The two-dose MVA-BN mpox vaccine induces a nondurable and low avidity MPXV-specific antibody response

Abstract

The 2022 global outbreak of clade IIb mpox was the first major outbreak of mpox outside of African nations. To control the outbreak, public health officials began vaccination campaigns using the third-generation orthopoxvirus vaccine modified vaccinia Ankara from Bavarian Nordic (MVA-BN). Prior to this outbreak, the durability of monkeypox virus (MPXV)-specific immunity induced by MVA-BN was poorly understood. In 2022, we launched the New York City Observational Study of Mpox Immunity (NYC OSMI, NCT05654883), a longitudinal study of 171 participants comprising MVA-BN vaccines and mpox convalescent individuals. Peripheral blood sampling was performed at intervals including prior to vaccination, after one dose, and after the second dose. MVA-BN vaccinees with and without a history of smallpox vaccination demonstrated detectable MPXV-specific memory B cells at 1-year post-vaccination. Additionally, MVA-BN increased MPXV neutralizing titers in smallpox vaccine-naïve vaccinees, with a comparable maximum titer reached in naïve and smallpox vaccine-experienced vaccinees. However, neutralizing titers returned to baseline within 5-7 months for naïve individuals, while remaining elevated in those with prior smallpox vaccination. Both naïve and experienced individuals generated robust IgG responses against MPXV H3 and A35, but naïve vaccinees' IgG responses showed lower avidity than experienced vaccinees. These data highlight a low avidity antibody response elicited by MVA-BN that is short-lived in naïve vaccinees. This work supports the need for long-term studies on protection induced by MVA-BN, including the potential need for booster doses as well as the development of next-generation orthopoxvirus vaccines.

Importance: The ongoing outbreaks of mpox demonstrate the continuing threat of orthopoxviruses to global health. While previous orthopoxvirus vaccines generated lifelong antibody and cellular immunity, we show here that the current mpox vaccine, MVA-BN or JYNNEOS, fails to induce durable antibody immunity in individuals with no prior smallpox vaccination. This raises the important question of whether MVA-BN vaccinees have long-term protection from mpox. Our work highlights the need for further studies into the durability of protection generated by MVA-BN as well as whether subsequent booster doses are necessary to maintain protection.

Keywords: JYNNEOS; MPXV; MVA-BN; antibody response; mpox; orthopoxviruses; vaccination.

Fig 1

MPXV H3- and A35-specific memory B cells are detectable at 1 year after MVA-BN vaccination. IgG+ memory B cells (MBCs) were quantified via ELISpot. (A) Diagram of study visits. For each visit, the median days post-vaccination are shown along with the range of days used for each visit in parentheses. The windows of time denoted for each visit are used throughout the paper in analyses of specific visits. Diagram generated in BioRender. (B) Aggregate MPXV H3-specific MBCs as a proportion of total IgG+ MBCs. (C) Aggregate MPXV A35-specific MBCs as a proportion of total IgG+ MBCs. (D) Representative well images from ELISpot assay. Convalescent (conv) samples were taken at ~1 year post-mpox symptom onset. In panels B and C, colored/black bars indicate the mean. Black, dashed line in all panels is the positivity threshold for binding as determined by 12 pre-2022 negative controls (mean of controls plus two times the standard deviation). Statistical testing conducted by Kruskal-Wallis test with Dunn’s method for multiple comparisons for panels B and C. Samples with a value of 0 have been displayed as 0.001% to accommodate the log scale; statistics were calculated with actual value. SC, subcutaneous; ID, intradermal; Exp, experienced vaccinees (prior history of smallpox vaccination); KLH, keyhole limpet hemocyanin (negative control).

Fig 2

MPXV neutralizing titers are lower in naïve vaccinees compared to experienced individuals following MVA-BN vaccination, with no significant effect from the route of administration on titer. (A) Aggregate MPXV neutralizing titer data for each study visit sorted by prior smallpox vaccination status. Colored/black bars on each group indicate the geometric mean titer. Serum-neutralizing titers were measured by a fluorescence-based microneutralization assay using authentic clade IIb MPXV. Pre-2022 controls come from the NYU Langone Vaccine Center Biorepository. Convalescent (conv) participants are sorted by whether they received MVA-BN following MPXV infection; all samples are taken at ~1 year post-symptom onset. (B) Stratification of MPXV neutralization at V3 by route of administration in naïve vaccinees. (C) Stratification of MPXV neutralization at V5 by route of administration in naïve vaccinees. Horizontal, black dashed line in all panels indicates the positivity threshold for neutralization (ID₅₀ = 33.7) as based on the pre-2022 negative controls (same method as Fig. 1). Statistical testing conducted by Kruskal-Wallis test with Dunn’s method for multiple comparisons for all panels.

