henhaobio
Sign in

The Lancet Infectious DiseasesImmunologyPeer-reviewed journal article

Antigenic and Virological Characteristics of SARS-CoV-2 Variant BA.3.2, XFG, and NB.1.8.1

Caiwan GuoiD, Yuanling YuiD, Jingyi LiuiD, Fanchong JianiD, Sijie YangiD, Weiliang Song, Lingling YuiD, Fei ShaoiD, Yunlong CaoiD

PDFDOIPreprint
In-depth full-text digest available · Jump to digest ↓

Abstract

The emergence of the SARS-CoV-2 saltation variant BA.3.2, which harbors over 50 mutations relative to its ancestral BA.3 lineage, has raised concerns about its potential to drive outbreaks similar to BA.2.86/JN.1. Concurrently, variants such as NB.1.8.1, LF.7.9, XEC.25.1, XFH, and XFG exhibit enhanced growth advantages over LP.8.1.1, necessitating a comparative analysis of their antigenic and virological characteristics. Here, we evaluated the infectivity, ACE2-binding efficiency, and immune evasion of these variants. Pseudovirus assays revealed BA.3.2s robust antibody evasion, including resistance to Class 1/4 monoclonal antibodies; however, its ACE2 engagement efficiency was markedly reduced due to a closed spike conformation, leading to low infectivity. While XFG and LF.7.9 demonstrated strong immune escape associated with A475V and N487D mutations, their reduced receptor-binding efficiency suggested a need for compensatory adaptations. In contrast, NB.1.8.1 retained high ACE2 affinity and humoral immune evasion, supporting its potential for future dominance. Collectively, BA.3.2s current profile limits its ability to compete with emerging variants like NB.1.8.1. Sustained monitoring of BA.3.2s evolution-- particularly for mutations stabilizing an open RBD conformation or enhancing escape from Class 1 antibodies--is essential to assess its outbreak potential.

In-depth summary

Claude Opus 5.5 · xhigh effort · 8 Nov 2026 · from the full text of version 1 (1 May 2025) · CC BY-NC 4.0

SARS-CoV-2 variant BA.3.2 strongly evades neutralizing antibodies but exhibits low infectivity due to a closed spike conformation, whereas NB.1.8.1 combines high ACE2 affinity with humoral immune escape.

The emergence of the saltation SARS-CoV-2 variant BA.3.2 (carrying >50 mutations relative to BA.3 and 44 relative to LP.8.1.1) alongside fast-growing JN.1 descendants raised concern over a potential BA.2.86/JN.1-like global wave. Comparative receptor-binding, infectivity, and neutralization profiling shows that BA.3.2 is currently constrained by poor spike-ACE2 engagement, whereas NB.1.8.1 achieves a balanced fitness profile.

Key results

  1. ShownBA.3.2 exhibits the strongest humoral immune evasion among tested variants in human convalescent plasma and mouse vaccine sera, while XFG, LF.7.9, and NB.1.8.1 show moderate escape relative to LP.8.1.1.

    In pseudovirus neutralization assays using human convalescent plasma from BA.5/BF.7-breakthrough cohorts reinfected with JN.1 (n = 29) or JN.1/XDV+F456L (n = 21), BA.3.2 reduced geometric mean 50% neutralizing titers (NT50) by 11-fold compared to BA.3 and 3- to 4-fold compared to LP.8.1.1, while XFG reduced NT50 nearly 2-fold and LF.7.9 and NB.1.8.1 reduced NT50 by 1.5- to 1.6-fold relative to LP.8.1.1. · Main Text; Patient recruitment and plasma isolation; Figure legends

  2. ShownBA.3.2 suffers from severely reduced spike-mediated human ACE2 engagement and low pseudovirus infectivity despite preserved monomeric RBD-ACE2 affinity, indicating a closed spike conformation.

