Antiviral efficacy of favipiravir against two prominent etiological agents of hantavirus pulmonary syndrome

Abstract

Hantavirus pulmonary syndrome (HPS) is caused by infection with several Sigmodontinae- and Neotominae-borne hantaviruses and has a case fatality rate of 30 to 50%. Humans often become infected by inhalation of materials contaminated with virus-laden rodent urine or saliva, although human-to-human transmission has also been documented for Andes virus (ANDV). The ability to transmit via aerosolization, coupled with the high mortality rates and lack of therapeutic options, makes the development of medical countermeasures against HPS imperative. In the present study, we evaluated the efficacy of the broad-spectrum antiviral agent favipiravir (T-705) against Sin Nombre virus (SNV) and ANDV, the predominant causes of HPS in North and South America, respectively. In vitro, T-705 potently inhibited SNV and ANDV, as evidenced by decreased detection of viral RNA and reduced infectious titers. For both viruses, the 90% effective concentration was estimated at ≤5 μg/ml (≤31.8 μM). In the lethal ANDV hamster model, daily administration of oral T-705 at 50 or 100 mg/kg of body weight diminished the detection of viral RNA and antigen in tissue specimens and significantly improved survival rates. Oral T-705 therapy remained protective against HPS when treatment was initiated prior to the onset of viremia. No disease model for SNV exists; however, using a hamster-adapted SNV, we found that daily administration of oral T-705 significantly reduced the detection of SNV RNA and antigen in tissue specimens, suggesting that the compound would also be effective against HPS in North America. Combined, these results suggest that T-705 treatment is beneficial for postexposure prophylaxis against HPS-causing viruses and should be considered for probable exposures.

Fig 1

In vitro efficacy of T-705 against Andes and Sin Nombre viruses. Vero E6 cells were infected with Andes virus or Sin Nombre virus (MOI, 0.01) and were cultured in the presence of varying concentrations of T-705. Samples (cell pellets and supernatants) were collected on days 3, 5, and 7 postinfection to determine the extent of viral replication. (A through D) For both viruses, treatment with T-705 at concentrations of 25 μg/ml or greater resulted in 100- to 1,000-fold reductions in viral RNA levels in supernatants (A and B) and cell pellets (C and D) from those for matched samples collected from infected, mock-treated cells. (E) In agreement with these data, T-705 greatly reduced the levels of progeny virus, with estimated EC₉₀s between 2.5 and 5 μg/ml for both viruses. Shown are the infectious titers determined in clarified supernatant samples collected 7 days postinfection. Each point represents the average value obtained from triplicate samples. Error bars represent the standard errors of the means. The limit of detection for the focus assay was 25 infectious particles per ml.

Fig 2

In vivo efficacy of T-705 against Andes virus. Seven groups of 9 hamsters were infected with a lethal dose of Andes virus and were treated daily with either T-705 (1, 5, 20, 50, or 100 mg/kg), 20 mg/kg ribavirin (designated R20mg), or a placebo (0 mg/kg; vehicle only) beginning 1 day postinfection. For each group, 6 hamsters were monitored for disease progression and survival, while 3 hamsters per group were euthanized for sample collection on day 8 postinfection. (A) Treatment with 50 or 100 mg of T-705/kg/day resulted in significant increases in the survival rate over that for placebo-treated control animals. (B) In agreement with the survival curves, higher lung weight-to-body weight ratios were noted for placebo-treated hamsters than for mock-infected hamsters (gray dashed line) or for groups treated daily with T-705 at 50 to 100 mg/kg. (C and D) Quantitative real-time RT-PCR was utilized to measure viral loads in tissue samples. Standards were derived from samples with known infectious titers. Dose-dependent reductions in the detection of viral RNA in blood (C) and lung (D) samples were observed, most notably for hamsters receiving T-705 at 100 mg/kg/day. Error bars represent the standard errors of the means. *, P < 0.05; **, P < 0.01.

Fig 3

Immunohistochemical analysis of lungs from Andes virus-infected hamsters treated daily with T-705, ribavirin, or a placebo. Lung specimens collected at 8 days postinfection from hamsters infected with a lethal dose of Andes virus and treated daily beginning 1 day postinfection with varying concentrations of T-705, 20 mg ribavirin/kg (R20 mg), or a placebo (0 mg/kg) were tested for the presence of Andes viral antigen (nucleocapsid protein) by standard immunohistochemistry techniques. In agreement with the RT-PCR analysis, treatment with 50 or, most notably, 100 mg T-705/kg/day resulted in reduced detection of Andes viral antigen.

Fig 4

Efficacy of delayed T-705 therapy. Five groups of 6 hamsters were infected with a lethal dose of Andes virus. On each of days 3, 4, 5, and 6 postinfection, T-705 therapy (a 300-mg/kg loading dose, followed by daily treatment with 200 mg/kg) was initiated for a single group of animals. The fifth group received daily placebo (vehicle-only) treatments beginning 3 days postinfection. T-705 therapy remained 100% effective at preventing lethal HPS disease in hamsters when therapy was initiated on or before day 4 postinfection. (Inset) The dramatic loss of efficacy observed when therapy was initiated after day 4 coincided with the onset of viremia, as suggested by the detection of Andes virus RNA in blood samples collected from individual groups immediately prior to the commencement of treatment. Error bars represent the standard errors of the means.

Fig 5

In vivo efficacy of T-705 against Sin Nombre virus. Three groups of 6 hamsters were infected with a hamster-adapted Sin Nombre virus and were treated daily, beginning 1 day postinfection, with T-705 (100 mg/kg/day), ribavirin (20 mg/kg/day), or a placebo. (A and B) Examination of blood (A) and lung (B) samples collected at 8 days postinfection revealed reductions in virus RNA levels as determined by quantitative real-time RT-PCR using standards derived from samples with known infectious titers. Error bars represent the standard errors of the means. ***, P < 0.0001. (C) In agreement with these findings, Sin Nombre viral nucleocapsid antigen was virtually undetectable in lung specimens from T-705-treated hamsters, in contrast to matched samples from placebo- or ribavirin-treated animals.