measles virus Edmonston B strain (ATCC VR‐24) and Rubella virus (RA27/3, ATCC CCL‐171) were propagated on cells before being titrated. Measles and Rubella Virus Titration Briefly, the supernatant of infected Vero cells (ATCC‐CCL‐81) was 10‐fold diluted. Therefore, 50 μL of each dilution was added in quadruplicate to a 96 well‐plate together with an equal volume (50 μL) containing 2 × 105 cells. After 5 days, the presence of cytopathic effect (CPE) was observed under a microscope (Olympus IX51; Olympus, Tokyo, Japan), and the 50% endpoint titers were calculated by using the Reed and Muench method [22]. UV‐Inactivation of Measles and Rubella Virus Measles (Edmonston B strain) and rubella (M33 strain) viruses were UV‐inactivated at a dose of about 1.5 × 102 µW/mm [2, 23]. The protein concentration of viruses obtained from Vero cell cultures was determined by the Bradford assay. To confirm the complete viral inactivation, the UV‐treated virus was titrated on Vero cells, showing no cytopathic effect by microscopical examination after 1 week. Measles IgM/IgG Antibody Detection Sera obtained from subjects were analyzed for the presence of specific IgM and/or IgG antibodies to the recombinant nucleoprotein of measles (Liaison DiaSorin, Saluggia, Italy), a chemiluminescence (CMIA) analyzer using paramagnetic solid phase microparticles. Measles threshold IgG values, regarded as positive immune status, were > 16.5 AU (arbitrary unit/mL) with a detection limit of 13.5 AU/mL, while the IgM threshold was > 1.1 AU/mL, with a detection limit of 0.9 AU/mL. Rubella IgG and IgM antibodies were tested by CMIA (Alinity, Abbott, USA). Peripheral Blood Mononuclear Cells Isolation and Stimulation For the evaluation of T, B, and NK cell responses after stimulation with UV‐inactivated measles or rubella virus, peripheral blood mononuclear cells (PBMCs) were isolated from the whole blood of selected subjects using Lympholyte Cell Separation Media (Cedarlane, Burlington, Canada). Afterward, PBMCs of each subject were washed with RBC (red blood cell) lysis buffer and seeded in quadruplicate at a density of 1×106 cells in 500 µL, in 24‐well plates with RPMI‐1640 medium (Euroclone, Milan, Italy), supplemented with 10% heat‐inactivated human serum (Euroclone, Milan, Italy), IL‐2 (20 U/mL), IL‐10 (50 ng/mL), GM‐CSF (50 ng/mL), and IL‐4 (0.5 ng/mL). After 72 h, 1 MOI of UV‐inactivated measles/rubella virus was added in three wells (for T, B, and NK analysis), and one well was kept as an unstimulated cell control. NK‐cell populations were harvested and analyzed by flow cytometry 24 h after antigen stimulation, while T and B cells were analyzed 96 h after stimulation. Multiparameter Flow Cytometry Analysis After stimulation, PBMCs of selected subjects were harvested and washed, using PBS supplemented with 3% FCS, then stained with a fixable viability dye (LIVE/DEAD Fixable Aqua Dead Cell Stain Kit, Thermo Fisher Scientific, Waltham, MA, USA) for 20 min at 4°C. Surface staining with antibodies binding to CD3 (clone SK7), CD4 (clone SK3), CD8 (clone RPA‐T8), CCR7 (clone 150503), CD45RA (clone HI100), CD38 (clone HIT2), HLA‐DR (clone G46‐6), CXCR3 (clone 1C6), and CCR6 (clone 11A9) was performed to analyze T cell helper and memory subsets, according to the manufacturer's instructions. Surface staining was also performed with CD3 (clone SK7), HLA‐DR (clone G46‐6), CD19 (clone HIB19), CD27 (clone L128), IgD (clone IA6‐2), CD20 (clone 2H7), CD24 (clone ML5), and CD38 (clone HIT2) for B cell analysis. All antibodies for T and B cell analysis were supplied from BD Biosciences (New York, NY, USA). Finally, CD3 (clone HIT3a), CD14 (clone 6ID3), CD19 (clone HIB19), CD16 (clone eBioCB16), and CD56 (clone CMSSB) were used for NK cell analysis. Antibodies were supplied by Thermo Fisher Scientific (Waltham, MA, USA). After fixation with PBS + 2% paraformaldehyde for 20 min at 4°C, cells were washed and resuspended in PBS, supplemented with 0.5 mM EDTA before being acquired with SO LSRFortessa X20 flow cytometer (BD Biosciences, New York, NY, USA). Data analysis was performed by using FlowJo v10 (TreeStar, Ashland, OR, USA). Granzyme B and IFN‐γ ELISA To further evaluate both T and NK cell response following stimulation with UV‐inactivated measles and rubella viruses (the latter serving as a positive control), supernatants of stimulated PBMCs were analyzed for IFN‐γ and Granzyme B expression using the Human IFN‐γ Quantikine and Human Granzyme B Duoset ELISA Kits (R&D Systems, MN, USA). Absorbance was measured at 450 nm using a GloMax Discover Microplate Reader (Promega Corp., WI, USA). The analyte concentration was calculated by using a standard curve and reported as picograms per milliliter (pg/mL). All experiments were performed in triplicate. Measles Microneutralization Assay A measles virus neutralization assay was carried out on Vero cells in a 96‐well microplate. Twenty‐five microliters of twofold serial dilutions (1:8 to 1:1024) of