The research team collected RSV-positive upper respiratory tract samples from patients who tested positive between October 2023 and March 2025, spanning two consecutive seasons that straddled the deployment of nirsevimab. Whole viral genomes were recovered using an amplicon-based sequencing protocol carried out on Illumina platforms, with assemblies generated through a consensus-building pipeline known as vcf-consensus-builder. This approach allowed the investigators to capture near-complete genomes from a substantial fraction of positive samples, providing the resolution needed to place each virus within a fine-grained phylogenetic framework and to examine amino acid variation in the proteins most relevant to prevention and treatment.

In total, the study retrieved 302 RSV genome sequences, with 129 derived from the 2023–2024 season and 173 from the 2024–2025 season. The clinical composition of the cohort reflects the burden of RSV disease in the region: most infections occurred in infants aged one year or younger, who accounted for 56.0 percent of cases, and males made up 57.3 percent of the sampled population. Lower respiratory tract infections were documented in 72.7 percent of cases, 67.2 percent of patients required hospitalization, and 15.0 percent were admitted to an intensive care unit. Notably, the dataset included 17 nirsevimab breakthrough infections, cases in which infants who had received the monoclonal antibody nonetheless developed symptomatic RSV disease.

Phylogenetic analysis using maximum-likelihood methods implemented in IQ-TREE, with a GTR plus F, invariant sites, and gamma model of substitution, assigned the circulating viruses to established clades within the two major RSV subgroups. The most frequently detected clades were A.D.3, representing 26.8 percent of sequences, and B.D.E.1, representing 20.2 percent, followed by A.D.1.2 at 17.5 percent and A.D.5.2 at 14.2 percent. This distribution confirms that multiple lineages of both RSV-A and RSV-B co-circulated in Lombardy throughout the study period, with no single clade achieving complete dominance across the two seasons.

Beyond clade assignment, the researchers applied Bayesian coalescent methods under a constant population size prior and a strict molecular clock to define what they termed large phylogenetic clusters, or LPCs. These were groups of more than ten sequences supported by a posterior probability of one, indicating tightly linked transmission chains or rapidly expanding lineages. Four such clusters were identified, and two of them proved particularly striking: a cluster of 56 A.D.3 strains and another of 17 A.D.1.2 strains were detected exclusively in the 2024–2025 season, the first season of routine nirsevimab use. The emergence of season-specific clusters suggests that new viral lineages established themselves in the population after the prophylaxis program began.

A central concern of the study was the monitoring of mutations associated with resistance to the preventive and therapeutic agents now deployed against RSV. The fusion protein, or F, is the target of the monoclonal antibodies palivizumab, nirsevimab, and clesrovimab, while the nucleoprotein, or N, is the target of the small-molecule inhibitor zelicapavir. Resistance-associated mutations were identified in two samples from the 2023–2024 season and two from the 2024–2025 season. These included K272E in an A.D.5.3 strain and N262Y in an A.D.3 strain, both associated with reduced susceptibility to palivizumab, and N208D in a B.D.E.1 strain as well as the combination of K68I and L204S in a B.D.E.1 nirsevimab breakthrough infection, mutations linked to nirsevimab resistance.

Among the 17 nirsevimab breakthrough infections, the investigators also detected the S213R substitution at antigenic site Ø of the F protein in an A.D.1.2 sample. Site Ø is the principal binding site for nirsevimab, and changes at this region can diminish antibody binding. By contrast, no mutations conferring resistance to clesrovimab, a newer monoclonal antibody that binds a different region of F, were detected during the study. However, four substitutions were observed at the clesrovimab binding site: K433R in a B.D.4.1.1 strain, C439R in an A.D.1.2 strain, N437S in an A.D.5.2 strain, and N444S in an A.D.3 strain, with three of the four appearing during the 2024–2025 season. These variants were not classified as resistance mutations, but their presence at a pharmacologically critical epitope warrants continued attention.

The nucleoprotein analysis likewise revealed signals relevant to antiviral therapy. The I129M substitution in the N protein, which is associated with resistance to zelicapavir, was detected in a B.D.E.1 sample from the 2024–2025 season at an intrapatient relative abundance of 0.05, indicating that the variant existed as a minor species within the quasispecies of that infection rather than as the dominant viral population. Such low-frequency variants are typically invisible to conventional diagnostic assays but can be resolved by deep sequencing, and their detection underscores the value of high-resolution genomic surveillance in identifying resistance determinants before they become established at the population level.

Multivariable analysis comparing the two seasons revealed several statistically significant differences associated with the 2024–2025 period. That season was characterized by an increased number of sequences belonging to large phylogenetic clusters, greater amino acid variability at antigenic site Ø of the F protein, and higher Ct values on diagnostic testing, which correspond to lower viral loads in respiratory samples. Hospital intensive care unit admissions were also fewer in the second season. While the observational design of the study precludes drawing direct causal links between nirsevimab deployment and these shifts, the concurrence of prophylaxis introduction with reduced disease severity, altered viral population structure, and expanded F protein variability provides a compelling rationale for sustained monitoring.

The overall message from the Lombardy surveillance program is one of cautious reassurance combined with vigilance. Nirsevimab breakthrough infections remained rare relative to the scale of the prophylaxis campaign, and bona fide resistance mutations were detected in only a handful of samples across two seasons. At the same time, the virus is demonstrably diversifying, with season-specific phylogenetic clusters and accumulating variation at antibody binding sites on the F protein. As monoclonal antibodies and vaccines become mainstays of RSV prevention worldwide, the Lombardy experience illustrates how whole-genome sequencing, phylogenetic clustering, and mutation-level analysis can be woven into routine public health practice, ensuring that shifts in viral evolution are detected early enough to inform and adjust prophylaxis strategies before they compromise protection of the most vulnerable patients.

Subject of Research: Genomic evolution and monoclonal antibody resistance of respiratory syncytial virus in Lombardy before and after nirsevimab introduction

Article Title: Genomic characterization of circulating respiratory syncytial virus strains in Lombardy before and after the introduction of nirsevimab prophylaxis

Article References: Uceda Renteria, S., Galli, C., Ferrari, G., Cossu, M. V., Lombardi, A., Tagliabue, C., Galli, G., Parisi, A., Pellegrinelli, L., Romano, G., Bono, P., Pitrolo, A. M. G., Salpini, R., Salvini, F., Svicher, V., Bosis, S., Callegaro, A., Gori, A., Baldanti, F., … Alteri, C. (2026). Genomic characterization of circulating respiratory syncytial virus strains in Lombardy before and after the introduction of nirsevimab prophylaxis. Genome Medicine. https://doi.org/10.1186/s13073-026-01768-x

Image Credits: AI Generated

DOI: 10.1186/s13073-026-01768-x

Keywords: respiratory syncytial virus, nirsevimab, genomic surveillance, viral evolution, monoclonal antibodies, fusion protein, phylogenetics, whole-genome sequencing, antiviral resistance, breakthrough infections, Lombardy, public health