Fig 3

MPXV neutralizing titers begin to wane after 3 weeks post-second vaccination (V3) in naïve individuals but only after 3 months post-second vaccination (V4) in experienced vaccinees. (A) Longitudinal neutralizing titers from naïve participants across all five study visits. The bolded line indicates the mean neutralizing titer for participants at each time point. (B) Longitudinal neutralizing titers from experienced participants across all five study visits. The bolded line indicates the mean neutralizing titer for participants at each time point. Horizontal, black dashed line in all panels indicates the positivity threshold for neutralization (ID₅₀ = 33.7) as based on the pre-2022 negative controls (same method as Fig. 1). Statistical testing conducted by Wilcoxon matched-pairs signed rank test. N below each panel indicates the number of visit-to-visit comparisons for a given pair of visits.

Fig 4

Modeling of neutralizing titer durability shows naïve individuals revert to negative titers by ~6 months, while experienced individuals remain positive beyond a year after MVA-BN vaccination. (A) Linear mixed-effects regression of naïve participants using all data points after the second dose of MVA-BN. Regressions of natural log-transformed neutralizing titers use days post-dose two and dosing interval as fixed effects and account for random subject effects by allowing for random intercepts. Colored regression lines with transparent ribbons indicate the predicted titers with a 95% confidence interval. Model parameters are listed in Table S7; 207 observations from 100 participants. (B) Linear mixed-effects regression of experienced participants using all data points after the second dose of MVA-BN. Regression of natural log-transformed neutralizing titers uses days post-dose two as a fixed effect and accounts for random subject effects by allowing for random intercepts. The solid black regression line with transparent ribbon indicates the predicted titers with 95% confidence interval. Model parameters are listed in Table S8; 106 observations from 48 participants. Horizontal, black dashed line in each panel indicates the positivity threshold for neutralization (Fig. 2).

Fig 5

Anti-MPXV H3 and -A35 IgG reach comparable peak titers in naïve and experienced participants, but wane more quickly in naïve participants following MVA-BN vaccination. (A) Aggregate anti-MPXV H3 IgG titers as measured by multiplexed immunoassay. Titers are reported as AUC from two serum dilutions. (B) Aggregate anti-MPXV A35 IgG titers as measured by multiplexed immunoassay. Titers are reported as AUC from two serum dilutions. (C) Longitudinal anti-MPXV H3 IgG titers for naïve participants at V1 and V3–5 as measured by multiplexed immunoassay. Titers are reported as AUC from two serum dilutions. (D) Same as C, but for experienced participants across V3–5. (E) Longitudinal anti-MPXV A35 IgG titers for naïve participants at V1 and V3–5 as measured by multiplexed immunoassay. Titers are reported as AUC from two serum dilutions. (F) Same as E, but for experienced participants across V3–5. MPXV convalescent (conv) samples were all taken at ~1 year post-symptom onset. In panels A and B, colored/black bars indicate the geometric mean. For panels C–F, the bolded line indicates the mean IgG titer for participants at each time point. Black, dashed line in all panels is the positivity threshold for binding as determined by pre-2022 negative controls (same samples and method as in Fig. 2). Statistical testing conducted by Kruskal-Wallis test with Dunn’s method for multiple comparisons for panels A and B, and by Wilcoxon matched-pairs signed rank test for panels C–F.

Fig 6

Naïve vaccinees generate low avidity IgG against MPXV H3 and A35 after vaccination with MVA-BN. (A) Aggregate avidity indices for anti-MPXV H3 IgG. (B) Aggregate avidity indices for anti-MPXV A35 IgG. (C) Longitudinal avidity indices for anti-MPXV H3 IgG antibodies for V3–5 of naïve participants. (D) Same as C, but for experienced participants. (E) Longitudinal avidity indices for anti-MPXV A35 IgG antibodies for V3–5 of naïve participants. (F) Same as E, but for experienced participants. Avidity was measured in a multiplexed immunoassay using a 2 M ammonium thiocyanate (NH₄SCN) wash and expressed as the ratio of the AUC from the NH₄SCN condition over the AUC from a PBS control condition. Convalescent (conv) samples were all taken at ~1 year post-mpox symptom onset. In panels A and B, colored/black bars indicate the mean. Statistical testing conducted by Kruskal-Wallis test with Dunn’s method for multiple comparisons for panels A and B, and by Wilcoxon matched-pairs signed rank test for panels C–F. Exp, experienced vaccinees (prior history of smallpox vaccination).