Genomic Sequencing Reveals Hidden Influenza A(H3N2) Subclade K School Outbreak in Burundi

In January 2026, health authorities in Burundi confronted a familiar and unsettling scenario. A secondary school in Bujumbura reported a sharp increase in respiratory illness among its students, and the immediate suspicion fell on SARS-CoV-2, the virus responsible for COVID-19. The prospect of a resurgence in a densely populated school setting was a serious concern for the national public health system. Yet when symptomatic students were tested for SARS-CoV-2, the results came back negative, leaving clinicians and epidemiologists with a cluster of unexplained severe acute respiratory infections and no identified cause.

That diagnostic dead end is precisely where the investigation described in a study published in BMC Infectious Diseases begins. A team led by Néhémie Nzoyikorera of the National Reference Laboratory at the National Public Health Institute in Burundi, working with colleagues at Erasmus University Medical Center in Rotterdam and partners across East Africa, Morocco, and Germany, applied targeted viral enrichment sequencing to nasopharyngeal samples collected from symptomatic students. The approach was designed to cast a wide net: rather than testing for one pathogen at a time, the method enriches and sequences viral genetic material present in a sample, allowing researchers to identify whatever viruses are there, known or unknown.

The technical foundation of the investigation rests on the Illumina Viral Surveillance Panel version 2, run on the compact iSeq 100 sequencing platform. Eight samples from SARS-CoV-2-negative students were processed through this targeted enrichment workflow. The panel uses probes designed to capture viral genomes across a broad range of respiratory and other viruses, concentrating scarce viral nucleic acid against a background of human DNA and other genetic noise. After sequencing, the reads were quality-checked, screened for viral identification, and then assembled into genome sequences where coverage permitted.

The results were decisive. Influenza A virus was detected in seven of the eight samples analyzed, and four of those samples yielded near-complete influenza A genomes of sufficient quality for detailed genomic characterization. When the researchers assigned clades to the reconstructed genomes, every characterized virus fell into the same category: Influenza A(H3N2) subclade K. This was the first genomic detection and characterization of that subclade in Burundi, a finding that carries significance well beyond the school gates because H3N2 subclade K has been drawing attention internationally as an emerging variant of the seasonal influenza virus.

Influenza A(H3N2) is one of the seasonal influenza A subtypes that circulates in human populations and is notorious for its capacity to evolve rapidly. The hemagglutinin protein, the HA of H3N2, is the primary target of protective antibodies, and mutations accumulating in the HA gene allow the virus to escape immunity built up through previous infections and vaccination. Clade and subclade designations, such as subclade K, are assigned on the basis of characteristic amino acid signatures in HA, and they help global surveillance laboratories track which genetic variants are spreading, which are vaccine-matched, and which signal a need for updated vaccine composition.

Phylogenetic analysis of the Burundian sequences added a layer of complexity to the outbreak picture. The sequences did not form a single tight cluster. Instead, they split into two distinct groups on the phylogenetic tree, suggesting that at least two separate viral lineages of H3N2 subclade K were circulating during the outbreak. This observation matters for outbreak interpretation. A single introduction followed by onward transmission would typically produce a compact cluster of closely related genomes, whereas the presence of divergent lineages points to multiple introductions or to undetected circulation in the community before the school cluster was recognized. Genomic analysis was therefore able to reveal structure in the outbreak that routine testing alone could not.

The researchers also examined the neuraminidase sequences for markers associated with resistance to neuraminidase inhibitors, the class of antiviral drugs that includes oseltamivir. No resistance-associated markers were detected, an important clinical finding because it indicates that antiviral treatment would be expected to remain effective against the circulating viruses if severe cases required it. The absence of resistance markers also provides reassurance for national treatment guidelines and for any future response to H3N2 subclade K in the region.

To ensure the findings contributed to the global picture, the influenza genome sequences were shared through GISAID, the Global Initiative on Sharing All Influenza Data, which serves as the central repository for influenza sequence data used by the World Health Organization and surveillance laboratories worldwide. Sharing genomes from Burundi addresses a longstanding gap in global influenza surveillance, which has historically been dominated by data from high-income countries in temperate regions. Sequences from African settings, particularly from outbreaks identified through unexplained severe acute respiratory infection surveillance, help clarify how subclade K and other variants move between regions and how they evolve in different population contexts.

The study also underscores the value of severe acute respiratory infection surveillance as a pillar of emerging disease preparedness. The samples in this investigation had been collected initially for SARI surveillance, and the national ethics framework, including approval by the National Ethics Committee of Burundi with a consent waiver, allowed their use for pathogen identification. When the initial hypothesis of a COVID-19 resurgence collapsed, the surveillance infrastructure and the stored samples provided the raw material for a rapid pivot to unbiased pathogen detection. The Burundi Public Health Emergency Operation Center supported the epidemiological investigations, and reagents were provided through the East African Community Secretariat’s Mobile Lab Project, illustrating how regional cooperation underpins laboratory response capacity.

The broader lesson from the Bujumbura school outbreak is that targeted viral enrichment sequencing is a practical tool for investigating acute respiratory outbreaks when the cause is unknown, particularly when initial testing is negative and routine testing for other respiratory viruses is limited. In settings where multiplex PCR panels are expensive or unavailable, a single sequencing run can identify the pathogen, characterize its genome, assign its clade, screen for resistance markers, and place it in a global phylogenetic context. The first genomic characterization of Influenza A(H3N2) subclade K in Burundi demonstrates that genomic surveillance capacity, strengthened through projects such as GREAT-LIFE at the national public health institute, can deliver actionable findings from a modest number of samples, and that the identification of distinct viral lineages within a single school outbreak highlights the value of genomic approaches to complement routine laboratory testing and support respiratory pathogen surveillance in Burundi and across the region.

Subject of Research: Genomic investigation of an Influenza A(H3N2) subclade K outbreak in a secondary school in Burundi, 2026

Article Title: Genomic investigation of an Influenza A(H3N2) subclade K outbreak in a secondary school in Burundi, 2026

Article References: Nzoyikorera, N., Schuele, L., Nieuwenhuijse, D. F., Nduwimana, C., Ndihokubwayo, A., Ihorimbere, T., Niyomwungere, D., Cassidy, H., Boter, M., Niyomwungere, A., Nizigiyimana, D., Uwineza, M. N., Diawara, I., Nzeyimana, E., Lagu, H. I., Gehre, F., Affara, M., Koopmans, M., Nyandwi, J., & Munnink, B. B. O. (2026). Genomic investigation of an Influenza A(H3N2) subclade K outbreak in a secondary school in Burundi, 2026. BMC Infectious Diseases. https://doi.org/10.1186/s12879-026-14678-5

Image Credits: AI Generated

DOI: 10.1186/s12879-026-14678-5

Keywords: Genomic, investigation, Influenza, H3N2, subclade, outbreak, secondary, school, Burundi, scientific research