Tracking BA.3.2, a Highly Divergent SARS-CoV-2 Variant, 2024–2026

This report describes the variant's spread as of February and March 2026.

New SARS-CoV-2 variants with changes in the spike protein keep appearing, and because the spike is the main target of the antibodies produced by infection and vaccination, those changes can affect how easily the virus spreads and escapes immunity. That is why COVID-19 vaccines are periodically updated. The Omicron variant (B.1.1.529), with about 32 spike mutations, began replacing earlier strains in December 2021 and caused a surge in cases and hospitalizations; the XBB (2022) and BA.2.86 (2023) variants later prompted vaccine updates.

BA.3.2 descends from BA.3, which circulated briefly alongside BA.1 and BA.2 in late 2021 and 2022. It carries about 70 to 75 substitutions and deletions in the spike gene compared with JN.1 and its descendant LP.8.1, the strains used in the 2025–26 COVID-19 vaccines, and in laboratory tests it escapes antibodies from current vaccines more readily.

How CDC watches for variants

CDC combines several kinds of genomic surveillance:

• Digital surveillance of sequences posted to open repositories such as GISAID and NCBI GenBank, plus media, GitHub, preprints and social media reports;
• National genomic surveillance of patient samples, to estimate how common each variant is;
• Traveler-Based Genomic Surveillance (TGS), which tests airplane wastewater and nasal swabs that volunteer international travelers collect themselves;
• Wastewater surveillance through the National Wastewater Surveillance System (about 1,300 U.S. sites) and WastewaterSCAN (150 sites).

A distinct new lineage

BA.3.2 is genetically distinct from the JN.1 lineages, including LP.8.1 and XFG, that have circulated in the United States since January 2024. Compared with LP.8.1, it has 20 differences in the spike's receptor-binding domain and 35 in its N-terminal domain, including deletions at sites 136–147 and 243–244 and a four-amino-acid insertion after site 214. Its spike differs from LP.8.1 by more than JN.1 differed from XBB.1.5, the 2023–2024 vaccine strain. BA.3.2 has split into two branches, BA.3.2.1 and BA.3.2.2, showing it is still evolving. U.S. sequences are scattered across its family tree, consistent with several separate introductions.

Spread around the world

• November 22, 2024: the first BA.3.2 sequence came from a respiratory sample in South Africa; investigators there isolated the virus from a 5-year-old boy.
• Spring 2025: detections followed in Mozambique (March 17), the Netherlands (April 12) and Germany (April 29).
• From September 2025: after months of scattered detections, reports rose, peaking in the week beginning December 7, 2025.
• November 2025–January 2026: BA.3.2 reached about 30% of sequences reported in Denmark, Germany and the Netherlands, though overall COVID-19 incidence there was not much higher than in earlier years.
• By February 11, 2026: it had been reported in at least 23 countries in Africa, Asia, Europe, North America and Oceania. Because many countries have limited sequencing, the true spread is likely wider.

The median time from sample collection to posting a sequence, for each country's first respiratory detection, was 18 days (range 3–123).

In the United States

• June 27, 2025: the first U.S. detection, in a traveler from the Netherlands tested through TGS at San Francisco International Airport. CDC cultured and characterized the virus and shared it with researchers.
• November 11, 2025: first detection in U.S. wastewater, in Rhode Island.
• December 4, 2025–January 4, 2026: the first three patient samples, from three states: two hospitalized older adults with other health conditions, one admitted for heart care four days earlier, and a young child treated as an outpatient. All survived.

As of February 11, 2026, BA.3.2 had turned up in nasal swabs from four travelers (returning from Japan, Kenya, the Netherlands and the United Kingdom), three airplane wastewater samples, samples from five patients in four states, and 132 wastewater samples from 25 states. It made up 0.19% of 2,579 sequences in national surveillance from December 1, 2025, to February 11, 2026. By March 12, the counts had grown to six travelers, 29 patients and 260 wastewater samples in 29 states and Puerto Rico, and 0.55% of 5,238 sequences.

In most states, wastewater picked up the variant many weeks before any patient sample did, making it an effective early warning. Few patient detections likely reflect the drop in sequencing since 2023. Finding BA.3.2 in hospitalized patients does not mean it causes more severe illness; those patients are more likely to be tested and sequenced, and the numbers are small.

Will it take over?

Earlier variants that escaped immunity replaced their predecessors: Omicron in late 2021 and early 2022, XBB lineages replacing BA.4/5 in late 2022 and early 2023, and JN.1, which grew out of BA.2.86 after a single L455S mutation, replacing XBB. No such replacement happened in 2024–25 or 2025–26.

In a laboratory study of seven variants, the 2025–2026 LP.8.1-based mRNA vaccine produced the weakest antibody neutralization against BA.3.2, which may reduce protection, though real-world data are still needed. BA.3.2's strong antibody evasion may explain its rise in northern Europe. Yet unlike earlier variants, it has not quickly overtaken the others; in several European countries it has circulated alongside JN.1 descendants at roughly 10% to 40%. Two laboratory studies found BA.3.2.1 and BA.3.2.2 bind less well to the ACE2 receptor and enter lung cells less readily than XFG and NB.1.8.1, which may hold it back. Seasonal increases or new mutations that restore that ability could let it spread more widely, but that had not been seen.

Why it matters

The 2025–2026 vaccines, an LP.8.1-adapted mRNA vaccine and a JN.1-adapted protein vaccine, protect against the variants that predominate in the United States. Immunity from infection and vaccination has reduced severe illness, but COVID-19 still caused an estimated 390,000 to 550,000 hospitalizations and 45,000 to 64,000 deaths in the 2024–25 respiratory season. Variants that can evade immunity could drive seasonal increases, so the authors call for continued genomic surveillance and real-world studies of how well vaccines and antivirals work against them.

The findings are limited by uneven international sequencing and reporting, delays between sample collection and sequence posting, the lack of standard methods for submitting sequences (especially from wastewater), the decline in U.S. sequencing, and the fact that BA.3.2's effect on illness severity is not yet known.

Sources

• Shakya M, Ma KC, Hughes LJ, et al. Early Detection and Surveillance of the SARS-CoV-2 Variant BA.3.2 — Worldwide, November 2024–February 2026. MMWR Morb Mortal Wkly Rep 75(10). https://www.cdc.gov/mmwr/volumes/75/wr/mm7510a1.htm
• CDC, preliminary estimates of COVID-19 burden for 2024–2025: https://www.cdc.gov/covid/php/surveillance/burden-estimates.html
• The report was written by CDC with university and wastewater surveillance partners; its figures are not reproduced here.