Rethinking Remdesivir: Synthesis, Antiviral Activity, and Pharmacokinetics of Oral Lipid Prodrugs

Abstract

Remdesivir (RDV; GS-5734) is currently the only FDA-approved antiviral drug for the treatment of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection. The drug is approved for use in adults or children 12 years or older who are hospitalized for the treatment of COVID-19 on the basis of an acceleration of clinical recovery for inpatients with this disease. Unfortunately, the drug must be administered intravenously, restricting its use to those requiring hospitalization for relatively advanced disease. RDV is also unstable in plasma and has a complex activation pathway which may contribute to its highly variable antiviral efficacy in SARS-CoV-2-infected cells. Potent orally bioavailable antiviral drugs for early treatment of SARS-CoV-2 infection are urgently needed, and several, including molnupiravir and PF-07321332, are currently in clinical development. We focused on making simple, orally bioavailable lipid analogs of remdesivir nucleoside (RVn; GS-441524) that are processed to RVn monophosphate, the precursor of the active RVn triphosphate, by a single-step intracellular cleavage. In addition to high oral bioavailability, stability in plasma, and simpler metabolic activation, new oral lipid prodrugs of RVn had submicromolar anti-SARS-CoV-2 activity in a variety of cell types, including Vero E6, Calu-3, Caco-2, human pluripotent stem cell (PSC)-derived lung cells, and Huh7.5 cells. In Syrian hamsters, oral treatment with 1-O-octadecyl-2-O-benzyl-glycero-3-phosphate RVn (ODBG-P-RVn) was well tolerated and achieved therapeutic levels in plasma above the 90% effective concentration (EC₉₀) for SARS-CoV-2. The results suggest further evaluation as an early oral treatment for SARS-CoV-2 infection to minimize severe disease and reduce hospitalizations.

Keywords: Caco-2 cells; Calu-3 cells; Huh7.5 cells; PSC-derived human lung cells; SARS-CoV-2; Vero E6 cells; antiviral agents; lipid prodrugs; remdesivir; remdesivir nucleoside.

FIG 1

Structures of remdesivir and related intermediates.

FIG 2

Synthesis of antiviral prodrugs 5a to 5c. a, 2′,3′-isopropylidene RVn (2a), DCC, DMAP, and pyridine, 90°C, 24 to 72 h; b, 37% HCl, THF, 3 to 18 h.

FIG 3

ODBG-P-RVn stability in human plasma at 37°C. Plasma was spiked with 2-μg/ml concentrations of ODBG-P-RVn and incubated at 37°C for 24 h. Samples were taken at 0.5, 1, 2, 4, 8, and 24 h and frozen for later analysis by LC-MS/MS.

FIG 4

SARS-CoV-2 inhibitory activity replicate experiments. (A and B) Antiviral dose-response curves for three remdesivir analogs, remdesivir (GS-5734), and remdesivir nucleoside (GS-441524) against SARS-CoV-2 infection in multiple cell types. SARS-CoV-2 relative viral RNA reduction (A) and percent inhibition (B) in the indicated cell types. Cells were pretreated with the indicated dose of the indicated drug for 30 min and then infected with SARS-CoV-2 isolate USA-WA1/2020 for 48 h. The relative SARS-CoV-2 spike RNA expression was determined by qRT-PCR. (C) Cytotoxicity in the indicated cells incubated in the presence of the indicated drug at the indicated concentration for 48 h, after which, cell viability was measured by the CellTiter-Glo assay. (A to C) Each antiviral and cytotoxicity dose-response data point indicates the average from 3 independent experiments performed in duplicates, except as indicated in Table 1. Error bars represent the standard errors of the means (SEMs).

FIG 5

ODBG-P-RVn antiviral activity and cytotoxicity is highly reproducible across cell lines. (A) EC₅₀ values for RDV, RVn, and ODBG-P-RVn in 5 different cell lines derived from lung (human PSC-derived, Calu-3), kidney (Vero E6), colon (Caco-2), or liver (Huh7.5). Cytotoxicity of ODBG-P-RVn (B) or RDV (C) in the indicated cells at the indicated concentration for 48 h, as measured by the CellTiter-Glo assay.

FIG 6

ODBG-P-RVn inhibits the human Alphacoronavirus 229E. (A) Antiviral dose-response curves for remdesivir (GS-5734) and ODBG-P-RVn against the human coronavirus 229E in MRC-5 cells. Cells were infected with 229E for 2 h followed by treatment with the indicated dose of the indicated drug for 72 h. The relative CPE was determined by measuring cell viability using an MTT assay. (B) Cytotoxicity in MRC-5 cells incubated in the presence of the indicated drug at the indicated concentration for 72 h, after which cell viability was measured by the MTT assay. Data points indicate the averages from 3 independent experiments performed in duplicates. Error bars represent the standard errors of the means (SEMs).

FIG 7

Seven-day oral pharmacokinetics in Syrian hamsters. Syrian hamsters were given vehicle or ODBG-P-RVn by oral gavage every 12 h for 7 days. Groups of 3 animals received vehicle or drug at doses of 16.9 and 13.2 mg/kg body weight. Animals were weighed daily and monitored for clinical signs. Plasma samples were obtained at 1, 3, 6, and 12 h on day 1 and day 7 and frozen for analysis of ODBG-P-RVn (A) and RVn (B) by LC-MS/MS. The EC₉₀ values in Vero E6 and PSC-lung cells for ODBG-P-RVn (A) and RVn (B) are shown.