Avian Flu Diary: IJID: Lack of Respiratory Droplet Transmission of Two Recent Human Influenza A(H5N1) Viruses in Female Ferrets

#19,076

Today we've a study that reminds us how much we've yet to learn about what it takes for a novel zoonotic virus - like HPAI H5Nx - to turn into a genuine pandemic threat.

First, a bit of history.

In the months following the discovery of first human infection with the Bovine B3.13 H5N1 virus (aka TX/37) in March of 2024, we saw a number of conflicting reports on the virulence and transmissibility of the HPAI virus in ferrets (see CDC: Updated Results On Texas H5N1 Virus In Ferrets).

Differences in methodology, equipment used, and endpoints, the host source (human vs bovine), viral evolution during isolation (cell culture vs eggs), and the use of a relatively small number of test animals all could have contributed to these mixed results.

A reminder that no study design is perfect, that research is often incremental, and that we should view each new finding in the context of what has previously been reported by others.

I mention all of these limitations because today's study, once again, provides some unexpected results.

Researchers at the Kawaoka lab at the University of Wisconsin-Madison, along with international partners took two human isolates (1) B3.13 from a Dairy Worker in Michigan (MI90) and (2) a D1.1 from the B.C. adolescent (BC2032) who was critically ill, and tested their transmissibility in female ferrets.

Interestingly, the B3.13 virus (MI90) - which only caused mild conjunctivitis in its human host - caused severe disease, extra-respiratory spread, and was lethal in test ferrets.

While the D1.1 virus (BC2032) - which put a teenage girl on ECMO - produced milder illness and no lethality in ferrets.

And neither virus transmitted via respiratory droplets in ferrets.

In addition to these non-intuitive findings, the authors also report that previous studies found the D1.1 (BC2032) strain led to 100% lethality in ferrets, while earlier studies found the MI90 strain caused `. . . sub-lethal disease in ferrets, with respiratory droplet transmission detected in 50% of naïve exposed animals.'

As we've discussed often, we have only a partial understanding of what genetic changes are needed to increase the virulence, transmissibility, or host range of a novel influenza A virus like HPAI H5.

Scientists often look for a handful of known amino acid changes (e.g. PB2 mutations like E627K, D701N, Q591K, and M631L and HA mutations like Q226L and E190D) which may favor mammalian adaptation, but new ones (see Sci. Adv.: PB2 and NP of North American H5N1 Virus Drive Immune Cell Replication and Systemic Infections) continue to be discovered.

And many minor, seemingly innocuous genetic changes - when stacked in the right combination - can greatly increase or decrease, their impact.

Exactly what part these (and other) factors may have played in today's unexpected results remains to be seen. I've just posted the abstract, and some brief excerpts, so follow the link to read the study in its entirety.

I'll have a brief postscript after the break.

Lack of Respiratory Droplet Transmission of Two Recent Human Influenza A(H5N1) Viruses in Female Ferrets

Tong Wang 1, Chunyang Gu 1, Lizheng Guan 1, Asim Biswas 1, Tadashi Maemura 1, Hassanein H. Abozeid 1 2, Peter J. Halfmann 1, Gabriele Neumann 1, Amie J. Eisfeld 1, Yoshihiro Kawaoka 1 3 4 5
https://doi.org/10.1016/j.ijid.2026.108514 Get rights and content
Under a Creative Commons license

Highlights

Two human clade 2.3.4.4b H5N1 viruses (B3.13 and D1.1) were tested in ferrets.

MI90-H5N1 (B3.13) caused severe disease, extra-respiratory spread, and lethality.

BC2032-H5N1 (D1.1) caused milder disease with no lethality in ferrets.

Neither virus transmitted via respiratory droplets in ferrets.

Respiratory droplet transmissibility of clade 2.3.4.4b H5N1 viruses appears variable.

Abstract

Background

Clade 2.3.4.4b highly pathogenic avian influenza A(H5N1) (HPAI H5N1) viruses are widespread globally and have transmitted from birds to dairy cattle at least four times in the United States, including once by a genotype B3.13 virus and three times by genotype D1.1 viruses. Despite their prevalence and known ability to infect humans, only a few studies have examined respiratory droplet transmission capabilities of clade 2.3.4.4b viruses in mammalian models of influenza infection.

Methods

Here, we assessed respiratory droplet transmission of two recent human clade 2.3.4.4b HPAI H5N1 viruses – A/Michigan/90/2024 (‘MI90-H5N1’), a B3.13 isolate, and plaque-purified A/British Columbia/PHL2032/2024 (‘BC2032-H5N1’), a D1.1 isolate – in the ferret model.

Findings

We found that MI90-H5N1, in contrast to earlier findings, causes severe disease and partial lethality in ferrets, with virus spread to extra-respiratory organs and no respiratory droplet transmission. BC2032-H5N1 caused less severe disease with no lethality in ferrets and, consistent with a recent report, failed to transmit via respiratory droplets.

Interpretation

Together with other reports, our results suggest that respiratory droplet transmissibility of clade 2.3.4.4b viruses is variable. Therefore, continued monitoring and risk assessment for emerging HPAI H5N1 viruses is essential to better understand their pandemic potential.

(SNIP)

A limitation of this study is that transmission was assessed in a single mammalian model under defined laboratory conditions, which may not fully capture variability in natural exposures or host species.

Only single isolates of each genotype were evaluated,and the BC2032-H5N1 virus used here was plaque-purified, capturing only one variant from a clinically heterogeneous sample.

Additionally, female ferrets were used in our experiments, whereas the CDC study employed male ferrets; potential effects of sex on disease severity or transmission efficiency were not assessed.

Finally, modest sample sizes could limit detection of rare transmission events.

Overall, these findings underscore that while some clade 2.3.4.4b H5N1 viruses can transmit via respiratory droplets, transmission efficiency is variable and may be dependent on virus genotype, pathogenicity, and other host and/or environmental factors. Therefore, it is critical to continue monitoring and risk assessment for emerging HPAI H5N1 viruses to better understand their pandemic potential.

Obviously, there is still a great deal of uncertainty in influenza research. This isn't the first time we've confronted conflicting data, nor will it be the last. A few past blogs include:

When Studies Collide (COVID-19 Edition)

When Flu Vaccine Studies Collide

When Studies Collide (Revisited)

Why Preprints Are Only Preprints

A reminder that gaining scientific knowledge is a process . . . one that evolves over time and often involves detours, setbacks, and constant reevaluation.