Given that virtually all seasonal H3N2 and 2009 H1N1 strains are resistant to amantadine, and virtually all currently circulating seasonal H1N1 strains are resistant to oseltamivir, the pharmacologic rationale for the development of a triple combination antiviral drug (TCAD) composed of amantadine, ribavirin, oseltamivir is that at least two, and possibly three drugs, in the TCAD regimen will be active against all of these viruses. A number of studies have evaluated double combinations of antivirals 
against influenza A infection in vitro
, and Hayden et al. have tested a triple combination of two antivirals with human interferon α 
. However, there have been few reports on the effects of drug combinations on resistant influenza viruses 
. Recently, Smee et al. evaluated the effects of double combinations of amantadine, ribavirin, and oseltamivir against the same amantadine-resistant H5N1 virus used in this study (A/Duck/MN/1525/81) 
. The authors found that the presence of amantadine in double combinations did not provide added benefit over the second drug alone, either in cell culture or in mouse models.
In the present study, we examined the efficacy and synergy of the TCAD regimen against viruses which were resistant to oseltamivir or amantadine, including 2009 H1N1. Consistent with the previous findings 
, we found that the 2009 H1N1 strains were susceptible to NAIs (oseltamivir carboxylate, zanamivir, and peramivir) and ribavirin, but were resistant to adamantanes (amantadine and rimantadine). Surprisingly, we found that amantadine as a single agent retained partial activity against these viruses (Table S2
), albeit the activity was reduced by 100-fold compared to a susceptible virus, whereas rimantadine had no activity below the 50% cytotoxic concentration. This observation suggests that phenotypic testing, in addition to determination of the genotype, may be necessary in order to fully understand the susceptibility profile of a novel virus and may have important implications in guiding the choice of antivirals for use in combinations.
Against the 2009 H1N1 strains, the interactions of oseltamivir carboxylate, peramivir, and zanamivir in double combinations ranged from additive to moderately antagonistic, indicative that the activity of these drugs was not enhanced in combination compared to their activity as single agents. These results suggest that double combinations of NAIs may not provide any added benefit over the drugs as single agents. Given that all NAIs bind in the same substrate binding pocket in NA, the use of these drugs in combination in the absence of enhanced activity raises the risk of selecting for a single mutation that could confer resistance to both neuraminidase inhibitors simultaneously. Indeed, cross-resistance to oseltamivir and zanamivir resulting from a single amino acid change has been documented for seasonal influenza A and B viruses 
. If this were to occur in the 2009 H1N1 background, the resulting virus would be resistant to all approved anti-influenza drugs.
In total, we tested the activity and synergy of the TCAD regimen against six amantadine-resistant viruses, including three strains of 2009 H1N1, and two oseltamivir-resistant viruses. The viruses tested in this study come from the three subtypes that cause significant morbidity and mortality in humans (H1N1, H3N2, and H5N1), and include seasonal, avian, and pandemic strains. The double combinations of amantadine/oseltamivir carboxylate, amantadine/ribavirin, and ribavirin/oseltamivir carboxylate were additive against 2009 H1N1, and ranged from additive to moderately synergistic against the other viruses (data not shown). In contrast, with the exception of the duck H5N1 virus, we found that the TCAD regimen was synergistic at clinically achievable concentrations of all three drugs, and that the synergy of the TCAD regimen was greater than that of any double antiviral drug combination. These data suggest that the TCAD regimen may have broad-spectrum antiviral activity against circulating influenza A viruses, including strains that are resistant to either classes of antivirals. To date, most influenza A strains in circulation (~99%) are resistant to either the adamantanes or oseltamivir, and not to both 
, and thus are expected to be susceptible to the TCAD regimen. Currently, rapid diagnostic tests are not available to determine the susceptibility profile of influenza viruses in real time, and thus clinicians do not often have the necessary information with which to guide appropriate antiviral use. The availability of a broad-spectrum antiviral therapy that would be effective against the majority of circulating strains regardless of the susceptibility would be of high clinical utility.
