BMC Public Health: 2018–2025 Serological study on avian influenza among occupationally exposed populations in the Poyang Lake area

 

#19,372

In 2025 a Chinese research study reported on testing environmental samples collected from southern and northern Chinese live poultry markets during 2019–2023 for evidence of avian influenza viruses. 

They examined > 64,000 samples, and found the incidence of H5, H7, and H9 influenza A detections substantially higher in southern provinces. Jiangxi, Yunnan, and Guangxi showed the highest level of environmental contamination, with positive rates of  82.3%, 49.41%, and 45.47%. 

Today we have a research study into the seroprevalence of  influenza A viruses among occupationally exposed individuals in the Poyang Lake area - located in Jiangxi province - which reported the highest environmental readings in the above report. 

Poyang Lake is the largest freshwater lake in China, and lies beneath the East Asian-Australasian Flyway. Poyang Lake also provides a habitat for more than a million migratory birds, including endangered Siberian cranes. 

Researchers tested blood from 2,437 poultry-exposed people around Poyang Lake, China, from 2018 to 2025 using hemagglutination inhibition assays. Predefined HI positivity thresholds were ≥ 1:20 for H3N8, H5N8, H5N1, H7N9, and H10N3 and ≥ 1:40 for H5N6 and H9N2.

Participants were mostly backyard poultry keepers and their households (90.3%) and live-market workers (9.7%). The median age was 62.  

While HI reactivity can't conclusively prove that an individual was previously infected by a specific virus (there's always a possibility of cross-reactivity with other viruses), it does suggest that a subject might have had a mild, or subclinical, infection with that virus sometime in the past. 

Admittedly, it would have been nice to see confirmatory Microneutralization (MN) testing of these samples, but they are considerably more resource intensive and technically demanding. HI titers also tend to fall faster than MN titers, and may drop below the positive cutoff within the span of a couple of years, which means that HI assays may miss older infections.  

With those limitations in mind, the authors found samples meeting the study's HI thresholds were 4.43% for H5N6 and 4.39% for H9N2, with lower rates for H7N9 (2.74%), H3N8 (1.44%), H5N8 (1.27%) and H10N3 (0.26%). 

First the link and abstract from the open-access study, after which I'll return with a postscript:

2018–2025 Serological study on avian influenza among occupationally exposed populations in the Poyang Lake area — the region with the highest environmental surveillance positive rate in China

Research Open access
Published: 14 October 2026 Cite this article
 
Download PDF
 
Chuang Nie, Yawei Li,Wenjing Xie, Ke Qian, Ting Xu, Fenglan He, Zhiqiang Deng, Wanting Wei, Yujie Hou, Xiang Tu, Tianxin Xiang & Wentao Song

We’re sharing this article early to provide faster access to peer-reviewed, accepted research. It is citable and carries a permanent DOI. This version is subject to further edits and will be replaced automatically by the final Version of Record. All legal disclaimers apply.

Abstract

Background

An avian influenza virus positivity rate as high as 82.3% has been reported in Jiangxi Province, China, underscoring this region as a potential hotspot for avian influenza virus circulation. As a crucial hub for migratory birds and a region where wild birds, domestic poultry, and humans frequently overlap, Poyang Lake provides favorable ecological conditions for cross-species transmission of avian influenza viruses. This study aimed to describe the seroprevalence of antibodies against selected avian influenza virus subtypes among occupationally exposed populations at surveillance sites around Poyang Lake from 2018 to 2025, and to provide data to inform ongoing surveillance and risk assessment.

Methods

A cross-sectional serological surveillance was conducted annually during winter and spring from 2018 to 2025 in Nanchang County (Nanchang) and Duchang and Wuning Counties (Jiujiang), Jiangxi Province. Serum samples were collected from occupationally exposed individuals, including backyard poultry keepers and their household members and live poultry market workers with direct poultry contact.

Hemagglutination inhibition assays were performed by the National Influenza Center to assess antibody reactivity against the test antigens representing H5N6, H9N2, H7N9, H3N8, H10N3, H5N8, and H5N1. Predefined HI positivity thresholds were ≥ 1:20 for H3N8, H5N8, H5N1, H7N9, and H10N3 and ≥ 1:40 for H5N6 and H9N2.

Results

A total of 2,437 valid serum samples were analyzed (1,599 from Jiujiang and 838 from Nanchang). Overall, 108 samples (4.43%) met the predefined HI positivity threshold for H5N6, and 107 (4.39%) met the threshold for H9N2. During 2018–2020, 25 of 914 samples (2.74%) met the threshold for H7N9.

During 2021–2025, 22 of 1,523 samples (1.44%) met the threshold for H3N8, including 9 of 100 samples (9.00%) collected in Nanchang in the 2024 surveillance year. Four of 1,523 samples (0.26%) met the threshold for H10N3; all four were from Jiujiang in the 2024 surveillance year. Four of 314 samples (1.27%) met the threshold for H5N8 in 2020. None of the 600 samples tested for H5N1 in the 2024 and 2025 surveillance years met the predefined threshold.

Conclusions

HI antibody reactivity varied across subtypes, surveillance locations, and years among occupationally exposed participants around Poyang Lake. The H3N8 finding in Nanchang and the H10N3 finding in Jiujiang in the 2024 surveillance year warrant further investigation. H5N6 peaks in Jiujiang in 2021 and in Nanchang in 2022; H9N2 peaks in 2021 in Jiujiang and 2023 in Nanchang. HI reactivity alone does not establish subtype-specific human infection or local virus circulation. Continued serological surveillance, interpreted alongside poultry and environmental virological data, may inform risk assessment.

        (Continue . . . ) 

We've previously looked at a number of human seroprevalence studies which have also suggested more human spillovers occur than have been reported, including:

• 2024's Nature Preprint: Serological Analysis in Humans in Malaysian Borneo Suggests Prior Exposure to H5 Avian Influenza despite no human  infections reported in the region. 

• 2018's J. Infect & Public Health: High Seroprevalence Of Avian Influenza H9 Among Poultry Professionals In Pakistan which found > 50% of 332 participants tested showed reactivity to H9.

