Artificial intelligence for toxicokinetic parameterization in environmental exposure assessment

Abstract

Toxicokinetics (TK) characterizes the absorption, distribution, metabolism, and excretion (ADME) of chemicals in the body and connects exposure with internal dose and potential toxicity. Because in vivo and in vitro TK assays are costly and low-throughput, and demand for emerging chemicals is growing, artificial intelligence and machine learning (AI/ML) offer alternative tools to predict ADME profiles for data-poor chemicals. This review examines recent developments in AI-based TK parameterization with relevance to environmental exposure assessment. First, we outline the essential elements of AI-based TK parameterization paradigms, including data sources, molecular representations, and commonly used ML algorithms. Second, the recent applications of AI/ML models across four ADME processes are introduced. Third, we discuss current limitations and future perspectives, including model reliability, interpretability, data availability for environmental chemicals, and integrating AI-predicted TK parameters into mechanistic models, such as physiologically based pharmacokinetic (PBPK) models and other downstream modeling frameworks. Although many existing studies rely on pharmaceutical or mixed chemical datasets, these approaches could support TK parameterization and exposure assessment for environmental chemicals. Further progress will depend on improved data quality, broader environmental chemical coverage, and more rigorous model validation and uncertainty evaluation.

Keywords

Toxicokinetics / exposure assessment / artificial intelligence / ADME

References

[1] National Research Council. Toxicity testing in the 21st century: a vision and a strategy. National Academies Press; 2007.

[2] Chen Q,Lin Z.Integration of toxicogenomics and physiologically based pharmacokinetic modeling in human health risk assessment of perfluorooctane sulfonate.Environ Sci Technol2022;56:3623-33

[3] Wambaugh JF,Ring CL.Evaluating in vitro-in vivo extrapolation of toxicokinetics.Toxicol Sci2018;163:152-69 PMCID:PMC5920326

[4] Hartung T.Toxicology for the twenty-first century.Nature2009;460:208-12

[5] U.S. EPA. TSCA chemical substance inventory. https://www.epa.gov/tsca-inventory. (accessed 2026-09-21)

[6] Wambaugh JF,Pearce R.Toxicokinetic triage for environmental chemicals.Toxicol Sci2015;147:55-67 PMCID:PMC4560038

[7] Breen M,Kreutz A,Wambaugh JF.High-throughput PBTK models for in vitro to in vivo extrapolation.Expert Opin Drug Metab Toxicol2021;17:903-21 PMCID:PMC9703392

[8] Isaacs KK,Dionisio KL.The chemical landscape of high-throughput new approach methodologies for exposure.J Expo Sci Environ Epidemiol2022;32:820-32 PMCID:PMC9882966

[9] Vamathevan J,Czodrowski P.Applications of machine learning in drug discovery and development.Nat Rev Drug Discov2019;18:463-77 PMCID:PMC6552674

[10] Koirala M,Mohamed Z.AI-integrated QSAR modeling for enhanced drug discovery: from classical approaches to deep learning and structural insight.Int J Mol Sci2025;26:9384 PMCID:PMC12525248

[11] Chou WC,Lin Z.Chapter 4 - Application of machine learning and artificial intelligence methods in physiologically based pharmacokinetic modeling. In Machine learning and artificial intelligence in toxicology and environmental health. Elsevier; 2026. pp. 99-138.

[12] Ajisafe OM,Egbon E,Olawade DB.The role of machine learning in predictive toxicology: a review of current trends and future perspectives.Life Sci2025;378:123821

[13] Lones MA.Avoiding common machine learning pitfalls.Patterns2024;5:101046 PMCID:PMC11573893

[14] Wang H,Liu W.Using machine learning for green substitution of industrial chemicals: integrating functionality, hazard, and life cycle impact.Chem Rev2026;126:841-94

[15] Fu X,Neagu D,Travis K.Data governance in predictive toxicology: a review.J Cheminform2011;3:24 PMCID:PMC3584675

[16] Deepika D.The role of “physiologically based pharmacokinetic model (pbpk)” new approach methodology (nam) in pharmaceuticals and environmental chemical risk assessment.Int J Environ Res Public Health2023;20:3473 PMCID:PMC9966583

[17] Claire T.Integrating toxicokinetics into toxicology studies and the human health risk assessment process for chemicals: Reduced uncertainty, better health protection.Regul Toxicol Pharmacol2022;128:105092

[18] Kim SJ,Lee DS,Lee YB.Gender differences in pharmacokinetics and tissue distribution of 3 perfluoroalkyl and polyfluoroalkyl substances in rats.Food Chem Toxicol2016;97:243-55

[19] Karmaus AL,Oyetade O.Perspectives on variability of in vivo toxicology studies: considerations for next-generation toxicology.Front Toxicol2026;8:1778353 PMCID:PMC12989281

[20] Napoli JA,Brandl P,Hert J.Multitask deep learning models of combined industrial absorption, distribution, metabolism, and excretion datasets to improve generalization.Mol Pharm2025;22:1892-900

[21] Kreutz A,Hogberg HT.Advancing understanding of human variability through toxicokinetic modeling, in vitro-in vivo extrapolation, and new approach methodologies.Hum Genomics2024;18:129 PMCID:PMC11580331

