Therapeutic efficacy of the small molecule GS-5734 against Ebola virus in rhesus monkeys

Abstract

The most recent Ebola virus outbreak in West Africa, which was unprecedented in the number of cases and fatalities, geographic distribution, and number of nations affected, highlights the need for safe, effective, and readily available antiviral agents for treatment and prevention of acute Ebola virus (EBOV) disease (EVD) or sequelae. No antiviral therapeutics have yet received regulatory approval or demonstrated clinical efficacy. Here we report the discovery of a novel small molecule GS-5734, a monophosphoramidate prodrug of an adenosine analogue, with antiviral activity against EBOV. GS-5734 exhibits antiviral activity against multiple variants of EBOV and other filoviruses in cell-based assays. The pharmacologically active nucleoside triphosphate (NTP) is efficiently formed in multiple human cell types incubated with GS-5734 in vitro, and the NTP acts as an alternative substrate and RNA-chain terminator in primer-extension assays using a surrogate respiratory syncytial virus RNA polymerase. Intravenous administration of GS-5734 to nonhuman primates resulted in persistent NTP levels in peripheral blood mononuclear cells (half-life, 14 h) and distribution to sanctuary sites for viral replication including testes, eyes, and brain. In a rhesus monkey model of EVD, once-daily intravenous administration of 10 mg kg(-1) GS-5734 for 12 days resulted in profound suppression of EBOV replication and protected 100% of EBOV-infected animals against lethal disease, ameliorating clinical disease signs and pathophysiological markers, even when treatments were initiated three days after virus exposure when systemic viral RNA was detected in two out of six treated animals. These results show the first substantive post-exposure protection by a small-molecule antiviral compound against EBOV in nonhuman primates. The broad-spectrum antiviral activity of GS-5734 in vitro against other pathogenic RNA viruses, including filoviruses, arenaviruses, and coronaviruses, suggests the potential for wider medical use. GS-5734 is amenable to large-scale manufacturing, and clinical studies investigating the drug safety and pharmacokinetics are ongoing.

Figure 1. Metabolism and mechanism of antiviral activity of GS-5734.

a, Chemical structures of GS-5734 and metabolic conversion to NTP. b, NTP formation in human monocyte-derived macrophages following 72-h incubation with 1 μM GS-5734 (black) or Nuc (red); mean ± s.d., from 3 donors. c, Antiviral activity of GS-5734 in HeLa cells against EBOV Makona (black symbols), EBOV Kikwit (open symbols), Marburg (red), Bundibugyo (orange), Sudan (blue) viruses; mean ± s.d. from triplicates. d, Amino acid sequence homology of EBOV and RSV strain A2 RdRp active site. Asterisks indicate residues predicted to contact NTP. e, Inhibition of RSV RdRp (blue), but not human RNA Pol II (black) or mitochondrial RNA (red) polymerases by NTP; mean ± s.d., n = 3 biological replicates. f, NTP-induced RNA chain termination in RSV RdRp primer-extension assay. For gel source data, see Supplementary Fig. 1. PowerPoint slide

Figure 2. GS-5734 pharmacokinetics and post-exposure protection against EBOV in rhesus monkeys.

a, Pharmacokinetics following intravenous administration of 10 mg kg⁻¹ GS-5734 dose in healthy rhesus macaques (mean ± s.d., n = 3). Plasma GS-5734 (black), alanine metabolite (red), and Nuc (blue); NTP in PBMCs (green). b, Tissue distribution of [¹⁴C]GS-5734 and metabolites at 4 h (blue) and 168 h (red) following intravenous 10 mg kg⁻¹ GS-5734 dose in healthy cynomolgus macaques (mean ± s.d., n = 3). c, Experimental design for GS-5734 efficacy evaluations in rhesus monkeys. No T_x, no treatment. d, Kaplan–Meier survival curves. *P < 0.05 for treatment versus vehicle groups assessed by log-rank analysis using Dunnett–Hsu procedure to adjust for multiple comparisons. e, Group geometric mean of plasma viral RNA concentrations; LLOQ, lower limit of quantitation; LOD, limit of detection. f, Individual plasma viral RNA in vehicle (blue) or 10 mg kg⁻¹ GS-5734 (black) groups. g, Group average clinical disease score. d, e, g, Black (open symbols), vehicle; red, 3 mg kg⁻¹ d0; green, 3 mg kg⁻¹ d2; blue, 10/3 mg kg⁻¹ d2; orange, 10/3 mg kg⁻¹ d3; black (closed symbols), 10 mg kg⁻¹ d3; n = 6 animals per group. Error bars omitted for clarity (e, g); x axes truncated to emphasize acute disease phase (f, g). PowerPoint slide Source data

