Artificial Neural Network Modeling for Drug Dialyzability Prediction

Abstract

-Purpose. The purpose of this study was to develop an artificial neural network (ANN) model to predict drug removal during dialysis based on drug properties and dialysis conditions. Nine antihypertensive drugs were chosen as model for this study. Methods. Drugs were dissolved in a physiologic buffer and dialysed in vitro in different dialysis conditions (UFRmin/UFRmax, with/without BSA). Samples were taken at regular intervals and frozen at -20ºC until analysis. Extraction methods were developed for drugs that were dialysed with BSA in the buffer. Drug concentrations were quantified by high performance liquid chromatography (HPLC) or mass spectrometry (LC/MS/MS). Dialysis clearances (CLDs) were calculated using the obtained drug concentrations. An ANOVA with Scheffe's pairwise adjustments was performed on the collected data in order to investigate the impact of drug plasma protein binding and ultrafiltration rate (UFR) on CLD. The software Neurosolutions ® was used to build ANNs that would be able to predict drug CLD (output). The inputs consisted of dialysis UFR and the herein drug properties: molecular weight (MW), logD and plasma protein binding. Results. Observed CLDs were very high for the majority of the drugs studied. The addition of BSA in the physiologic buffer statistically significantly decreased CLD for carvedilol (p= 0.002) and labetalol (p<0.001), but made no significant difference for atenolol (p= 0.100). In contrast, varying UFR does not significantly affect CLD (p>0.025). Multiple ANNs were built and compared, the best model was a Jordan and Elman network which showed learning stability and good predictive results (MSE testing = 129). Conclusion. In this study, we have developed an ANN-model which is able to predict drug removal during dialysis. Since experimental determination of all existing drug CLDs is not realistic, ANNs represent a promising tool for the prediction of drug CLD using drug properties and dialysis conditions.

Chronic kidney disease and arrhythmias: conclusions from a Kidney Disease: Improving Global Outcomes (KDIGO) Controversies Conference

“… Modified from Potpara et al 101 and Weir et al 102 Metoprolol elimination data from Hoffman et al 103 Labetalol protein binding data from Drugbank.ca 104 and dialyzability data from in vitro data by Daheb et al 105 All other dialyzability data from Frishman. 106 CKD, chronic kidney disease; CrCl, creatinine clearance; GFR, glomerular filtration rate.…”

“… Labetalol protein binding data from Drugbank.ca 104 and dialyzability data from in vitro data by Daheb et al 105 …”

Extracorporeal treatment for poisoning to beta-adrenergic antagonists: systematic review and recommendations from the EXTRIP workgroup

“…In the final analysis, 76 studies were included for qualitative analysis, including 4 in vitro experiments [ 58 – 61 ], 2 animal experiments [ 62 , 63 ], 1 pharmacokinetic simulation study [ 64 ], 37 pharmacokinetic studies [ 65 – 101 ], and 32 case reports/series [ 13 , 15 , 35 , 102 – 130 ]. No comparative studies or randomized controlled trials were identified.…”

Potential Mechanism of Action of Cyclosporin A in Human Dermal Fibroblasts—Transcriptomic Analysis of CYPs

“…The results were analyzed with the use of neural networks and a standard statistic method to calculate the impact of CsA on gene expression. The use of an artificial neural network has a wide range of usage in drug research, enabling the prediction of various processes [ 25 ]. In microarray analysis, the large amount of results can be used to create Kohonen self-organizing maps, which give the opportunity to cluster the analyzed gene expression data, showing possible interaction between them based on a mathematical algorithm [ 26 ].…”