Sunday, October 01, 2023

Eurosurveillance: A community Cluster of Influenza A(H3N2) Virus infection with Reduced Susceptibility to Baloxavir - Japan 2023

Credit NIAID












#17,703

Prior to the emergence of the COVID pandemic in early 2020, we'd been following sporadic reports of  resistance to a relatively new influenza antiviral  - one used in Japan since early 2018, but approved in the United States in October of that year - called Baloxavir marboxil (trade name Xofluza®). 

As a new class of antiviral, Baloxavir was a particularly welcome addition to our armamentarium since the two most common neuraminidase inhibitors - zanamivir and oseltamivir (Tamiflu ®) - had been in use for 20 years, and there were growing concerns over creeping resistance.

In March of 2019, however, we saw several reports - and a Eurosurveillance Rapid Communications - on the detection of a small number of Baloxavir resistant flu viruses among patients in Japan - including three that had not received the antiviral drug.

In August 2019, in EID Journal: H-2-H Transmission Of A(H3N2) with Reduced Susceptibility to Baloxavir, Japan, we looked at a study that reported:

During the 2018–19 influenza season in Japan, we detected 32 mutant influenza A(H3N2) viruses carrying various types of PA I38 substitutions, 4 of which were isolated from children < 12 years of age without prior baloxavir exposure.

The researchers concluded: `These 4 children were probably infected with mutant viruses acquired from hosts previously treated with baloxavir.'

The spontaneous development of resistant viruses in a patient who is already receiving antiviral drugs is a known - but relatively rare (roughly 1%) - complication. Usually, these mutated viruses suffer a fitness penalty' - making them less likely to be transmitted on to others.

But a September 2019 report, J.I.D: Replicative Fitness of Seasonal Influenza A Viruses with Decreased Susceptibility to Baloxavir, evaluated the biological fitness of baloxavir resistant viruses collected the previous winter, and found them only mildly impaired.

Even though the incidence of resistance remained low, public health agencies in the United States and Japan announced enhanced surveillance for the 2019-2020 flu season. That season was interrupted by the arrival of COVID in January 2020, and influenza - and Baloxavir consumption - plummeted over the next two years.  

Influenza returned for the 2022-2023 flu season, however, and with it came increased uptake of Baloxavir in Japan.  Today we have a report, published last week in the journal Eurosurveillance, which looks at a small community cluster of Baloxavir resistant viruses detected last spring. 

Due to its length, I've only posted the link, abstract, and some excerpts. Follow the link to read it in its entirety.  I'll have a bit more when you return. 

(Edit:  I accidentally copied the title from an earlier study which I had open in my browser. I've corrected it below).

A community cluster of influenza A(H3N2) virus infection with reduced susceptibility to baloxavir due to a PAE199G substitution in Japan, February to March 2023 

Baloxavir marboxil, a cap-dependent endonuclease inhibitor, was approved in Japan in 2018 for the treatment and prophylaxis of influenza A and B virus infections in patients 12 years and older and children younger than 12 years weighing at least 10 kg. Since the 2017/18 influenza season, we have been monitoring the susceptibility of circulating influenza viruses to baloxavir and the neuraminidase (NA) inhibitors oseltamivir, peramivir, zanamivir, and laninamivir. Approximately 10-15% of total isolates are randomly selected and analysed.

During our nationwide monitoring, we detected sporadic cases of influenza A(H1N1)pdm09 and A(H3N2) viruses with reduced susceptibility to baloxavir in the 2018/19 and 2019/20 seasons [1-3]. These viruses were detected in children younger than 10 years and possessed E23K or I38T amino acid substitutions in their polymerase acidic subunit (PA). No community clusters of such mutant viruses have been identified to date.

Here, we report a community cluster of influenza A(H3N2) virus with reduced susceptibility to baloxavir due to a PA E199G substitution.

(SNIP)

During the season 2022/23, we isolated 12 influenza A(H3N2) viruses in Nara, a town located near Osaka and Kyoto, as part of our nationwide monitoring (Table 1). In Nara, influenza A(H3N2) viruses with haemagglutinin (HA) genes from phylogenetic clades 2a.1, 2a.1b, 2a.3a and 2b co-circulated (https://nextstrain.org/flu/seasonal/h3n2/ha/2y). In Japan, 159 (48%) of the 329 influenza A(H3N2) viruses tested in this season were found to belong to clade 2b, followed by 76 (23%) that belonged to clade 2a.3a.

We found that three of the 12 influenza A(H3N2) viruses possessed an E199G substitution in their PA protein, which is associated with reduced susceptibility to baloxavir [6,7]. These PA E199G mutant viruses (A/Nara/10/2023, A/Nara/12/2023 and A/Nara/14/2023) belonging to clade 2a.3a were detected in three outpatients in their 30s, 50s and 60s who presented at the same hospital between February and March 2023 with fever over 38 °C. None of them had received baloxavir before specimen collection.

          (SNIP)

Discussion

The PA E199G mutant influenza A(H3N2) virus was previously detected during treatment-emergent variant monitoring in a paediatric study of baloxavir and showed reduced susceptibility to baloxavir [6]. In the study presented here, three PA E199G mutant A(H3N2) viruses (A/Nara/10/2023, A/Nara/12/2023 and A/Nara/14/2023) belonging to clade 2a.3a were detected between 20 February and 3 March 2023 in three patients without prior baloxavir treatment. The primary case of these mutant virus infections could not be identified. However, two wild-type influenza A(H3N2) viruses (A/Nara/3/2023 and A/Nara/8/2023) belonging to the same clade were detected between 30 January and 13 February 2023 in patients who received baloxavir on the day of specimen collection in the same hospital as the patients with the PA E199G mutant virus (Table 1). The whole genome sequences of A/Nara/3/2023, isolated from a teenage child, and the three PA E199G mutant viruses were almost identical except for HA I156M and NA S335G substitutions of A/Nara/3/2023. However, A/Nara/8/2023, isolated from an adult patient, possessed six substitutions compared with the PA E199G mutant viruses: two in polymerase basic protein 2 (PB2), one in PB1, two in PA and one in HA.

An influenza outbreak occurred in the school attended by the patient infected with A/Nara/3/2023. We previously reported that influenza viruses with reduced susceptibility to baloxavir were detected 3–6 days after baloxavir administration [2]. These results suggest the possibility that the PA E199G mutant A(H3N2) virus emerged due to the selective pressure of baloxavir and started to spread earlier in the season. In addition, A/Nara/10/2023 and A/Nara/12/2023 belonged to family clusters where the transmission chain started with children, and A/Nara/14/2023 was detected in a sporadic case. As specimens from other outbreaks were unavailable, we cannot rule out that the PA E199G mutant virus circulated among children.

Of the substitutions that emerged after baloxavir treatment in the past, PA I38 substitutions were the most frequent and had the greatest impact on baloxavir susceptibility [9]. The frequency of emergence of PA I38 mutant viruses in children younger than 12 years is higher than that in patients aged 12–64 years [2,6,10]. We previously reported human-to-human transmission of PA I38T and PA E23K mutant viruses in children younger than 10 years [1-3], but no community clusters of these viruses have been identified. The higher frequency of antiviral-resistant influenza viruses in children appears to be associated with longer periods of viral shedding and higher viral titres than in adults [11].

In our current study, however, a community cluster of the PA E199G mutant virus was detected among patients in their 30s, 50s and 60s. Further studies are required to evaluate the replication and transmission fitness of the PA E199G mutant viruses.

Although patients infected with the PA I38 mutant viruses exhibited prolonged virus shedding, a rebound in virus titres after treatment and delayed symptom alleviation [10,12-15], the significance of the PA E199G substitution and the effectiveness of baloxavir in patients infected with the PA E199G mutant viruses have not been established. Our data suggest that the PA E199G mutant viruses have retained viral fitness and have the potential for human-to-human transmission. Furthermore, accumulation of multiple antiviral-resistant substitutions has a synergistic effect, resulting in a further reduction in the susceptibility to inhibitors [8]. Therefore, monitoring amino acid substitutions is essential to protect public health and support clinical management.

Conclusion

The PA E199G mutant influenza A(H3N2) viruses showed reduced susceptibility to baloxavir in vitro; however, the clinical significance of the PA E199G substitution remains unclear. Therefore, baloxavir may be effective in patients infected with these mutant viruses.


Whether this is a blip in the data, or the start of a trend, is impossible to tell at this point.  We do know that widely used antiviral medications - much like antibiotics - can have a limited useful life, as viruses are often able to develop resistance over time. 

Our oldest class of influenza antivirals - M2 ion channel blockers (e.g. Amantadine, Rimantadine) - were developed in the 1950s. But by late 2005 Amantadine began to lose its effectiveness against the H3N2 seasonal flu virus and some strains of the H5N1 bird flu.

In January of 2006 the CDC issued a warning to doctors not to rely on Amantadine or Rimantadine to treat influenza.

Tamiflu (Oseltamivir) which had been approved in 1999, while far more expensive, became the new treatment standard.  

Within a couple of years seasonal H1N1 began to show growing resistance to Tamiflu as well (although H3N2 remained sensitive). This resistance was due to a mutation, known as H274Y, where a single amino acid substitution (histidine (H) to tyrosine (Y)) occurs at the neuraminidase position 274.

By the winter of 2008 - in the space of just about a year – seasonal H1N1 had gone from almost 100% sensitive to the drug to nearly 100% resistant 
(see CIDRAP On the CDC Change Of Advice On Tamiflu).

It seemed as if antiviral crisis was unavoidable, when in a Deus Ex Machina moment - a new swine-origin H1N1 virus - one that happened to be sensitive to Tamiflu, emerged as a pandemic strain in the spring of 2009.

In a matter of months pdmH1N1 supplanted the old H1N1 virus, and our antiviral crisis was averted. Since then, resistance to Tamiflu has been low (roughly 2%), but we continue to monitor it closely. 

For some other blogs on influenza antiviral resistance, you may wish to revisit: