Recovery of Antibody Immunity After a Resurgence of Respiratory Syncytial Virus Infections

Abstract

Longitudinal measurements of respiratory syncytial virus (RSV) immunity over 4 winter seasons reveal that viral neutralization titers, RSV prefusion F protein (pre-F)–specific immunoglobulin M and immunoglobulin G (IgG) levels, and RSV antibody–dependent cellular phagocytosis function gradually returned to prepandemic levels in female healthcare and school workers of childbearing age after 2 winter seasons, following the resurgence of RSV cases in the Vancouver metropolitan region (British Columbia, Canada). In contrast, pre-F IgG avidity profiles remained unchanged. These findings support the notion that repeated viral infections are necessary to maintain high RSV antibody levels in the population.

Longitudinal antibody measurements in women of childbearing age show that respiratory syncytial virus (RSV) immunity returned to prepandemic levels following a resurgence of cases, supporting the conclusion that repeated viral infections are necessary to maintain high RSV antibody levels.

Respiratory syncytial virus (RSV) is the main cause of acute respiratory infection in young children [1]. The seasonal epidemiology of RSV has been well documented in British Columbia (BC), Canada's most western province, where cases occur typically between October and April [1]. This offers a natural setting to monitor the relationships between RSV cases and changes in immunity to the virus at the population level. With the implementation of coronavirus disease 2019 (COVID-19) pandemic measures, RSV cases nearly disappeared in BC between March 2020 and August 2021 [1]. Over the same period, we reported a significant waning of RSV antibody immunity in women of childbearing age and infants in Vancouver, BC's metropolitan area [2, 3]. As the pandemic measures were gradually relaxed, BC experienced 2 intense RSV seasons, with increased infections in older children between September 2021 and March 2022, and a dramatic increase in RSV hospitalizations during the following RSV season between October 2022 and February 2023 [1]. The reasons for these epidemiological perturbations are not completely understood but may have been related to changes in levels of immunity to RSV in the population, resulting in an increased pool of vulnerable children [1, 4].

Newborns are largely immunologically naive, except for maternal immunoglobulins transferred through the placenta in the last trimester of pregnancy. After birth, maternal immunity wanes rapidly, within months. Beyond this period, susceptibility to RSV gradually decreases as children develop immunity de novo following their first RSV exposure. By 2 years of age, most children have acquired active T-cell immunity against the virus, so their risk of severe RSV disease decreases significantly [5]. In 2023, an RSV vaccine was approved during pregnancy, to boost maternal antibodies transferred through the placenta to the fetus and decrease the severity of RSV infections in infants [6]. These data confirmed the importance of maternal antibody levels for optimal protection of young infants. We report temporal changes in RSV antibody immunity in women of childbearing age after a prolonged lack of viral circulation and resumption of RSV cases.

METHODS

Study Design, Setting, and Participants

RSV antibody measures (see RSV Antibody Measures section) were measured in 2 independent cohorts of women of childbearing age from the Vancouver metropolitan area in whom serial blood samples were available from the same individuals after each RSV season in 2020, 2021, 2022, and 2023. The characteristics and sample collection periods from May 2020 to September 2023 for the 2 cohorts are detailed in Supplementary Table 1. Healthcare workers were enrolled by email at BC Children's and Women's hospitals and affiliated research institute, as described elsewhere [3], with inclusion of samples from female participants 18–51 years old who had at least 2 serial samples. For the school worker cohort, samples were randomly selected from a larger cohort of females aged 18–51 years enrolled from the Vancouver metropolitan area in whom all serial annual blood samples were collected in 2021, 2022, and 2023 [7]. Written consent was obtained from all participants. Sample size was based on the availability of samples. The epidemiology of RSV cases in BC was obtained from the Public Health Agency of Canada's publicly available weekly reports [8], and data on RSV hospitalizations in infants under 6 months until May 2023 were obtained from a prior analysis [1], with data added herein to April 2024.

RSV Antibody Measures

All blood samples were collected in gold-top serum separator tubes with polymer gel (BD Biosciences), clotted at room temperature for 30 minutes, followed by centrifugation at 1400g for 10 minutes, aliquoting, and freezing (−80°C) within 4 hours.

Immunoglobulin M (IgM) and immunoglobulin G (IgG) levels against the RSV prefusion (pre-F) protein were quantified using the VPLEX Respiratory Panel 1 Kit (K15365U, for IgG and K15366U for IgM, from Meso Scale Diagnostics) at dilutions of 1:10 000, after blocking with an excess (2.0 µg/mL) of soluble postfusion protein (RSF-V52H6, ACROBiosystems), as this was shown to be a better correlate of neutralization compared to measuring pre-F IgG levels without blocking [9]. Data were reported as arbitrary units per milliliter (AU/mL). RSV microneutralization was measured in all samples using a live-virus in vitro plaque reduction assay utilizing a green fluorescent protein–expressing recombinant RSV-A subtype, as described previously [3]. Results were expressed as serum titers to prevent 95% viral cell infection (syncytial formation) (NT₉₅). Palivizumab (positive control) results showed interbatch consistency with titers between 1:64 and 1:128, and RSV back titration consistently remained between 3850 and 4800 plaque-forming units/mL. RSV F antibody–dependent cellular phagocytosis (ADCP) scores were determined from a random subset of samples from each cohorts, diluted 1:100, as described previously [2]. For RSV IgG antibody avidity specific to the RSV F protein, the relative fractions of “low,” “intermediate,” “high,” and “very high” avidity of IgG against RSV prefusion F were determined after blocking with soluble RSV postfusion F, using the VPLEX Respiratory Panel 1 IgG kit, at serum dilutions of 1:10 000, and using ammonium thiocyanate concentrations between 0 and 3.0 mol/L (M), as described previously [10–12]. Additional methodological details of the RSV assays, and of the validation of the corresponding data, are provided in the Supplementary Material.

Statistical Analysis

Antibody measures over time were plotted as time-specific log₁₀-transformed values, fit with locally estimated scatterplot smoothing with corresponding 95% confidence intervals (CIs). To analyze temporal differences statistically, a mixed-effects model (to account for repeated samples per participant) was fitted, including year of sampling and cohort as fixed effects, and a random intercept for participant ID, on log-transformed data. Results are presented for temporal changes in IgM, IgG, viral neutralization titers, and ADCP scores, as adjusted mean differences between years, with significance represented as corresponding 95% CIs. Spearman correlations were used to examine relationships between RSV immunity measures. Analyses were conducted in R version 4.03 software. RSV case counts were descriptively plotted by season.

Ethics Statement

The study was approved by the University of British Columbia Children's and Women's Research Ethics Board (certificate numbers H20-03593 and H20-01205).

RESULTS

As shown in Figure 1, the 2021 samples were collected after the RSV season at similar times in both the eligible healthcare (n = 10–18, depending on time period) and school worker (n = 125) cohorts after a prolonged period of lack of RSV circulation, whereas for 2022 and 2023, samples were collected after the RSV season peak in school workers and later after the season in the healthcare workers. Geometric means and standard deviation for pre-F IgM, IgG, neutralization titers, and ADCP scores—with number of samples used for each antibody measure—are shown in Supplementary Table 2. Estimates for changes between sampling periods with 95% CIs are provided in Supplementary Table 3. In both cohorts, RSV neutralization and ADCP scores (which measure Fc-mediated antibody functions) increased over 2 years following the atypical resumption of RSV infections in 2021–2022 and 2022–2023. RSV pre-F–specific IgM and IgG showed similar increasing trends across periods (Figure 2; Supplementary Table 3). More specifically, RSV-neutralizing titers (NT₉₅) increased in both cohorts from 2021 to 2023, right after RSV started circulating again in BC in October 2021. ADPC scores also increased in healthcare workers from 2021 to 2023. However, this increase was delayed until after the RSV season peak in 2023 in school workers (Figure 1). Correspondingly, weekly RSV hospitalizations in infants aged <6 months at BC Children's Hospital, the main pediatric hospital in BC, exceeded prepandemic levels in 2022–2023, as RSV antibody measures were gradually restored in the 2 cohorts (Figure 1). For the RSV prefusion F IgG avidity profiles, no differences were detected between periods, although a significant portion of IgGs displayed high or very high avidity antibody fractions (Supplementary Figure 1).

Figure 1.

Temporal trends in respiratory syncytial virus (RSV) cases in British Columbia (BC) and serum antibody measures in women of childbearing age, and RSV hospitalizations in young infants. A, Total number of weekly reported RSV cases in BC, Canada, between 1 October 2019 and 24 April 2024 obtained from the Public Health Agency of Canada (https://www.canada.ca/en/public-health/services/surveillance/respiratory-virus-detections-canada.html). B, Virus neutralization titers (expressed as serum dilution that blocked 95% of viral infection of cultured cells in vitro [NT₉₅], with a lower limit of detection of 8 [dotted line]). C, RSV F protein–specific antibody–dependent cellular phagocytosis (ADCP) scores, measured in healthcare workers (black circles) and school workers (open gray circles) with locally estimated scatterplot smoothing fit and 95% confidence intervals. D, RSV hospitalizations in children <6 months of age at BC Children's Hospital, the main pediatric hospital in BC that also services the Vancouver city catchment area, between 1 October 2019 and 24 April 2024.

Figure 2.

Temporal trends in respiratory syncytial virus (RSV) prefusion F (pre-F)–specific immunoglobulin M (IgM) and immunoglobulin G (IgG) levels. Pre-F–specific IgM (A) and IgG (B) levels were quantified after blocking sera with an excess of RSV postfusion F protein, in healthcare workers (black circles) and school workers (open gray circles) in 2020, 2021, 2022, and 2023 with locally estimated scatterplot smoothing fit and 95% confidence intervals.

DISCUSSION

This study showed a gradual recovery of RSV antibody immunity after a resumption of RSV circulation, in women of childbearing age. A strength of this study is that the RSV antibody measures were assessed using multiple assays, in longitudinally obtained blood samples from the same individuals in 2 independent cohorts of women residing in BC's Vancouver metropolitan area where the epidemiology of RSV has also been well characterized [1]. RSV epidemics follow a biennial pattern, in northern, temperate regions like Canada [13]. From this pattern, authors have estimated the half-life of population “immunity” against RSV to be between 6 and 12 months [13]. The waning kinetics of RSV antibody levels in absence of infections corroborated these estimates [3]. In the current study, the RSV antibody measures gradually recovered within 2 years after the resumption of cases. These data are consistent with recent epidemiological modeling showing that a period of 2 years was required for RSV epidemics to return to a prepandemic pattern in the Netherlands [14], and suggest that repeated outbreaks are required to maintain high RSV antibody levels in adults.

This study also suggests that measuring only serum RSV F protein IgG concentrations provides an incomplete picture of changes resulting from natural exposures to the virus in populations. The profound changes in RSV neutralization but marginal changes in pre-F (specific) IgG levels despite high correlation between these 2 measures likely reflects their nonlinear relationship and possibly, the saturated levels in adults compared to infants [3]. Alternatively, the discrepant trends may reflect epitope differences between these 2 measures. Furthermore, while the neutralization assay used RSV-A, other antibody measures, titers of IgG against the RSV F protein and ADCP, are much less affected by differences in RSV subtypes due to high sequence homology (>90%) between RSV-A and RSV-B for the F protein [15]. In RSV challenge experiments, virus neutralization assays correlated positively with symptomatic RSV infections, whereas Fc-mediated functions, such as ADCP, correlated more closely with viral titers and severe disease [4]. In the current study, both antibody measures showed significant recovery after RSV cases resurged. On the other hand, pre-F IgG levels marginally changed, and avidity profiles remained stable across years. This may not be surprising given that RSV is a relatively stable virus, which could limit the drive for antibody affinity maturation.

This study has limitations. First, the lack of measures in mother–infant dyads and the limited sample size preclude detailed quantitative modeling of the direct relationship between RSV immune measures and clinical outcomes in young children. Nonetheless, this study provided specific antibody measures for modeling in larger studies. Second, the increased RSV hospitalizations in infants aged <6 months suggested that the prolonged lack of RSV may have increased the pool of immunologically vulnerable infants [1]. However, the temporal changes in hospitalizations in infants are likely confounded by multiple other testing, behavioral and environmental factors. For example, in 2021–2022, RSV hospitalizations in BC abruptly dropped in January 2022 as Omicron emerged, and while public health authorities recommended strict masking again. This may explain why the first season after the resurgence was not as intense as the following season in 2022–2023. Third, it is important to point out that this study was not designed to establish correlates of protection; therefore, it remains unclear whether these measures can predict the severity of RSV epidemics.

In conclusion, this study showed that RSV neutralization and ADCP recovered gradually over 2 seasons after RSV cases resurged in BC. These data could inform the design of larger-scale studies, specifically identifying measures that could be useful to model the relationship between population immunity and how these measures may drive the intensity of RSV epidemics in children. Understanding these temporal relationships may help define correlates of protection to predict the timing and intensity of RSV epidemics, and guide the seasonal implementation of RSV immunizations.

Supplementary Data

Supplementary materials are available at The Journal of Infectious Diseases online (http://jid.oxfordjournals.org/). Supplementary materials consist of data provided by the author that are published to benefit the reader. The posted materials are not copyedited. The contents of all supplementary data are the sole responsibility of the authors. Questions or messages regarding errors should be addressed to the author.

Notes

Acknowledgments. RSV-A green fluorescent protein was provided by Mark Peeples from Nationwide Children's Hospital (Columbus, Ohio). We thank Lauren Muttucomaroe, Tisha Montgomery, and Bethany Poon for their assistance with recruiting healthcare workers; the leadership teams of the Richmond, Delta, and Vancouver school boards; and the educators who contributed samples for this study.

Author contributions. Conceptualization: F. R. and P. M. L. Methodology: F. R., S. L., A. L., M. V. P., L. G., B. A.-R., and A. W. Data analysis and visualization: F. R., J. B., M. V. P., L. G., A. J., B. A.-R., and P. M. L. Data interpretation: F. R., J. B., M. V. P., L. G., A. J., B. A.-R. and P. M. L. Writing–original draft: F. R. and P. M. L. Writing–review & editing and approval of the submitted manuscript: All authors.

Financial support. The study was funded by the Canadian Institutes of Health Research (grant number PJT-166103 to P. M. L.) and the government of Canada via its COVID-19 Immunity Task Force (to P. M. L.). F. R. was funded by the German Research Foundation (Deutsche Forschungsgemeinschaft, RE 4598/1-2). B. A.-R. received a postdoctoral award from Michael Smith Health Research. P. M. L. received grant salary support from the British Columbia Children's Hospital Foundation.