vaccinated or naturally infected people sera were added to an equal volume of the Edmonston B strain MV containing 250 TCID50 and incubated for 90 min at 37°C. Finally, 50 μL of Vero cells suspension (2 × 105 cells/mL) prepared in a complete DMEM (Euroclone, Milan, Italy) medium was added to each well. Five days after incubation at 37°C, the cultures were microscopically examined for the presence of CPE. Neutralization titers were expressed as the reciprocal of the highest serum dilution inhibiting at least 90% (MNT90) of CPE. Serum samples with neutralizing titers of less than 8 were considered negative. Anti‐Measles serum (3rd International Standard) 97/648 (NIBSC, Blanche Ln, South Mimms, Potters Bar EN6 3QG, UK) was included in each assay as a positive control, showing a 90% neutralizing endpoint titer corresponding to a concentration of 37.5 mIU/mL. Statistical Analysis Statistical significances were assessed with the two‐tailed χ 2 test, paired or unpaired depending on need. Results were considered statistically significant at p < 0.05. When appropriate, ± standard deviation (SD) and 95% confidence interval (95% CI) were calculated and reported. All analyses were performed by using Graph Pad Prism software (v.7.0). To represent data and to perform results comparisons, each single cell population analyzed by cytofluorimetry was normalized to the amounts of cells of interest for the specific subset of the cell population. With regard to Fluorescence‐activated Cell Sorting (FACS) data, values were reported as population percentage (%) with respect to the ancestral population. Finally, differences concerning stimulation were reported in estimation plots, as an increase/decrease in the population percentage compared to the unstimulated control. Results Cellular Immune Response in Nonresponders We tried to assess whether, albeit responding against rubella and mumps, subjects who had not responded against measles after two cycles of MMR vaccination were still able to preserve an immunological memory against the virus in case of re‐exposure. For this purpose, PBMCs isolated from these subjects were in vitro stimulated with the measles virus. Subsets of T, B, and NK cell populations were analyzed. PBMCs were also stimulated with UV‐inactivated rubella virus since previous serological tests on sera of the same subjects had revealed a good IgG response against this virus. Regarding T cell subsets, data showed little or no substantial alterations in the CD3+ T cell population (both CD4+ and CD8+) 96 h after stimulation with measles antigen. On the contrary, the same cell populations showed a different behavior when stimulated with UV‐inactivated rubella virus. A more pronounced response emerged with significant induction of both CD4+(p = 0.03) and CD8+naïve cells (CCR7−, CD45RA+) (p = 0.04) (Figure 1), compared to the unstimulated samples. As far as CD4 subsets are concerned, a significant induction of Th17 cells (p = 0.04 vs. nonstimulated, p = 0.004 vs. measles stimulation) was observed after stimulation with rubella virus. Additionally, IFN‐γ levels in supernatants of nonstimulated and stimulated PBMCs with UV‐inactivated measles virus showed no significant changes, whereas a significant increase in IFN‐γ production was observed after stimulation with UV‐inactivated rubella virus (p = 0.001). No significant differences in any cytokine were observed between the two groups of subjects. Representation of normalized parameters related to immune CD4+ (a, b) and CD8+ (c, d) T cell populations before and after stimulation with UV‐inactivated measles (a, c) or rubella (b, d) virus with relative IFN‐γ induction (e, f). Nonstimulated samples (blue) are linked to the stimulated counterparts (measles, red; rubella, green) through connection lines. Among both populations, Naïve, Central Memory (CM), Effector Memory (EM), and Terminally Differentiated Effector Memory (TEMRA) cells were analyzed. Moreover, among CD4+ cells, T helper (Th) 1, 2, and 17 were analyzed. Results were expressed as the variation in the population percentage (with respect to CD4+ and CD8+ cells). Statistical significance was assessed using the two‐tailed χ 2 test. The p values are reported in the figures, where * stands for p < 0.05 and ** for p < 0.01. No differences were reported in the CD4+ and CD8+ populations, as well as in IFN‐γ induction, between the nonstimulated and measles‐stimulated groups. Activated naïve, activated switched memory, and unswitched memory B cell subsets also revealed no relevant differences after measles stimulation, compared to the nonstimulated cells. Significant induction of both activated naïve and memory‐switched cells to rubella antigen was observed (p < 0.001) instead, thus confirming responsiveness to this virus (Figure 2). Representation of activated (Activ) Naïve and Switched (Sw) Memory, together with Unswitched (Unsw) Memory subsets variations (%) in nonstimulated (blue circles), stimulated with measles (red circles) or rubella (green circles) cells. Results are reported in horizontal lines as mean values. Individual data points are shown. The p values of the differences in cell populations between nonstimulated and stimulated groups are reported in the figures, where * stands for p < 0.05 and *** for p < 0.0001. Finally, natural killer cells (NK) were evaluated 24 h following in vitro stimulation, revealing a difference in behavior between MeV and RV. As shown in Figure 3, no alterations in both CD56dim and CD56bright populations, with respect to the unstimulated control, were observed after measles exposure. On the other hand, stimulation with RV induced a significant drop in the CD56dim population (p = 0.005) and a significant increase in the CD56bright subset (p = 0.01). These findings were further supported by the quantitation of Granzyme B in the supernatants of both nonstimulated and stimulated PBMCs. Specifically, no significant differences in Granzyme B concentrations were detected after stimulation with UV‐inactivated measles virus, while a significant increase in Granzyme B production was observed after stimulation with UV‐inactivated rubella virus (p = 0.001). These results confirm the reliability of the method employed and demonstrate that the measles virus did not induce any measurable alterations in these NK cell subsets. Representation of the relative NK cell subsets (CD56 dim, CD56 bright) variations (%) in nonstimulated (blue circles), stimulated with measles (section a, red circles), or rubella (section b, green circles), together with the relative Granzyme B production (c, d). Results are reported in horizontal lines as mean values. Individual data points are shown. The p values are reported in the figures, where * stands for p < 0.05 and ** for p < 0.01. Anti‐Measles IgG Response in MMR Booster Recipients To evaluate the efficacy and durability of immune response after booster dose administration among vaccinated subjects, who became IgG negative to measles over time, we selected 24 subjects (9 males, 15 females; median age 29.1 years; average elapsed time from second dose administration 15.3 years;). Of these, 11 (four males, seven females) seroconverted after the administration of a further single booster dose, while the remaining 13 (five males, nine females) needed an additional booster to seroconvert, administered 1 month after the previous one. Subsequently, the antibody titers were evaluated 1 year and 3 years later for single and double booster receivers to evaluate the trend of antibody response. At the first timepoint, all subjects showed specific IgG response after one or two booster administrations, although at a low level (median titers 58.7 AU/mL for single‐booster recipients and 59.9 AU/mL for two‐dose recipients; p > 0.05). Then, subjects were monitored again 3 years later to evaluate the specific antibody level. Results showed a significant decline of median IgG titers in both single‐booster (50.2 vs. 29.1 AU/mL; p = 0.004) and double‐booster (59.9 vs. 30.2 AU/mL; p = 0.01) recipients (Figure 4). No significant differences in median IgG titers were found in the two groups with respect to gender. Representation of anti‐measles circulating IgG titers in single‐ and double‐bolstered populations 1 year and 3 years post‐boost, respectively (a, c). Individual data points are shown as circles (1 year) or squares (3 years). In each plot, the top of the bar represents the median, while the top and bottom lines show the 95% confidence interval (95% CI). The p values are reported in the figures, where * stands for p < 0.05 and ** for p < 0.01 estimation plots were reported to show the decrease of antibody titer on each subject, both for single (b) and double‐bolstered groups (d). Mean values ± SD were reported in horizontal lines. Within estimation plots, samples with the same value cannot be individually identified as they overlap. Finally, we aimed to verify whether the low levels of circulating anti‐measles IgG at 3 years corresponded to a lack of neutralizing antibodies, considered the best correlate of protection against infection (Figure 5). Results obtained from both groups were compared to those obtained from a group of vaccine responders, represented by 10 subjects (five males, five females, mean age 26.8 years), whose neutralizing GMTs were evaluated 3 years after a two‐dose MMR vaccination schedule. Results showed low neutralizing titers in both the boosted groups, with a geometric mean titer (GMT) of 28.53 (95% CI: 19.56–41.62) for the double booster and 25.66 (95% CI: 15.40–42.78) for the single‐booster group, significantly lower than that one observed in the primary vaccinated group (GMT = 380.3, 95% CI: 183.5–788.1, p < 0.001), whose antibody titer was considered protective [24]. No significant differences related to gender were found in this analysis. Comparison of neutralizing antibody titers 3 years after a single or double MMR