Adult RSV Vaccination: What Is the Role of RSV Subgroup in Disease Prevention?

Abstract

Respiratory syncytial virus (RSV) cocirculates as two antigenic subgroups, RSV A and RSV B, with irregular dominance by season. Despite highly conserved fusion (F) protein epitopes, sequence, and antigenic differences between A and B, especially within prefusion F, can modulate neutralization and susceptibility. First-in-class adult RSV vaccines licensed in 2023 adopt two antigen strategies: monovalent stabilized prefusion F based on sequences from the RSV A subgroup, and bivalent stabilized prefusion F antigens from both RSV A and RSV B subgroups to decrease reliance on RSV A cross-protection. Clinical data from bivalent vaccine studies support high efficacy against disease from both subgroups, while monovalent vaccine studies suggest that RSV B cross-protection may wane more quickly. Emerging and continued real-world vaccine effectiveness monitoring of both monovalent and bivalent RSV vaccines will be needed to assess this early signal.

Introduction

Respiratory syncytial virus (RSV) is a leading cause of lower respiratory tract disease (LRTD) in infants and older adults [1–3]. Two antigenic subgroups, RSV A and RSV B, have been recognized for decades and cocirculate globally with variable seasonal predominance [4]. Recent evidence supports that both subgroups can cause severe disease [5], and multiple vaccines have been developed to protect the highest risk populations [6–8]. Sequence diversity in the F and G surface glycoproteins between subgroups may influence neutralization by recently developed vaccines [9], and it is uncertain whether bivalent vaccine platforms offer broader protection across subgroups compared with monovalent options. Here, we provide an overview of the key aspects of RSV subgroups in relation to structure, epidemiology, and vaccine protection with currently available data.

Overview of RSV A and B Structure

RSV is an enveloped virus with several membrane proteins (Fig. 1) and is categorized into two cocirculating subgroups, A and B [10]. The F protein is responsible for viral fusion to the membrane of the host cell during virus entry and is crucial for infection [11]. Both F and G proteins exhibit major antigenic capacity for vaccine protection [12]. Because the F protein mediates viral fusion, antibodies to stabilized prefusion F (preF) are the most potent neutralizers [11]. While many neutralizing epitopes are conserved across the subgroups, multiple analyses show sequence differences between A and B in key antigenic sites on the F protein (Fig. 2) [11, 13]. These sequence differences can affect antibody affinity, access, and receptor binding [14].

Fig. 1.

Enveloped RSV structure with several membrane proteins

Fig. 2.

Subgroup sequence differences cluster in prefusion specific sites, with many differences located in binding sites with highest neutralizing potency

Of the two RSV proteins with the highest ability to trigger an immune response, the G protein, which is mucin-like and highly glycosylated, shows about ten times more antigenic variation than the F protein. The immune system places strong pressure on RSV, causing ongoing genetic and antigenic changes. The high variability of the G protein points to positive selection, helping the virus escape host immunity and highlighting the G protein’s role in evading the immune system [14]. The G protein is abundant on the surface of RSV and is highly variable between A and B subgroups. By contrast, the F protein is much more stable, showing only about 10% difference between the A and B subgroups, although genetic and antigenic changes, especially in RSV B viruses, are still observed. As a result, continued genetic drift and growing differences between A and B must be kept in mind when adapting future preventive measures. Including both A and B variants of the F protein in vaccines helps protect against subgroup-specific changes, preventing decreased antibody effectiveness due to other shifts in the virus structure [15].

Clinical, Epidemiological, and Seasonal Variation of RSV Subgroups

Contemporary reviews have concluded that there is varying predominance of RSV A or B when evaluating global RSV circulation patterns year over year [16]. The predominant RSV subgroup may also vary geographically during the same season, or alternate widely from season to season within one region. As an example, during the 2021–2022 season, RSV B was the predominant subgroup in the USA, while RSV A was predominant in Europe. The following season (2022–2023), the pattern flipped, with RSV A predominance in the USA and RSV B in Europe [17, 18].