• 2016's JID: High Seroprevalence Of H5N1 Antibodies In Indonesian Poultry Workers found 84% of sera tested positive

• In 2012, in H5N1 Seroprevalence Among Jiangsu Province Poultry Workers, we saw a study that found across three locations tested (Gaochun, Jianhu and Gaoyou counties) the percentage of workers testing positive ranged from zero (Gaochun) to 5.38% (95%CI, 2.19%–10.78%) in Gaoyou.

The good news is, this strongly suggests the 50% CFR (Case fatality rate) of H5N1 is badly overstated, since many mild infections are never counted.  

Conversely, it also suggests that novel flu infections may occur far more often than previously thought, which increases the risks of viral adaptation to a human host, and provides additional opportunities for novel/seasonal coinfection, which could lead to a reassortment. 

Either way, this is a reminder that a lot more goes on with the evolution and spread of novel flu viruses than our limited surveillance might suggest.

ISRV: Nomenclature Updates to the Hemagglutinin Gene Clade Designations Resulting From the Continued Evolution of High Pathogenicity Avian Influenza A(H5) Virus Clades 2.3.2.1c and 2.3.4.4

(Credit CDC)

#19,367

Thirteen years ago in 2013's EID Journal: The Expanding Variants Of H5N1, we looked at the continued evolution and diversification of H5N1 viruses around the globe.  At that time I wrote:

H5N1 virus was first identified in 1996 it has expanded into more than 20 different clades and subclades, and various versions of the virus now circulate in different parts of the world. You can see the evolution of the virus through 2011 in the chart below. 

NOTE: Not all of these clades continue to circulate.

 Clade 2.3.2 (and now 2.3.2.1) are very common in South East Asia, clades 2.2.1 and 2.2 are endemic in Egypt and clades 2.1.1, 2.1.2. and 2.1.3 are found in Indonesia.

This snapshot was taken before the emergence, and eventual world domination, of H5N8  clade 2.3.4.4, which would begin decimating South Korean poultry just two months after this was published, and of an H5N6 clade 2.3.4.4a virus which emerged in Sichuan, China a few months after that. 

In many ways, 2014 was a pivotal year for H5Nx viruses, with H5N1's dominance soon to be supplanted by a less deadly (to humans), but more easily spread by migratory birds, H5N8 clade 2.3.4.4x lineage.  

Older clades, such as Cambodia's 2.3.2.1c H5N1 virus went to ground (but would return roughly a decade later).  H5N8 dominated until 2020-2021, when an H5N1 clade 2.3.4.4b came roaring back, largely supplanting H5N8. 

In the spring of 2025, Cambodia's clade 2.3.2.1c was provisionally redesignated clade 2.3.2.1e  after analysis showed it had drifted away from the older clade. 

Similar evolutionary shifts have been happening in other clades over the past dozen years, which is why the WHO/WOAH/FAO H5 Evolution Working Group has updated the naming system for H5 bird flu viruses for the first time since their 2014 update.

Today's report from Influenza and Other Respiratory Viruses updates the official nomenclature for H5 viruses up to July 4, 2024.  

The working group added 12 new subclade designations:

• Clade 2.3.2.1c was split into 2.3.2.1c–g. 
• 2.3.2.1d: China, 2014–2016.
• 2.3.2.1e: Vietnam and Cambodia, 2020–2024. 
• 2.3.2.1f: India, the Middle East and Africa.
• 2.3.2.1g: Indonesia, 2012–2016.

• Clade 2.3.4.4 was split into 2.3.4.4a–h 
• plus a "2.3.4.4-like" label for 125 outlier viruses that don't fit neatly

Some of these designations have been used provisionally for years, but are now made official. 

Although there was enough divergence among clade 2.3.4.4b viruses to technically warrant additional subclade splits, that decision has been deferred for now:  The Authors explained:

Furthermore, several of the newly designated 2.3.4.4 subclades had intra-clade APDs over the 1.5% threshold required for further splitting of clades due to the rapid expansion of these clades into new regions (e.g., 1.98% intra-clade APD of 2.3.4.4b representing over 13,000 sequences). Due to sampling biases stemming from the overrepresentation of sequences from specific outbreaks and insufficient data from regions with recent introductions, further subdivision of subclades with intra-clade APDs greater than 1.5% was deferred. Additional genetic data are needed to better assess the endemicity of specific clusters and justify further splits beyond the current subclades 2.3.4.4a–h.

Due to the highly technical and niche nature of today's study, I've only posted the link and abstract. Those wanting a deeper dive will want to follow the link to read it in its entirety. 

I'll have a brief postscript after the break.

Nomenclature Updates to the Hemagglutinin Gene Clade Designations Resulting From the Continued Evolution of High Pathogenicity Avian Influenza A(H5) Virus Clades 2.3.2.1c and 2.3.4.4

Tommy T. Lam, C. Todd Davis … See all authors 

First published: 07 October 2026 https://doi.org/10.1111/irv.70317

Group members and collaborators are provided in the Supporting Information (Data S1). 

 

ABSTRACT

Background and Methods

The evolutionary divergence of the A(H5) hemagglutinin (HA) gene of high pathogenicity avian influenza (HPAI) viruses (A/goose/Guangdong/96 lineage) was analyzed by phylogenetic and average pairwise distance methods to identify clades that merit nomenclature updates.

Results and Conclusions

Based on this assessment, 12 new clade designations were recommended based on division of clade 2.3.2.1c and 2.3.4.4 viruses, which were reported in Africa, Antarctica, Asia, Europe, the Middle East, the Americas, and Oceania since the most recent WHO/WOAH/FAO H5 Evolution Working Group update.

        (Continue . . . )

H5's evolution continues, and we will undoubtedly see additional `provisional' clade and subclade designations added in the years ahead.  

The bottom line is that H5 viruses have not only grown more diverse, they are a far cry from what emerged in China in the 1990s, and sparked pandemic fears in the early 2000s. 

Whether any of that gives them the ability to spark a human pandemic remains to be seen.  But it would be reckless to assume that since their ancestors didn't, their descendants never will.  

Poultry Sci & Mgt: Partial Protection of Commercial H5 Avian Influenza Vaccines to clade 2.3.4·4b H5N1 and H5N8 Avian Influenza Infections in Chickens

Poultry Vaccination - Photo Credit OIE

#19,327

Nearly 14 years ago, in Egypt: A Paltry Poultry Vaccine, we looked at an analysis by the researchers from the Virology department at St. Jude Children’s Research Hospital - that studied the effectiveness of six commercially available H5 poultry vaccines then deployed in Egypt.

