Comment on “Adopting Race‐Free Estimated Glomerular Filtration Rate for Unifying Medication‐Related Decision‐Making: An Opinion of the Nephrology Practice and Research Network of the American College of Clinical Pharmacy”

Nemecek et al. and the American College of Clinical Pharmacy Nephrology Practice and Research Network (Nephrology PRN) call for “education and adoption of non–race-based CKD-EPI” as has been recommended previously by the National Kidney Foundation-American Society of Nephrology (NKF-ASN) Task Force [1-3]. We disagree and believe that the authors, Nephrology PRN, and the NKF-ASN task force have failed to recognize prior warnings and concerns expressed in using estimated glomerular filtration (eGFR) equations for renal drug dosing. First, injudicious use of eGFR equations as a substitute for CrCl has legitimate safety implications. Significant dosing errors, especially for drugs with narrow therapeutic indices such as antithrombotics, have resulted in severe bleeding [4]. These eGFR equations, including the MDRD equation and newer CKD-EPI variants, were developed for the purpose of staging chronic kidney disease (CKD) and not for drug dosing. Such population-based equations, including the recently proposed race-free CKD-EPI equation and their variants, have not been validated or approved by the US Food and Drug Administration (FDA) for use in drug dosing. Of note, Nemecek misinterprets the 2024 FDA guidance by implying that the FDA is promoting the use of CKD-EPI over CrCl [5]. This reflects a misunderstanding of FDA guidance documents, which are intended for use by manufacturers when designing studies during drug development. However, it is to be noted that this FDA guidance recognized the importance of expressing eGFR as mL/min rather than normalizing to a BSA value of 1.73 m2. To imply that it is a guide for pharmacy clinicians to employ CKD-EPI for widespread use in clinical care, misrepresents the FDAs intent. The recommendation by Nemecek is particularly concerning in the context of older adults, where eGFR equations have been shown to overestimate measured kidney function that could result in excessive doses being administered. This has been demonstrated in prior studies investigating the MDRD study equation leading to 30%–60% higher doses of gabapentin, amantadine, and various antimicrobials when compared to CrCl [6-8]. More recently, studies have shown that using the CKD-EPI for drug dosing can lead to antimicrobial dosing errors in older adults with renal impairment [9]. Failure to cite these references is a significant oversight of the authors' understanding of this topic [2]. The key finding that pharmacists are not using the CKD-EPI equation is not surprising or unexpected. Between1976 and 2009, 3026 new drug applications were approved by the FDA [10]. Conservatively estimating that 50% or more of these drugs undergo substantial renal elimination (30% or more), as many as 1513 medications were studied using CrCl for dosing adjustment in patients with renal impairment. Pharmacists practicing in accordance with FDA labeling would use the CrCl equation appropriately for drug dosing. This is consistent with how pharmacokinetic studies have historically been designed during drug development to evaluate the impact of kidney function on drug exposure. This finding may indicate pharmacists' lack of awareness or hesitancies with the NKF Task Force recommendation to use CKD-EPI eGFR equations outside their intended purpose. Aside from not being validated for drug dosing, the eGFR equations are complex, requiring additional steps such as back-correcting an automated eGFR result. Here, the eGFR is reported in units of milliliters per minute per 1.73 m2 (mL/min/1.73 m2) and requires calculation of body surface area (BSA) to yield units of milliliters per minute (mL/min) in order to be consistent with CrCl [5]. While there are clinical calculators to estimate renal function using CKD-EPI creatinine, cystatin C, and combined creatinine-cystatin C, for drugs where this method has not been validated by pharmacokinetic data or the FDA-approved product label, validation is not likely to occur. Further, the BSA-modified, race-free CKD-EPI equation has not been validated in pharmacokinetic studies in special populations such as frail older adults where mean (SD) BSA is 1.64 ± 0.19 m2 [11]. In conclusion, we believe that the recommendations made by Nemecek, the Nephrology PRN, and the NKF-ASN Task Force should be interpreted with extreme caution, particularly as they relate to application in adults 85 years of age and older, the most rapidly growing segment of the US population. Until studies are conducted to assess the relationship between eGFR equations, including CKD-EPI variants, and a drug's pharmacokinetic parameters and/or pharmacodynamic (toxicity) end points, traditional methods such as CrCl should continue to be employed in accordance with FDA-approved drug safety information. Dr. Barbara Zarowitz is a member of the JACCP Editorial Board. The authors declare no conflicts of interest.