(S-059) Population Pharmacokinetic, Exposure-Response, and Time-To-Event Analyses of Probenecid in Symptomatic, Non-Hospitalized Patients with COVID-19

Benjamin Kennard – Allucent; David Martin – TrippBio; Lisa Benincosa – Allucent; Jessica Roberts – Allucent

Objectives: Probenecid, an oral uricosuric agent, is being developed as a broad-spectrum antiviral and was evaluated for potential suppressive effects of SARS-Cov-2 replication in a Phase 2 study in patients with symptomatic, mild-to-moderate COVID-19 The aims of this study were to develop a population pharmacokinetic (popPK) model to characterize the pharmacokinetics (PK) of probenecid, evaluate the exposure-response (E-R) relationships between steady-state exposures (area under the concentration time curve for the dosing interval [AUCtau,ss] and maximal concentration [Cmax,ss]) versus viral load and body temperature, and conduct time-to-event (TTE) modeling to assess the relationship between steady-state exposures and time to viral clearance.

Methods: The analyses included data from a Phase 2 clinical efficacy study in patients who received placebo or 500 mg or 1000 mg probenecid twice daily for 5 days. Monolix 2024R1 was used for the popPK analysis. Individual patient empirical Bayes estimates of PK parameters were used to simulate rich curves in Simulx 2024R1 for steady-state exposure parameters after the first dose on Day 5 for E-R and TTE analyses. The E-R relationship between steady-state exposure parameters or dose versus body temperature and viral load were assessed via a linear regression model where the 95% confidence intervals and F-statistics were evaluated for significance in R 4.0.2. TTE analyses performed in R 4.0.2 were stratified by dose and steady-state exposure quartiles.

Results: The final popPK dataset contained 103 observations from 17 patients. A 1-compartment model with first order absorption, lag-time, linear elimination, and a proportional residual error model best characterized probenecid’s PK where model parameters were all well estimated with good agreement between observed, population, and individual predicted concentrations. No statistically significant covariates were identified.

The E-R analysis included a linear regression model that was able to identify a statistically significant relationship between both steady-state exposures (AUCtau,ss and Cmax,ss) and dose versus body temperature.

TTE survival analysis demonstrated patients who received 500 mg or 1000 mg BID reach viral clearance up to 2 and 4 days sooner, respectively, compared to placebo-treated patients. Additionally, TTE survival analysis demonstrated an exposure-dependent relationship between probenecid concentrations and time to viral clearance when stratified by steady-state exposure quartiles.

Conclusions: The popPK model successfully described probenecid disposition, generating reliable model-predicted steady state exposures for subsequent E-R and TTE analyses. The results of the E-R and TTE analyses suggest there is an exposure dependent relationship between probenecid dose or concentrations and time to viral clearance.

Citations: [1] Murray J, Hogan RJ, Martin DE, et al. Probenecid inhibits SARS-CoV-2 replication in vivo and in vitro. Sci Rep. 2021;11(1):18085. Published 2021 Sep 10. doi:10.1038/s441598-021-97658-w

Keywords: COVID-19, population pharmacokinetics, exposure-response