They reported - out of the 6 vaccines tested - only one (based on a locally acquired H5N1 seed virus) actually appeared to offer genuine protection.

This was at a time when H5 vaccination was only utilized by a handful of countries (primarily China, Indonesia, Egypt, and Vietnam), and the OIE (now WOAH) routinely warned that vaccination of poultry `. . . cannot be considered a long-term solution to combating the avian flu virus'.

The authors of this study urged: 

In light of our findings, we recommend that the H5N1 prevention and control strategy in Egypt be updated and reinforced. Special consideration should be given to the vaccination strategy, and the use of vaccines based on currently circulating viruses is advisable.

Fast-forward 6 years (2018) and the St. Jude research team revisited Egypt's poultry vaccination program (see Sci. Reports: Efficacy Of AI Vaccines Against The H5N8 Virus in Egypt), and found many of the same problems persisted. 

They wrote:

Our results indicate that most of the commercial poultry H5 vaccines used in the present study were ineffective because the seed viruses in these vaccines are genetically distinct from the H5N8 viruses currently circulating in Egypt.

Although some of the commercial vaccines protected chickens from mortality, they failed to prevent chickens from shedding the virus. Accordingly, we recommend updating and reinforcing the H5N8 prevention and control strategies in Egypt. The vaccination strategy should be reconsidered based on currently circulating viruses.

While the reduced mortality among poultry may have seemed like a `win' for farmers, theses avian flu viruses could continue to spread - often unnoticed - due to the `masking' of symptoms by the vaccine.

Of course, Egypt wasn't alone in the continued use of older, mismatched, poultry vaccines.  Earlier in 2018, in PLoS One: Effectiveness of HPAI H5N1 Vaccination in Poultry - Indonesia, we looked at an assessment of that country's poultry vaccination campaign. 

HPAI vaccination, intensively applied in Sector 3 layers in Indonesia, had highly variable outcome, including vaccination failures and did not provide sufficiently long protective immunity in the majority of flocks. Indonesia adopted HPAI vaccination in 2004 with the aim of reducing the incidence of H5N1 infections in poultry, with the ultimate objective of achieving eradication of the virus.

The concern is that vaccine mismatches, and/or poor or inconsistent application, can allow avian flu to spread silently among flocks, and may allow AI viruses to continue to reassort and evolve, potentially leading to the emergence of new variants of avian flu.

 A few earlier blogs addressing those concerns include:

Subclinical Highly Pathogenic Avian Influenza Virus Infection among Vaccinated Chickens, China.

Study: Recombinant H5N2 Avian Influenza Virus Strains In Vaccinated Chickens

EID Journal: Subclinical HPAI In Vaccinated Poultry – China

While there have certainly been successful avian flu vaccination campaigns (see EID Journal: China's H5+H7 Poultry Vaccination Program, Guangdong 2017-18), vaccines must be continually updated - and flocks tested for breakthrough infections - in order to remain successful.  

In 2025, in NPJ Vaccines: Impact of Inactivated Vaccine on Transmission and Evolution of H9N2 Avian Influenza Virus in Chickens, we saw evidence that not only had inactivated vaccines failed to prevent - or even reduce - LPAI H9N2 in China's poultry, they may have driven viral evolution (including mammalian adaptations).

None of this is to suggest that an effective poultry vaccine campaign can't be mounted, or effectively maintained.  But it does require more than just a `vaccinate & forget' strategy. 

All of which brings us to a new study which finds that despite the introduction of newer vaccine formulations, many of the earlier concerns about Egypt's vaccination program persist.  I've only posted the Abstract, and some excerpts, so follow the link to read it in its entirety. 

I'll have a bit more after the break. 

Partial protection of commercial H5 avian influenza vaccines to clade 2.3.4·4b H5N1 and H5N8 avian influenza infections in chickens

Research
Open access
Published: 08 September 2026
Volume 3, article number 22 (2026)

Download PDF

Ahmed El Taweel, Mokhtar Gomaa, Yassmin Moatasim, Omnia Kutkat, Mina Nabil Kamel, Mohamed El Sayes, Mohamed GabAllah, Mohamed Ahmed Ali, Ghazi Kayali & Rabeh El-Shesheny

Abstract

Background

Avian influenza viruses (AIVs) continue to cause substantial economic losses to the poultry industry and pose a significant burden on global public health systems. The increasing diversity and geographical spread of different subtypes of influenza viruses in poultry populations increases the risk for reassortment events and the emergence of novel strains. Vaccination remains a key strategy for preventing influenza infection and its complications, especially high pathogenic avian influenza (HPAI), as it can reduce morbidity, mortality, and viral transmission.

However, vaccine effectiveness is closely tied to the antigenic match between vaccine strains and circulating viruses. In this study, we tested six commonly used commercial poultry H5 vaccines in Egypt to evaluate mortality, and virus shedding following experimental infection of specific pathogen free chickens with clade 2.3.4·4b H5N1 and H5N8 HPAI viruses.

Results

Our results showed that most of the tested vaccines provided detectable antibody titers and protected against mortality but were not able to control virus shedding. ValleyVac vaccines induced high HI titers (up to 9 log2) against recent clade 2.3.4·4b A(H5N1) and A(H5N8) strains, while Zoetis and Egy-flu R 2-in-1 induced lower HI titers (<6 log2). Although some of the commercial vaccines did not elicit considerable HI titers, vaccines provided 71%–100% protection against clade 2.3.4·4b viruses. Importantly, virus titers were detected till seven days post-infection (dpi). This incomplete protection is likely due to genetic and antigenic mismatches between the vaccine seed strains and the H5N1 and H5N8 viruses detected in Egypt.

Conclusions

These results highlight the urgent need to revise H5Nx prevention and control strategies in Egypt and update vaccine seed strains to reflect currently circulating HPAI viruses.

(SNIP)

The prolonged use of suboptimal vaccines or inadequate vaccination programs may contribute to the selection of antigenic escape mutants through immune pressure, potentially reducing vaccine effectiveness.

        (SNIP)

In summary, our data indicate that vaccines designed on strains that differ from circulating strains are not able to provide optimal protection. Although an immune response was elicited and survival rates were high, viral shedding continued. This indicates that the use of those vaccines will reduce morbidity and mortality but will allow virus transmission and potentially the emergence of vaccine escape mutants.

