Drug Clearance in Neonates: A Combination of Population Pharmacokinetic Modelling and Machine Learning Approaches to Improve Individual Prediction.
Journal
Clinical pharmacokinetics
ISSN: 1179-1926
Titre abrégé: Clin Pharmacokinet
Pays: Switzerland
ID NLM: 7606849
Informations de publication
Date de publication:
11 2021
11 2021
Historique:
accepted:
28
04
2021
pubmed:
28
5
2021
medline:
26
11
2021
entrez:
27
5
2021
Statut:
ppublish
Résumé
Population pharmacokinetic evaluations have been widely used in neonatal pharmacokinetic studies, while machine learning has become a popular approach to solving complex problems in the current era of big data. The aim of this proof-of-concept study was to evaluate whether combining population pharmacokinetic and machine learning approaches could provide a more accurate prediction of the clearance of renally eliminated drugs in individual neonates. Six drugs that are primarily eliminated by the kidneys were selected (vancomycin, latamoxef, cefepime, azlocillin, ceftazidime, and amoxicillin) as 'proof of concept' compounds. Individual estimates of clearance obtained from population pharmacokinetic models were used as reference clearances, and diverse machine learning methods and nested cross-validation were adopted and evaluated against these reference clearances. The predictive performance of these combined methods was compared with the performance of two other predictive methods: a covariate-based maturation model and a postmenstrual age and body weight scaling model. Relative error was used to evaluate the different methods. The extra tree regressor was selected as the best-fit machine learning method. Using the combined method, more than 95% of predictions for all six drugs had a relative error of < 50% and the mean relative error was reduced by an average of 44.3% and 71.3% compared with the other two predictive methods. A combined population pharmacokinetic and machine learning approach provided improved predictions of individual clearances of renally cleared drugs in neonates. For a new patient treated in clinical practice, individual clearance can be predicted a priori using our model code combined with demographic data.
Sections du résumé
BACKGROUND
Population pharmacokinetic evaluations have been widely used in neonatal pharmacokinetic studies, while machine learning has become a popular approach to solving complex problems in the current era of big data.
OBJECTIVE
The aim of this proof-of-concept study was to evaluate whether combining population pharmacokinetic and machine learning approaches could provide a more accurate prediction of the clearance of renally eliminated drugs in individual neonates.
METHODS
Six drugs that are primarily eliminated by the kidneys were selected (vancomycin, latamoxef, cefepime, azlocillin, ceftazidime, and amoxicillin) as 'proof of concept' compounds. Individual estimates of clearance obtained from population pharmacokinetic models were used as reference clearances, and diverse machine learning methods and nested cross-validation were adopted and evaluated against these reference clearances. The predictive performance of these combined methods was compared with the performance of two other predictive methods: a covariate-based maturation model and a postmenstrual age and body weight scaling model. Relative error was used to evaluate the different methods.
RESULTS
The extra tree regressor was selected as the best-fit machine learning method. Using the combined method, more than 95% of predictions for all six drugs had a relative error of < 50% and the mean relative error was reduced by an average of 44.3% and 71.3% compared with the other two predictive methods.
CONCLUSION
A combined population pharmacokinetic and machine learning approach provided improved predictions of individual clearances of renally cleared drugs in neonates. For a new patient treated in clinical practice, individual clearance can be predicted a priori using our model code combined with demographic data.
Identifiants
pubmed: 34041714
doi: 10.1007/s40262-021-01033-x
pii: 10.1007/s40262-021-01033-x
doi:
Substances chimiques
Vancomycin
6Q205EH1VU
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
1435-1448Subventions
Organisme : Beijing Obstetrics and Gynecology Hospital affiliated to Capital Medical University
ID : FCYY201715
Organisme : National Natural Science Foundation of China
ID : Grant 81803433
Organisme : National Science and Technology Major Projects for Major New Drugs Innovation and Development
ID : 2017ZX09304029-001
Organisme : National Science and Technology Major Projects for Major New Drugs Innovation and Development
ID : 2017ZX09304029-002
Informations de copyright
© 2021. The Author(s), under exclusive licence to Springer Nature Switzerland AG.
