Dataset Information

Application of Machine-Learning Models to Predict Tacrolimus Stable Dose in Renal Transplant Recipients.

ABSTRACT: Tacrolimus has a narrow therapeutic window and considerable variability in clinical use. Our goal was to compare the performance of multiple linear regression (MLR) and eight machine learning techniques in pharmacogenetic algorithm-based prediction of tacrolimus stable dose (TSD) in a large Chinese cohort. A total of 1,045 renal transplant patients were recruited, 80% of which were randomly selected as the "derivation cohort" to develop dose-prediction algorithm, while the remaining 20% constituted the "validation cohort" to test the final selected algorithm. MLR, artificial neural network (ANN), regression tree (RT), multivariate adaptive regression splines (MARS), boosted regression tree (BRT), support vector regression (SVR), random forest regression (RFR), lasso regression (LAR) and Bayesian additive regression trees (BART) were applied and their performances were compared in this work. Among all the machine learning models, RT performed best in both derivation [0.71 (0.67-0.76)] and validation cohorts [0.73 (0.63-0.82)]. In addition, the ideal rate of RT was 4% higher than that of MLR. To our knowledge, this is the first study to use machine learning models to predict TSD, which will further facilitate personalized medicine in tacrolimus administration in the future.

SUBMITTER: Tang J

PROVIDER: S-EPMC5296901 | biostudies-literature | 2017 Feb

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Application of Machine-Learning Models to Predict Tacrolimus Stable Dose in Renal Transplant Recipients.

Tang Jie J Liu Rong R Zhang Yue-Li YL Liu Mou-Ze MZ Hu Yong-Fang YF Shao Ming-Jie MJ Zhu Li-Jun LJ Xin Hua-Wen HW Feng Gui-Wen GW Shang Wen-Jun WJ Meng Xiang-Guang XG Zhang Li-Rong LR Ming Ying-Zi YZ Zhang Wei W

Scientific reports 20170208

Tacrolimus has a narrow therapeutic window and considerable variability in clinical use. Our goal was to compare the performance of multiple linear regression (MLR) and eight machine learning techniques in pharmacogenetic algorithm-based prediction of tacrolimus stable dose (TSD) in a large Chinese cohort. A total of 1,045 renal transplant patients were recruited, 80% of which were randomly selected as the "derivation cohort" to develop dose-prediction algorithm, while the remaining 20% constitu ...[more]

PMID: 28176850

Similar Datasets

Project description:Tacrolimus, an immunosuppressant used in solid organ transplantation, has a narrow therapeutic index and exhibits inter-individual pharmacokinetic variability. Achieving and maintaining a therapeutic level of the drug by giving appropriate doses is crucial for successful immunosuppression, especially during the initial post-transplant period. We studied the effect of CYP3A5, CYP3A4, and ABCB1 gene polymorphisms on tacrolimus trough concentrations in South Indian renal transplant recipients from Kerala to formulate a genotype-based dosing equation to calculate the required starting daily dose of tacrolimus to be given to each patient to attain optimal initial post-transplant period drug level. We also investigated the effect of these genes on drug-induced adverse effects and rejection episodes and looked into the global distribution of allele frequencies of these polymorphisms. One hundred forty-five renal transplant recipients on a triple immunosuppressive regimen of tacrolimus, mycophenolate mofetil, and steroid were included in this study. Clinical data including tacrolimus daily doses, trough levels (C0) and dose-adjusted tacrolimus trough concentration (C0/D) in blood at three time points (day 6, 6 months, and 1-year post-transplantation), adverse drug effects, rejection episodes, serum creatinine levels, etc., were recorded. The patients were genotyped for CYP3A5*3, CYP3A4*1B, CYP3A4*1G, ABCB1 G2677T, and ABCB1 C3435T polymorphisms by the PCR-RFLP method. We found that CYP3A5*3 polymorphism was the single most strongly associated factor determining the tacrolimus C0/D in blood at all three time points (p < 0.001). Using multiple linear regression, we formulated a simple and easy to compute equation that will help the clinician calculate the starting tacrolimus dose per kg body weight to be administered to a patient to attain optimal initial post-transplant period tacrolimus level. CYP3A5 expressors had an increased chance of rejection than non-expressors (p = 0.028), while non-expressors had an increased risk for new-onset diabetes mellitus after transplantation (NODAT) than expressors (p = 0.018). Genotype-guided initial tacrolimus dosing would help transplant recipients achieve optimal initial post-transplant period tacrolimus levels and thus prevent the adverse effects due to overdose and rejection due to inadequate dose. We observed inter-population differences in allele frequencies of drug metabolizer and transporter genes, emphasizing the importance of formulating population-specific dose prediction models to draw results of clinical relevance.

