ABSTRACT
Background
Globally, RSV declined to historically low levels following stay‐at‐home orders in 2020 and re‐emerged as restrictions lifted, with large inter‐seasonal outbreaks and other shifts in epidemiology. The effect of the pandemic restrictions on genomic diversity has not been well studied.
Methods
Individuals with respiratory symptoms were enrolled into community‐based surveillance studies and provided nasal swabs over three respiratory virus seasons from 2019 to 2022 in the Seattle metropolitan area in Washington State, USA. Samples were tested for RSV by RT‐qPCR, with whole genome sequencing performed for a subset.
Results
Among children and adults, respectively, 1.8% (232/13,014) and 0.2% (86/46,042) of samples tested positive for RSV‐A and 1.1% (139/13016) and 0.3% (119/46043) for RSV‐B. Positive cases were observed in childcare (n = 39), clinic (n = 270), university (n = 20), workplace (n = 32) or nursing facility (n = 1) sites and through at‐home self‐administered testing (n = 253). Median age for all participants was 33 years (IQR: 23, 48) and 3 years (IQR: 3, 33) for those who tested positive. Cough (≥ 85%) and rhinorrhea (≥ 85%) were the most common symptoms for RSV cases. During 2019–2020, six RSV‐A and three RSV‐B lineages circulated (majority lineages a.d.1 and B.D.4.1.1). Almost no RSV was detected during 2020–2021. During 2021–2022, RSV‐B lineage B.D.E.1 was dominant.
Conclusions
In a community surveillance study spanning several years, RSV was commonly detected in young children across several sites. Circulation of RSV was disrupted by pandemic mitigation measures, with shifts in genomic diversity detected after lifting of pandemic restrictions.
Keywords: community surveillance, molecular epidemiology, respiratory syncytial virus, whole genome sequencing
1. Background
Respiratory syncytial virus (RSV) is an important cause of morbidity and mortality among young children, immunocompromised adults, and adults ≥ 60 years [1, 2]. In the United States, an estimated 1%–2% of infants are hospitalized with RSV in their first year of life, and nearly all children are infected with RSV by 2 years of age [3, 4]. Reinfections with RSV are common throughout childhood and into adulthood but decrease in severity; however, disease severity increases again among adults ≥ 60 years [4]. Recently, RSV prefusion F protein‐based vaccines targeting maternal and older adult populations and two infant population monoclonal antibody products were licensed and recommended for use in the United States [5, 6, 7]. Several other RSV vaccine and monoclonal antibody candidates targeting maternal, pediatric, and adult populations continue to be evaluated in clinical trials [8].
Following stay‐at‐home orders and non‐pharmaceutical interventions implemented in 2020 to reduce transmission of SARS‐CoV‐2, RSV declined to historically low levels. Expected seasonal peaks in disease did not occur globally during the 2020–2021 respiratory virus season [9]. As pandemic restrictions were lifted, RSV re‐emerged, causing larger than expected inter‐seasonal outbreaks in disease in 2021–2022. The pandemic may also have altered RSV genomic diversity. Few studies have assessed RSV genomic diversity from the pre‐ to post‐COVID‐19 pandemic periods, and no studies to our knowledge have assessed changes in RSV genomic diversity over this period among community‐based samples. In Australia, reductions in RSV genomic diversity were observed among clinical cases from the pre‐COVID‐19 pandemic period to March 2021, although it is unclear if these changes were sustained over time [10]. In Japan, outpatient data from 2019 to 2020 showed a decline in the number of RSV‐A clusters circulating, but no change in the diversity of RSV‐B [11]. In the United States, the 2022 off‐season surge in RSV has been shown to be due to several existing RSV lineages and not the emergence of a new, more transmissible RSV lineage among clinical cases in Massachusetts, Washington, and Chicago [12, 13, 14]. Only one of these US studies specifically analyzed RSV genomic diversity over time and found evidence of a decline in diversity for RSV‐B, but not RSV‐A, following the pandemic [12].
Analyses of RSV genomic data prior to the widespread introduction of vaccines and monoclonal antibodies, including describing impact of the COVID‐19 pandemic, are needed to establish baseline data for post‐licensure vaccine and monoclonal antibody impact studies. It remains unknown how selective pressure of vaccines and monoclonal antibodies, which target the F glycoprotein of RSV, will impact genomic diversity of RSV [15, 16]. In addition, a better understanding of RSV transmission patterns among individuals of all ages in community settings is needed, as most RSV genomic surveillance captures clinical cases.
Using publicly available RSV genomes, we described RSV genomic diversity prior to, during, and following implementation of COVID‐19 pandemic restrictions. We also used data from the Seattle Flu Study (SFS), a community‐based respiratory virus surveillance study, to describe genomic diversity in community settings over time and by age.
2. Methods
2.1. Study Design, Data Collection, and Laboratory Methods
We used data from three community‐based respiratory virus surveillance studies conducted as part of SFS from 2018 to 2022 (Figure 1). First, participants were cross‐sectionally enrolled at community testing kiosks located throughout the city, including childcare centers, a university campus, public spaces of clinical settings, skilled nursing facilities, workplaces, transit centers, and other public places (referred to here as “Kiosks”) from 2018 to 2020. For the second study, participants were cross‐sectionally enrolled at home/online (“Swab & Send”) from 2019 to 2020. In the last study, participants were cross‐sectionally enrolled through the Seattle Coronavirus Assessment Network (“SCAN”) from 2020 to 2022. Sequences and metadata from these three sources were combined and will be referred to as the SFS dataset.
FIGURE 1.
SFS community‐based respiratory virus surveillance study timelines and nasal swab collection methods.
The study design, data collection, and laboratory methods have been previously described [17]. In brief, respiratory specimens were collected from participants who reported two or more symptoms (fever, cough, sore through, headache, diarrhea, nausea or vomiting, runny/stuffy nose, rash, fatigue, myalgias, trouble breathing, and ear pain/discharge) of acute respiratory infection (ARI) or cough alone for Kiosks and Swab & Send and at enrollment regardless of symptoms for SCAN. Midturbinate swabs were collected by research staff, self‐collected and observed by study staff, or self‐collected at home (unobserved) and returned to the laboratory by mail. Demographic, household, and illness data were self‐reported by participants for all SFS studies, and data were collected and managed using Project REDCap [18, 19]. All SFS participants completed informed consent (or assent and parental consent for participants < 18 years of age) at enrollment.
Nasal swabs were tested using TaqMan RT‐qPCR on OpenArray (Thermo Fisher) for the presence of several respiratory pathogens, including RSV, at the University of Washington Northwest Genomics Center. Swab & Send and SCAN samples were mailed at ambient temperature in either universal transport media, viral transport media, or as dry swabs (depending on study year). Whole genome sequencing (WGS) was attempted on all RSV‐positive samples with relative cycle threshold (Crt) values < 20 either using a hybrid capture approach [20] or using molecular inversion probes as described previously, and sequenced on Illumina instruments [21, 22]. Consensus genomes were assembled using the Seattle Flu Study Assembly pipeline (https://github.com/seattleflu/assembly) or a modified iVar‐based bioinformatics pipeline [23]; template consensus genomes were created using a custom R script and finished consensus genomes were created using Bowtie. All genome sequences used in this study were submitted to GenBank (BioProjects PRJNA988152 and PRJNA988154).