Références
Hsieh EM, Hornik CP, Clark RH, Laughon MM, Benjamin DK Jr, Smith PB, et al. Medication use in the neonatal intensive care unit. Am J Perinatol. 2014;31(9):811–21. https://doi.org/10.1055/s-0033-1361933 .
doi: 10.1055/s-0033-1361933
pubmed: 24347262
Riou S, Plaisant F, Maucort Boulch D, Kassai B, Claris O, Nguyen K-A. Unlicensed and off-label drug use: a prospective study in French NICU. Acta Paediatr. 2015;104(5):e228–31. https://doi.org/10.1111/apa.12924 .
doi: 10.1111/apa.12924
pubmed: 25669964
Coppini R, Simons SHP, Mugelli A, Allegaert K. Clinical research in neonates and infants: challenges and perspectives. Pharmacol Res. 2016;108:80–7. https://doi.org/10.1016/j.phrs.2016.04.025 .
doi: 10.1016/j.phrs.2016.04.025
pubmed: 27142783
Jacqz-Aigrain E, Leroux S, Thomson AH, Allegaert K, Capparelli EV, Biran V, et al. Population pharmacokinetic meta-analysis of individual data to design the first randomized efficacy trial of vancomycin in neonates and young infants. J Antimicrob Chemother. 2019;74(8):2128–38. https://doi.org/10.1093/jac/dkz158 .
doi: 10.1093/jac/dkz158
pubmed: 31049551
Tang BH, Wu YE, Kou C, Qi YJ, Qi H, Xu HY, et al. Population pharmacokinetics and dosing optimization of amoxicillin in neonates and young infants. Antimicrob Agents Chemother. 2019. https://doi.org/10.1128/AAC.02336-18 .
doi: 10.1128/AAC.02336-18
pubmed: 31138577
pmcid: 6879210
Bradley JS, Sauberan JB, Ambrose PG, Bhavnani SM, Rasmussen MR, Capparelli EV. Meropenem pharmacokinetics, pharmacodynamics, and Monte Carlo simulation in the neonate. Pediatr Infect Dis J. 2008;27(9):794–9. https://doi.org/10.1097/INF.0b013e318170f8d2 .
doi: 10.1097/INF.0b013e318170f8d2
pubmed: 18645546
Murphy KP. Machine learning : a probabilistic perspective. Cambridge: MIT Press; 2012.
Al’Aref SJ, Anchouche K, Singh G, Slomka PJ, Kolli KK, Kumar A, et al. Clinical applications of machine learning in cardiovascular disease and its relevance to cardiac imaging. Eur Heart J. 2019;40(24):1975–86. https://doi.org/10.1093/eurheartj/ehy404 .
doi: 10.1093/eurheartj/ehy404
pubmed: 30060039
Choy G, Khalilzadeh O, Michalski M, Do S, Samir AE, Pianykh OS, et al. Current applications and future impact of machine learning in radiology. Radiology. 2018;288(2):318–28. https://doi.org/10.1148/radiol.2018171820 .
doi: 10.1148/radiol.2018171820
pubmed: 29944078
Rutledge RB, Chekroud AM, Huys QJ. Machine learning and big data in psychiatry: toward clinical applications. Curr Opin Neurobiol. 2019;55:152–9. https://doi.org/10.1016/j.conb.2019.02.006 .
doi: 10.1016/j.conb.2019.02.006
pubmed: 30999271
Smith NM, Lenhard JR, Boissonneault KR, Landersdorfer CB, Bulitta JB, Holden PN, et al. Using machine learning to optimize antibiotic combinations: dosing strategies for meropenem and polymyxin B against carbapenem-resistant Acinetobacter baumannii. Clin Microbiol Infect. 2020;26(9):1207–13. https://doi.org/10.1016/j.cmi.2020.02.004 .
doi: 10.1016/j.cmi.2020.02.004
pubmed: 32061797
pmcid: 7587610
Zhu H, Huang SM, Madabushi R, Strauss DG, Wang Y, Zineh I. Model-informed drug development: a regulatory perspective on progress. Clin Pharmacol Ther. 2019;106(1):91–3. https://doi.org/10.1002/cpt.1475 .
doi: 10.1002/cpt.1475
pubmed: 31162631
Goulooze SC, Zwep LB, Vogt JE, Krekels EHJ, Hankemeier T, van den Anker JN, et al. Beyond the randomized clinical trial: innovative data science to close the pediatric evidence gap. Clin Pharmacol Ther. 2020;107(4):786–95. https://doi.org/10.1002/cpt.1744 .
doi: 10.1002/cpt.1744
pubmed: 31863465
Wilbaux M, Fuchs A, Samardzic J, Rodieux F, Csajka C, Allegaert K, et al. Pharmacometric approaches to personalize use of primarily renally eliminated antibiotics in preterm and term neonates. J Clin Pharmacol. 2016;56(8):909–35. https://doi.org/10.1002/jcph.705 .
doi: 10.1002/jcph.705
pubmed: 26766774
Zhao Y, Yao BF, Kou C, Xu HY, Tang BH, Wu YE, et al. Developmental Population Pharmacokinetics And Dosing Optimization Of Cefepime In Neonates And Young Infants. Front Pharmacol. 2020;11:14. https://doi.org/10.3389/fphar.2020.00014 .
doi: 10.3389/fphar.2020.00014
pubmed: 32116695
pmcid: 7010644
Qi H, Kou C, Qi YJ, Tang BH, Wu YE, Jin F, et al. Population pharmacokinetics and dosing optimization of latamoxef in neonates and young infants. Int J Antimicrob Agents. 2019;53(3):347–51. https://doi.org/10.1016/j.ijantimicag.2018.11.017 .
doi: 10.1016/j.ijantimicag.2018.11.017
pubmed: 30472290
Wu YE, Wang T, Yang HL, Tang BH, Kong L, Li X, et al. Population pharmacokinetics and dosing optimization of azlocillin in neonates with early-onset sepsis: a real-world study. J Antimicrob Chemother. 2021;76(3):699–709. https://doi.org/10.1093/jac/dkaa468 .
doi: 10.1093/jac/dkaa468
pubmed: 33188385
Li X, Qi H, Jin F, Yao B-F, Wu Y-E, Qi Y-J, et al. Population pharmacokinetics–pharmacodynamics of ceftazidime in neonates and young infants: dosing optimization for neonatal sepsis. Eur J Pharm Sci. 2021. https://doi.org/10.1016/j.ejps.2021.105868 .
doi: 10.1016/j.ejps.2021.105868
pubmed: 34656775
Sahigara F, Ballabio D, Todeschini R, Consonni V. Defining a novelk-nearest neighbours approach to assess the applicability domain of a QSAR model for reliable predictions. J Cheminform. 2013;5(1):27.
doi: 10.1186/1758-2946-5-27
Kamiński B, Jakubczyk M, Szufel P. A framework for sensitivity analysis of decision trees. CEJOR. 2017;26(1):135–59. https://doi.org/10.1007/s10100-017-0479-6 .
doi: 10.1007/s10100-017-0479-6
pubmed: 29375266
Kégl B. The return of AdaBoost.MH: multi-class Hamming trees. CoRR 2014. 1312.6086
Geurts P, Ernst D, Wehenkel L. Extremely randomized trees. Mach Learn. 2006;63(1):3–42.
doi: 10.1007/s10994-006-6226-1
Svetnik V. Random forest : a classification and regression tool for compound classification and QSAR Modeling. J Chem Inf Comput Sci. 2003;43(6):1947–58.
doi: 10.1021/ci034160g
Friedman JH. Greedy function approximation: a gradient boosting machine. Ann Stat. 2001;29(5):1189–232.
doi: 10.1214/aos/1013203451
Dorugade AV, Kashid DN. Alternative method for choosing ridge parameter for regression. Appl Math Sci. 2010;4(9):447–56.
Tibshirani R. Regression shrinkage and selection via the Lasso. J R Stat Soc Ser B Methodol. 1996;58(1):267–88.
Zou H, Hastie T. Regularization and variable selection via the elastic net. J R Stat Soc B. 2005;67(2):301–20.
doi: 10.1111/j.1467-9868.2005.00503.x
Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, et al. Scikit-learn: machine learning in Python. J Mach Learn Res. 2011;12:2825–30.
Krstajic D, Buturovic LJ, Leahy DE, Thomas S. Cross-validation pitfalls when selecting and assessing regression and classification models. J Cheminform. 2014;6(1):10. https://doi.org/10.1186/1758-2946-6-10 .
doi: 10.1186/1758-2946-6-10
pubmed: 24678909
pmcid: 3994246
Varma S, Simon R. Bias in error estimation when using cross-validation for model selection. BMC Bioinform. 2006;7:91. https://doi.org/10.1186/1471-2105-7-91 .
doi: 10.1186/1471-2105-7-91
Brownlee J. Data preparation for machine learning: data cleaning, feature selection, and data transforms in Python, Machine Learning Mastery. 2020.