Mass vaccination remains a crucial component of a multifaceted approach aimed at preventing AIV illness and its spread in poultry. The ongoing threat of an AIV pandemic highlights the need to design novel vaccines that are easier to manufacture, more streamlined, and capable of inducing better and broader immune protection against emerging strains.

 

        (Continue . . . )

 

These, and other (often trade related) concerns - are why many countries (including the United States, Canada and the UK) have yet to authorize the use of HPAI poultry vaccines, although many are studying the matter (see UK Defra Announces New Avian Influenza Vaccine Trials Begin in UK).

But as HPAI increases its presence in the environment -  and continues to spill over into livestock - the need for well mounted vaccination campaigns only increases. 

We've discussed some of the ongoing biosecurity measures that would be required to mount a safe and effective poultry vaccination campaign (see UK Joint Taskforce Policy Paper: Vaccination of Birds Against HPAIV (bird flu)), including:

• Most captive birds would likely require more than one vaccination over their lifetime, and regular testing that can differentiate infected from vaccinated animals (DIVA) would be needed to prevent `subclinical spread' of the virus. 
•  Culling and/or quarantine would still be needed for `breakthrough' infections
• A different vaccine, schedule and testing regimen would likely be required for non-avian livestock.
• And vaccine seed viruses would have to be regularly updated to match circulating strains

None of this would be cheap or easy, but - given the growing endemicity of the H5Nx virus - vaccination may be our best option going forward.

But only if we take the time and effort to do it right. 

Pathogen's Progress

H5 Affected States & Provinces June 2015

#19,326

Prior to 2021, North America had only had one brush with the Asian-origin HPAI H5 virus; a reassorted H5N8 subtype which appeared in South Korea in January of 2014 before beginning a world tour. 

It crossed the north Pacific - carried by migratory birds - the following fall, arriving in the Pacific Northwest in December 2014, sparking the first HPAI H5 epizootic in the United States and Canada (see map above).

That H5Nx virus, however, disappeared completely the following summer, apparently still ill-suited for long-term carriage by wild and migratory birds (see PNAS: The Enigma Of Disappearing HPAI H5 In North American Migratory Waterfowl).

That limitation would evaporate in the summer of 2016 when H5N8 reassorted into a far more robust threat (probably in China or Siberia) - arriving in Europe in October - and sparking their biggest (at that time) epizootic on record. EID Journal: Reassorted HPAI H5N8 Clade 2.3.4.4. - Germany 2016). 

This new H5N8 was not only deadlier to some bird species, it was more easily carried by others, increasing its range and impact dramatically over the next 6 months.  

We also saw this virus spin off several short-lived subtypes, including H5N5, H5N3, and H5N9. 

Over the winter of 2016-17, this H5N8 virus spread from Europe, into the Middle East, and then North Africa. By late spring, 2017, it had crossed the equator for the first time,  and set up shop in South Africa, making it the most successfully disseminated HPAI H5 virus we'd seen.

A European H5N6 briefly appeared in 2018, but H5N8 held on until a new, reassorted H5N1 virus appeared in Europe in 2020, and began to supplant H5N8. 

Unlike previous H5Nx incarnations, this new H5N1 showed an increased affinity for infecting mammals, and a much wider avian host range (see DEFRA: The Unprecedented `Order Shift' In Wild Bird H5N1 Positives In Europe & The UK).

By 2022 this `new and improved' H5N1 virus had crossed both the Atlantic and Pacific oceans, spreading first to North America, followed by South America, the Antarctic, and most recently (2026) into Australia/New Zealand.   

While avian flu activity still decreased during the hotter months, no longer did HPAI H5 disappear completely each summer. It could be maintained (at low levels) in local birds, livestock, small mammals, and likely even the environment. 

All of which makes the HPAI H5 threat of today far different from what it was even a decade ago. The H5N1 virus of today has expanded not only its geographic range, but its host range as well.  

And it isn't done evolving. 

Since its arrival in 2021, the H5Nx virus has diversified into more than 100 distinct genotypes in North America alone, with scores of others emerging around the globe.  A few notable North American examples include:

• In late 2023 a new B3.13 genotype emerged that could efficiently infect dairy cattle while mildly infecting dozens of humans

• Last fall we saw 3 new genotypes emerge (D1.1, D1.2, D1.3), with D1.1 producing more severe illness in humans. 

• In October 2024 Canada experienced a multi-farm outbreak of emerging N1 genotype of H5N1 that was highly resistant to NAI antivirals (carrying the H275Y mutation)

These variants illustrate just how quickly (and dramatically) the avian `fluscape' can change, and we generally see those changes in the fall, when migratory birds - which spent their summers in their high latitude roosting areas - return. 

A 2016 study (see Sci Repts.: Southward Autumn Migration Of Waterfowl Facilitates Transmission Of HPAI H5N1), suggests that waterfowl pick up new HPAI viruses in the spring (likely from poultry or terrestrial birds) on their northbound trip to their summer breeding spots - where they spread and potentially evolve - and then redistribute them on their southbound journey the following fall.

Our own Arctic Refuge, where more than 200 bird species spend their summers, serves as a central hub, and funnels migratory birds south via all four North American Flyways.

While this fall's HPAI viruses could be similar to what we saw last year - they might include new, potentially game-changing, genotypes - which could either increase or decrease their threat.

Evolution is a double-edged sword, and what it adds it can also take away. But the trajectory of H5N1 over the past 4+ years has been towards more mammalian adaptation, and increased host range.  

We've also seen reports of H5N5 activity above the Arctic Circle (see Norwegian Veterinary Institute Reports Another HPAI H5N5 Infected Polar Bear on Svalbard Island), and H5N5 claimed the life of a Washington state bird keeper last fall. 

The APPA estimates `Eleven (11) million U.S. households own backyard chickens (a 28% increase from 2023)', yet a recent survey (see MMWR: Knowledge, Attitudes, and Practices Regarding Avian Influenza Among Owners of Backyard Flocks) found significant gaps in their knowledge of dealing with avian flu.

While no one can predict with any confidence what HPAI will look like later this fall, the one thing we can be sure of is the 2026 fall southbound migration has already begun, and the time to prepare for its arrival grows shorter with every passing day. 

And if recent history is any guide, we should be preparing for additional surprises. 

Norwegian Veterinary Institute Reports Another HPAI H5N5 Infected Polar Bear on Svalbard Island

 

 

Credit: Multiple transatlantic incursions of HPAI 

clade 2.3.4.4b A(H5N5) virus into North America 

and spillover to mammals

#19,310

Although most of the world is primarily concerned with HPAI H5N1, in northern Europe, Iceland, and parts of Canada, a different avian flu subtype - HPAI H5N5 - has been making inroads as well. 

Additionally, here in the United States HPAI H5N5 infected and killed a resident of Washington State last November, following contact with backyard birds.

In the summer of 2022, the Norwegian Veterinary Institute reported both H5N1 and H5N5 for the first time in wild birds on Svalbard, which lies above the Arctic circle (see More HPAI (H5N5 & H5N1) Detected In Arctic (Svalbard).

Last May, in Norway Veterinary Institute Reports HPAI H5N5 In Polar Bears on Svalbard Island, we looked at a report of the recent detection of HPAI H5N5 virus in both a dead walrus and a polar bear on Svalbard Island.

Although no other animals were confirmed infected during this investigation, there were eye witness accounts of two other polar bears exhibiting potential neurological symptoms.

Five weeks ago, in Preprint: HPAI H5N5 in a Polar Bear and Atlantic Walrus, Svalbard, 2026, with Widespread Seroconversion in Polar Bears, we saw a report indicating that H5N5 exposure is far more common among polar bears on Svalbard island than previously thought.

Somewhat surprisingly, most polar bears appear to survive H5N5 infection, although severe (and likely neurotropic) illness can occur.

Yesterday the Norwegian Veterinary Institute reported on a second polar bear - recovered in June some 23 km distant from the one found in May - has tested positive for H5N5. 

(translation)
Bird flu detected in another polar bear in Svalbard
Published 25.08.2026

The Norwegian Veterinary Institute has detected avian influenza in a young polar bear found dead on the northern tip of Spitsbergen. This is the second time the virus has been detected in polar bears in Norway and Europe.

The detected highly pathogenic avian influenza (HPAI) virus is of the subtype H5N5, a type of the virus that has been previously detected in Svalbard: In wild birds for the first time in 2022 , in a walrus in 2023 , in arctic foxes in 2025 , and in a polar bear and walrus in the spring of 2026. This latest detection is the second in polar bears in Norway and Europe.

Highly pathogenic avian influenza viruses have also previously been detected in carnivorous mammals such as red foxes, otters and lynxes on mainland Norway after infection from wild birds.

Sampled in the field by the District Governor

The polar bear, a young individual, was found dead on Æøya, near Virgohamna. The animal was sampled by the Governor in June 2026, after avian influenza virus was detected in a dead polar bear and a dead walrus in Raudfjorden in May 2026. The distance between the two findings of dead animals is approximately 23 kilometers.

- After the findings on the polar bear and walrus in Raudfjorden, we also recommended sampling the polar bear that was on Æøya, says veterinarian and wildlife health specialist at the Veterinary Institute Knut Madslien.

The samples from the polar bear were then sent to the Norwegian Veterinary Institute and tested for avian influenza virus and rabies. Highly pathogenic avian influenza virus was detected in samples from both the brain and nose. Rabies virus was not detected.

Bird flu virus is circulating in the highlands

"The findings are part of a trend where highly pathogenic avian influenza virus is increasingly being detected in mammals in Europe. At the same time, the virus has spread to new areas in recent years, including the Arctic, where it may have consequences for vulnerable populations and ecosystems," explains Ragnhild Tønnessen, avian influenza coordinator at the Norwegian Veterinary Institute.

- Our researchers, in close collaboration with the Norwegian Polar Institute and other partners, have found that a large proportion of the polar bears examined in Svalbard after 2022 have been exposed to previous avian influenza infection. Although we currently have limited knowledge of the consequences this has for the polar bears, the results indicate that the virus is circulating in the population and that exposure is more widespread than detections in dead animals alone would indicate, says Tønnessen.

Increased infection pressure in mammals

Most avian influenza viruses are best adapted to be transmitted between birds. However, mammals can sometimes be infected through direct contact with sick or dead birds or other mammals infected with the virus.

The Veterinary Institute will conduct further investigations of the detected virus to characterize it more closely and clarify whether it shows signs of being specifically adapted to mammals.

Second detection in polar bears in Europe

In other predators, such as red foxes and lynxes, avian influenza can cause clinical signs of brain disease. Neurological signs such as circling gait, tilted head position, paralysis and reduced shyness towards humans are common.

Avian influenza virus in polar bears has been scientifically published twice previously: In a young male polar bear found dead in Alaska in August 2023 and in a young male polar bear found dead in Raudfjorden on Svalbard in May 2026. However, it is not known how the virus affects individual individuals or the population of polar bears.

– This finding strengthens the indications that highly pathogenic avian influenza virus can also affect polar bears, and that young individuals are particularly vulnerable. This is also consistent with findings in other carnivorous mammals, where the virus mainly affects younger individuals. However, the knowledge base is still very limited and it is important to gain more knowledge about how the virus can affect polar bears, says Madslien.

Report if you observe dead or sick animals and birds

Both highly pathogenic avian influenza and rabies are serious diseases found in Svalbard and can be transmitted to humans. Contact with sick and dead animals should therefore be avoided.

If sick or dead animals are observed on Svalbard, it is important that the findings are reported to the Governor . This is important to obtain an overview of disease incidence and mortality, although it does not always trigger sampling. If avian influenza is suspected in birds and other animals on the mainland, the Norwegian Food Safety Authority must be notified .

The Norwegian Veterinary Institute is the national reference laboratory for avian influenza and has molecular methods for the detection and characterization of avian influenza viruses.

Admittedly, HPAI H5N5 remains a minor player in the world of avian flu, but it has shown remarkable tenacity, persisting and even thriving in some high latitude regions of the Northern Hemisphere. 

A recent study (see J. Virology: Risk assessment of avian influenza A(H5N5) virus from the first human case using the ferret model) found that while the Washington State isolate replicated robustly in mammalian cells and caused severe disease, it still lacked adaptations supporting efficient airborne spread. 

Six years ago HPAI H5N8 subtype dominated the world's avian flyways, only to be abruptly supplanted by a new and improved H5N1 in 2021. 

