Preclinical characterization of an intravenous coronavirus 3CL protease inhibitor for the potential treatment of COVID-19

The COVID-19 pandemic, caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has presented an unprecedented global health crisis. While vaccines offer a path towards eradication, the need for effective antiviral treatments remains critical, particularly given the potential emergence of future coronaviruses. Currently, remdesivir is the only FDA-approved antiviral for COVID-19, but its efficacy has been variable. Monoclonal antibodies targeting the viral spike protein have also shown promise, but their use is limited to early stages of infection. Therefore, the development of novel antiviral therapies targeting different stages of the viral life cycle is paramount. The 3C-like protease (3CL protease or 3CLpro), a crucial enzyme for SARS-CoV-2 replication, presents an attractive drug target due to its essential role in viral polyprotein processing and the absence of close human homologs, suggesting potential for selective inhibition. Previous work with SARS-CoV, a closely related coronavirus, identified PF-00835231 as a potent 3CL protease inhibitor. Given the high sequence similarity between SARS-CoV and SARS-CoV-2 3CL proteases, particularly in the active site, PF-00835231 became a promising candidate for COVID-19 treatment. However, its poor oral bioavailability necessitated the development of a suitable delivery method and prompted the development of PF-07304814, a phosphate prodrug designed to enhance systemic exposure and therapeutic efficacy. This study investigates the preclinical characteristics of PF-07304814 and its active moiety PF-00835231, focusing on its antiviral activity, pharmacokinetic properties, and safety profile to support its clinical development as a potential treatment for COVID-19.

Extensive research has focused on developing effective antiviral treatments for coronaviruses. The main protease (Mpro), also known as 3CLpro, has emerged as a critical target due to its essential role in viral replication and its unique sequence compared to human proteases. Studies have demonstrated the efficacy of protease inhibitors against other viruses like HIV and HCV. The structure of the SARS-CoV Mpro has been determined, which facilitated structure-based drug design approaches. Remdesivir, a nucleoside analog targeting the viral RNA-dependent RNA polymerase (RdRp), has been approved for COVID-19, but its efficacy is debated. Monoclonal antibodies targeting the spike protein have also shown clinical success, but are limited to early-stage infections. There is a clear unmet need for effective broad-spectrum antiviral agents with diverse mechanisms of action, and 3CL protease inhibition remains an important area of exploration. Several studies have previously identified and characterized small-molecule inhibitors of the SARS-CoV Mpro. These compounds have provided valuable insights into the structural features required for effective inhibition and have served as the foundation for the development of new inhibitors for SARS-CoV-2.

The study utilized a multi-faceted approach to evaluate PF-00835231 and its prodrug PF-07304814. In vitro studies included thermal shift assays to assess binding affinity to SARS-CoV-2 3CLpro, biochemical assays to determine inhibitory activity against a panel of coronavirus 3CLpro enzymes, and cell-based assays to measure antiviral activity against SARS-CoV-2 in various cell lines (VeroE6-enACE2, VeroE6-EGFP, A549-ACE2, and MRC-5). The impact of P-glycoprotein (P-gp) efflux on antiviral activity was also investigated using a P-gp inhibitor (CP-100356). Combination studies with remdesivir were performed using HeLa-ACE2 cells and high-content imaging, with synergy analysis conducted using Loewe, Bliss, and HSA models, as well as SynergyFinder. In vivo efficacy studies were conducted in mouse models of SARS-CoV (MA15 strain) and SARS-CoV-2 infection. For the SARS-CoV-2 model, mice were transduced with Ad5-hACE2 to enable infection. PF-00835231 was administered subcutaneously (s.c.), and viral titers and weight changes were monitored. Extensive ADME (absorption, distribution, metabolism, excretion) studies were performed, including in vitro assessments of metabolic stability, cytochrome P450 enzyme inhibition, and transporter inhibition, and in vivo pharmacokinetic (PK) analyses in rats, dogs, and monkeys. Human PK parameters were predicted using in vitro and in vivo data. Safety pharmacology assessments included genetic toxicity, secondary pharmacology profiling, and hERG inhibition assays. A 24-hour continuous intravenous infusion GLP toxicity study was performed in rats, along with an exploratory non-GLP toxicity study in rats for PF-00835231. The synthesis of PF-07304814 is described, including detailed methods for characterization and assessment of its physicochemical properties (LogD, solubility, pKa) using shake-flask, LC-UV and NMR techniques. In vitro assays were used to determine metabolic stability of PF-07304814 and the extent of its conversion to PF-00835231. High performance liquid chromatography tandem mass spectrometry (HPLC-MS) was used to assess compounds and metabolites in metabolic stability, enzyme kinetics and PK studies. Statistical analyses were employed to determine significance.