A recent review by Nuttens et al. [5] evaluated the structural differences between RSV subgroups, their epidemiology, and genomic diversity. Among studies from 1985 to 2023, 2014 marked a distinct shift in overall subgroup predominance. Although predominant subgroup still varied annually, RSV A was the most identified subgroup prior to 2014, with predominance documented in 76% of seasons. After 2014, RSV A was predominant in only 40% of the seasons. Vaccine performance must be resilient to such epidemiologic dynamics, underscoring the need for comparable vaccine effectiveness for both subgroups.

Nuttens et al. also highlighted a decrease in RSV genotype diversity over the most recent two decades (2000–2020). As of 2020, the Ontario (ON) and Buenos Aires (BA) subgroups were the sole predominant lineages seen globally.

There are limited data describing clinical differences between subgroups, including disease severity, and most data are derived from pediatric populations. Studies in the USA and Finland have shown that bronchiolitis occurred at lower rates in children with RSV B compared with RSV A [19–21], but these findings were not replicated in other populations [5]. Data from Taiwan between September 2018 to August 2023 [22] found that patients with subgroup A disease had significantly more fever (91% versus 75%, p = 0.005) and longer fever duration (median 2 versus 1 day, p = 0.03) than patients with subgroup B disease, while patients with subgroup B disease had higher rates of underlying medical conditions (44% versus 29%, p = 0.047). Other clinical severity indicators and outcomes (e.g., dyspnea, intensive care unit admission rates, intubation, and mortality) did not differ significantly between the two subgroups. Currently available evidence does not support a severity difference by subgroup after 2014, despite previous assertions that, clinically, RSV A is associated with more severe disease [23].

Rationale of RSVPreF Vaccine Development

The RSVPreF vaccine was developed as a bivalent vaccine to increase direct antigenic immunity against both RSV A and B subgroups rather than depend on RSV A cross-protection against RSV B. Cross-protection refers to the phenomenon whereby a vaccine designed for one strain or type of pathogen provides partial immunity against related but different strains or types [24]. For some vaccines (e.g., influenza), cross-protection is short lasting because the immune response generated may not fully recognize or neutralize divergent strains, especially if they frequently mutate or have different surface proteins [25]. Over time, immune memory may wane and the effectiveness against these related strains may diminish [26].

All available RSV vaccines target the F protein in the prefusion state. Although the F protein contains multiple prefusion and postfusion antigenic sites, only prefusion F can bind host cells to allow RSV infection. Antibodies specific to the prefusion form are most effective at blocking infection. Antibodies that bind to sites ø and V, found only on prefusion F, have significantly higher neutralizing capacity than antibodies that bind to sites I, II, and IV (Fig. 2). Further, many of the differences between RSV A and RSV B subgroups are located within the prefusion ø antigenic site.

Other available vaccines (e.g., RSVPreF3-AS01E and mRNA-1345) [7, 27] including in clinical development [28] utilize only the RSV A prefusion antigen because of the homology in antigenic structure between the two subgroups. Early immunogenicity data across vaccines demonstrated equally robust RSV A and RSV B neutralizing responses with the bivalent RSVpreF, whereas monovalent preF vaccines showed a more robust neutralizing response against RSV A than RSV B [29–31]. Clinical studies of such vaccines have shown strong cross-protection against RSV A and B in the initial season of use, but emerging data suggest that subgroup-specific RSV B efficacy may wane more quickly than for RSV A with monovalent vaccines (Fig. 3), although differences in the involved seasons of surveillance, study design, and case definitions limit the strength of this association.

Fig. 3.

Efficacy of RSV vaccines for LRTD by season and RSV subgroup

Limited data are available on the effect of revaccination by RSV subgroup. To date, bivalent RSVpreF and adjuvanted RSVpreF3 have not seen immunogenicity titers return to levels seen following the primary dose after revaccination 1–3 years after the initial dose [32, 33], presumably owing to residual titers from the first dose because these protein subunit vaccines have a longer duration of protection. RSV A and B titers were comparable. For monovalent mRNA-1345 with shorter duration of protection, revaccination 12 months after the primary dose found RSV-subgroup-specific antibody responses similar to results after the primary dose at 29 days following the revaccination dose, with comparable titers for the RSV A and B subtype [34].

RSVPreF Phase 1–3 Clinical Data

Pre-licensure studies of the RSVPreF vaccine were conducted to evaluate dose, the impact of adjuvant inclusion, and safety [33, 35]. Overall, the studies found that the RSVPreF vaccine was well tolerated and highly immunogenic in adults and geometric mean titers remained elevated for 12 months after vaccination. The results of functional antibody assays were robust for both RSV A and B subgroups, and inclusion of an aluminum hydroxide adjuvant did not boost these immune responses further. Inclusion of various adjuvants had no impact on immune response, thus the unadjuvanted 120 μg bivalent dose was selected to move forward in testing. A human challenge study was then conducted, which found that the RSVpreF vaccine was effective at providing protection against the RSV A challenge virus (VE 86.7% for symptomatic RSV infection) [36]. The vaccine, compared with placebo, reliably prevented all reverse-transcription quantitative polymerase chain reaction (RT-qPCR)-confirmed infections, whether or not symptoms appeared. It also both lowered the highest viral loads and shortened the duration of viral shedding. After RSV A exposure, 73% of those who received the placebo experienced at least a fourfold increase in antibodies against RSV antigens within 12 days, while none of the RSVpreF vaccine recipients showed such an increase. This further demonstrated the vaccine’s effectiveness in protecting against RSV infection.

The pivotal phase 3 safety and efficacy clinical trial (RENOIR) was completed in 2023 and enrolled 36,967 participants (aged 60 years and older, healthy or with stable comorbid conditions) randomized 1:1 to receipt of 120 μg RSVPreF vaccine dose versus placebo [6]. Participants were followed over two full RSV seasons (mean follow-up time 16.4 months) [37, 38]. The study supported that the vaccine was highly efficacious against lower respiratory tract illness (LRTI) in seasons 1 and 2, and overall across the two seasons. Importantly, the efficacy remained high against both RSV subgroups, with modest waning by season or overall (Fig. 3). Corresponding to the efficacy results, high RSV neutralizing titers were observed 1 month after RSVpreF vaccination in persons aged ≥ 60 years, with similarly robust responses across subgroups.

Real-World Vaccine Effectiveness Data for RSV Vaccines by Subgroup

Recent observations from real-world studies suggest potentially greater waning of RSV B subgroup effectiveness with monovalent vaccines versus bivalent RSVpreF (Table 1). A recent publication from the US Centers for Disease Control evaluated both monovalent and bivalent RSV vaccine effectiveness (VE) against RSV-associated acute respiratory illness (ARI) hospitalization among adults aged 60 years or older during two RSV seasons, i.e., the Investigating Respiratory Viruses in the Acutely Ill (IVY) study [39]. This was a test-negative case–control study of US adults hospitalized with ARI. Among the RSV-positive cases, most patients received bivalent RSVPreF (39.7%) or monovalent RSVPreF3 (42.9%). The study also reported RSV subgroup of cases, with 42.9% identified as RSV B, 36.4% as RSV A, and 15.9% unknown. When evaluating the effectiveness of both vaccines against the outcome of RSV-related ARI hospitalization, overall vaccine effectiveness appears similar (monovalent 64% (95% confidence interval (CI): 47–76); bivalent 61% (95% CI: 41–74)). When evaluating effectiveness by subgroup, the bivalent vaccine appears to provide greater protection than the monovalent vaccine for RSV B cases (81% (95% CI: 50–93) and 65% (95% CI: 34–82), respectively), although the confidence intervals were overlapping and not all of the events could be subtyped. Effectiveness against RSV A cases was similar for both vaccines (monovalent: 59% (95% CI: 25–77); bivalent: 57% (95% CI: 19–77)). RSV A and B VE by time since vaccination was not reported. Another real-world vaccine effectiveness study of bivalent RSVPreF vaccination reported vaccine effectiveness over time from Kaiser Permanente Southern California (KPSC), indicating strong vaccine effectiveness against both RSV A- and B-related lower respiratory tract disease (LRTD) hospitalization or emergency department visits during the first (RSV A: 78 (59, 88); RSV B: 78 (51, 90) and second (RSV A: 75 (39,89); RSV B: 77 (-1, 95)) RSV seasons following vaccination in the USA (2023–2025) (Table 2) [40]. Differences in vaccine effectiveness between the two studies could be due to differences in severity of comorbidities or immunocompromising conditions between the two populations and/or differences in quality of the vaccine exposure data (i.e., KPSC included comprehensive electronic health record vaccination documentation supplemented by linkage with its state vaccine registry, whereas the IVY study relied on various immunization record sources including patient self-report). Additional published data evaluating the vaccine effectiveness of both monovalent and bivalent vaccines against RSV-associated disease by subgroup are expected, which will allow further evaluation of potential differences in vaccine effectiveness by subgroup over time for monovalent and bivalent vaccines.

Table 1.

RSV vaccine effectiveness estimated in two separate studies using real-world data across two seasons, overall, and by RSV subgroup

Vaccine/study/outcome	Overall VE %VE (95% CI)	RSV B VE %VE (95% CI)	RSV A VE %VE (95% CI)	Relative difference RSV B versus RSV A (%)
IVY (RSV-related ARI hospitalizations)
Bivalent RSVPreF [39]	61 (41,74)	81 (50,93)	57 (19,77)	42
Adjuvanted RSVPreF3 [39]	64 (47,76)	65 (34,82)	59 (25,77)	10
KPSC (RSV-related LRTD hospitalizations/ED visits)
Bivalent RSVPreF [40]	80 (69,87)	79 (57,90)	79 (65,87)	0
KPSC (RSV-related ARI hospitalizations/ED visits)
Bivalent RSVPreF [40]	80 (70,86)	82 (64,91)	78 (66,86)	5

Overall VE was estimated using all RSV cases, while subgroup-specific VE estimates used only those RSV cases with known subgroup: 79% (651/821) cases in the IVY study and 100% (775/775) cases in the KPSC study

Relative difference between RSV A VE and RSV B VE was calculated as (RSV A VE – RSV B VE)/(RSV A VE)

RSV, respiratory syncytial virus; LRTD, lower respiratory tract disease; ED, emergency department; VE, vaccine efficacy; KPSC, Kaiser Permanente Southern California; ARI, acute respiratory illness; IVY, Investigating Respiratory Viruses in the Acutely Ill

Table 2.

Bivalent RSVpreF vaccine effectiveness by RSV subgroup, measured using real-world data across two seasons—Kaiser Permanente Southern California, 2023–2025

Vaccine/study/outcome	Season 1			Season 2			Relative change in point estimate across two seasons
	Overall %VE (95% CI)	RSV A %VE (95% CI)	RSV B %VE (95% CI)	Overall %VE (95% CI)	RSV A %VE (95% CI)	RSV B %VE (95% CI)	Δ VE RSV A (%)	Δ VE RSV B (%)
KPSC (RSV-related LRTD hospitalizations/ED visits)
Bivalent RSVPreF [40]	81 (68,88)	78 (59, 88)	78 (51,90)	75 (47,89)	75 (39,89)	77 (-1, 95)	↓ 4	↓ 1
KPSC (RSV-related ARI hospitalizations/ED visits)
Bivalent RSVPreF [40]	83 (72–89)	83 (70–90)	81 (61–91)	69 (43–84)	65 (31–82)	81 (20, 95)	↓ 22	no Δ

Relative change in point estimate calculation using RSV A as an example: (season 1 RSV A VE − season 2 RSV A VE)/(season 1 RSV A VE)

RSV, respiratory syncytial virus; LRTD, lower respiratory tract disease; ED, emergency department; VE, vaccine efficacy; KPSC, Kaiser Permanente Southern California; ARI, acute respiratory illness

Summary of Support for Bivalent Formulation

For bivalent RSVPreF versus monovalent RSV vaccine designs, vaccine efficacy for season 1 suggests similar subgroup A and B efficacy (Fig. 3). However, longer observation periods and independent publications now point to lower point estimates against RSV B over time for monovalent designs, with statistically significant gaps seen in some mRNA-based data [7, 41]. The strength of these conclusions is limited by overlapping confidence intervals in some cases. It is important to note that each of the vaccines have different definitions for the primary outcome (lower respiratory tract illness or lower respiratory tract disease) and different follow-up time points, and the results are not directly comparable [42]. For RSVPreF and mRNA-1345, data presented are for a primary outcome of LRTI/LRTD with three or more symptoms because this endpoint corresponds to the most true lower tract illness and is most comparable to the LRTD definition used for the RSVPreF3 vaccine [43]. To our knowledge, none of the current trials were powered to evaluate RSV subgroup-specific efficacy.

These findings are biologically plausible given subgroup sequence differences in neutralization sites (Fig. 2) [44] and fit the broader pattern in vaccinology that cross-protection outside the vaccine’s matched antigen can be less robust and shorter lived. By including RSVpreF from both subgroups, bivalent RSVPreF was designed to avoid reliance on cross-subgroup immunity and targets the inherent differences between the subgroups, providing balanced neutralizing responses against both RSV A and RSV B [5]. While head-to-head studies would be required to confirm this, the practical implication is that bivalency may reduce the risk of subgroup-specific drops in protection, particularly during RSV B-dominant periods or in later seasons post-vaccination. This understanding supports the need for broad, durable coverage across subgroups as a practical goal of vaccination.

Conclusions

The recent development and approval of several monovalent and bivalent RSV vaccines offer an opportunity to prevent substantial morbidity and mortality in adults. Current data show that all available vaccines are effective against LRTI/LRTD, but the platform and/or RSVPreF source varies by vaccine. Both subgroups, RSV A and RSV B, can cause severe disease, and the predominant RSV subgroup varies year over year. RSV A-based vaccines rely on cross-protection for RSV B coverage. Evidence from efficacy studies and the first two seasons of real-world effectiveness data suggest that the bivalent RSVPreF vaccine maintains protection against both subgroups. There is early evidence that monovalent formulations may risk having reduced cross-protection over time. Additional studies and follow-up are needed to assess the consistency of the observed improved RSV B efficacy over time with higher-valency RSV vaccine.

Author Contributions

All authors (S. Elizabeth Williams, Giovanni Checcucci Lisi, Kate Luisi, Kumar Ilangovan, Cassandra Hall-Murray, Kyla Hayford, and Elizabeth Begier) contributed to the study conception and design. The first draft of the manuscript was written by S. Elizabeth Williams, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

This report was sponsored by Pfizer, Inc. The journal’s Rapid Service Fee is funded by Pfizer, Inc.

Data Availability

Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.

Declarations

Conflicts of Interest

All authors (S. Elizabeth Williams, Giovanni Checcucci Lisi, Kate Luisi, Kumar Ilangovan, Cassandra Hall-Murray, Kyla Hayford, Elizabeth Begier) are employees of Pfizer, Inc, and own Pfizer stock.

Ethics/Ethical Approval

This article is based on previously conducted studies and does not contain any new studies with human participants or animals performed by any of the authors.