Remdesivir and GS-441524 plasma concentrations in patients with end-stage renal disease on haemodialysis

Sir, Remdesivir is a monophosphoramidate prodrug of a nucleoside analogue that possesses antiviral activity against human and zoonotic coronaviruses including SARS-CoV-2, SARS-CoV and MERS-CoV.1,2 The nucleoside core (GS-441524) is the predominant circulating plasma metabolite.1,3 Pharmacokinetic studies in healthy volunteers demonstrated a rapid decline in remdesivir concentration [maximum concentration (Cmax) 2610 ng/mL; half-life 0.98 h] with a subsequent increase in GS-441524 concentration (Cmax 140 ng/mL; half-life 25.3 h).1 Less than 10% of remdesivir is excreted renally, but 49% is recovered in urine as GS-441524, suggesting accumulation is possible in renal failure.1,3 Remdesivir is formulated with the solubilizing agent sulfobutylether-β-cyclodextrin (SBECD). Due to nephrotoxicity concerns, patients with moderate to severe renal dysfunction were excluded in clinical trials.4–7 Notably, rates of nephrotoxic adverse events in placebo-controlled trials have not demonstrated signals of increased nephrotoxicity with remdesivir.4,8 Indeed, SBECD-mediated nephrotoxicity appears to be more of a theoretical concern than one substantiated by clinical evidence.9 At University of California Los Angeles Ronald Reagan Medical Center and University of California Los Angeles Medical Center Santa Monica, 5 day courses of remdesivir were initiated for three patients with end-stage renal disease (ESRD) requiring haemodialysis for severe COVID-19. We describe their clinical courses, as well as plasma concentrations of remdesivir and GS-441524. Samples for remdesivir and GS-441524 were collected as part of routine clinical care and analysed using LC-MS/MS in accordance with FDA guidance.10 Concentrations were collected at times coinciding with nursing room entry for usual care to avoid increased personal protective equipment (‘PPE’) waste. Pre- and post-haemodialysis concentrations were assessed. Per institutional policy, Institutional Review Board (‘IRB’) approval was not required or obtained. Patient 1 is a 44-year-old man with a past medical history significant for ESRD on haemodialysis, hypertension and hyperlipidaemia. Upon presentation, he was hypoxic (SpO2 89%) and was placed on low-flow oxygen supplementation. Remdesivir was initiated on hospital day (HD) 2. Plasma concentrations of remdesivir and GS-441524 were measured following the second dose. Remdesivir plasma concentrations were not elevated compared with reference data, but GS-441524 concentrations were 3–4.5-fold higher than day 5 Cmax [see Figure 1 and Table S1 (available as Supplementary data at JAC Online)]. Mild transaminitis was noted on HD5, which peaked on HD7 (AST 71 U/L; ALT 89 U/L), with subsequent decline until discharge (HD13). His hospitalization was otherwise uneventful. Plasma concentrations of remdesivir and GS-441524. Time corresponds to time since initiation of remdesivir therapy. GS, GS-441524; RDV, remdesivir. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Patient 2 is an 89-year-old man with a past medical history significant for hypertension and ESRD on haemodialysis. Upon presentation, he was hypoxic (SpO2 91%), requiring low-flow oxygen supplementation, and remdesivir was initiated on HD3. Plasma remdesivir and GS-441524 concentrations were assessed prior to and following the third dose. An approximately 2 h half-life for remdesivir was observed. GS-441524 concentrations were 3.5–6.3-fold higher than reference Cmax and a post-haemodialysis concentration was 45% lower than the preceding concentration. No adverse effects were observed during therapy. He was transitioned to ambient air on HD12 and discharged on HD13. Patient 3 is a 75-year-old man with a past medical history significant for hypertension, diabetes mellitus, coronary artery disease and ESRD on haemodialysis. Upon presentation, he was hypotensive (blood pressure 79/45) and hypoxic (SpO2 74%), requiring low-flow oxygen supplementation, and remdesivir was started on HD2. Plasma remdesivir and GS-441524 concentrations were checked prior to and following the fifth dose. Remdesivir half-life was approximately 2 h with undetectable concentrations by the end of the dosing interval. GS-441524 concentrations were considerably higher than reference values (5.2–10.4-fold). No adverse effects occurred while on remdesivir therapy, and he was discharged on HD7. Remdesivir given at standard doses for a 5 day course with concomitant monitoring was undertaken for three patients. Remdesivir half-life in two of the patients was approximately twice as long as in healthy volunteers (∼2 h versus ∼1 h). A decline in remdesivir concentration in each patient was accompanied by a simultaneous increase in GS-441524 concentration, as conversion from prodrug into metabolite occurred. Notably, the highest measured GS-441524 concentration was ∼10-fold higher than day 5 Cmax in healthy volunteers (1470 ng/mL in patient 3; day 5 Cmax 142 ng/mL). Considering its renal elimination and long half-life, this was likely due to accumulation. Despite this, remdesivir was well tolerated overall. It is unknown if toxicities associated with remdesivir therapy are driven by the prodrug, GS-441524, another intermediate or a combination of these. GS-441524 appears to be removed by haemodialysis, with post-haemodialysis concentrations 45%–49% lower than the most recent pre-haemodialysis measurement. Our case series has several limitations. Firstly, our sample size was small and these results should be interpreted cautiously. Secondly, concentration measurement was not standardized and, as such, pharmacokinetic data were not richly sampled, preventing AUC calculation. Additionally, post-haemodialysis might have been too early to assess redistribution. Lastly, no effluent dialysate concentrations were measured. To the best of our knowledge, these are the first reported data on remdesivir and GS-441524 exposure in haemodialysis-dependent patients. Additional data are needed for tolerability in this vulnerable population. Future comprehensive dose-finding studies are needed to assess strategies for adjustment for varying degrees of renal impairment, such as reduced frequency or reduced doses. This study was carried out as part of our routine work.