Antiviral Efficacy and Host Innate Immunity Associated with SB 9200 Treatment in the Woodchuck Model of Chronic Hepatitis B

Abstract

SB 9200, an oral prodrug of the dinucleotide SB 9000, is being developed for the treatment of chronic hepatitis B virus (HBV) infection and represents a novel class of antivirals. SB 9200 is thought to activate the viral sensor proteins, retinoic acid-inducible gene 1 (RIG-I) and nucleotide-binding oligomerization domain-containing protein 2 (NOD2) resulting in interferon (IFN) mediated antiviral immune responses in virus-infected cells. Additionally, the binding of SB 9200 to these sensor proteins could also sterically block the ability of the viral polymerase to access pre-genomic RNA for nucleic acid synthesis. The immune stimulating and direct antiviral properties of SB 9200 were evaluated in woodchucks chronically infected with woodchuck hepatitis virus (WHV) by daily, oral dosing at 15 and 30 mg/kg for 12 weeks. Prolonged treatment resulted in 2.2 and 3.7 log10 reductions in serum WHV DNA and in 0.5 and 1.6 log10 declines in serum WHV surface antigen from pretreatment level with the lower or higher dose of SB 9200, respectively. SB 9200 treatment also resulted in lower hepatic levels of WHV nucleic acids and antigen and reduced liver inflammation. Following treatment cessation, recrudescence of viral replication was observed but with dose-dependent delays in viral relapse. The antiviral effects were associated with dose-dependent and long-lasting induction of IFN-α, IFN-β and IFN-stimulated genes in blood and liver, which correlated with the prolonged activation of the RIG-I/NOD2 pathway and hepatic presence of elevated RIG-I protein levels. These results suggest that in addition to a direct antiviral activity, SB 9200 induces antiviral immunity during chronic hepadnaviral infection via activation of the viral sensor pathway.

Fig 1. SB 9200 treatment induces dose-dependent, transient suppression of serum viremia and antigenemia.

Changes in WHV DNA (left panels) and WHsAg levels (right panels) relative to T₀ (pretreatment baseline) during daily, oral treatment with SB 9200 for 12 weeks in individual woodchucks administered a low (A, D) or high dose (B, E) and mean of each group (C, F). At T₀, mean WHV DNA levels were 4.78x10¹⁰ and 5.27x10¹⁰ genomic equivalents/mL serum and mean WHsAg levels were 3.30x10⁵ and 4.32x10⁵ ng/mL serum in the low or high dose groups, respectively. Error bars represent the standard error of the mean.

Fig 2. SB 9200 treatment results in dose-dependent increases in total plasma exposure of SB 9000.

Changes in mean plasma levels of SB 9000 during daily, oral treatment with SB 9200 for 12 weeks in woodchucks administered a low (A) or high dose (B). Changes in mean serum WHV DNA relative to T₀ is plotted on the left y-axis. Error bars represent the standard error of the mean. Because plasma values represent single time-point values at 2 hours post-dosing, it is likely that the peak individual plasma levels do not represent the actual Cmax since the Tmax in individual woodchucks could vary. Peak levels in individual woodchucks were between 25–150 and 75–170 ng/mL in the low and high dose groups, respectively.

Fig 3. SB 9200 treatment induces dose-dependent, transient reduction in hepatic WHV nucleic acids.

Changes in mean hepatic levels of WHV cccDNA (left panels), WHV RI DNA (middle panels) and WHV RNA (right panels) relative to week -1 (pretreatment baseline) in response to SB 9200 treatment at a low (A, D, G) or high dose (B, E, H) and mean of each group (C, F, G). At week -1, WHV DNA RI levels were 1830.9 and 1965.7 pg/μg total DNA and WHV RNA levels were 42.9 and 44.3 pg/μg total RNA in the low and high dose groups. Error bars represent the standard error of the mean.

Fig 4. SB 9200 treatment results in transient reduction of WHcAg expression and inflammation in liver.

Changes in mean immunohistochemistry (IHC) score for cytoplasmic WHcAg expression in liver (top panels) and liver histology score for portal and lobular sinusoidal hepatitis (bottom panels) in response to SB 9200 treatment at a low (A, C) or high dose (B, D). Changes in mean serum WHV DNA relative to T₀ is plotted on the left y-axis. Error bars represent the standard error of the mean.

Fig 5. SB 9200 treatment is associated with dose-dependent, transient increases of SDH but not of other liver enzymes.

Changes in mean serum levels of SDH, AST and ALT in response to SB 9200 treatment at a low (A) or high dose (B). Changes in mean serum WHV DNA relative to T₀ is plotted on the left y-axis. Error bars represent the standard error of the mean.

Fig 6. SB 9200 treatment induces dose-dependent and sometimes long-lasting expression increases of type I IFNs, cytokine, and ISGs in peripheral blood.

Changes in mean blood transcript levels of IFN-α, IFN-β, and IL-6 (top panels) and of CXCL10, OAS1 and ISG15 (bottom panels) in response to SB 9200 treatment at a low (A, C) or high dose (B, D). Changes in mean serum WHV DNA relative to T₀ is plotted on the left y-axis. Error bars represent the standard error of the mean.

Fig 7. SB 9200 treatment results in comparable and long-lasting expression increases of type I IFNs, cytokine, and ISGs in liver.

Changes in mean liver transcript levels of IFN-α, IFN-β, and IL-6 (top panels) and of CXCL10, OAS1 and ISG15 (bottom panels) in response to SB 9200 treatment at a low (A, C) or high dose (B, D). Changes in mean serum WHV DNA relative to T₀ is plotted on the left y-axis. Error bars represent the standard error of the mean.

Fig 8. SB 9200 treatment induces long-lasting activation of the RIG-I/NOD2 pathway and presence of elevated RIG-1 protein levels in liver.

Changes in mean liver transcript levels of RIG-I, NOD2, STING, and IRF3 in a subset of woodchucks in response to SB 9200 treatment at a low dose (n = 3) (A) or high dose (n = 3) (B). Changes in mean IHC scores for cytoplasmic RIG-I expression in liver of all woodchucks treated with SB 9200 at a low (n = 5) (C) or high dose (n = 5) (D). Changes in mean serum WHV DNA relative to T₀ is plotted on the left y-axis. Error bars represent the standard error of the mean.