Zhendong Chen1, Qian Dong1, Charalambos Dokos1, Jana Boland1, Professor Uwe Fuhr1, Max Taubert1
1Department I of Pharmacology, Center for Pharmacology, Faculty of Medicine and University Hospital Cologne, University of Cologne
Introduction: Accurately assessing glomerular filtration rate (GFR) based on plasma creatinine concentrations is challenging in patients with unstable renal function. Compartmental nonlinear mixed-effects modelling offers a feasible method to more accurately assess unstable creatinine clearance (CrCL) and account for measurement errors[1-3]. A proposed creatinine population pharmacokinetic (popPK) model effectively described renal function changes and outperformed standard methods[4]. However, a further evaluation of this model by an independent assessment of GFR is desirable before application. Additionally, this study assumed a creatinine volume of distribution (Vd) of 60% of total body weight (TBW)[4], which neglected variability and may result in biased estimates of other kinetic parameters. Beyond glomerular filtration of creatinine, the net creatinine tubular secretion (nCTS) also accounts for 10-40% of its renal excretion[5]. The transport pathway responsible for creatinine secretion has been identified as basolateral organic cation transporter 2 (OCT2) and apical multidrug and toxin extrusion proteins (MATEs) through both in vitro experiments and clinical studies[6]. Therefore, creatinine can also serve as an endogenous marker of transporter activity. Objectives: The present study aims at precisely estimate creatinine Vd in order to refine the creatinine popPK model and improving the description of creatinine kinetics as a prerequisite for more reliable creatinine-based assessment of renal function and/or OCT2/MATE activity in humans. Methods: In a clinical study, iohexol was administered intravenously as a reference GFR marker, and creatinine through a meat meal. A high dose of 3235 mg and a low dose of 259 mg iohexol were both investigated. Individual creatinine dose was calculated based on measured creatinine concentration in boiled meat and weight of consumed meat. Two iohexol dose levels were evaluated: a high dose (3235 mg) and a low dose (259 mg). The individual creatinine dose was determined based on the measured creatinine concentration in boiled meat and the weight of meat consumed by each subject. A joint pharmacometric model for iohexol and creatinine was developed using dense plasma and urine sampling. The popPK modeling was conducted using a nonlinear mixed-effects approach with NONMEM version 7.4.0 (ICON Development Solutions, USA), Perl-speaks-NONMEM (PsN) version 5.3.0 (Uppsala University, Sweden), using the first-order conditional estimation with interaction method throughout model development. The following assumptions were made for all participants: 1) no changes in typical value of iohexol clearance (IoCL), CrCL, and creatinine generation rate throughout the study; 2) iohexol and creatinine are solely eliminated via kidney; and 3) the circadian rhythms of IoCL and CrCL follow a consistent sine function pattern within a day[7]. Simulations were used to evaluate the effect of different creatinine Vd values on CrCL estimation after acute kidney injury (AKI) and to assess the impact of limited sampling strategies on GFR and nCTS estimation. Results: Fourteen participants, mean age 33 years (range: 23–48), were enrolled, including 2 in the pilot and 12 in the main study. The dataset includes 771 iohexol and 826 creatinine plasma concentrations, and 439 measurements for both iohexol and creatinine in urine. A joint model for iohexol and creatinine was developed to accurately describe plasma and urine concentrations, incorporating CrCL as the sum of GFR (equivalent to IoCL) and nCTS, while accounting for TBW effects and circadian variation. Pharmacokinetic parameters for iohexol and creatinine aligned with reported values, but a lower Vd of 41% of TBW and a nCTS fraction of 31% relative to overall CrCL were observed. Additionally, sex was identified as a statistically significant covariate on nCTS. Commonly used equations based on single-point creatinine measurement all overestimated GFR, with the Modification of Diet in Renal Disease (MDRD) equation performing best, followed by Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) 2009 equation. Simulations demonstrate the effect of Vd estimate accuracy on detecting AKI from creatinine plasma concentrations only. Following low-dose iohexol administration of 259 mg, a single plasma sample at 5 hours and a urine sample from 0–5 hours provided accurate estimates of both GFR and nCTS using the joint model and enabled adequate correction for incomplete urine collection. Conclusion: The joint model of iohexol and creatinine can serve as an effective tool for estimating unstable GFR and nCTS, with nCTS potentially acting as a marker for assessing renal OCT2/MATE activity.
[1] Bragadottir G, Redfors B, Ricksten SE. Assessing glomerular filtration rate (GFR) in critically ill patients with acute kidney injury–true GFR versus urinary creatinine clearance and estimating equations. Crit Care. 2013. 17:R108. [2] Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron. 1976. 16:31-41. [3] Scappaticci GB, Regal RE. Cockcroft-Gault revisited: new deliverance on recommendations for use in cirrhosis. World J Hepatol. 2017. 9:131-138. [4] Ullah S, Zoller M, Jaehde U, et al. A model-based approach to assess unstable creatinine clearance in critically ill patients. Clin Pharmacol Ther. 2021. 110:1240-1249. [5] Levey AS, Perrone RD, Madias NE. Serum creatinine and renal function. Annu Rev Med. 1988. 39:465-90. [6] Mathialagan S, Feng B, Rodrigues AD, et al. Drug-Drug Interactions involving renal OCT2/MATE transporters: clinical risk assessment may require endogenous biomarker-informed approach. Clin Pharmacol Ther. 2021. 110:855-859. [7] Koopman MG, Koomen GC, Krediet RT, et al. Circadian rhythm of glomerular filtration rate in normal individuals. Clin Sci (Lond). 1989. 77:105-11.
Reference: PAGE 33 (2025) Abstr 11370 [www.page-meeting.org/?abstract=11370]
Poster: Clinical Applications