Wang J, Kumar SS, Sherwin CM, Ward R, Baer G, Burckart GJ, et al. Renal clearance in newborns and infants: predictive performance of population-based modeling for drug development. Clin Pharmacol Ther. 2019;105(6):1462–70. https://doi.org/10.1002/cpt.1332 .
doi: 10.1002/cpt.1332
pubmed: 30565653
Sheiner LB, Rosenberg B, Marathe VV. Estimation of population characteristics of pharmacokinetic parameters from routine clinical data. J Pharmacokinet Biopharm. 1977;5(5):445–79. https://doi.org/10.1007/bf01061728 .
doi: 10.1007/bf01061728
pubmed: 925881
Brussee JM, Calvier EA, Krekels EH, Valitalo PA, Tibboel D, Allegaert K, et al. Children in clinical trials: towards evidence-based pediatric pharmacotherapy using pharmacokinetic–pharmacodynamic modeling. Expert Rev Clin Pharmacol. 2016;9(9):1235–44. https://doi.org/10.1080/17512433.2016.1198256 .
doi: 10.1080/17512433.2016.1198256
pubmed: 27269200
Koch G, Pfister M, Daunhawer I, Wilbaux M, Wellmann S, Vogt JE. Pharmacometrics and machine learning partner to advance clinical data analysis. Clin Pharmacol Ther. 2020;107(4):926–33. https://doi.org/10.1002/cpt.1774 .
doi: 10.1002/cpt.1774
pubmed: 31930487
pmcid: 7158220
Graaf PH. Introduction to population pharmacokinetic/pharmacodynamic analysis with nonlinear mixed effects models. CPT Pharmacomet Syst Pharmacol. 2014;3:e153. https://doi.org/10.1038/psp.2014.51 .
doi: 10.1038/psp.2014.51
Meibohm B, Laer S, Panetta JC, Barrett JS. Population pharmacokinetic studies in pediatrics: issues in design and analysis. AAPS J. 2005;7(2):E475–87. https://doi.org/10.1208/aapsj070248 .
doi: 10.1208/aapsj070248
pubmed: 16353925
pmcid: 2750985
Wade KC, Wu D, Kaufman DA, Ward RM, Benjamin DK Jr, Sullivan JE, et al. Population pharmacokinetics of fluconazole in young infants. Antimicrob Agents Chemother. 2008;52(11):4043–9. https://doi.org/10.1128/AAC.00569-08 .
doi: 10.1128/AAC.00569-08
pubmed: 18809946
pmcid: 2573107
Li Z, Chen Y, Li Q, Cao D, Shi W, Cao Y, et al. Population pharmacokinetics of piperacillin/tazobactam in neonates and young infants. Eur J Clin Pharmacol. 2013;69(6):1223–33. https://doi.org/10.1007/s00228-012-1413-4 .
doi: 10.1007/s00228-012-1413-4
pubmed: 23354809
Cohen-Wolkowiez M, Watt KM, Zhou C, Bloom BT, Poindexter B, Castro L, et al. Developmental pharmacokinetics of piperacillin and tazobactam using plasma and dried blood spots from infants. Antimicrob Agents Chemother. 2014;58(5):2856–65. https://doi.org/10.1128/AAC.02139-13 .
doi: 10.1128/AAC.02139-13
pubmed: 24614369
pmcid: 3993246
Kuppens M, George I, Lewi L, Levtchenko E, Allegaert K. Creatinaemia at birth is equal to maternal creatinaemia at delivery: does this paradigm still hold? J Matern Fetal Neonatal Med. 2012;25(7):978–80. https://doi.org/10.3109/14767058.2011.602144 .
doi: 10.3109/14767058.2011.602144
pubmed: 21867404
Allegaert K, van de Velde M, van den Anker J. Neonatal clinical pharmacology. Paediatr Anaesth. 2014;24(1):30–8. https://doi.org/10.1111/pan.12176 .
doi: 10.1111/pan.12176
pubmed: 23617305
Obermeyer Z, Emanuel EJ. Predicting the future—big data, machine learning, and clinical medicine. N Engl J Med. 2016;375(13):1216–9. https://doi.org/10.1056/NEJMp1606181 .