PF-00835231 demonstrated tight and specific binding to SARS-CoV-2 3CLpro (Tm shift of 14.6 °C). It exhibited potent and broad-spectrum inhibitory activity against a panel of coronavirus 3CLpro enzymes (Ki values ranging from 30 pM to 4 nM), with selectivity over human proteases. In vitro cellular antiviral activity against SARS-CoV-2 was initially modest (EC50 values in the micromolar range), but significantly improved with the addition of a P-gp inhibitor, suggesting P-gp-mediated efflux. The increased potency with P-gp inhibition in Vero cells aligned with activity observed in human lung cell lines with lower P-gp expression. PF-00835231 showed additive/synergistic effects in combination with remdesivir. In vivo studies in mouse models of SARS-CoV and SARS-CoV-2 infection demonstrated dose-dependent reduction in lung viral titers and improved weight retention following PF-00835231 administration. Treatment efficacy was retained even when treatment was delayed by one day. ADME studies revealed that PF-00835231 is primarily metabolized by CYP3A4, with a moderate plasma clearance and short half-life. It showed low risk for drug-drug interactions through CYP inhibition and transporter inhibition. Oral bioavailability was low (<2%). PF-07304814, a phosphate prodrug of PF-00835231, exhibited greatly enhanced aqueous solubility (>200 mg/mL), enabling higher achievable doses compared to PF-00835231. In vivo, PF-07304814 was efficiently converted to PF-00835231, achieving high systemic exposure. The predicted human PK profile of PF-07304814 indicates the potential to achieve therapeutic concentrations of PF-00835231 through intravenous continuous infusion. Preclinical safety studies, including genetic toxicity, secondary pharmacology, and hERG inhibition, demonstrated a favorable safety profile for PF-07304814, with a NOAEL of 1000 mg/kg in rats. Minimal, non-adverse effects were observed in rats with high doses of PF-00835231.

The findings demonstrate that PF-07304814, a highly soluble phosphate prodrug of PF-00835231, holds considerable promise as a potential intravenous treatment for COVID-19. The efficient in vivo conversion of PF-07304814 to PF-00835231, coupled with its favorable PK properties and robust antiviral activity against SARS-CoV-2, suggests its suitability for clinical development. The observed synergy/additivity with remdesivir adds further therapeutic value. The extensive preclinical safety profile supports the advancement of PF-07304814 into clinical trials. The efficient conversion of PF-07304814 to PF-00835231 in vivo further supports the design strategy for improving the bioavailability of PF-00835231. The study highlights the importance of considering drug efflux transporters in in vitro antiviral assays, particularly when using Vero cells, which express high levels of P-gp. Future studies should explore the potential for resistance development and the long-term safety of PF-07304814. The potential for combination therapies involving PF-07304814 should also be investigated.

This study provides comprehensive preclinical data supporting the clinical development of PF-07304814 as a potential COVID-19 treatment. The phosphate prodrug strategy successfully addressed the poor solubility and bioavailability of the active moiety PF-00835231. The potent broad-spectrum antiviral activity, favorable safety profile, and potential for combination therapy warrant further investigation in clinical trials. Future research should focus on optimizing dosing regimens, evaluating long-term safety, and investigating potential resistance mechanisms.

The study primarily used in vitro and animal models, which may not fully reflect the complexity of human COVID-19 infection. While the mouse models utilized were designed to mimic human infection, there are inherent differences between animal and human physiology that could affect the translation of these findings to clinical settings. Further, human pharmacokinetic predictions are based on preclinical data, and there may be inter-individual variability in human response. The study didn't explicitly address resistance mechanisms.