Captisol and GS-704277, but Not GS-441524, Are Credible Mediators of Remdesivir’s Nephrotoxicity

LETTER

A recent paper published by Le and colleagues in Antimicrobial Agents and Chemotherapy describes the use of remdesivir (GS-5734) in a patient with a double lung transplant and afflicted with COVID-19 (1). The patient experienced a progressive decline in renal function 3 days after initiating remdesivir treatment, which prompted hemodialysis. Due to acute kidney impairment (AKI), remdesivir therapy was discontinued 1 week after initiation despite a lack of clinical improvement. The authors suggest that accumulation of the main remdesivir metabolite, GS-441524, and the remdesivir excipient, sulfobutylether-β-cyclodextrin (SBED; Captisol) may lead to AKI and offer hemodialysis as a possible solution to remove both GS-441524 and Captisol and maintain remdesivir treatment. While we agree with the authors’ well-supported assertion that renal clearance of Captisol (2, 3) likely contributes to the nephrotoxicity associated with remdesivir treatment (4), we refute the assumption that GS-441524 is a major driver of AKI. We further contend that removal of GS-441524 likely hampers the overall efficacy of remdesivir, given its documented ability to exert anti-SARS-CoV-2 activity (5, 6).

A central issue with the authors’ analysis lies in their incomplete description of the remdesivir bioactivation scheme. While it is true that GS-441524 is the major persistent metabolite following intravenous administration of remdesivir (7–11), the authors fail to consider the role of the intermediate l-alanine metabolite (GS-704277) in their discussion of renal toxicity, omitting GS-704277 from their bioactivation schematics. GS-704277 is a low-molecular-weight (MW = 441 g/mol) anionic molecule. Administration of remdesivir results in transient appearance of GS-704277 before further hydrolysis to GS-441524 (12). Studies conducted by Gilead Sciences in rats have demonstrated that GS-704277, but not GS-441524 or remdesivir, is an effective substrate of OAT3 and likely contributes to renal adverse events in rats (13). Though the interaction between GS-704277 and human OATs has not yet been characterized, pharmacokinetic (PK) studies in humans point to the extensive renal clearance of GS-704277 rather than GS-441524. A study by Tempestilli and colleagues documented the PK of remdesivir and GS-441524 in a patient with normal kidney function and in a patient with renal impairment (14). While the patient with renal impairment retained higher levels of GS-441524 throughout the 24-h experiment (Fig. 2 in reference 14), the identical decay kinetics (slope) of GS-441524 for both patients implies that the higher levels of GS-441524 result from reduced clearance of the intermediate GS-704277. If renal impairment resulted in reduced clearance of GS-441524, then the half-life of GS-441524 in the renally compromised patient would be much longer than that observed for the uncompromised patient. However, this is not the case. The rate of clearance and half-life of GS-441524 is essentially identical for both patients, implicating the ability for GS-704277 to be extensively cleared by the kidneys.

In addition to overlooking the significance of GS-704277, the authors erroneously claim that the 50% effective concentration (EC₅₀) of GS-441524 against SARS-CoV-2 has not been reported. Pruijssers and colleagues have demonstrated that GS-441524 exhibits similar or superior efficacy to remdesivir in Calu3 and VeroE6 cells infected with SARS-CoV-2 (5). The authors’ recommendation for hemodialysis assumes that remdesivir is the main contributor to antiviral activity despite the fact that its half-life pales in comparison to that of GS-441524 (11). We acknowledge that hemodialysis may alleviate remdesivir-related AKI by removing Captisol and GS-704277. However, we assert that GS-441524 is an unlikely culprit behind AKI. We thus caution the authors of potential reductions in treatment efficacy that could be associated with removing GS-441524, given its low micromolar anti-SARS-CoV-2 activity (5, 6). Hemodialytic removal of GS-441524 may further diminish the therapeutic effects of remdesivir treatment.