A reminder of just how quickly the avian flu-scape can change, and why we continue to watch this subtype with interest. 

Virology: Acquisition of specific human respiratory tract binding by 2.3.4.4b H5N1 hemagglutinins requires multiple mutations

 

#19,288

There remains much interest in exactly what changes the HPAI H5Nx virus would need in order to become a more `humanized' virus - like seasonal H3N2 and H1N1 - which have a strong affinity for the alpha 2,6 receptor cells most commonly found in the human respiratory system.

While it seems likely that it would require multiple concurrent mammalian adaptations to create a pandemic strain, the ability of avian flu viruses to bind to human α2-6 receptor cells is considered one of the biggest obstacles the virus must overcome in order to successfully spread in humans.

Complicating matters, as we've discussed previously (see here, here, and here), HPAI H5 viruses encompass a large and growing array of closely related - but genetically distinct subclades, subtypes, and genotypes - and a mutation that might impact one strain might not have the same effect on another. 

In December of 2024, a study in Science: A Single Mutation in Bovine Influenza H5N1 Hemagglutinin Switches Specificity to Human Receptors, suggested a single HA amino acid change (Q226L) could switch H5N1 B3.13 from binding preferentially to avian-type receptor cells to human-type receptor cells.

The authors wrote:

“Our experiments revealed that the Q226L mutation could significantly increase the virus’ ability to target and attach to human-type receptors,” explains Paulson. “This mutation gives the virus a foothold on human cells that it didn’t have before, which is why this finding is a red flag for possible adaptation to people.”

The shift alone, however, may not be enough to enable human-to-human transmission. Other genetic changes—such as mutations in polymerase basic 2 (E627K) that enhance viral replication and stability in human cells—would likely be necessary for the virus to spread efficiently among people.

The following month, in Preprint: The Q226L Mutation Can Convert a Highly Pathogenic H5 2.3.4.4e Virus to Bind Human-type Receptors, another study (later published in PNAS) found that adding Q226L to an older clade 2.3.4.4e H5N6 virus could switch receptor binding from avian to human-type receptors. 

While concerning, we've not seen any evidence of efficient or sustained spread of H5N1 in humans, suggesting there are additional enhancements required. 

Today we've a new study that offers an update on impact of the Q226L mutation in H5N1 B3.13, which finds that its impact varies between HPAI H5 strains. 

While Q226L did shift binding in laboratory tests towards human-type glycans, those glycans were not found on the human tracheal tissues tested, and the Q226L mutant failed to bind human trachea.

The addition of another HA mutation, N224K, increased human-tracheal binding, but only sparsely and mainly to goblet cells. All of which reinforces the notion that  receptor adaptation to humans in HPAI H5 is neither simple nor uniform.

Q226L appears to be decisive in some H5 strains, but the current bovine 2.3.4.4b H5N1 virus likely requires additional changes before it can efficiently bind to human receptors in the upper airway.  

I've only posted the link, abstract, and some excerpts from the study. Follow the link to read it in its entirety. 

Acquisition of specific human respiratory tract binding by 2.3.4.4b H5N1 hemagglutinins requires multiple mutations

Authors: María Ríos Carrasco, Mafalda F. Guerreiro Cabana, Eszter Kovács, Zoé Ducarne, Cindy G. J. Cleypool , Geert-Jan Boons, Robert P. de Vries  r.vries@uu.nlAuthors Info & Affiliations

https://doi.org/10.1128/jvi.01065-26

ABSTRACT

It has been suggested that the hemagglutinin of the human-infecting cattle-derived 2.3.4.4b virus A/Texas/34 H5N1 (H5TX) requires only one mutation, namely Q226L, to switch from binding avian-type to human-type receptor preference. In this study, we examined the binding of H5TX Q226L, along with other key mutations, to sections of human trachea. We conclude that, while H5TX Q226L can bind human-type receptors, more than a single mutation is required for this protein to bind to human respiratory tract tissue.

We also report changes in receptor-binding specificity of another 2.3.4.4b HA mutant, H5FR Q226L (from A/duck/France/161108/16 H5N8), associated with the presence of a multibasic cleavage site. This study offers insight into the determinants of evolution toward human-type receptor binding in currently circulating H5Nx viruses. It also emphasizes the importance of testing individual strains using additional methods, including tissue-based approaches, alongside synthetic glycans.

IMPORTANCE

Currently, H5N1 influenza A viruses are responsible for numerous zoonotic spillover events, from infecting birds to other mammals, including dairy cattle. Although no human-to-human transmission has been observed, several people have been infected. This host range expansion is typically linked to changes in one of the viral surface proteins, hemagglutinin, which can switch its preference from avian-type to human-type receptors. To better understand the potential of the currently circulating H5N1 virus to transmit among humans, we evaluated the effects of the Q226L mutation, in combination with other amino acid substitutions, on binding to the human trachea. We also studied the effect of the multibasic cleavage site, a specific motif present in highly pathogenic influenza strains, on receptor-binding properties. These findings provide insight into the role of receptor binding in influenza infections.

        (SNIP)

We conclude that more than a single mutation is required for the hemagglutinin of the currently circulating 2.3.4.4b viruses to bind human respiratory tissue. This aligns with previous work showing that multiple amino acid changes are required to switch from avian- to human-type receptors in H5 strains (3). One key takeaway from this study is the significance of glycan architecture. H5TX mutants, similarly to other HAs, distinguish not only between SIA linkage but are also selective according to glycan branch length and sialylation (31).

More research is needed on the presence of asymmetrical glycans in the human respiratory tract and their role in influenza A virus hemagglutinin attachment to cells. Moreover, we emphasize the importance of not underestimating the MBCS’s influence on hemagglutinin’s receptor-binding properties.

       (Continue . . . )

Preprint: The canine respiratory epithelium is a permissive ecosystem for influenza interspecies transmission and emergence

 

#19,196

Until 2004, dogs were considered relatively immune to influenza, but that changed abruptly when an equine H3N8 virus spilled over to dogs at a Florida racetrack, and began its world tour.

Three years later, an avian H3N2 virus spilled over to dogs in South Korea, and spread across Asia, eventually arriving in the United States in 2015.

In 2010, in Morens & Taubenberger on Influenza’s History, we looked a a fascinating 11-page historical review of Influenza outbreaks amongst a variety of hosts (human, avian, equine, porcine, canine, etc.) going back more than 3,000 years by Jeffrey K. Taubenberger and David Morens.

They cited a number of pandemics where anecdotal accounts mentioned dogs falling ill, either before - or concurrent to - when human cases emerged.

Since then we've looked at a lot of studies on dogs and flu, including: 

• In late 2012, in China: Avian-Origin Canine H3N2 Prevalence In Farmed Dogs, we looked at a study that found a newly emergent avian origin H3N2 virus prevalent in pet and farmed dog populations in southern China.

• In 2015, in Korea Finds More Dogs With H5N8 Antibodies, we saw evidence of HPAI infection in dogs. 

• Also in 2014, in Study: Dogs As Potential `Mixing Vessels’ For Influenza - we looked at the ability of different influenza strains (canine, equine and human) to infect, and replicate in, canine tracheal tissues. 

• In 2015, researchers reported finding Influenza A(H6N1) In Dogs, Taiwan

• Again in 2015, in Virology J: Human-like H3N2 Influenza Viruses In Dogs - Guangxi, China, we looked at the discovery two H3N2 CIVs possessing high homology with human/swine influenza viruses.

• In 2023 we looked at a lengthy (57-page) preprint called Increased Public Health Threat of Avian-origin H3N2 Influenza Virus During Evolution in Dogs from researchers in China who warned on the continued evolution of canine H3N2, which appears to be moving towards becoming a more `humanized' virus.

• More recently (2025) we looked at A Brief History of Influenza A in Canines & Canadian Report On A Recent Fatal H5N1 Dog Infection.

While dogs are more apt to have less severe flu illness than cats - making infections easier to miss - there is little doubt they are susceptible.  Unknown, however, is whether they are mostly a dead-end host, or if they are capable to transmitting the virus to others. 

Today we've got a preprint from researchers in the UK which attempts to better understand the risks. Using canine lung explants, they find that dogs are a plausible host for influenza reassortment, as they contain both human-type (α2,6) and avian-type (α2,3) receptor cells.

The caveats being that this is an ex vivo study, and is subject to a number of limitations (see below), and there is currently no evidence that novel influenza viruses are spreading efficiently among canines. 

But the only constant with influenza viruses is that they change, making today's status tenuous at best.  Due to its length, I've only posted some excerpts. Follow the link to read the report in its entirety.  

The canine respiratory epithelium is a permissive ecosystem for influenza interspecies transmission and emergence
Hanting Chen, Jack Hassard, Jiayun Yang, Callum Magill, Toby Carter, Jean-Remy Sadeyen, Aimi Ito, Clio Duerr,  Hannah Rose Montgomery, Savitha Raveendran, Kieran Dee,  Maximillian N J Woodall, Grace B Tyson, Maria M Afonso, Verena Schultz,  Claire Mary Smith,  Margaret J Hosie, Stuart Haslam, Munir J Iqbal,  Pablo R Murcia

doi: https://doi.org/10.64898/2026.06.04.730051
This article is a preprint and has not been certified by peer review 

Preview PDF

Abstract

The outcome of virus spillover ranges from dead-end infections to pandemics and is underpinned by host-pathogen interactions as well as evolutionary and epidemiological processes. The emergence of novel influenza A viruses (IAVs) has been associated with reassortment events involving multiple species, highlighting the importance of reservoir and intermediate hosts in viral emergence.

Highly pathogenic H5N1 IAVs of the 2.3.4.4b genotype have caused a panzootic affecting a broad range of mammals. The role of dogs -arguably the most popular companion animal and a natural host of IAVs- in the ecology of IAVs under this new zooepidemiological scenario is unknown. To address this, we characterised the glycome of the dog respiratory epithelium, infected canine tracheal explants with multiple IAVs (including canine H3N2 and H3N8, equine H3N8, avian H3N8 and H5N1, swine H1N1, human H1N1 and H3N2, and bovine H5N1 viruses), and determined their cellular tropism.

We show that the respiratory tract of dogs presents abundant sialylated glycans known to act as IAV receptors. Further, most IAVs (including 2.3.4.4b viruses) infected and replicated in dog tracheas, targeting mainly ciliated cells. Serological testing showed evidence of influenza spillover infections in dogs from the UK.

Overall, our results show that the canine respiratory tract can provide a suitable environment for the generation of new IAVs. Given the multi-host contact networks of dogs in nature, they could act as recipients, bridging hosts, and/or mixing vessels for multiple IAV lineages, playing a central role in the ecology of influenza emergence.

(SNIP) 

Our results suggest that dogs could act as a bridging species for the emergence of novel IAVs, including the H5N1 2.3.4.4b genotype. Serological studies in North America and Europe show that hunting dogs are routinely exposed to H5N1 IAVs 53,54. 

Vaccination of high-risk dog subpopulations could reduce the risk ofinter species infections (avian-to-dog and dog-to-human) and should be considered as preventative measures. 

This study has various limitations. Tracheal explants represent a significant portion of the airways, but this experimental system does not include the upper respiratory tract, the bronchial tree, nor the pulmonary parenchyma, which are likely to exhibit different virus/host interactions due to the presence of different cell types.

Further, explants lack systemic responses that are normally mounted during viral infections. Despite these limitations, our results are consistent with observations reported in field studies 59,60 and experimental in vivo infections 26,27

In sum, influenza spillover and emergence require a succession of processes that occur at different scales8. As the canine respiratory tract is a suitable ecosystem for IAV infection and reassortment and dogs routinely interact with multiple species that support endemic IAVs, dogs could facilitate the flow of virus between sympatric species. Targeted approaches to reduce the risk of IAV infections in dogs should be part of preparedness efforts to control influenza cross-species switching. 

       (Continue . . . ) 

 

OFID: Central Nervous System Involvement by Novel Clade 2.3.2.1e H5N1 Avian Influenza Virus in a Paediatric Patient

 

#19.161

While seasonal flu can occasionally cause neurological symptoms (see 2018's Neuroinfluenza: A Review Of Recently Published Studies) it is relatively rare, and usually only results in mild, and transient symptoms.

Similarly, avian influenza typically presents as a pulmonary infection, but over the past 2 decades we've seen growing evidence of neurological involvement as well.

