ABSTRACT
BACKGROUND: Nirsevimab is an extended half-life monoclonal antibody to the respiratory syncytial virus (RSV) fusion protein that has been developed to protect infants for an entire RSV season. Previous studies have shown that the nirsevimab binding site is highly conserved. However, investigations of the geotemporal evolution of potential escape variants in recent (ie, 2015-2021) RSV seasons have been minimal. Here, we examine prospective RSV surveillance data to assess the geotemporal prevalence of RSV A and B, and functionally characterise the effect of the nirsevimab binding-site substitutions identified between 2015 and 2021. METHOD(S): We assessed the geotemporal prevalence of RSV A and B and nirsevimab binding-site conservation between 2015 and 2021 from three prospective RSV molecular surveillance studies (the US-based OUTSMART-RSV, the global INFORM-RSV, and a pilot study in South Africa). Nirsevimab binding-site substitutions were assessed in an RSV microneutralisation susceptibility assay. We contextualised our findings by assessing fusion-protein sequence diversity from 1956 to 2021 relative to other respiratory-virus envelope glycoproteins using RSV fusion protein sequences published in NCBI GenBank. FINDINGS: We identified 5675 RSV A and RSV B fusion protein sequences (2875 RSV A and 2800 RSV B) from the three surveillance studies (2015-2021). Nearly all (25 [100%] of 25 positions of RSV A fusion proteins and 22 [88%] of 25 positions of RSV B fusion proteins) amino acids within the nirsevimab binding site remained highly conserved between 2015 and 2021. A highly prevalent (ie, >40.0% of all sequences) nirsevimab binding-site Ile206Met:Gln209Arg RSV B polymorphism arose between 2016 and 2021. Nirsevimab neutralised a diverse set of recombinant RSV viruses, including new variants containing binding-site substitutions. RSV B variants with reduced susceptibility to nirsevimab neutralisation were detected at low frequencies (ie, prevalence <1.0%) between 2015 and 2021. We used 3626 RSV fusion-protein sequences published in NCBI GenBank between 1956 and 2021 (2024 RSV and 1602 RSV B) to show that the RSV fusion protein had lower genetic diversity than influenza haemagglutinin and SARS-CoV-2 spike proteins. INTERPRETATION: The nirsevimab binding site was highly conserved between 1956 and 2021. Nirsevimab escape variants were rare and have not increased over time. FUNDING: AstraZeneca and Sanofi.Copyright © 2023 Elsevier Ltd. All rights reserved.
ABSTRACT
Background. AZD7442 is a combination of extended-half-life SARS-CoV-2- neutralizing monoclonal antibodies (tixagevimab/cilgavimab) that bind to distinct epitopes on the SARS-CoV-2 spike protein. In the PROVENT study, a single 300 mg intramuscular dose of AZD7442 demonstrated 77% efficacy for prevention of COVID-19 vs placebo at primary analysis, with 83% efficacy through 6-months follow-up, and was well-tolerated. We report conservation of AZD7442 binding sites and neutralizing activity against pseudotyped virus-like particles (VLPs) harboring spike substitutions identified in surveillance, and clinical SARS-CoV-2 isolates from the PROVENT study. Methods. Consensus SARS-CoV-2 whole genome sequences were analyzed from open source databases to identify prevalent spike substitutions within the AZD7442 binding site. Phenotypic analyses determined neutralization susceptibility of pseudotyped VLPs with identified spike substitutions. Genotypic analyses were also performed on SARS-CoV-2 spike sequences from PROVENT study (NCT04625725) participants with RT-PCR-positive symptomatic illness. Results. Most residues in tixagevimab (13/17) and cilgavimab (13/19) binding sites were >99% conserved among global SARS-CoV-2 isolates (N=8,373,740 through Apr 19, 2022). In 2021, AZD7442 binding site polymorphisms emerged among circulating strains (prevalence: R346K, 11%;N440K, 22%;G446S, 15%;S477N, 28%;L452R, 43%;T478K, 70%;E484A, 27%;E484K, 3%;Q493R, 27%), but these did not affect AZD7442 in vitro neutralization potency. AZD7442 retained neutralization activity against variants of concern or interest tested, including Omicron BA.2, with moderate reduction observed for Omicron BA.1. By median 6-months follow-up (Aug 29, 2021, data cut-off) in the PROVENT study, there were no AZD7442-resistant substitutions observed in breakthrough SARS-CoV-2 illness visits. Conclusion. AZD7442 retained neutralization activity against all SARS-CoV-2 variants of concern or interest evaluated. Binding site substitutions identified in circulation, and in breakthrough SARS-CoV-2 infections following a single 300 mg dose of AZD7442 in the PROVENT study, were not associated with AZD7442 escape.