Updated joint FAO/WHO/WOAH Public Health Assessment of Recent Influenza A(H5) Virus Events in Animals and People

 
#18,815

Yesterday the FAO/WHO/WOAH released an updated assessment (based on data as of July 1st) on the threat posed by various clades and subtypes of HPAI H5.

While they continue to assess the global public health risk of influenza A(H5) viruses to be low, the risk of infection for occupationally exposed persons is deemed to be low to moderate.

These updates are released roughly every 3 months (see March 2025 update), and are based primarily on reports from member nations, which may or may not be complete or up to date (see WHO Guidance: Surveillance for Human Infections with Avian Influenza A(‎H5)‎ Viruses).

This 13-page document contains information on recent spillovers into both animals and humans, and is well worth reading in its entirety. 

There is a lot to unpack here, and you'll want to read it in its entirety, but over the last 3 months, WOAH reports:

Between 1 March and 1 July 2025, an additional 807 A(H5N1) outbreaks in animals (including bird and mammal species) have been reported to WOAH. Of these, 268 outbreaks occurred in poultry (of any farming system), 389 outbreaks in wild bird and 92 outbreaks occurred in mammalian species. In Cambodia, 9 out of 14 outbreaks in poultry occurred in the vicinity of reported human cases.

In their discussion of clade 2.3.4.4b's viral characteristics, they address a number of topics we've discussed recently, including Antiviral resistance (see EID Journal: Antiviral Susceptibility of Influenza A(H5N1) Clade 2.3.2.1c and 2.3.4.4b Viruses from Humans, 2023–2024):

Available virus sequences from human cases have shown some genetic markers that may reduce susceptibility to neuraminidase inhibitors (antiviral medicines such as oseltamivir) or endonuclease inhibitors (such as baloxavir marboxil). While these changes may reduce antiviral susceptibility in laboratory testing, the clinical impact of these genetic changes requires further studies. 

And potential preexisting community immunity (see Study: Preexisting Immunity to the 2009 Pandemic H1N1 Virus Reduces Susceptibility to H5N1 Infection and Disease in Ferrets).

Based on limited seroprevalence information available on A(H5) viruses, human population immunity against the HA of A(H5) viruses is expected to be minimal; human population immunity targeting the N1 neuraminidase is found to be present although the impact of this immunity is yet to be understood. 

Experimental studies suggest prior A(H1N1) immunity reduced virus replication and disease severity of bovine-derived B3.13 genotype A(H5N1) virus in ferrets and that ferrets with this pre-existing immunity expressed A(H5N1) cross-reacting antibodies to the neuraminidase protein. 

However, the effectiveness of quadrivalent seasonal influenza vaccine (QIV) against influenza A(H5N1) virus remains a speculation, as a recent study observed no cross-neutralisation of H5N1 viruses by sera from patients vaccinated against seasonal influenza with QIV.4 

These risk assessments are only as good as the data that these agencies have access to, and as we've discussed previously (see From Here to Impunity), many nations continue to hold avian flu information close to their vest. 

Accordingly, the FAO/WHO/WOAH cab cite only medium confidence in their assessment:

Confidence level of the assessment

The overall confidence in the risk assessment is considered medium. The information used is derived from reports from national animal and human health authorities. There may be biases in surveillance, testing and reporting. Although the results and conclusions from peer-reviewed publications, pre-print publications and unpublished data informed this risk assessment, no systematic literature review was undertaken. Critical knowledge gaps remain in the understanding of the epidemiology.

We continue to see member nations encouraged to promptly report any spillovers of novel influenza to the appropriate agency, but it is unclear how much of the bird flu iceberg remains hidden.