A few early reports include: 

• A 2009 PNAS study (Highly pathogenic H5N1 influenza virus can enter the central nervous system and induce neuroinflammation and neurodegeneration) found that the H5N1 virus was highly neurotropic in lab mice, and in the words of the authors `could initiate CNS disorders of protein aggregation including Parkinson's and Alzheimer's diseases’.

• Six years later - following the 2014 death of the first imported H5N1 case in Canada - we saw a study (see CJ ID & MM: Case Study Of A Neurotropic H5N1 Infection - Canada), where the authors wrote: `These reports suggest the H5N1 virus is becoming more neurologically virulent and adapting to mammals'.

• In a 2015 Scientific Reports study on the genetics of the H5N1 clade 2.3.2.1c virus - Highly Pathogenic Avian Influenza A(H5N1) Virus Struck Migratory Birds in China in 2015 – the authors described its neurotropic effects, and warned that it could pose a ` . . . significant threat to humans if these viruses develop the ability to bind human-type receptors more effectively.'

HPAI H5Nx's threat largely receded between 2016-2020, but since 2021 has been on the ascendent again, primarily due to highly successful clades like 2.3.4.4b and 2.3.2.1c/e. Along with this renewed vigor we've also seen a rise in reports of neurological manifestations:

CDC EID Journal: Encephalitis and Death in Wild Mammals at An Animal Rehab Center From HPAI H5N8 - UK

Ontario: CWHC Reports HPAI H5 Infection With Severe Neurological Signs In Wild Foxes (Vulpes vulpes) 

In 2022, in Clinical Features of the First Critical Case of Acute Encephalitis Caused by Avian Influenza A (H5N6) Virus, we saw the first known case of neuroinfluenza in an H5N6 patient; a 6-year-old girl who was admitted to a hospital with mild pneumonia - but severe encephalitis - in January of that year. 

The following year, in Cell: The Neuropathogenesis of HPAI H5Nx Viruses in Mammalian Species Including Humans, we looked at an excellent review of recent surge in neurological infections reported in mammals and humans.

Highlights

• Highly pathogenic avian influenza (HPAI) H5Nx viruses can cause neurological complications in many mammalian species, including humans.

• Neurological disease induced by HPAI H5Nx viruses in mammals can manifest without clinical respiratory disease.

• HPAI H5Nx viruses are more neuropathogenic than other influenza A viruses in mammals.

• Severe neurological disease in mammals is related to the neuroinvasive and neurotropic potential of HPAI H5Nx viruses.

• Cranial nerves, especially the olfactory nerve, are important routes of neuroinvasion for HPAI H5Nx viruses.

• HPAI H5Nx viruses have a broad neurotropic potential and can efficiently infect and replicate in various CNS cell types.

• Vaccination and/or antiviral therapy might in part prevent neuroinvasion and neurological disease following HPAI H5Nx virus infection, although comprehensive studies in this area are lacking.

Even the relatively mild `bovine' H5N1 strain (B3.13) has been shown to have neurotropic qualities (see Preprint: Recent Bovine HPAI H5N1 Isolate is Highly Virulent for Mice, Rapidly Causing Acute Pulmonary and Neurologic Disease), at least in mice.

Thirteen months ago (April 2025) we saw a preliminary report on a neuroinvasive infection in an 8-y.o. girl (see Vietnam: Ho Chi Minh DOH Reports A Rare H5N1 Encephalitis Case In a Child), which reported:

As noted by infectious experts, this is a rare case in which the A/H5N1 avian influenza virus damages the central nervous system and does not attack the respiratory tract.

While much of the following report will be primarily of interest to clinicians, we have a detailed follow up on the Vietnamese case. Follow the link to read it in its entirety.

Central Nervous System Involvement by Novel Clade 2.3.2.1e H5N1 Avian Influenza Virus in a Paediatric Patient

Phung Nguyen The Nguyen , Nguyen Thanh Hung , Ngo Ngoc Quang Minh , Nguyen Thi Thu Hong , Nguyen Thi Thanh Huong , Cao Minh Hiep , Le Nguyen Thanh Nhan , Tran Van Dinh , Du Tuan Quy , Tran Thanh Thuc
Open Forum Infectious Diseases, ofag283, https://doi.org/10.1093/ofid/ofag283

Published: 07 May 2026

Novel clade 2.3.2.1e A(H5N1) virus was detected in cerebrospinal fluid but not in respiratory,rectal-swab and blood samples of an eight-year-old boy presenting with meningoencephalitis without respiratory symptoms. Cerebrospinal fluid A(H5N1)-hemagglutinin-specific antibody levels were higher than that of sera. Clinicians should be aware of emerging clade 2.3.2.1eA(H5N1) associated meningoencephalitis.

       (SNIP)

A(H5N1)-associated CNS infection in humans has rarely been reported but typically present as a complication, following respiratory symptoms [3-5]. Notably, our patient presented with meningoencephalitis in the absence of respiratory symptoms. Additionally, unlike the previously reported patients, who had viral RNA detected in both CSF and non-CSF samples [3-5], our patient only had viral RNA detected in serial CSF samples in the absence of viral RNA detected in urine, blood, rectal swab and respiratory samples.

Low respiratory-tract viral loads, transient viral replication in the respiratory tract, and/or delayed sample collection (illness day 6 onward) might explain the negative PCR findings in non-CSF samples, including the endotracheal aspirate sample. 

Notably, HPAI A(H5N1) viruses can infect human respiratory tissues by binding to receptors bearing sialic acids linked to galactose by α2,3-linkages, which are found in the lungs and lower respiratory tract, supported by the chest radiograph findings suggestive of lower left lung pneumonia.

        (SNIP)

In summary, we report on a HPAI A(H5N1) infection in a child presenting with  meningoencephalitis in the absence of respiratory symptoms. Viral RNA was detected in cerebrospinal fluid but not in respiratory, rectal-swab and blood samples.

Testing for IAV and  A(H5N1) virus should be considered in patients presenting with CNS infection with a history of exposure (e.g. dead poultry). Clinicians should be aware of meningoencephalitis associated with A(H5N1) infection in the absence of respiratory symptoms.

       (Continue . . . )

Not only can these neurological complications prove quite serious, atypical presentations can significantly delay proper diagnosis, isolation, and treatment. 

A reminder that HPAI H5 isn't your father's influenza. 

And we continue to treat it as such at our considerable peril.