booster dose among vaccinated seronegative subjects with those observed in naïve subjects, 3 years following the complete vaccination schedule. Individual data points are shown as black circles. In each plot, the top of the bar represents the geometric mean, while the top and bottom lines show the 95% confidence interval (95% CI). The p values are reported in the figures, where *** stands for p < 0.001, and ns stands for not significant. Discussion The study sheds light on crucial aspects of the immunological response to measles vaccination, particularly on individuals with nonresponsiveness or waning immunity. Previous studies have also evaluated the immune response to booster doses, offering valuable insights into their efficacy and durability in similar populations [24, 25]. The global resurgence of measles infections underscores the need to understand vaccine failure mechanisms and the efficacy of booster strategies. This investigation on the cellular immune responses among nonresponsive individuals to measles vaccination revealed intriguing insights into immunological memory maintenance after measles virus exposure. No significant alterations in T, B, and NK cell subsets were observed after measles virus stimulation. These results were further confirmed by the lack of induction of both IFN‐γ and Granzyme B in the supernatants of stimulated PBMCs compared to the nonstimulated control (p > 0.5). To validate the method and rule out stimulation protocol issues, samples were also stimulated with the rubella virus. Significant alterations in naïve CD4/8+, Th17+, NK, and B cell populations, along with strong inductions of both IFN‐γ and Granzyme B levels (p = 0.001 for both), were observed after rubella stimulation, confirming the validity of the procedure to justify a specific impairment in measles‐specific immunity rather than a general immune dysfunction in nonresponsive individuals. Booster dose in individuals with waning or undetectable antibody levels following a complete MMR vaccination schedule showed a weak humoral response against measles (about 50 AU/mL). Concerns arise on the durability of immune response elicited by booster doses and the antibody titer decline over time. While booster doses temporarily conferred a low level of immunity in individuals with waning antibody levels, their long‐term benefits appeared limited, due to antibody titer decline. Indeed, the antibody titers after a year (50.2 AU/mL) halved in 3 years (29.1 AU/mL), indicating that the response to further booster doses was not efficacious in nonresponders. It is worth noting that neutralizing antibody GMTs of 28.53 and 25.66 after double or single booster, respectively (with only three subjects showing a GMT > 64), may not be sufficient to provide adequate protection. A previous study by Müller in 1996 proposed a titer > 1:64 as the protective antibody level against measles [26]. Presumably, the above subjects might not be protected against measles. Therefore, the current vaccination strategies, which rely on booster doses, should be reconsidered in subjects who do not adequately respond after the complete cycles of vaccination. Similarly, the decline of immune response over time has also been recorded for mumps in vaccinated young adults. This has been mainly attributed to antigenic differences between the vaccine strain and the infecting strain [27, 28]. Additionally, lower memory B cell frequency and waning antibody levels contribute to reduced immunity [27]. Thus, enhancing the MMR vaccine formulation to elicit a stronger and more durable protection against measles and mumps, may offer a more sustainable strategy for controlling and eliminating these diseases. Therefore, continued research and innovation in vaccine development are crucial to address evolving challenges in infectious disease prevention and control. A mucosal vaccine would likely contribute to strengthening immunity against measles and provide a chance to induce long‐lasting immunity. Further research is needed to elucidate vaccine failure mechanisms, optimize booster strategies, and identify novel correlates of protection to prevent measles resurgence in response to the growing threats of infectious diseases. Author Contributions Maria G. Cusi contributed to the conceptualization, validation, supervision, and review of the manuscript. Gabriele Anichini prepared the first draft of the manuscript. Gabriele Anichini and Chiara Terrosi contributed to the design and optimization of experimental methods. Giovanni Battista Miceli, Emma Borrelli, and Simonetta Fabrizi provided the clinical samples. Experiments were conducted by Gabriele Anichini and Chiara Terrosi, and Giulia Alessandri and Gabriele Anichini contributed to formal analysis and data curation. All authors had full access to all the data in the study and accepted responsibility for the decision to submit for publication. Conflicts of Interest The authors declare no conflicts of interest.