Figure 3. Amelioration of EVD clinical pathology by GS-5734 in rhesus monkeys.

a–f, Group mean (n = 6 per group) values of platelets (a), thrombin time (b), aspartate aminotransferase (AST, c), alanine aminotransferase (ALT, d), blood urea nitrogen (BUN, e), and creatinine (f). Black (open symbols), vehicle; red, 3 mg kg⁻¹ d0; green, 3 mg kg⁻¹ d2; blue, 10/3 mg kg⁻¹ d2; orange, 10/3 mg kg⁻¹ d3; black (closed symbols), 10 mg kg⁻¹ d3. Error bars omitted for clarity; x axes truncated at day 15. *P < 0.05 for comparison of mean change from day 0 of vehicle and 10 mg kg⁻¹ d3 groups at day 7 using Wilcoxon rank-sum test without adjustment for multiple comparisons. PowerPoint slide Source data

Extended Data Figure 1. Intracellular metabolism of GS-5734.

a, Intracellular metabolite profile in human macrophages. Following a 2-h pulse incubation (black bar at top of y axis) of human monocyte-derived macrophages with 1 μM GS-5734 (mean ± s.d., from three donors). GS-5734 is rapidly metabolized and not detected in cells. Transient exposure to the intermediate alanine metabolite (Ala-Met) is observed, followed by persistent Nuc exposure. The pharmacologically active NTP is formed quickly, achieving a maximum intracellular concentration at 4 h and persisted with a half-life of 16 ± 1 h in the three donors. Intracellular concentration was estimated on the basis of an intracellular volume of 1 pl per cell. b, Efficiency of GS-5734 activation in human and rhesus cells in vitro. Intracellular NTP concentrations formed in human and rhesus PBMCs, monocytes, and monocyte-derived macrophages during a 2-h incubation with 1 μM GS-5734 (results are the mean ± s.d. of two (PBMC and monocyte) to six (macrophage) independent experiments performed with cells from different donors). Intracellular concentrations were estimated on the basis of a cell volume of 0.2 pl per cell for PBMCs and monocytes and 1 pl per cell for macrophages. c, Intracellular NTP levels required for inhibition of EBOV replication in cell culture. The mixture of GS-5734 and its diastereomer on phosphorous was incubated continuously for 72 h at 1 μM and levels of intracellular NTP were determined (results are the average of duplicate incubations performed in each cell type; two independent studies were performed in HMVEC isolated from different donors). The corresponding EBOV EC₅₀ values for the prodrug diastereomeric mixture were 100, 184, and 121 nM in human macrophages, HeLa, and HMVEC, respectively, suggesting that an average intracellular NTP concentration of approximately 5 μM is required for 50% inhibition of EBOV in vitro. Source data

Extended Data Figure 2. Inhibition of EBOV Makona replication by GS-5734.

Huh-7 cells infected with wild-type EBOV (Makona) were incubated for 3 days in the presence of serial dilutions of GS-5734. The amount of infectious virus produced was quantified by endpoint dilution assay of culture media on fresh Vero cell monolayers and the tissue culture infectious dose that caused 50% infection (TCID₅₀) was determined. Independently, total RNA was extracted from infected cells and EBOV RNA levels were quantified using a nucleoprotein (NP) gene-specific qRT–PCR. Values represent mean ± s.d. of log₁₀-transformed values, n = 4 replicates.

Extended Data Figure 3. Homology model of RSV A2 (cyan) and EBOV (coral) polymerase based on HIV reverse transcriptase (PDB: 1RTD) with NTP (green and red representing the nucleoside and triphosphate portion, respectively).

Extended Data Figure 4. Clinical signs of disease in individual rhesus monkeys exposed to Ebola virus.

Animals were observed multiple times each day and were subjectively assigned a clinical disease score ranging from 0 to 5 based on responsiveness, posture, and activity. Maximum daily scores were converted to colour code, with darker colours indicative of more severe disease signs. The schematic was truncated to emphasize clinical scores during the acute disease phase, and none of the animals exhibited clinical disease signs outside of the times that are shown.

Extended Data Figure 5. Amelioration of EVD clinical pathology by GS-5734 in rhesus monkeys.

a–c, Group mean (n = 6 per group) values of d-dimer (a), activated partial thromboplastin time (b), and lipase (c). Black (open symbols), vehicle; red, 3 mg kg⁻¹ d0; green, 3 mg kg⁻¹ d2; blue, 10/3 mg kg⁻¹ d2; orange, 10/3 mg kg⁻¹ d3; black (closed symbols), 10 mg kg⁻¹ d3. Error bars omitted for clarity; x axes truncated at day 15. *P < 0.05 for comparison of mean change from d0 of vehicle and 10 mg kg⁻¹ d3 groups at day 7. Source data