Vianney Tuloup, Laurent Bourguignon
Université Lyon, Université Lyon 1, UMR CNRS 5558, Laboratoire de Biométrie et Biologie Evolutive, Villeurbanne, France
Objectives: Chronic kidney disease (CKD) is a major public healthcare priority. Patients with CKD are also at higher risk of comorbidities like cardiovascular disease, diabetes, demineralising disease or infections. CKD is associated with a change in the pharmacokinetics of drugs mainly eliminated via the kidneys, but also for drugs with high metabolic clearance, through modification of the unbound fraction, bioavailability, enzymatic activity, etc.
Simples models based on an In vivo Mechanistic Static Model (IMSM) approach were published to describe the modifications in metabolic clearance observed in cirrhosis, as well as during drug interactions [1-3]. We hypothesized that this approach could also be used to describe the effect of renal failure on metabolic drug clearance.
The objective of this study was to build a single model able of predicting drug exposure variation in moderate or severe CDK patients, for drugs mainly eliminated by enzymatic pathway.
Methods: The metrics used to explain and characterize the impact of renal impairment on drug pharmacokinetics is the area under the concentration time curve (AUC), presented as the ratio of AUC of renal impaired patient to the AUC of a healthy volunteer.
A PUBMED research was performed to identify studies that compared drug exposition in healthy subjects and in CKD patients, for drugs mainly eliminated by enzymatic pathway. We retain articles with measured AUC or clearance in healthy subjects and renal impaired patients. Information collected were AUC, route of administration, type of dosing regimen (single or multiple administration), renal clearance estimation method and number of patients in each renal impairment stage.
Remaining fraction of cytochrome activity was estimated in each CKD stage with a constrained nonlinear multivariable function. Initial values were based on a literature dataset. Confidence intervals were calculated with a bootstrap resampling procedure. Predicted AUC ratio were considered as acceptable if they stood in the 50%-200% interval of predicted on observed AUC ratio.
Results: On 100 drugs reported on the ddi-predictor database with moderate or high-hepatic metabolism, 35 had exploitable published data on AUC ratio in renal impairment.
Remaining activity of CYP3A4 was estimated at 0.901 in stage 3 CKD, and 0.722 in stage 4. As well, CYP2D6, 2C9, 2C19 and 1A2 have a remaining fraction of 0.729, 0.795, 0.838, 0.846 respectively in stage 3 and 0.578, 0.66, 0.725 and 0.548 respectively in stage 4. Ninety-two percent (n=50) of AUC ratios were correctly predicted according to the criteria choosen for this study. Four outliers were found underpredicted.
Conclusions: We built and evaluated a model for prediction of drug exposure modification in CKD patients, for drugs mainly eliminated by enzymatic pathway. Our model correctly predicted 92% of included drugs’ AUC ratio. This AUC ratio can be used as an adjustment factor for drug dosage adaptation in CKD patient. Nevertheless, our model may help clinicians in adjusting drug dose regimens in this patient’s population.
References:
[1] Tod M, Pierrillas PB, Bourguignon L, Goutelle S. Comparison of the static in vivo approach to a physiologically based pharmacokinetic approach for metabolic drug–drug interactions prediction. Int J Pharmacokinet. 4 avr 2016;1(1):25-34.
[2] Steelandt J, Jean-Bart E, Goutelle S, Tod M. A Prediction Model of Drug Exposure in Cirrhotic Patients According to Child–Pugh Classification. Clin Pharmacokinet. 1 déc 2015;54(12):1245-58
[3] Tod M, Goutelle S, Clavel-Grabit F, Nicolas G, Charpiat B. Quantitative Prediction of Cytochrome P450 (CYP) 2D6-Mediated Drug Interactions. Clin Pharmacokinet. 1 août 2011;50(8):519-30.
Reference: PAGE 29 (2021) Abstr 9610 [www.page-meeting.org/?abstract=9610]
Poster: Methodology - Estimation Methods