doi: 10.1056/NEJMp1606181
pubmed: 27682033
pmcid: 5070532
Podda M, Bacciu D, Micheli A, Bellu R, Placidi G, Gagliardi L. A machine learning approach to estimating preterm infants survival: development of the Preterm Infants Survival Assessment (PISA) predictor. Sci Rep. 2018;8(1):13743. https://doi.org/10.1038/s41598-018-31920-6 .
doi: 10.1038/s41598-018-31920-6
pubmed: 30213963
pmcid: 6137213
Bartz-Kurycki MA, Green C, Anderson KT, Alder AC, Bucher BT, Cina RA, et al. Enhanced neonatal surgical site infection prediction model utilizing statistically and clinically significant variables in combination with a machine learning algorithm. Am J Surg. 2018;216(4):764–77. https://doi.org/10.1016/j.amjsurg.2018.07.041 .
doi: 10.1016/j.amjsurg.2018.07.041
pubmed: 30078669
Schreuder MF, Bueters RR, Allegaert K. The interplay between drugs and the kidney in premature neonates. Pediatr Nephrol. 2014;29(11):2083–91. https://doi.org/10.1007/s00467-013-2651-0 .
doi: 10.1007/s00467-013-2651-0
pubmed: 24217783
Van der Auwera P, Santella PJ. Pharmacokinetics of cefepime: a review. J Antimicrob Chemother. 1993;32(Suppl B):103–15. https://doi.org/10.1093/jac/32.suppl_b.103 .
doi: 10.1093/jac/32.suppl_b.103
pubmed: 8150753
Shepherd AM, Hardin TC, Ludden TM, Miner DJ, Coleman DL. Latamoxef (moxalactam) kinetics in volunteers studied by a specific HPLC assay technique. J Antimicrob Chemother. 1983;12(4):377–86. https://doi.org/10.1093/jac/12.4.377 .
doi: 10.1093/jac/12.4.377
pubmed: 6643333
Mastrandrea V, Ripa S, La Rosa F, Tarsi R. Human intravenous and intramuscular pharmacokinetics of amoxicillin. Int J Clin Pharmacol Res. 1984;4(3):209–12.
pubmed: 6490239
Singlas E, Haegel C. Clinical pharmacokinetics of azlocillin [in French]. Presse Med. 1984;13(13):788–96.
pubmed: 6231596
Gundert-Remy U, Weber E. Elimination of azlocillin in patients with biliary t-tube drainage. Eur J Clin Pharmacol. 1982;22(5):435–9. https://doi.org/10.1007/BF00542549 .
doi: 10.1007/BF00542549
pubmed: 7117356
Ljungberg B, Nilsson-Ehle I. Comparative pharmacokinetics of ceftazidime in young, healthy and elderly, acutely ill males. Eur J Clin Pharmacol. 1988;34(2):179–86. https://doi.org/10.1007/BF00614556 .
doi: 10.1007/BF00614556
pubmed: 3289950
Leroux S, Turner MA, Guellec CB, Hill H, van den Anker JN, Kearns GL, et al. Pharmacokinetic studies in neonates: the utility of an opportunistic sampling design. Clin Pharmacokinet. 2015;54(12):1273–85. https://doi.org/10.1007/s40262-015-0291-1 .
doi: 10.1007/s40262-015-0291-1
pubmed: 26063050
Zhao W, Kaguelidou F, Biran V, Zhang D, Allegaert K, Capparelli EV, et al. External evaluation of population pharmacokinetic models of vancomycin in neonates: the transferability of published models to different clinical settings. Br J Clin Pharmacol. 2013;75(4):1068–80. https://doi.org/10.1111/j.1365-2125.2012.04406.x .
doi: 10.1111/j.1365-2125.2012.04406.x
pubmed: 23148919
Abitbol CL, DeFreitas MJ, Strauss J. Assessment of kidney function in preterm infants: lifelong implications. Pediatr Nephrol. 2016;31(12):2213–22. https://doi.org/10.1007/s00467-016-3320-x .
doi: 10.1007/s00467-016-3320-x
pubmed: 26846786
Singlas E, Haegel C. Clinical pharmacokinetics of azlocillin. Presse Med. 1984;13(13):788–96.
pubmed: 6231596