To place SFS sequences in the context of global RSV genetic diversity, sequences were downloaded from NCBI Genbank—these included all publicly available full RSV genomes from 2005 to 2023 as of January 2024 (the global Genbank dataset). To place SFS sequences in the context of local RSV genetic diversity, the global Genbank dataset was subset by the Geo_location flag to only include samples from Washington State (the Washington State Genbank dataset). Of note, SFS sequences were not excluded from the Washington State Genbank dataset, and similarly Washington State sequences were not excluded from the global Genbank dataset. Therefore, the SFS dataset is a subset of the global and Washington State Genbank datasets. SFS‐generated genomes are identifiable within these datasets by accessions associated with BioProjects PRJNA988152 and PRJNA988154; all remaining Genbank sequences were contributed by other groups.
2.2. Genomic Analyses
Analyses were conducted separately for RSV‐A and RSV‐B and included both SFS genomes and publicly available full RSV genomes from GenBank. Genome quality was assessed, and RSV lineages were assigned using Nextclade [24]. Genomes with a “bad” quality rating from Nextclade (based on missing data, mixed sites, private mutations, mutation clusters, frameshifts, and premature codon stops), missing collection date, or missing > 10% of nucleotides were excluded.
Maximum likelihood phylogenetic trees for RSV‐A and RSV‐B were created using Nextstrain [25]. Sequences were aligned using multiple sequence alignment (MAFFT) relative to a reference genome (GenBank accession KU950499 for RSV‐A and KU950676 for RSV‐B) and trees included all quality‐filtered SFS genomes and a subsample of five publicly available GenBank genomes per country per calendar month from 2005 to 2023; this subsampling was applied only for phylogenetic tree reconstruction, whereas diversity and lineage‐frequency analyses used the full, quality‐filtered datasets. Time‐resolved phylogenies were estimated using TreeTime [26]. Trees were visualized using Auspice and the ggtree package in R [27]. Phylogenetic trees were used to assess the relationship between study genomes and publicly available genomes from around the world, placing SFS genomes in the context of contemporaneous global RSV diversity and supporting lineage assignments.
Analysis of genetic diversity was performed using the ape and pegas R packages [28, 29], which were used to calculate the nucleotide diversity metric, π, defined as the mean proportion of nucleotide differences per site across all pairs of sequences within a group to summarize the standing genetic variation in that group. We also computed root‐to‐tip divergence, the substitutions per site accumulated between each genome and the inferred root of the time‐resolved phylogeny from Nextstrain, to assess the rate at which diversity accumulated over the study period. Diversity metrics are reported descriptively; no statistical tests were performed to compare diversity between periods, subgroups, or age groups. All statistical analyses were conducted in R (version 4.1.1, [30]).
3. Results
From November 2018 to July 2022, 1.8% (232/13,014) and 0.2% (86/46,042) of SFS samples tested positive for RSV‐A and 1.1% (139/13,016) and 0.3% (119/46,043) for RSV‐B, among children (< 18 years of age) and adults (≥ 18 years of age), respectively (Figure 2); 48 (13 RSV‐A, 35 RSV‐B) had missing age information. RSV percent positivity was greater prior to the COVID‐19 pandemic (November 1, 2018, to March 22, 2020; 8.8% and 1.3% for RSV‐A and 2.8% and 0.8% for RSV‐B among children and adults, respectively) than following the Washington State stay‐at‐home order (March 23, 2020, to July 31, 2022; 0.1% and 0.02% for RSV‐A and 0.6% and 0.2% for RSV‐B among children and adults, respectively). Positive SFS samples came from childcare centers (n = 39), public clinical spaces (n = 270), university campuses (n = 20), workplaces (n = 32), or skilled nursing facility (n = 1) kiosk sites and at‐home self‐administered testing (n = 253). Median age for all SFS participants was 33 years (IQR: 23, 48, n = 59,059) and 3 (IQR: 3, 33, n = 567) for those who tested positive. Cough (≥ 85%) and rhinorrhea (≥ 85%) were the most common symptoms for RSV‐positive children and adults. Median relative cycle threshold values were lower among children (13.7, IQR: 10.5 19.6 for RSV‐A and 12.6, IQR: 9.5, 21.0 for RSV‐B) compared to adults (19.3, IQR: 14.5, 23.1 for RSV‐A and 18.2, IQR: 13.3, 23.1 for RSV‐B).
FIGURE 2.
RSV‐positive cases and RSV percent positivity among community SFS samples.
Of the 624 samples that were positive for RSV (331 for RSV‐A and 293 for RSV‐B), 186 RSV‐A (56.2%) and 157 RSV‐B (53.6%) met the Crt < 20 threshold and were submitted for whole genome sequencing. Overall, 279 (81.3%) yielded a genome deposited in GenBank and 273 passed quality filtering and were included in further analysis. Genome recovery among submitted samples was similar for the two subgroups and declined with increasing Crt, from 98% (41/42) and 96% (53/55) at Crt < 10 to 51% (28/55) and 47% (20/43) at Crt 15–20 for RSV‐A and RSV‐B, respectively. Overall yield, defined as the proportion of all RSV‐positive samples that ultimately produced a deposited genome, was similar across the three study arms: 44.1% (162/367) for Kiosks, 41.6% (42/101) for Swab & Send, and 48.1% (75/156) for SCAN (Figure S1). Median Crt was higher among Swab & Send samples, which were self‐collected at home and returned by mail (18.2, IQR 14.0–21.7), than among staff‐collected Kiosks samples (14.7, IQR 10.6–21.2) or SCAN samples (15.1, IQR 9.8–23.4). Within subgroups, yield varied by study arm (RSV‐A: 50.6% Kiosks, 36.9% Swab & Send, 17.6% SCAN; RSV‐B: 30.5% Kiosks, 50.0% Swab & Send, 51.8% SCAN), reflecting differences in the Crt distribution of positive samples by subgroup and collection method rather than differences in sequencing performance (Table S1).
Genomic analyses included 149 RSV‐A and 124 RSV‐B whole genome sequences from SFS (Table 1), in addition to 1674 RSV‐A and 1220 RSV‐B whole genome sequences from GenBank (106 and 118 excluded due to “bad” quality rating or missing > 10% of nucleotides, and 531 and 48 excluded due to missing collection date for RSV‐A and RSV‐B, respectively). An additional 1173 RSV‐A and 653 RSV‐B sequences were excluded from maximum likelihood phylogenetic trees during sampling by year, month, and country, resulting in final trees containing 1431 RSV‐A and 1495 RSV‐B sequences respectively.