    Soluble hACE2 inhibition of VSV pseudoviruses showed that full-length BA.3.2 spike had the lowest ACE2 engagement efficiency (significantly below BA.3 and LP.8.1.1) and markedly reduced infectivity in Vero cells relative to LP.8.1.1, whereas surface plasmon resonance (SPR) showed BA.3.2 RBD bound hACE2 with affinity comparable to LP.8.1.1. · Main Text; Surface Plasmon Resonance; Figure legends

  3. ShownNB.1.8.1 retains high ACE2 binding affinity and infectivity, whereas LF.7.9, XFG, and XFH exhibit impaired receptor engagement.

    Among all variants tested by SPR and soluble hACE2 pseudovirus inhibition (BA.3.2, NB.1.8.1, LF.7.9, XEC.25.1, XFH, XFG), NB.1.8.1 displayed the highest RBD-hACE2 binding affinity and maintained Vero-cell infectivity, whereas LF.7.9 (via A475V) and XFG (via N487D) had markedly reduced RBD-hACE2 affinity and XFH (via L335S) had reduced spike ACE2 engagement. · Main Text; Figure legends

  4. ShownBA.3.2 uniquely escapes broad-spectrum Class 1/4 RBD monoclonal antibodies, while convergent RBD mutations drive class-specific escape in competing variants.

    Against a panel of RBD-targeting neutralizing mAbs, BA.3.2 resisted Class 1/4 antibodies that remained potent against LF.7.9, XEC.25.1, XFG, XFH, and NB.1.8.1; Class 1 escape was driven by A475V in LF.7.9 and N487D/Q493E in XFG, Class 1/2 escape was enhanced by N487D (XFG), K478I (NB.1.8.1), and K478N (BA.3.2), and A435S reduced neutralization across all epitopes in XEC.25.1 and NB.1.8.1. · Main Text; Figure legends

Methods, data and reproducibility

Fitted logistic regression models to daily GISAID sequence frequencies (April 1, 2024 to April 11, 2025; 7-day generation time) to estimate relative growth advantages over LP.8.1.1. Measured RBD-hACE2 binding kinetics via Biacore 8K SPR, VSV pseudovirus infectivity in Vero cells, and neutralization IC50/NT50 in Huh-7 cells using soluble hACE2, RBD mAbs, vaccinated mouse sera, and human reinfection plasma.

  • Global daily SARS-CoV-2 variant sequence frequencies from GISAID (April 1, 2024 to April 11, 2025) · SARS-CoV-2 (virus)
  • Human convalescent plasma from individuals vaccinated with 2-3 doses of inactivated vaccine with BA.5/BF.7 breakthrough infection and reinfected with JN.1 or JN.1/XDV+F456L · human · n = n = 29 (JN.1 reinfection cohort) and n = 21 (JN.1/XDV+F456L reinfection cohort)
  • Sera from naive mice immunized with two doses of SARS-CoV-2 spike mRNA vaccines · mouse
Methods & tools
GISAID, Biacore 8K Control Software, Biacore 8K Evaluation Software, Racmacs, ggplot2
Reagents & environment
Biacore 8K Control Software (v4.0.8.20368), Biacore 8K Evaluation Software (v4.0.8.20368), Racmacs (v1.1.35), ggplot2 (v3.4.1)
Reproducibility gaps
No public repository links or accession identifiers for raw neutralization titers, SPR sensorgrams, or GISAID sequence sets are provided in the text.
Limitations (2)
  • Viral entry, receptor engagement, and neutralization were measured using VSV-based spike pseudoviruses in Vero and Huh-7 cell lines rather than authentic live SARS-CoV-2 isolates. · noted in this summary
  • Human convalescent plasma was sampled from Chinese cohorts primed with inactivated vaccines where Class 1/4 antibodies are prevalent and immune imprinting is less pronounced than in mRNA-vaccinated populations. · stated by the authors

Read this if you track SARS-CoV-2 variant evolution, antigenic cartography, or monoclonal antibody escape and need comparative data on BA.3.2, XFG, and NB.1.8.1.

Citations
103
FWCI
78.58
Field percentile
top 1%

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.

Related papers

Similarity Map
SelectedSame fieldCross-field
THISYang '24Liu '25Yang '23Li '24Yang '23Turner '26
81% semantic matchImmunology
S. Yang, Y. Yu, F. Jian+13 more·bioRxiv·Mar 7, 2024