Avian Flu Diary: A Review Of H5N1 Clade 2.3.2.1c Infections In Cambodia

#17,884

Over the past 48 hours we've seen two new H5N1 infections reported from Cambodia (see here and here), bringing the total number reported over the past year to 8. The two most recent cases are currently alive, and being treated in the ICU, but we know that 4 of the 6 earlier cases died (see CDC chart above).

All four of the fatalities in 2023 were children, while the two survivors were both adults (one was only mildly affected).

Although the reasons aren't clear, younger people (< 40) seem to be more susceptible to H5 infections while the opposite appears true for H7N9 (see H7N9: The Riddle Of The Ages). Childhood immune imprinting, however (see Nature: Declan Butler On How Your First Bout Of Flu Leaves A Lasting Impression), is believed to be a possible factor.

The newer clade 2.3.4.4b H5N1 virus - which has conquered much of the world and has killed tens of thousands of mammals since 2021 - has (thus far) proven far less lethal in humans.

Of the 13 cases reported since 2022 (see chart above), 7 were asymptomatic, 4 had critical illness, and 1 died.

The caveat being, that we only know about the cases that get reported. It is quite likely that some (perhaps, many) cases (both mild & severe) have occurred and have simply gone unrecognized or unreported around the world.

Previously we've seen seroprevalence studies on poultry workers in Asia (see PLoS One: Seroprevalence Of H5N1 Among Bangladeshi Poultry Workers) that suggest human infection with H5N1 is probably a lot more common than we know.

This idea was revisited again last August in a BMJ Global Health Journal commentary (see below) which reported on human seropositivity in 5 live bird markets in Egypt - a country which hasn't reported a human infection since 2017 - which found increasing evidence of previous infection with H5N1.

We are underestimating, again, the true burden of H5N1 in humans
Mokhtar Gomaa1, Yassmin Moatasim1, Ahmed El Taweel1, Sara H Mahmoud1, Amira S El Rifay1, Ahmed Kandeil1,2, Pamela P McKenzie2, Richard J Webby2, Rabeh El-Shesheny1,
Mohamed Ahmed Ali1,Ghazi Kayali3

https://doi.org/10.1136/bmjgh-2023-013146

Summary box

From mid-2021, a dramatic increase in activity and geographical distribution of clade 2.3.4.4b H5N1 viruses in animals occurred.
The extent of clade 2.3.4.4b human infections is unknown, but almost certainly more than the reported number, especially as this virus has shown an ability to infect several non-human mammals on a large scale.
Our study in five live bird markets showed that human seropositivity appeared to follow virus detection in wild and domestic birds sold live in the markets suggesting this as source of virus exposure.
Our estimate of 4.6% clade 2.3.4.4b H5N1 seroprevalence is more than double that of our earlier clade 2.2 seroprevalence estimate suggesting that the currently circulating viruses are more infectious to mammalian hosts.
The scale, geographical scope and increasing list of infected species with clade 2.3.4.4b viruses urge the conduct of additional seroprevalence studies.

The reality is, mild or even moderate cases are less likely to seek medical care, making it unlikely they will be tested, or reported. Some severe cases may go undetected as well, but on average, they are more likely to be picked up by surveillance.

Similarly, in China we've followed the spillover of HPAI H5N6 to humans, where roughly 90 cases have been reported since 2014 (see map below). Of those, roughly half (of the cases where we have an outcome) proved fatal.

But once again, we have no idea how many mild or moderate cases go unreported. Since we don't have a good handle on the denominator - the total number of people infected - we can't reliably calculate the CFR.

This chronic lack of data has sparked many debates over the years on the true lethality of H5 (see mBio: Mammalian-Transmissible H5N1 Influenza: Facts and Perspective).

Raw data suggests that H5N1 clade 2.3.2.1c and HPAI H5N6 currently pose more of a threat to human health than does H5N1 clade 2.3.4.4b, but we missing a lot of critical information.

The oft-quoted `factoid' that H5N1 kills 50% of those it infects is, thankfully, highly unlikely. Of course, even a 2% CFR would be disastrous in a pandemic.

What we can say is that some strains of HPAI H5 can kill up to 50% of those who are sick enough to be hospitalized. But there could easily be 10 - or perhaps 100 - milder cases for every severe case.

While we haven't seen any detailed reviews of the more recent H5N1 cases from Cambodia, last August the Journal Emerging Microbes & Infections published an analysis of the viruses detected in first (father/daughter) cluster reported nearly a year ago.

Reassuringly, they found that virus did not bind efficiently to human receptor cells, limiting its zoonotic potential.

Characterization of the haemagglutinin properties of the H5N1 avian influenza virus that caused human infections in Cambodia

Pengxiang Chang, Jiayun Yang, Thusitha K. Karunarathna, Mehnaz Qureshi, Jean-Remy Sadeyen & Munir Iqbal
Article: 2244091 | Received 13 Apr 2023, Accepted 30 Jul 2023, Published online: 25 Aug 2023

https://doi.org/10.1080/22221751.2023.2244091

ABSTRACT

High pathogenicity avian influenza (HPAI) H5N1 is a subtype of the influenza A virus primarily found in birds. The subtype emerged in China in 1996 and has spread globally, causing significant morbidity and mortality in birds and humans. In Cambodia, a lethal case was reported in February 2023 involving an 11-year-old girl, marking the first human HPAI H5N1 infection in the country since 2014.

This research examined the zoonotic potential of the human H5N1 isolate, A/Cambodia/NPH230032/2023 (KHM/23), by assessing its receptor binding, fusion pH, HA thermal stability, and antigenicity. Results showed that KHM/23 exhibits similar receptor binding and antigenicity as the early clade 2.3.2.1c HPAI H5N1 strain, and it does not bind to human-like receptors. Despite showing limited zoonotic risk, the increased thermal stability and reduced pH of fusion in KHM/23 indicate a potential threat to poultry, emphasizing the need for vigilant monitoring.