Prediction of Pediatric Drug Clearance Using Growth and Maturation Models for GFR

• Main Authors: TSENG, CHUAN-CHIH, 曾傳志
• Other Authors: HU, TEH-MIN
• Format: Others
• Language: zh-TW
• Published: 2017
• Online Access: http://ndltd.ncl.edu.tw/handle/8n5rjn

碩士 === 國防醫學院 === 藥學研究所 === 105 === Background: Pediatric dosing information is not routinely acquired during the drug development process, and empirical rules based on body weight or age are often used for estimating doses in pediatric patients. However, “children are not small adults”the difference between children and adults cannot be realized by a simple linear function of body weight or age. Recently, more complex mathematical models have been proposed to describe the changes of drug clearance at different ages from birth to adulthood. These models account for growth and maturation in child development, and have been applied successfully to a wide variety of drugs. Since drugs have diverse pharmacokinetic properties, with varied dependence on hepatic metabolism and renal excretion, the maturation model often includes drug-specific parameter values for a particular drug; and generalizability is difficult, or not impossible. Objectives: The objectives of the present study are (1) to investigate whether a general model can be used for predicting the clearance values of mainly renally excreted drugs based on the assumption that GFR maturation may account for the maturation of renal drug-eliminating ability; (2) to compare the prediction performance among various prediction models; and (3) to investigate the relationship between the maturation of renal function indices (e.g. GFR, renal blood flow rate, and active tubular secretion rate) and the maturation of drug elimination in the human body. Methods: Four GFR maturation models and two allometric scaling models (size models) were included to predict drug clearance (CL) at different child ages. Real, observed CL values were obtained from published studies through a PubMed-based text-mining procedure. Besides, simulations were conducted to generate simulated CL values (based on the WHO/CDC growth chart). Predictive performance was evaluated using various precision and accuracy measures (e.g. RMSE, AFE, and AAFE). Besides, two specific models (Hayton GFR vs. Mahmood GFR) were used to describe the relationship between maturation of renal function and that of drug elimination. Results: A total of 868 exact and 1628 simulated CL values were included for analysis. The exact CL values are from 39 drugs, spanning various therapeutic categories. The maturation models generally have better predictive performance than the size models in younger ages; however, at 2 years of age and above, the prediction accuracy seems to be similar, suggesting that the size model alone can be used for the prediction of drug clearance at ages above 2 years. Remarkably, among all the maturation models, the Hayton equation provides the best predictions. The advantage and attractiveness of using a GFR maturation model based on the Hayton equation is that a general equation with a limited number of parameters can be used in predicting drug CL values for a wide variety of drugs at different human developmental stages. Besides, we found that among the three indicators of renal function (GFR, tubular active secretion, renal blood flow), GFR maybe the best parameter to describe the development of drug-eliminating ability. Conclusions: The study demonstrates that the GFR maturation models may be used as a general model for predicting the total clearance of drugs that are mainly excreted from the kidneys. The study further shows that the development of human drug elimination capability is closely related to the maturation process of GFR.