Early reports of H7N9 out of China have indicated a small, but worrisome number of oseltamivir (Tamiflu ®) resistant strains detected among those infected during the first wave (see The Lancet: Antiviral Resistance In Two H7N9 Patients).
In most cases, patients were believed to have developed resistance after being placed on the antiviral medication.
What is commonly called `spontaneous resistance’.
The Lancet study found that a mutation R292K (Arginine to Lysine at position 292 in the NA) – also known as Arg292Lys – already known to confer antiviral resistance to seasonal flu (see Resistant influenza A viruses in children treated with oseltamivir: descriptive study), appeared in two patients after several days of oseltamivir therapy.
Today we’ve a new study appearing in the open access journal mBio that finds that the standard laboratory tests for antiviral susceptibility can miss `mixed population’ infections (comprised of both resistant and susceptible strains), and that antiviral treatment could suppress the susceptible strains while allowing resistant strains to flourish.
And this particular mutation can provide resistance not only to oseltamivir, but to to zanamivir and peramivir as well.
The corresponding author of the study is Dr. Robert Webster of St. Jude Children’s Hospital, considered the dean of influenza virology. In a press release from the American Society for Microbiology, we get a summary of what this study discovered.
In the mBio study, the authors tested antiviral susceptibility of an H7N9 strain isolated from the first confirmed human case of avian H7N9 influenza using a method that tests the activity of the neuraminidase enzyme. The reassuring results were, unfortunately, misleading: the enzyme-based test indicated that the flu strain was susceptible to NA inhibiting antiviral drugs, but it is not.
A closer look at the viral isolate revealed it is actually made up of two distinct types of H7N9 viruses. Roughly 35% of the viruses carry the R292K mutation, making them resistant to NA inhibitors, and 65% are sensitive to these same drugs. The enzyme-based testing gave misleading results, says Webster, because the functioning wild-type enzymes masked the presence of the non-functioning mutant enzymes.
Using NA inhibitors to treat a patient infected with a resistant strain of H7N9 only encourages the virus to proliferate and can lead to enhanced spread of the resistant strain. The authors write that these results prove that it is crucial to use a gene-based surveillance technique that can detect these resistant influenza strains in a mixed infection.
But the news isn't all bad. Webster also points out that antiviral resistance is something of a burden for influenza viruses, and that fitter wild-type H7N9 strains may eventually win out over resistant strains. In the absence of a drug like Tamiflu, Webster says, it seems unlikely that these resistant viruses would acquire epidemic characteristics.
"The great need at the moment are additional drugs aimed at additional sites in the influenza genome. There are some [drugs] in the pipeline, but they are still under testing at the moment," says Webster. "We'd better get some vaccine seed stocks up and ready. The antiviral option for controlling H7N9 isn't too good."
Here is a link to the mBio study, after which I’ll be back with a cautionary tale about our previous experiences with evolving antiviral resistance and seasonal flu.
H.-L. Yen, J. L. McKimm-Breschkin, K.-T. Choy, D. D. Y. Wong, P. P. H. Cheung, J. Zhou, I. H. Ng, H. Zhu, R. J. Webby, Y. Guan, R. G. Webster and J. S. M. Peiris
Our results confirmed that the NA R292K mutation confers resistance to oseltamivir, peramivir, and zanamivir in the novel human H7N9 viruses. Importantly, detection of the resistance phenotype may be masked in the clinical samples containing a mixed population of R/K at NA residue 292 in the enzyme-based NA inhibition assay.
IMPORTANCE The neuraminidase (NA) inhibitors oseltamivir and zanamivir are currently the front-line therapeutic options against the novel H7N9 influenza viruses, which possess an S31N mutation that confers resistance to the M2 ion channel blockers. It is therefore important to evaluate the sensitivity of the clinical isolates to NA inhibitors and to monitor for the emergence of resistant variants.
We characterized the A/Shanghai/1/2013 (H7N9) isolate which contained a mixed population of R/K at NA residue 292. While the clinical isolate exhibited a phenotype of sensitivity to NA inhibitors using the enzyme-based NA inhibition assay, the plaque-purified A/Shanghai/1/2013 virus with dominant K292 was resistant to zanamivir, peramivir, and oseltamivir.
Resistance to NA inhibitors conferred by the R292K mutation in a human influenza virus H7N9 isolate can be masked by a mixed R/K viral population, and this should be taken into consideration while monitoring antiviral resistance in patients with H7N9 infection.
While it is true that resistant strains of influenza A tend to be less transmissible than their susceptible `wild type’ counterparts, we have some examples where resistant strains have been `fit’ enough spread globally.
By 2005, nearly all of the influenza samples around the globe had developed resistant to the older M2 ion channel blockers like Amantadine, leaving oseltamivir and zanamivir (neuraminidase inhibitors) as the two remaining treatments.
In 2006 we saw a smattering of oseltamivir resistant seasonal H1N1 cases, almost always attributed to `spontaneous mutations’ within a patient receiving the drug. While of concern to the patient being treated, it appeared to be poorly transmissible.
In the 2006-2007 flu season, laboratories found no resistant strains in Europe or Japan, and in less than 1% of samples from the United States.
This resistance in this viral strain was mostly caused by a mutation called H275Y, where a single amino acid substitution (histidine (H) to tyrosine (Y)) occurs at the neuraminidase position 275.
(Note: some scientists use 'N2 numbering' (H274Y) and some use 'N1 numbering' (H275Y))
The following year, during the 2007-2008 flu season, oseltamivir resistant viruses suddenly took flight, and by the spring of 2008 roughly 25% of European samples tested showed the H275Y mutation (see Increased Tamiflu Resistance In Seasonal Influenza).
By December of 2008, the CDC was forced to issue major new guidance for the use of antivirals for the second time in just three years (see CIDRAP article With H1N1 resistance, CDC changes advice on flu drugs).
The `replacement’ 2009 pandemic virus, that supplanted the largely resistant seasonal strain the following spring, was fortunately sensitive to Tamiflu. While we’ve seen some scattered cases of resistance in the new H1N1 strain, so far they are relatively few, and highly scattered (see Antiviral Resistance In 2009 H1N1 Influenza A Strain).
Granted, H7N9 is a different influenza A strain, and it carries a different mutation than the old H1N1 virus, but this bit of history shows how quickly our pharmaceutical options can evaporate.
The concern with today’s study is the use of antivirals on patients who have both susceptible and resistant viruses could help promote the replication and spread of resistant strains.
Which could inadvertently help drive the evolution of the H7N9 virus towards a predominantly resistant strain.