Project description:Study objectiveThis study investigated race and sex differences in tacrolimus pharmacokinetics and pharmacodynamics in stable kidney transplant recipients.Design and settingA cross-sectional, open-label, single center, 12-h pharmacokinetic-pharmacodynamic study was conducted. Tacrolimus pharmacokinetic parameters included area under the concentration-time curve (AUC0-12 ), AUC0-4 , 12-h troughs (C12 h ), maximum concentrations (Cmax ), oral clearance (Cl), with dose-normalized AUC0-12 , troughs, and Cmax with standardized adverse effect scores. Statistical models were used to analyze end points with individual covariate-adjustment including clinical factors, genotypic variants CYP3A5*3, CYP3A5*6, CYP3A5*7(CYP3A5*3*6*7) metabolic composite, and ATP binding cassette gene subfamily B member 1 (ABCB1) polymorphisms.Patients65 stable, female and male, Black and White kidney transplant recipients receiving tacrolimus and mycophenolic acid ≥6 months post-transplant were evaluated.Measurements and main resultsBlack recipients exhibited higher tacrolimus AUC0-12 (Race: p = 0.005), lower AUC* (Race: p < 0.001; Race × Sex: p = 0.068), and higher Cl (Race: p < 0.001; Sex: p = 0.066). Greater cumulative (Sex: p < 0.001; Race × Sex: p = 0.014), neurologic (Sex: p = 0.021; Race × Sex: p = 0.005), and aesthetic (Sex: p = 0.002) adverse effects were found in females, with highest scores in Black women. In 84.8% of Black and 68.8% of White patients, the target AUC0-12 was achieved (p = 0.027). In 31.3% of White and 9.1% of Black recipients, AUC0-12 was <100 ng‧h/ml despite tacrolimus troughs in the target range (p = 0.027). The novel CYP3A5*3*6*7 metabolic composite was the significant covariate accounting for 15%-19% of tacrolimus variability in dose (p = 0.002); AUC0-12 h * (p < 0.001), and Cl (p < 0.001).ConclusionsTacrolimus pharmacokinetics and adverse effects were different among stable kidney transplant recipient groups based upon race and sex with interpatient variability associated with the CYP3A5*3*6*7 metabolic composite. More cumulative, neurologic, and aesthetic adverse effects were noted among females. Tacrolimus regimens that consider race and sex may reduce adverse effects and enhance allograft outcomes by facilitating more patients to achieve the targeted AUC0-12 h .

Project description:BackgroundRenal replacement therapy (RRT) is a public health problem worldwide. Kidney transplantation (KT) is the best treatment for elderly patients' longevity and quality of life.ObjectivesThe primary endpoint was to compare elderly versus younger KT recipients by analyzing the risk covariables involved in worsening renal function, proteinuria, graft loss, and death one year after KT. The secondary endpoint was to create a robot based on logistic regression capable of predicting the likelihood that elderly recipients will develop worse renal function one year after KT.MethodUnicentric retrospective analysis of a cohort was performed with individuals aged ≥60 and <60 years old. We analysed medical records of KT recipients from January to December 2017, with a follow-up time of one year after KT. We used multivariable logistic regression to estimate odds ratios for elderly vs younger recipients, controlled for demographic, clinical, laboratory, data pre- and post-KT, and death.Results18 elderly and 100 younger KT recipients were included. Pretransplant immune variables were similar between two groups. No significant differences (P > 0.05) between groups were observed after KT on laboratory data means and for the prevalences of diabetes mellitus, hypertension, acute rejection, cytomegalovirus, polyomavirus, and urinary infections. One year after KT, the creatinine clearance was higher (P = 0.006) in youngers (70.9 ± 25.2 mL/min/1.73 m2) versus elderlies (53.3 ± 21.1 mL/min/1.73 m2). There was no difference in death outcome comparison. Multivariable analysis among covariables predisposing chronic kidney disease epidemiology collaboration (CKD-EPI) equation <60 mL/min/1.73 m2 presented a statistical significance for age ≥60 years (P = 0.01) and reduction in serum haemoglobin (P = 0.03). The model presented goodness-fit in the evaluation of artificial intelligence metrics (precision: 90%; sensitivity: 71%; and F 1 score: 0.79).ConclusionRenal function in elderly KT recipients was lower than in younger KT recipients. However, patients aged ≥60 years maintained enough renal function to remain off dialysis. Moreover, a learning machine application built a robot (Elderly KTbot) to predict in the elderly populations the likelihood of worse renal function one year after KT.

Dataset Information

Application of Machine-Learning Models to Predict Tacrolimus Stable Dose in Renal Transplant Recipients.

Publications

Application of Machine-Learning Models to Predict Tacrolimus Stable Dose in Renal Transplant Recipients.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets