PSI-7851, a Pronucleotide of β-d-2′-Deoxy-2′-Fluoro-2′-C-Methyluridine Monophosphate, Is a Potent and Pan-Genotype Inhibitor of Hepatitis C Virus Replication

Hepatitis C virus (HCV) currently affects more than 170 million people worldwide. Approximately 70% of infected individuals develop chronic hepatitis, among whom about 20% will develop liver cirrhosis and fibrosis and up to 5% will progress to hepatocellular carcinoma (2). The current standard of care (SOC), which combines pegylated alpha interferon (PegIFN-α) and ribavirin (RBV), has limited efficacy in providing a sustained virological response (SVR), especially in individuals with HCV genotype 1 (∼50%), the most prevalent genotype in Western countries (8, 11, 35). The impact of genetic diversity of HCV in patients receiving SOC therapy has been reviewed (26): SVR rates are higher in patients infected with genotype 2 or 3 (∼80%), patients infected with genotype 4 appear to have a slightly better SVR rate (∼60%) than patients infected with genotype 1, and patients infected with genotypes 5 and 6 may achieve an SVR at a level between those of genotypes 1 and 2/3. In addition to the variability in efficacy, the lengthy treatment (24 to 48 weeks) with SOC is frequently associated with undesirable side effects that may include anemia, fatigue, and depression (7). There is an urgent medical need to develop anti-HCV therapies that are safer and more effective. Direct-acting antivirals (DAAs) are compounds that target a specific viral protein. Currently, four major classes of DAAs are being investigated in phase II or III clinical trials: NS3 protease inhibitors, NS5A inhibitors, allosteric nonnucleoside NS5B polymerase inhibitors, and nucleoside/-tide NS5B polymerase inhibitors (21, 27, 46). Challenges for these DAAs include safety, pan-genotypic activity, and/or emergence of resistant viruses. An effective antiviral therapy against hepatitis C should encompass a broad spectrum of activity against all HCV genotypes, shorten treatment duration, have minimal side effects, and have a high barrier to resistance. The HCV NS5B RNA-dependent RNA polymerase (Pol) is a critical component of the replicase complex and is responsible for initiating and catalyzing viral RNA synthesis (16, 32, 58). There is no human homolog of this protein, and it is absolutely required for viral infectivity (19). As a result, the HCV NS5B is an attractive target for the development of antiviral compounds. There are two major classes of NS5B inhibitors: nucleoside analogs, which are anabolized to their active triphosphates and act as alternative substrates for the polymerase, and nonnucleoside inhibitors (NNIs), which bind to allosteric regions on the protein. Two major drawbacks associated with NNIs are that the activity appears to vary significantly among different HCV genotypes and even subtypes (15, 33) and that there is a relatively low barrier for resistance as evidenced by the numerous naturally occurring resistant variants reported in the literature (18). In contrast, nucleoside analogs are similarly active across HCV genotypes (13, 15, 33) and have a higher barrier of resistance compared to the NNIs and NS3 protease inhibitors (36). To date only two amino acid changes within the NS5B polymerase that confer resistance to nucleoside inhibitors have been identified: S96T and S282T (1, 29). The S96T mutation confers resistance to 4′-azidocytidine (R1479), while the S282T mutation is resistant to a number of 2′-C-methyl-modified nucleoside inhibitors (1, 29, 38, 43). In order for nucleoside analogs to be active as alternative substrates, they must first be phosphorylated by cellular kinases to their corresponding 5′-triphosphates, which are active alternative substrate inhibitors for the NS5B polymerase. The efficiency of these metabolic steps, the stability of the triphosphates, and the affinity of the triphosphates for the NS5B polymerase are all important factors in determining the antiviral activities of nucleoside inhibitors. PSI-6130, 2′-F-2′-C-methylcytidine, was previously shown to be a specific inhibitor of HCV RNA replication in the replicon assay system (52). However, when the uridine analog, 2′-F-2′-C-methyluridine (referred to as PSI-6206), was tested in the replicon assay, it failed to inhibit HCV RNA synthesis due to the inability of cellular enzymes to metabolize PSI-6206 to its triphosphate, PSI-7409 (5, 34, 42). Biochemical studies with PSI-7409 showed that this compound was able to inhibit RNA synthesis mediated by the HCV replicase complex and by purified recombinant HCV NS5B polymerase (34, 42). Furthermore, in vitro stability studies using primary human hepatocytes demonstrated that PSI-7409 has a significantly longer half-life (t1/2, 38 h) than PSI-6130-TP (t1/2, 4.7 h), which could be a desirable pharmacologic benefit (34). In order to bypass the initial nonproductive phosphorylation step of PSI-6206, the phosphoramidate prodrug methodology was explored as an approach to deliver 2′-F-2′-C-methyluridine monophosphate (47, 48). An extensive series of phosphoramidate prodrugs were synthesized, and PSI-7851 demonstrated the desired characteristics with regard to activity and in vitro toxicity. Herein we present the results of in vitro studies characterizing PSI-7851, a potent and specific anti-HCV compound with pan-genotype activity.