TABLE 1.
Description of RSV‐positive sequenced SFS samples during the pre‐ and post‐COVID‐19 periods.
| Pre‐COVID‐19 restrictions a (N = 194) | Post‐COVID‐19 restrictions b (N = 79) | Total (N = 273) | |
|---|---|---|---|
| n (%) | |||
| RSV subgroup and lineage | |||
| RSV‐A | 146 (75%) | 3 (4%) | 149 (55%) |
| A.D | 8 (4%) | 0 (0%) | 8 (3%) |
| a.d.1 | 109 (56%) | 1 (1%) | 110 (40%) |
| A.D.2.1 | 4 (2%) | 0 (0%) | 4 (1%) |
| A.D.2.2 | 4 (2%) | 0 (0%) | 4 (1%) |
| A.D.3 | 20 (10%) | 1 (1%) | 21 (8%) |
| A.D.5 | 1 (1%) | 0 (0%) | 1 (0%) |
| A.D.5.2 | 0 (0%) | 1 (1%) | 1 (0%) |
| RSV‐B | 48 (25%) | 76 (96%) | 124 (45%) |
| B.D.4.1 | 1 (1%) | 0 (0%) | 1 (0%) |
| B.D.4.1.1 | 44 (23%) | 8 (10%) | 52 (19%) |
| B.D.E.1 | 0 (0%) | 68 (86%) | 68 (25%) |
| B.D.E.2 | 3 (2%) | 0 (0%) | 3 (1%) |
| Collection location | |||
| Childcare center | 1 (1%) | 0 (0%) | 1 (0%) |
| College campus | 4 (2%) | 0 (0%) | 4 (1%) |
| Home collection returned by mail | 42 (22%) | 74 (94%) | 116 (42%) |
| Public space of a clinical building | 146 (75%) | 0 (0%) | 146 (53%) |
| Workplace | 1 (1%) | 5 (6%) | 6 (2%) |
| Age | |||
| < 18 years | 142 (73%) | 45 (57%) | 187 (68%) |
| ≥ 18 years | 39 (20%) | 29 (37%) | 68 (25%) |
| Unknown | 13 (7%) | 5 (6%) | 18 (7%) |
| Sex, N (%) | |||
| Female | 92 (47%) | 42 (53%) | 134 (49%) |
| Male | 89 (46%) | 32 (41%) | 121 (44%) |
| Unknown | 13 (7%) | 5 (6%) | 18 (7%) |
| Symptom status c | |||
| Asymptomatic | 0 (0%) | 2 (3%) | 2 (1%) |
| Symptomatic | 180 (93%) | 72 (91%) | 252 (92%) |
| Number of symptoms, median (IQR) | 6 (4, 8) | 5 (3, 7) | 6 (4, 8) |
| Cough | 177 (98%) | 67 (93%) | 244 (97%) |
| Runny or stuffy nose | 161 (89%) | 67 (93%) | 228 (90%) |
| Fatigue | 129 (72%) | 36 (50%) | 165 (65%) |
| Feeling feverish | 132 (73%) | 33 (46%) | 165 (65%) |
| Increased trouble breathing | 116 (64%) | 10 (14%) | 126 (50%) |
| Sore throat | 79 (44%) | 43 (60%) | 122 (48%) |
| Headache | 52 (29%) | 24 (33%) | 76 (30%) |
| Nausea or vomiting | 54 (30%) | 15 (21%) | 69 (27%) |
| Muscle or body aches | 52 (29%) | 13 (18%) | 65 (26%) |
| Chills or shivering | 47 (26%) | 16 (22%) | 63 (25%) |
| Sweats | 48 (27%) | 15 (21%) | 63 (25%) |
| Ear pain or discharge | 39 (22%) | 9 (12%) | 48 (19%) |
| Diarrhea | 22 (12%) | 11 (15%) | 33 (13%) |
| Rash | 12 (7%) | 5 (7%) | 17 (7%) |
| Unknown | 14 (7%) | 5 (6%) | 19 (7%) |
Prior to March 23, 2020, when the Washington State stay‐at‐home order went into effect.
Post‐March 23, 2020.
Number of symptoms and specific symptom types are reported among symptomatic cases only.
During the 2019–2020 respiratory virus season in Washington State, RSV‐A and RSV‐B co‐circulated with six RSV‐A and three RSV‐B lineages (predominantly a.d.1 and B.D.4.1.1) in Washington State. This analysis used all Washington State GenBank sequences inclusive of the 149 RSV‐A and 124 RSV‐B QC‐filtered SFS sequences (Table 2, Figures 3 and 4). Overall, SFS sequences accounted for 43.8% (194/443) of Washington State RSV‐A and RSV‐B sequences in GenBank before COVID‐19 restrictions (146/319 RSV‐A, 48/124 RSV‐B) and 52.3% (79/151) after restrictions were imposed (3/6 RSV‐A, 76/145 RSV‐B); the remainder were submitted by other Washington State laboratories and lack associated demographic or clinical metadata (Table 2). Almost no RSV was detected in our study cohorts during the 2020–2021 respiratory virus season. In the first season after COVID‐19 restrictions were lifted (2021–2022), RSV‐B lineage B.D.E.1 was dominant among both SFS genomes and Washington State genomes. In the 2022–2023 respiratory virus season, after the conclusion of SFS, RSV‐A lineage a.d.5.2 was dominant in Washington State, with fewer cases of other RSV‐A lineages and RSV‐B lineage B.D.E.2. All RSV lineages observed in Washington State from November 2018 to February 2023, that is, nearly a year beyond the end of the analysis period of SFS, were also observed in SFS. In the phylogenies, SFS genomes grouped with genomes of the same lineage sampled elsewhere and were interspersed among globally sampled genomes rather than forming locally restricted clusters (Figures 3B,C and 4B,C).
TABLE 2.
Summary of RSV lineages observed in SFS, Washington State, and globally, March 2018 to March 2022.
| RSV‐A | RSV‐B | |||
|---|---|---|---|---|
| Pre‐COVID‐19 restrictions a | Post‐COVID‐19 restrictions b | Pre‐COVID‐19 restrictions a | Post‐COVID‐19 restrictions b | |
| RSV sequences from Seattle Flu Study | ||||
| Sequences, N | 146 | 3 | 48 | 76 |
| Lineages, N | 6 | 3 | 3 | 2 |
| RSV sequences from Washington State, USA, in GenBank c | ||||
| Sequences, N | 319 | 6 | 124 | 145 |
| Lineages, N | 6 | 3 | 3 | 2 |
| RSV sequences globally in GenBank c | ||||
| Sequences, N | 1505 | 472 | 1157 | 271 |
| Lineages, N | 13 | 12 | 5 | 6 |
March 1, 2018 to March 22, 2020.
March 23, 2020 (when the Washington State stay‐at‐home order went into effect) to March 31, 2022.