Importantly, we found that amantadine and oseltamivir contributed to the synergy of the TCAD regimen against amantadine-resistant and oseltamivir-resistant viruses. The contributions from both drugs to the synergy of the TCAD regimen were significant at clinically achievable concentrations where they had little or no antiviral activity as a single agent. For instance, a comparison of the synergy volume of the TCAD regimen at 0.32 µg/mL amantadine to the synergy volume of the ribavirin/oseltamivir carboxylate double combination (no amantadine) revealed that amantadine contributed 39%, 24%, and 44% to the total synergy of the TCAD regimen against CA04, CA05, and CA10, respectively. Similarly, against the oseltamivir-resistant viruses, oseltamivir carboxylate at 0.32 µg/mL contributed 76% and 83% to the total synergy of the TCAD regimen against MS H274Y and HI H274Y, respectively. Thus, all three drugs contributed to the synergy and activity of the TCAD regimen against amantadine- and oseltamivir-resistant viruses, and the activities of amantadine and oseltamivir were restored in the context of the TCAD regimen against influenza strains that were resistant to these drugs, thereby maximizing the clinical utility of these drugs.
The mechanism(s) by which amantadine and oseltamivir carboxylate contribute to the synergy of the TCAD regimen against resistant strains is unclear. The interactions between M2, HA, and NA on the surface of the influenza particle are complex and not well understood, and a number of studies have demonstrated that HA-M2 and HA-NA interactions can affect the susceptibility to amantadine and oseltamivir, respectively 
. Furthermore, amantadine has been demonstrated to exert antiviral activity via interactions with HA at higher concentrations 
. It is conceivable that, as the result of protein-protein interactions between M2, HA, and NA, the binding of a drug at one site may affect the conformation and therefore affinity for another drug at another site. The mechanism by which ribavirin contributes to the synergy of the TCAD regimen is also unclear. Ribavirin has been documented to act through multiple mechanisms affecting both virus replication and host immune response 
, and it remains to be elucidated which of these mechanisms are responsible for the synergy with amantadine and oseltamivir.
Finally, we evaluated the activity and inhibitory quotient (IQ) of TCAD against susceptible and resistant viruses representing the currently circulating strains. While the correlation between IQ and clinical efficacy has not been demonstrated for influenza, it is valuable to construct a relative ranking of the IQ of different antiviral regimens against susceptible and resistant viruses in order to assess the spectrum of their activity. When tested against a seasonal susceptible H1N1 virus, an amantadine-resistant 2009 H1N1, and a seasonal oseltamivir-resistant H1N1 virus, TCAD was uniformly active against all three viruses with significantly high IQs (8.33 to 17.24; ). This suggests that TCAD may have broad antiviral activity against all currently circulating influenza strains and may have good efficacy in the clinical setting against these strains.
Our data suggests that a triple combination antiviral drug (TCAD) composed of amantadine, ribavirin, and oseltamivir may be an effective and viable therapeutic option for the treatment of pandemic and seasonal influenza infection. The body of data presented in this report validates the TCAD hypothesis, which states that for any given susceptible or resistant circulating influenza virus, at least two, and in some cases all three, drugs in TCAD will be active. Furthermore, the TCAD regimen appears to overcome baseline drug resistance and thus may represent a highly active antiviral therapy for seasonal and pandemic influenza. The safety, pharmacokinetics, distribution, and metabolism of amantadine, ribavirin, and oseltamivir as single agents are well understood, and it is not expected that co-administration of the three drugs will result in substantially increased risk to patients compared to the administration of the individual drugs. In addition, all three double combinations have been tested in humans without adverse effects, including amantadine plus oseltamivir 
, amantadine plus ribavirin 
, and ribavirin plus oseltamivir 
. Clinical trials to assess the efficacy and safety of TCAD for the treatment of influenza have been initiated, and will provide important data on the use of TCAD against both pandemic and seasonal influenza.