Effect of Baloxavir and Oseltamivir in Combination on Infection with Influenza Viruses with PA/I38T or PA/E23K Substitutions in the Ferret Model

Abstract

Amino acid substitutions I38T and E23K in the influenza polymerase acidic (PA) protein lead to reduced susceptibility to the influenza antiviral drug baloxavir. The in vivo effectiveness of baloxavir and oseltamivir for treatment of these viruses is currently unknown. Using patient-derived influenza isolates, combination therapy was equally effective as monotherapy in reducing viral titers in the upper respiratory tract of ferrets infected with A(H1N1pdm09)-PA/E23K or A(H3N2)-PA/I38T. When treated with baloxavir plus oseltamivir, infection with a mixture of PA/I38T or PA/E23K and corresponding wild-type virus was characterized by a lower selection of viruses with reduced baloxavir susceptibility over the course of infection compared to baloxavir monotherapy. De novo emergence of the oseltamivir resistance mutation NA/H275Y occurred in ferrets treated with oseltamivir alone but not in ferrets treated with baloxavir plus oseltamivir. Our data suggest that combination therapy with influenza drugs with different mechanisms of action decreased the selection pressure for viruses with reduced drug susceptibility. IMPORTANCE Influenza viruses cause significant morbidity and mortality worldwide but can be treated with antiviral drugs. In 2018, a highly effective antiviral drug, baloxavir marboxil, was licensed. However, the selection of viruses with baloxavir resistance was relatively high following treatment, which may compromise the effectiveness of the drug. Here, we took two different influenza viruses that are resistant to baloxavir and tested the effectiveness alone and in combination with oseltamivir (a second influenza antiviral drug) in the ferret model. Our findings suggest that combination treatment may be a more effective method than monotherapy to reduce the selection of resistant viruses. These results may have important clinical implications for the treatment of influenza.

Keywords: antiviral agents; antiviral resistance; ferret; influenza; virus.

FIG 1

Surface plot to show interactions between baloxavir and oseltamivir against the growth of A(H3N2)-WT or PA/I38T and A(H1N1pdm09)-WT or PA/E23K viruses using the Bliss independence synergy model. The z-axis depicts the percent inhibition of viral growth relative to no drug (0 nM baloxavir plus 0 nM oseltamivir). The baloxavir and oseltamivir concentrations were tested in full factorial combination, and the colored shading on the plot depicts additive (green), synergistic (blue), and antagonistic (red) effects.

FIG 2

Effect of antiviral treatment on viral shedding in ferrets infected with A(H3N2)-WT or A(H3N2)-PA/I38T. Ferrets were inoculated intranasally with 10⁵ TCID₅₀/500 μL with a pure population of A(H3N2)-WT (a) or a pure population of A(H3N2)-PA/I38T (b). Antiviral treatment was commenced 24 h postinfection with 4 mL/kg placebo (subcutaneous single dose, methylcellulose vehicle), 10 mg/kg/day oseltamivir monotherapy (oral, BID), 4 mg/kg baloxavir monotherapy (subcutaneous, single dose), or a combination of baloxavir and oseltamivir (doses as described for each monotherapy). Nasal washes were collected daily for 10 days, and the infectious virus titer was determined in MDCK-SIAT cells. The viral titers in the nasal washes of ferrets in each antiviral treatment group are represented as means ± the standard deviations, and the area under the curve for each group is shown above the bar graph. Statistical analysis was used to compare the area under the curve for each antiviral treatment to the corresponding placebo (unpaired t test with Welch’s correction; ns, nonsignificant; *, P < 0.05; **, P < 0.01). The LLOD for the assay is 10¹ TCID₅₀/mL.

FIG 3

Effect of antiviral treatment on viral shedding in ferrets infected with A(H1N1pdm09)-WT or A(H1N1pdm09)-PA/E23K. Ferrets were inoculated intranasally with 10⁴ TCID₅₀/500 μL with a pure population of A(H1N1pdm09)-WT (a) or a pure population of A(H1N1pdm09)-WT (b). Antiviral treatment was commenced 24 h postinfection with 4 mL/kg placebo (subcutaneous single dose, methylcellulose vehicle), 10 mg/kg/day oseltamivir monotherapy (oral, BID), 4 mg/kg baloxavir monotherapy (subcutaneous, single dose), or a combination of baloxavir and oseltamivir (doses as described for each monotherapy). Nasal washes were collected daily for 10 days, and infectious virus titers were determined in MDCK-SIAT cells. The viral titers in the nasal washes of ferrets in each antiviral treatment group are represented as means ± the standard deviations, and the area under the curve for each group is shown above the bar graph. Statistical analysis was used to compare the area under the curve for each antiviral treatment to the corresponding placebo (unpaired t test with Welch’s correction; ns, nonsignificant; *, P < 0.05; **, P < 0.01). The LLOD for the assay is 10¹ TCID₅₀/mL.

FIG 4

Effect of antiviral treatment on the relative proportion of A(H3N2)-PA/I38T in ferrets infected with a competitive mixture of WT:PA/I38T. Ferrets were intranasally inoculated with 10⁵ TCID₅₀/500 μL of 20% A(H3N2)-PA/I38T:80% A(H3N2)-WT. Antiviral treatment was commenced 24 h postinfection with placebo (subcutaneous single dose, methylcellulose vehicle) (a), oseltamivir monotherapy (oral, BID) (b), baloxavir monotherapy (subcutaneous, single dose) (c), or combination therapy with oseltamivir and baloxavir (doses as described for each monotherapy) (d). Nasal washes were collected daily for 10 days. The infectious virus titers were determined by titration in MDCK cells (left panel), and the percentages of PA/I38T in the nasal wash were determined by pyrosequencing for the duration of viral shedding (right panel). The average area under the curve is represented above the viral shedding for each treatment group. The pyrosequencing plots are shown for each individual ferret, and the colors correspond to the virus titer in TCID₅₀. Whole-genome sequencing was performed on samples obtained from day 5 or the final day of viral shedding. The LLOD for the TCID₅₀ assay is 10¹ TCID₅₀/mL.

FIG 5

Effect of antiviral treatment on the relative proportion of A(H1N1pdm09)-PA/E23K in ferrets infected with a competitive mixture of WT and PA/E23K. Ferrets were intranasally inoculated with 10⁴ TCID₅₀/500 μL of 20% A(H1N1pdm09)-PA/E23K:80% A(H1N1pdm09)-WT. Antiviral treatment was commenced 24 h postinfection with placebo (subcutaneous single dose, methylcellulose vehicle) (a), oseltamivir monotherapy (oral, BID) (b), baloxavir monotherapy (subcutaneous, single dose) (c), or combination therapy with oseltamivir and baloxavir (doses as described for each monotherapy) (d). Nasal washes were collected daily for 10 days. The infectious virus titers were determined by titration in MDCK cells (left panel), and the percentages of PA/I38T in the nasal wash were determined by pyrosequencing for the duration of viral shedding (right panel). The average area under the curve is represented above the viral shedding for each treatment group. The pyrosequencing plots are shown for each individual ferret, and the colors correspond to the TCID₅₀ shedding. Whole-genome sequencing was performed on samples obtained from day 5. The LLOD for the TCID₅₀ assay is 10¹ TCID₅₀/mL.