Washington state sequences include sequences from SFS, and global sequences include Washington sequences.
FIGURE 3.
(A) RSV‐A lineages observed in SFS by age and over time; (B) maximum likelihood phylogenetic tree over calendar time; dashed vertical line indicates March 23, 2020, when the Washington State stay‐at‐home order went into effect; and (C) maximum likelihood phylogenetic tree over genetic divergence. Maximum likelihood phylogenetic trees include all SFS sequences and a sample of samples from GenBank.
FIGURE 4.
(A) RSV‐B lineages observed in SFS by age and over time, (B) maximum likelihood phylogenetic tree over calendar time; dashed vertical line indicates March 23, 2020, when the Washington State stay‐at‐home order went into effect; and (C) maximum likelihood phylogenetic tree over genetic divergence. Maximum likelihood phylogenetic trees include all SFS sequences and a sample of samples from GenBank.
Among the 273 quality‐filtered SFS genomes, 11 lineages were represented (7 RSV‐A, 4 RSV‐B). OpenArray Crt values (range: 4.8, 19.1) were compared across lineages to determine whether certain lineages might be associated with higher viral loads (Figure S2). Only two lineages per subgroup were observed frequently enough to support this comparison (RSV‐A: a.d.1 Crt 11.9 [IQR 9.7, 14.5, n = 110] and a.d.3 10.8 [10.2, 13.1, n = 21]; RSV‐B: B.D.4.1.1 11.2 [9.0, 14.6, n = 52] and B.D.E.1 10.0 [8.2, 12.6, n = 68]); the remaining seven lineages had n ≤ 8. Crt did not differ significantly across lineages within either subgroup (Kruskal–Wallis p = 0.28 for RSV‐A, p = 0.07 for RSV‐B). On stratification by age, the RSV‐B difference was confined to adults (children p = 0.96; adults p = 0.002, B.D.4.1.1 14.8 vs. B.D.E.1 10.6) but was confounded by other variables such as collection period, study arm, and swab type rather than a lineage effect alone. For example, B.D.E.1 was detected exclusively after the stay‐at‐home order and almost entirely in SCAN (at‐home self‐collected anterior nasal swabs), whereas B.D.4.1.1 was predominantly pre‐restriction and largely staff‐collected.
Patterns of RSV genomic diversity globally from November 2018 to February 2022 appeared broadly similar to those in Washington State, with a few notable differences (Figure 5). During the 2019–2020 respiratory virus season, the majority of genomes were also due to RSV‐A lineage a.d.1 and RSV‐B lineage B.D.4.1.1. During 2020–2021, when virtually no RSV was observed in Washington State, 150 RSV‐A lineage a.d.1.3 and eight RSV‐B lineage B.D.E.4 cases were observed in Australia, two lineages not observed in this study in Washington State. Similar to Washington State, the predominant lineages observed globally in 2021–2022 and 2022–2023 were RSV‐B lineage B.D.E.1 and RSV‐A lineage a.d.5.2, respectively.
FIGURE 5.
RSV lineages observed globally (A, N = 3405 sequences from the global GenBank dataset) and in Washington State, USA (B, N = 594 sequences from the Washington State GenBank dataset, which includes SFS genomes). Panels display all RSV genomes in GenBank with collection dates from March 2018 to March 2022.
Comparing sequences from 2‐year windows before and after the COVID‐19 lockdown in Washington State (March 23, 2020), we observed numerically lower nucleotide diversity (π) among RSV‐B sequences (0.006 vs. 0.003 for n = 124 pre‐COVID‐19 vs. n = 145 post‐COVID‐19 sequences). However, global RSV‐B sequences from pre‐COVID‐19 and post‐COVID‐19 periods had numerically similar π (0.005 for n = 1157 and 271, respectively). For RSV‐A, the number of post‐COVID‐19 sequences from Washington State (n = 6) was too low to perform a similar comparison, but mean pairwise nucleotide diversity among global RSV‐A sequences was numerically lower in the post‐lockdown period (π = 0.012 vs. 0.010 for n = 1505 pre‐COVID‐19 and n = 472 post‐COVID‐19 sequences, respectively); these differences were not evaluated statistically. To assess whether the observed change in diversity differed from expectation, we compared the divergence estimated using root‐to‐tip distances using substitution rates estimated within the Nexstrain tree (Figures 3 and 4). The overall divergence during the study period was similar for SFS sequences versus global sequences (7.16 vs. 7.20 × 10−4 and 6.99 vs. 6.93 × 10−4 substitutions per site per year for RSV‐A and RSV‐B, respectively).
Among SFS genomes, diversity estimates were of similar magnitude in children (n = 187 sequenced samples) and adults (n = 68), although not compared statistically. Less prevalent lineages (a.d.2.1, a.d.2.2, a.d.5, B.D.E.2, n < 5 each) were only observed among children, possibly due to fewer positives among adults (Figures 3A and 4A). Multiple sets of identical sequences including both children and adults were observed to cluster closely in the phylogenetic tree, but there was insufficient data to perform a detailed analysis of transmission chains.
4. Discussion
Using data from a community‐based respiratory virus surveillance study in Seattle, USA, in addition to publicly available genomes, we observed that RSV genomic diversity shifted from the pre‐COVID‐19 pandemic period to the pandemic‐related decline in RSV transmission, followed by the post‐pandemic restrictions off‐season re‐emergence. Genomic patterns observed globally were largely similar to those in Washington State, USA, and SFS. Our study contributed RSV whole genome sequences from community settings over three respiratory virus seasons spanning pre‐ and post‐COVID‐19 pandemic periods, which may have allowed us to better capture changes in RSV transmission compared to surveillance of clinical cases alone. Our community‐based study also included individuals of all ages, allowing us to capture a broader proportion of the population than that which is typically at high risk for clinical RSV disease.
The shift from RSV‐A predominance before the pandemic to RSV‐B predominance during the 2021–2022 re‐emergence is consistent with the alternating subgroup predominance that has long characterized RSV epidemiology, in which one subgroup tends to dominate for one or more seasons before being displaced by the other [12]. Within each subgroup, there are also multiple subclades that cocirculate. The near‐absence of RSV circulation during 2020 likely amplified this shift: with transmission interrupted for a full season, the susceptible population expanded across age groups, and RSV‐B re‐expanded preferentially into that enlarged susceptible pool as non‐pharmaceutical interventions were relaxed. The subsequent return to RSV‐A predominance (lineage a.d.5.2) in Washington State in 2022–2023 is consistent with resumption of the usual alternating pattern rather than with a durable change in subgroup dominance.
