Effect Measure

To investigate the fitness (pathogenicity and replication efficiency) of NAI-resistant [neuraminidase inhibitor resistant] H5N1, Yen and colleagues used reverse genetics to create recombinant H5N1 influenza viruses with either wild-type neuraminidase (NA) or one of four single amino acid changes found in nature and known to be associated with NAI resistance: E119G, H274Y, R292K, and N294S. For comparison, they created recombinant H1N1 viruses with the same mutations.1

The researchers found that two of the four mutations (E119G and R292K) resulted in H5N1 and H1N1 viruses that did not grow well in vitro and could not be stably maintained. The other two mutations (H274Y and N294S) resulted in viruses that replicated as efficiently in vitro as wild-type H5N1 and H1N1 viruses. H274Y resulted in markedly reduced susceptibility to oseltamivir but not zanamivir in both H5N1 and H1N1. N294S resulted in moderately reduced susceptibility to oseltamivir in both viruses but also slightly reduced susceptibility to zanamivir.

The H5N1 viruses with either of the two mutations were just as lethal to mice as wild-type H5N1; however, the H1N1 viruses with either mutation did show decreased lethality. The explanation for this may be that although both mutations resulted in decreased NA enzymatic activity in both H5N1 and H1N1, the NA enzymatic activity of the wild-type H5N1 was significantly higher than that of the wild-type H1N1. Therefore, the reduced NA function did not affect the H5N1 virus fitness but did affect that of the H1N1. The function of NA is to enable release of the newly created virus from the host cell; therefore, reduced NA activity would be expected to lead to a reduced ability to replicate and cause disease. (CBN Report)

We recently did an exhaustive set of posts (here) on a mathematical model that made the assumption a fully or mostly fit resistant strain was possible. The modeling paper, by Lipsitch et al., showed that even if resistance develops, use of Tamiflu prophylactically and therapeutically was still a winning strategy in the first wave, but the resistance would eventually spread and dominate the circulating strains. From the Lipsitch paper:

Important predictions that we believe to be robust to model structure are that (1) antiviral use will favor the spread of resistance even if such use rarely generates de novo [spontaneously arising] resistant strains; (2) despite the spread of resistance, prophylaxis and treatment can both delay and reduce the size of the epidemic; (3) nondrug interventions (if effective) and antiviral use - which will likely be used together in the response to a pandemic - generally have synergistic benefits, despite the fact that nondrug interventions may promote resistance; and (4) relatively minor differences in fitness cost may make large differences in outcomes, even when emergence probabilities are low (Figure 4). These results extend those of previous models, which showed (like our model) that the fitness cost of resistance strongly influences the ability of resistant strains to spread during an epidemic. (Lipsitch et al., PLoS Medicine)

Oseltamivir (Tamiflu) seems to be more affected than zanamivir (Relenza), but the former can be taken orally in capsule form while the latter uses an inhaler for administration. A handful of patients with either H274Y or N294S mutations have been reported in the literature. Not good news, but not unexpected.

Source paper: Yen HL, Ilyushina NA, Salomon R, et al. Neuraminidase inhibitor-resistant recombinant a/vietnam/1203/04 (H5N1) influenza viruses retain their replication efficiency and pathogenicity in vitro and in vivo. J Vir 2007:81(22); 12418-12426. http://jvi.asm.org/cgi/content/full/81/22/12418.