Our finding that the COVID‐19 pandemic disrupted trends in RSV genomic diversity was also observed in Australia, Japan, and in other US studies, but specific observed changes vary. In addition to RSV diversity declining in the first year of the COVID‐19 pandemic, two distinct, new RSV‐A lineages were observed in Australia [10]. These lineages were categorized under lineage a.d.1.3 in this study, which has only been observed in Australia from November 2020 onward. Conversely, in this study and other US studies, all RSV lineages observed in the post‐COVID‐19 pandemic period were also observed in the pre‐pandemic period [12, 13, 14]. In Japan, seven clusters of RSV‐A were observed in 2019–2020 prior to the COVID‐19 pandemic versus only two clusters observed following the pandemic through 2022, but only a single RSV‐B cluster was included over the entire study period [11]. In the one US study that specifically evaluated changes in genomic diversity following the COVID‐19 pandemic, evidence of a more homogenous population of RSV‐B viruses was observed in 2021–2022, but this was not observed for RSV‐A. However, the 2021–2022 respiratory virus season in the US was dominated by RSV‐B, with very little RSV‐A. Therefore, this study may not have been able to adequately assess changes in diversity of RSV‐A using data only through 2022 as SFS community samples were only collected through July 2022. However, publicly available Washington State data from 2022–2023, does suggest a shift from RSV‐A lineage a.d.1 prior to the COVID‐19 pandemic to RSV‐A lineage a.d.5.2. We know of no other studies to date specifically investigating the impact of the pandemic on RSV genomic diversity, likely because RSV genomic surveillance is conducted less frequently than that for other respiratory viruses, including influenza, for which this “bottleneck” effect on genomic diversity during the COVID‐19 pandemic has been well documented [31, 32, 33, 34].
While RSV‐B predominated and showed numerically lower genomic diversity during the 2021–2022 season, more recent genomic surveillance suggests this contraction was transient. Genomic surveillance in Sydney, Australia documented a rebound in RSV diversity in the seasons following the pandemic, with re‐diversification of circulating lineages [35]; and surveillance in Minnesota, USA similarly observed continued lineage turnover and diversification through 2023–2024 [35, 36]. Because SFS community sampling ended in July 2022, our findings represent a snapshot of RSV population structure during a distinct post‐pandemic period, and continued genomic surveillance will be needed to characterize the longer‐term evolutionary trajectory of RSV‐B and RSV‐A lineages.
Efforts to draw comparisons across studies are complicated by a lack of standardized nomenclature for RSV lineages and varying methods used to assess changes in RSV genomic diversity over time used across studies [37, 38, 39, 40, 41]. For example, the Australia study used the historical RSV‐A ON1 strain designation and defined lineages within study sequences [10]. The RSV‐A ON1 strain captured all RSV‐A in the Australian study and is comparable to the Goya et al. GA2 lineage and sub‐lineages, which were reported in Japan and other US studies described above [11, 37]. Therefore, conclusions regarding whether genomic diversity changed based solely on the observed number of lineages circulating pre‐ and post‐COVID‐19 pandemic periods may partially depend on the nomenclature system used. Substantial progress has since been made toward a standardized classification of RSV genomic diversity. Following an initial proposal for unified genotype definition based on the G gene and a subsequent proposal for nomenclature below the species level, whole genome‐based genotyping systems were developed, and a consensus lineage classification was established by the RSV Genotyping Consensus Consortium. Lineage definitions under this system are maintained as living designations in public repositories and are implemented in Nextclade, which we used for lineage assignment in this study. The value of this standardization is evident in our own data: under the older G gene‐based genotyping system, all genomes in this study would be assigned to single GA and GB genotypes, whereas the consensus system resolves the same sequences into seven RSV‐A and four RSV‐B lineages. We nonetheless report distance‐based diversity metrics alongside lineage frequencies because these are independent of classification boundaries and capture variation arising within a lineage. As lineage definitions continue to be updated as new diversity is designated, we recommend that tools return standardized version numbers so that lineage‐based comparisons remain interpretable over time. Additional metrics such as nucleotide diversity and distance‐based metrics should be employed to complement lineage‐based analyses.
Although we did not test differences statistically, our analysis showed numerically lower genetic diversity post‐lockdown, a pattern more consistent with a population bottleneck following interrupted transmission rather than with selective pressure acting on the viral population. Several non‐mutually exclusive factors may contribute to this reduced diversity. The marked reduction in RSV circulation during the period of COVID‐19 mitigation measures including non‐pharmaceutical interventions curtailed RSV transmission narrowing the number of circulating lineages that could seed subsequent seasons. Changes in transmission dynamics and population susceptibility may favor more rapid expansion of the lineages that survived the bottleneck. Sampling differences over time, geography, and stochastic effects of sampling from a smaller post‐pandemic pool could also contribute to certain lineages being overrepresented.
Such bottlenecks may carry public health and clinical implications. In the short term, a less diverse, more antigenically homogeneous virus population could be more uniformly targeted by interventions directed at the highly conserved prefusion F glycoprotein including maternal vaccines and long‐acting monoclonal antibodies such as nirsevimab and clesrovimab, since reduced standing variation offers fewer opportunities for immune escape. The reduced diversity we observed would therefore not be expected to compromise the activity of currently licensed F‐directed interventions, with the caveat that we did not perform a detailed analysis of escape mutations. Conversely, bottlenecks can facilitate the stochastic fixation of novel variants and rapid re‐diversification once transmission resumes, as has been observed for RSV‐B in subsequent seasons [35] and for influenza following analogous pandemic‐era suppression [33, 34]. This is precisely why baseline, pre‐intervention diversity data of the kind reported here are valuable. To date, available data indicate that these interventions retain broad activity against circulating strains, though ongoing genomic surveillance remains important to monitor for viral evolution that could affect antigenicity or the emergence of escape‐associated mutations [42, 43]. More broadly, our characterization of baseline diversity and its disruption during the pandemic provides a reference point against which future selective pressures can be monitored.
This analysis has several limitations. First, SFS used convenience sampling at public places in Seattle, USA, and may not be representative of the entire metropolitan area. Second, most SFS samples were collected from individuals with symptoms of acute respiratory infection (ARI), and we were unable to compare trends in genomic diversity between individuals with symptomatic and asymptomatic infections. Enrollment criteria also differed across the study period: post‐restriction sequences (2021–2022) were derived exclusively from SCAN, in which participants were not required to be symptomatic at the time of swabbing, whereas pre‐restriction collection targeted individuals with symptoms of ARI. Differences in surveillance approach before and after the pandemic may therefore have influenced both the observed distribution of RSV‐A and RSV‐B and estimates of genomic diversity. While inclusion of asymptomatic or pauci‐symptomatic infections could shift the relative proportions of RSV‐A and RSV‐B if disease severity differed between subgroups, there is limited evidence to suggest differences in the asymptomatic proportion of RSV infections by subgroup. We expect the impact on within‐subgroup genetic diversity to be limited, as the reduced diversity observed for RSV‐B during the 2021–2022 season is more consistent with limited viral introductions following COVID‐19 mitigation measures than with differences based on symptom status. Third, only a subset of RSV‐positive samples captured by SFS could be sequenced and specimens with high Crt values (corresponding to a low viral load) could not be sequenced. Therefore, genomic diversity among lower viral load infections is unobserved in this study. Among samples that were sequenced, we were able to compare Crt values by lineage, but comparisons were limited to this restricted range of Crt values and to lineages that had sufficient representation. The analysis was largely exploratory due to insufficient sample sizes to account for covariates, so lineage‐specific viral load could not be separated from period‐ and collection‐related effects. Overall sequencing yield was similar across the three SFS study arms, but because the arms were deployed sequentially (Swab & Send exclusively before the stay‐at‐home order and SCAN exclusively after it), differences in yield by study arm cannot be separated from calendar period. The similarity in overall yield across arms makes it unlikely that differential genome recovery accounts for the shift in lineage composition we observed, although a contribution cannot be excluded. Fourth, there is a lack of demographic and clinical data associated with publicly available genomes in GenBank. Therefore, analyses by age were limited to SFS genomes only, and we could not identify whether non‐SFS sequences from Washington State in GenBank represented clinical or community cases. Fifth, comparisons of genomic diversity across periods and age groups were descriptive and we did not perform formal statistical testing. Because of differences over time in sample numbers, viral load distribution, cohort composition, and public sequence availability, which can each affect measured diversity, we cannot attribute the lower post‐lockdown diversity to any single mechanism. Therefore, small differences should not be over‐interpreted. Sixth, the global subsample used for tree reconstruction reflects the composition of publicly available sequence data and is weighted toward countries and periods with greater sequencing capacity. Therefore, it may not fully represent the global genetic diversity of RSV. In addition, because SFS community sampling ended in July 2022, our findings capture a single post‐pandemic season and provide a snapshot of RSV population structure rather than a trajectory; more recent surveillance from other settings suggests that the contraction in RSV‐B diversity was transient, and our results should not be read as describing a durable narrowing of RSV genetic diversity. Lastly, although our study contributed a large proportion of the RSV genomes during the study period, the sampling strategy was not designed to interrogate transmission chains and we did not analyze genomic clusters or introduction events; such phylodynamic analyses would require denser and more systematically sampled data than were available for this study. Future studies should prioritize multi‐year analyses using consistent surveillance approaches across periods, so that changes in RSV subgroup distribution and genomic diversity can be distinguished from changes in sampling design.
Despite these limitations, this is the first study to our knowledge to assess changes in RSV genomic diversity from pre‐ to post‐COVID‐19 pandemic periods specifically among community cases. Our study was also strengthened by inclusion of demographic data associated with genomes, such as age. Finally, we generated whole genome sequences for our study samples, versus sequences of the F‐ or G‐gene only, allowing better assessment of genomic diversity, transmission, and also examination of individual genes targeted by currently available vaccines and monoclonal antibodies.
5. Conclusions
The COVID‐19 pandemic was associated with a shift in RSV genomic diversity, although there was geographic heterogeneity in specific changes observed globally. In Washington State, USA, RSV‐A and RSV‐B (majority lineages a.d.1 and B.D.4.1.1) co‐circulated in 2019–2020. Following implementation of COVID‐19 pandemic restrictions, no RSV was observed during the 2020–2021 respiratory virus season. In 2021–2022, RSV re‐emerged off‐season and shifted to predominantly RSV‐B (majority lineage B.D.E.1) and subsequently RSV‐A (majority lineage a.d.5.2) in 2022–2023. SFS surveillance data contributed community‐based RSV whole genome sequences to the literature and documented transmission between children and adults in community settings.
Author Contributions
Pavitra Roychoudhury: data curation, formal analysis, writing – review and editing, conceptualization, methodology, investigation, visualization, writing – original draft, supervision. Julia C. Bennett: conceptualization, investigation, methodology, visualization, writing – original draft, formal analysis, data curation. Anna Elias‐Warren: methodology. Erica Wetzler: writing – review and editing. Lea M. Starita: data curation, resources. Jeremy Stone: software, data curation, resources. Ben Capodanno: data curation, resources. Robin Prentice: project administration. Peter D. Han: data curation, resources. Sally B. Grindstaff: data curation, resources. David Reinhart: data curation, resources. Jennifer K. Logue: project administration. Caitlin R. Wolf: project administration. Michael Boeckh: data curation, investigation. Christina M. Lockwood: investigation. Amanda C. Perofsky: methodology, writing – review and editing. Cécile Viboud: methodology. Jay Shendure: investigation, data curation. Trevor Bedford: data curation, investigation. Marco Carone: formal analysis, methodology. Janet A. Englund: investigation, writing – review and editing, conceptualization. Helen Y. Chu: writing – review and editing, funding acquisition, investigation, conceptualization, methodology, validation, supervision.
Funding
The analysis reported in this publication was supported by Gates Ventures. The content is solely the responsibility of the authors and does not necessarily represent the official views of the funder. Award/grant number is not applicable.
Disclosure
No copyrighted material from other sources that required permission to reproduce is included in the manuscript.
Ethics Statement
This study was approved by the University of Washington Human Subjects Institutional Review Board (IRB).
Consent
All participants completed informed consent (or parental or guardian consent for participants < 18 years of age at enrollment).
Conflicts of Interest
H.Y.C. has served on advisory boards for Vir and Roche. J.A.E. has served as a consultant with AbbVie, AstraZeneca, GlaxoSmithKline, Merck, Meissa, Pfizer, Shionogi, and Cidarra, outside of the submitted work, and has received research funding from AstraZeneca, Moderna, GlaxoSmithKline, and Pfizer. M.B. has served as a consultant with AstraZeneca outside of the submitted work and has received research funding from AstraZeneca and Pulmotect. All other authors declare no conflicts of interest.
Supporting information
Figure S1: Sequencing yield among RSV‐positive SFS samples by study arm. (A) Sequencing outcomes by study arm. (B) Genome recovery by binned Crt values to allow comparison at different Crt levels. (C) Distribution of Crt values by study arm.
Figure S2: OpenArray relative cycle threshold (Crt) among quality‐filtered Seattle Flu Study RSV genomes (n = 273). (A) Crt by RSV lineage, separately for RSV‐A and RSV‐B; boxes show median and interquartile range with whiskers to 1.5× IQR, points are individual genomes colored by age group, and lineages are ordered by median Crt. Boxes are omitted for lineages with fewer than five genomes. (B) The same data for the four lineages with ≥ 10 genomes, stratified by age group.
Figure S3: Age and clinical presentation by RSV lineage. (A) Age distribution by lineage (n = 262 sequenced cases with age; labels indicate genome counts). (B) Percentage of symptomatic cases reporting each symptom by lineage, with 95% confidence intervals. (C) The same symptoms among all 46,015 symptomatic SFS participants by study arm irrespective of RSV status (left) and among 251 symptomatic sequenced RSV cases (right). (D) Number of symptoms reported by study arm, among all symptomatic participants and among sequenced RSV cases; median (IQR) and n below each box.
Table S1: Sequencing yield.
Data Availability Statement
All data used in the analysis are included in this published article and its Supporting Information. Additional materials are available from the corresponding author (helenchu@uw.edu) upon reasonable request.
References
- 1. Falsey A. R., Hennessey P. A., Formica M. A., Cox C., and Walsh E. E., “Respiratory Syncytial Virus Infection in Elderly and High‐Risk Adults,” New England Journal of Medicine 352, no. 17 (2005): 1749–1759, 10.1056/NEJMoa043951. [DOI] [PubMed] [Google Scholar]
- 2. Shi T., McAllister D. A., O'Brien K. L., et al., “Global, Regional, and National Disease Burden Estimates of Acute Lower Respiratory Infections due to Respiratory Syncytial Virus in Young Children in 2015: A Systematic Review and Modelling Study,” Lancet 390, no. 10098 (2015): 946–958, 10.1016/S0140-6736(17)30938-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Hall C. B., Weinberg G. A., Iwane M. K., et al., “The Burden of Respiratory Syncytial Virus Infection in Young Children,” New England Journal of Medicine 360, no. 6 (2009): 588–598, 10.1056/NEJMoa0804877. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Walsh E. E., “Respiratory Syncytial Virus Infection: An Illness for All Ages,” Clinics in Chest Medicine 38, no. 1 (2017): 29–36, 10.1016/j.ccm.2016.11.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.“ABRYSVO,” FDA, retrieved July 10, 2026, https://www.fda.gov/vaccines‐blood‐biologics/abrysvo.
- 6.“AREXVY,” FDA, retrieved July 28, 2026, https://www.fda.gov/vaccines‐blood‐biologics/arexvy.
- 7.“FDA Approves New Drug to Prevent RSV in Babies and Toddlers,” FDA, August 9, 2024, https://www.fda.gov/news‐events/press‐announcements/fda‐approves‐new‐drug‐prevent‐rsv‐babies‐and‐toddlers.
- 8.“RSV Vaccine and mAb Snapshot,” retrieved August 17, 2026, https://www.path.org/our‐impact/resources/rsv‐vaccine‐and‐mab‐snapshot/.
- 9. Olsen S. J., “Changes in Influenza and Other Respiratory Virus Activity During the COVID‐19 Pandemic—United States, 2020–2021,” MMWR. Morbidity and Mortality Weekly Report 70 (2020): 1013–1019, 10.15585/mmwr.mm7029a1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Eden J. S., Sikazwe C., Xie R., et al., “Off‐Season RSV Epidemics in Australia After Easing of COVID‐19 Restrictions,” Nature Communications 13, no. 1 (2022): 2884, 10.1038/s41467-022-30485-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Yoshioka S., Phyu W. W., Wagatsuma K., et al., “Molecular Epidemiology of Respiratory Syncytial Virus During 2019–2022 and Surviving Genotypes After the COVID‐19 Pandemic in Japan,” Viruses 15, no. 12 (2023): 2382, 10.3390/v15122382. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Rios‐Guzman E., Simons L. M., Dean T. J., et al., “Deviations in RSV Epidemiological Patterns and Population Structures in the United States Following the COVID‐19 Pandemic,” Nature Communications 15, no. 1 (2024): 3374, 10.1038/s41467-024-47757-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Goya S., Sereewit J., Pfalmer D., et al., “Genomic Characterization of Respiratory Syncytial Virus During 2022–23 Outbreak, Washington, USA,” Emerging Infectious Diseases 29, no. 4 (2023): 865–868, 10.3201/eid2904.221834. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Adams G., Moreno G. K., Petros B. A., et al., “Viral Lineages in the 2022 RSV Surge in the United States,” New England Journal of Medicine 388, no. 14 (2023): 1335–1337, 10.1056/NEJMc2216153. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Mas V., Nair H., Campbell H., Melero J. A., and Williams T. C., “Antigenic and Sequence Variability of the Human Respiratory Syncytial Virus F Glycoprotein Compared to Related Viruses in a Comprehensive Dataset,” Vaccine 36, no. 45 (2018): 6660–6673, 10.1016/j.vaccine.2018.09.056. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Kim S., Williams T. C., Viboud C., Campbell H., Chen J., and Spiro D. J., “RSV Genomic Diversity and the Development of a Globally Effective RSV Intervention,” Vaccine 39, no. 21 (2021): 2811–2820, 10.1016/j.vaccine.2021.03.096. [DOI] [PubMed] [Google Scholar]
- 17. Chu H. Y., Chu H. Y., Boeckh M., et al., “LB21. The Seattle Flu Study: A Community‐Based Study of Influenza,” Open Forum Infectious Diseases 6, no. Supplement_2 (2019): S1002, 10.1093/ofid/ofz415.2504. [DOI] [Google Scholar]
- 18. Harris P. A., Taylor R., Thielke R., Payne J., Gonzalez N., and Conde J. G., “Research Electronic Data Capture (REDCap)—A Metadata‐Driven Methodology and Workflow Process for Providing Translational Research Informatics Support,” Journal of Biomedical Informatics 42, no. 2 (2009): 377–381, 10.1016/j.jbi.2008.08.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Harris P. A., Taylor R., Minor B. L., et al., “The REDCap Consortium: Building an International Community of Software Platform Partners,” Journal of Biomedical Informatics 95 (2019): 103208, 10.1016/j.jbi.2019.103208. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Müller N. F., Wagner C., Frazar C. D., et al., “Viral Genomes Reveal Patterns of the SARS‐CoV‐2 Outbreak in Washington State,” Science Translational Medicine 13, no. 595 (2021): eabf0202, 10.1126/scitranslmed.abf0202. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Chu H. Y., Boeckh M., Englund J. A., et al., “The Seattle Flu Study: A Multiarm Community‐Based Prospective Study Protocol for Assessing Influenza Prevalence, Transmission and Genomic Epidemiology,” BMJ Open 10, no. 10 (2020): e037295, 10.1136/bmjopen-2020-037295. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Cox S. N., Roychoudhury P., Frivold C., et al., “Household Transmission and Genomic Diversity of Respiratory Syncytial Virus in the United States, 2022–2023,” Clinical Infectious Diseases: An Official Publication of the Infectious Diseases Society of America 81, no. 6 (2022): 1159–1169, 10.1093/cid/ciaf048. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Grubaugh N. D., Gangavarapu K., Quick J., et al., “An Amplicon‐Based Sequencing Framework for Accurately Measuring Intrahost Virus Diversity Using PrimalSeq and iVar,” Genome Biology 20, no. 1 (2019): 8, 10.1186/s13059-018-1618-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Aksamentov I., Roemer C., Hodcroft E. B., and Neher R. A., “Nextclade: Clade Assignment, Mutation Calling and Quality Control for Viral Genomes,” Journal of Open Source Software 6, no. 67 (2021): 3773, 10.21105/joss.03773. [DOI] [Google Scholar]
- 25. Hadfield J., Megill C., Bell S. M., et al., “Nextstrain: Real‐Time Tracking of Pathogen Evolution,” Bioinformatics 34, no. 23 (2018): 4121–4123, 10.1093/bioinformatics/bty407. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Sagulenko P., Puller V., and Neher R. A., “TreeTime: Maximum‐Likelihood Phylodynamic Analysis,” Virus Evolution 4, no. 1 (2018): vex042, 10.1093/ve/vex042. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Xu S., Li L., Luo X., et al., “Ggtree: A Serialized Data Object for Visualization of a Phylogenetic Tree and Annotation Data,” iMeta 1, no. 4 (2022): e56, 10.1002/imt2.56. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Paradis E., “pegas: An R Package for Population Genetics With an Integrated–Modular Approach,” Bioinformatics 26, no. 3 (2010): 419–420, 10.1093/bioinformatics/btp696. [DOI] [PubMed] [Google Scholar]
- 29. Paradis E., Claude J., and Strimmer K., “APE: Analyses of Phylogenetics and Evolution in R Language,” Bioinformatics 20, no. 2 (2004): 289–290, 10.1093/bioinformatics/btg412. [DOI] [PubMed] [Google Scholar]
- 30. R Core Team , “R: A Language and Environment for Statistical Computing,” (R Foundation for Statistical Computing, 2021). [Google Scholar]
- 31. Rios Guzman E. and Hultquist J. F., “Clinical and Biological Consequences of Respiratory Syncytial Virus Genetic Diversity,” Therapeutic Advances in Infectious Disease 9 (2022): 20499361221128091, 10.1177/20499361221128091. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Langedijk A. C., Harding E. R., Konya B., et al., “A Systematic Review on Global RSV Genetic Data: Identification of Knowledge Gaps,” Reviews in Medical Virology 32, no. 3 (2022): e2284, 10.1002/rmv.2284. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Dhanasekaran V., Sullivan S., Edwards K. M., et al., “Human Seasonal Influenza Under COVID‐19 and the Potential Consequences of Influenza Lineage Elimination,” Nature Communications 13, no. 1 (2022): 1721, 10.1038/s41467-022-29402-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Paget J., Caini S., Riccio M. D., van Waarden W., and Meijer A., “Has Influenza B/Yamagata Become Extinct and What Implications Might This Have for Quadrivalent Influenza Vaccines?,” Eurosurveillance 27, no. 39 (2022): 2200753, 10.2807/1560-7917.ES.2022.27.39.2200753. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Herkes J. M. S., Foster C. S. P., Michie A., Bartlett A. W., Rawlinson W. D., and Walker G. J., “Genomic Surveillance of Respiratory Syncytial Virus in Sydney Reveals Rebound in Postpandemic Viral Diversity,” Open Forum Infectious Diseases 13, no. 4 (2026): ofag192, 10.1093/ofid/ofag192. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Evans D., Kunerth H., Mumm E., et al., “Genomic Epidemiology of Human Respiratory Syncytial Virus, Minnesota, USA, July 2023–February 2024,” Emerging Infectious Diseases Journal 30, no. 11 (2024): 2414–2418, 10.3201/eid3011.241000. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Goya S., Galiano M., Nauwelaers I., et al., “Toward Unified Molecular Surveillance of RSV: A Proposal for Genotype Definition,” Influenza and Other Respiratory Viruses 14, no. 3 (2020): 274–285, 10.1111/irv.12715. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Ramaekers K., Rector A., Cuypers L., Lemey P., Keyaerts E., and Van Ranst M., “Towards a Unified Classification for Human Respiratory Syncytial Virus Genotypes,” Virus Evolution 6, no. 2 (2020): veaa052, 10.1093/ve/veaa052. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Chen J., Qiu X., Avadhanula V., et al., “Novel and Extendable Genotyping System for Human Respiratory Syncytial Virus Based on Whole‐Genome Sequence Analysis,” Influenza and Other Respiratory Viruses 16, no. 3 (2022): 492–500, 10.1111/irv.12936. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Salimi V., Viegas M., Trento A., et al., “Proposal for Human Respiratory Syncytial Virus Nomenclature Below the Species Level,” Emerging Infectious Diseases journal 27, no. 6 (2021): 1–9, 10.3201/eid2706.204608. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Goya S., Ruis C., Neher R. A., et al., “Standardized Phylogenetic Classification of Human Respiratory Syncytial Virus Below the Subgroup Level,” Emerging Infectious Diseases Journal 30, no. 8 (2024): 1631–1641, 10.3201/eid3008.240209. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Mahool K. A., Kijak G. H., Ahani B., et al., “Nirsevimab Binding Site Conservation in RSV F Protein Between 2019 and 2024: Analysis of Sequencing Data From the US OUTSMART‐RSV Surveillance Study and Public Databases,” Open Forum Infectious Diseases 13, no. Suppl 1 (2026): ofaf695.063, 10.1093/ofid/ofaf695.063. [DOI] [Google Scholar]
- 43. Wilkins D., Langedijk A. C., Lebbink R. J., et al., “Nirsevimab Binding‐Site Conservation in Respiratory Syncytial Virus Fusion Glycoprotein Worldwide Between 1956 and 2021: An Analysis of Observational Study Sequencing Data,” Lancet Infectious Diseases 23, no. 7 (2023): 856–866, 10.1016/S1473-3099(23)00062-2. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1: Sequencing yield among RSV‐positive SFS samples by study arm. (A) Sequencing outcomes by study arm. (B) Genome recovery by binned Crt values to allow comparison at different Crt levels. (C) Distribution of Crt values by study arm.
Figure S2: OpenArray relative cycle threshold (Crt) among quality‐filtered Seattle Flu Study RSV genomes (n = 273). (A) Crt by RSV lineage, separately for RSV‐A and RSV‐B; boxes show median and interquartile range with whiskers to 1.5× IQR, points are individual genomes colored by age group, and lineages are ordered by median Crt. Boxes are omitted for lineages with fewer than five genomes. (B) The same data for the four lineages with ≥ 10 genomes, stratified by age group.
Figure S3: Age and clinical presentation by RSV lineage. (A) Age distribution by lineage (n = 262 sequenced cases with age; labels indicate genome counts). (B) Percentage of symptomatic cases reporting each symptom by lineage, with 95% confidence intervals. (C) The same symptoms among all 46,015 symptomatic SFS participants by study arm irrespective of RSV status (left) and among 251 symptomatic sequenced RSV cases (right). (D) Number of symptoms reported by study arm, among all symptomatic participants and among sequenced RSV cases; median (IQR) and n below each box.
Table S1: Sequencing yield.
Data Availability Statement
All data used in the analysis are included in this published article and its Supporting Information. Additional materials are available from the corresponding author (helenchu@uw.edu) upon reasonable request.