[22] Alves VM,Kleinstreuer N.Curated data in - trustworthy in silico models out: the impact of data quality on the reliability of artificial intelligence models as alternatives to animal testing.Altern Lab Anim2021;49:73-82 PMCID:PMC8609471

[23] Pearce RG,Strope CL,Sipes NS.httk: R package for high-throughput toxicokinetics.J Stat Softw2017;79:1-26 PMCID:PMC6134854

[24] Wambaugh JF,Ring CL.Assessing toxicokinetic uncertainty and variability in risk prioritization.Toxicol Sci2019;172:235-51 PMCID:PMC8136471

[25] Mansouri K,Chang X,Allen D. 4.18.T-01 - OPERA: open-source QSAR models for regulatory support. 2023. https://studio.m-anage.com/setac/sna2023/meetingapp.cgi/Paper/17094. (accessed 2026-09-21)

[26] Chakraborty S,Mitta R., In silico toxicokinetics. In Computer simulations in the pharmaceutical industry. CRC Press; 2026. pp. 199-220.

[27] Chen CY.Exploring the potential and challenges of developing physiologically-based toxicokinetic models to support human health risk assessment of microplastic and nanoplastic particles.Environ Int2024;186:108617

[28] Wilhelm S,Dai Q.Analysis of nanoparticle delivery to tumours.Nat Rev Mater2016;1:16014

[29] Chen Q,Chou WC.Meta-analysis of nanoparticle distribution in tumors and major organs in tumor-bearing mice.ACS Nano2023;17:19810-31 PMCID:PMC10604101

[30] Mi K,Yuan L.Analysis of pharmacokinetic-pharmacodynamic relationships of nanoparticles against tumors.ACS Nano2026;20:22485-504 PMCID:PMC13488496

[31] Alexandropoulos SAN,Vrahatis MN.Data preprocessing in predictive data mining.Knowl Eng Rev2019;34:e1

[32] Cabello-Solorzano K,Peña M,Tallón-Ballesteros AJ.The impact of data normalization on the accuracy of machine learning algorithms: a comparative analysis. In 18th International Conference on Soft Computing Models in Industrial and Environmental Applications (SOCO 2023). Springer, Cham; 2023. pp. 344-53.

[33] David L,Mercado R.Molecular representations in AI-driven drug discovery: a review and practical guide.J Cheminform2020;12:56 PMCID:PMC7495975

[34] Deng J,Wang H,Samaras D.A systematic study of key elements underlying molecular property prediction.Nat Commun2023;14:6395 PMCID:PMC10575948

[35] Handa K,Kageyama M.Computational approaches to DMPK: a realistic assessment of current methods and their practical impact. Part I: Physicochemical and in vitro properties.Drug Discov Today2025;30:104422

[36] Ryu JY,Jang J.PredAOT: a computational framework for prediction of acute oral toxicity based on multiple random forest models.BMC Bioinformatics2023;24:66 PMCID:PMC9951537

[37] Xu Y,Lai L.Deep learning based regression and multiclass models for acute oral toxicity prediction with automatic chemical feature extraction.J Chem Inf Model2017;57:2672-85

[38] Jiang J,Wei GW.GGL-Tox: geometric graph learning for toxicity prediction.J Chem Inf Model2021;61:1691-700 PMCID:PMC8155789

[39] Noga M.Preliminary prediction of toxicologically relevant physicochemical properties of Novichoks: the first comparative in silico studies.Chem Biol Interact2025;419:111644

[40] Li T,Tong W.Bridging organ transcriptomics for advancing multiple organ toxicity assessment with a generative AI approach.NPJ Digit Med2024;7:310 PMCID:PMC11538515

[41] Führer F,Diedam H,Menz S.A deep neural network: mechanistic hybrid model to predict pharmacokinetics in rat.J Comput Aided Mol Des2024;38:7

[42] Wu X,Chou WC,Lin Z.A machine learning-empowered quantitative structure-activity relationship model for predicting the plasma half-life of drugs in dogs.AAPS J2025;28:22

[43] Fan N,Wang J,Yang Y.Bridging data and drug development: machine learning approaches for next-generation ADMET prediction.Drug Discov Today2025;30:104487

[44] Ng SSS.Evaluating the use of graph neural networks and transfer learning for oral bioavailability prediction.J Chem Inf Model2023;63:5035-44 PMCID:PMC10467575

[45] Chou WC.Machine learning and artificial intelligence in physiologically based pharmacokinetic modeling.Toxicol Sci2023;191:1-14 PMCID:PMC9887681

[46] Svetnik V,Tong C,Sheridan RP.Random forest: a classification and regression tool for compound classification and QSAR modeling.J Chem Inf Comput Sci2003;43:1947-58

[47] Chen T. XGBoost: a scalable tree boosting system. arXiv 2016, arXiv:1603.02754. Available online: https://doi.org/10.48550/arXiv.1603.02754. (accessed 2026-09-21)

[48] Rodríguez-Pérez R.Evolution of support vector machine and regression modeling in chemoinformatics and drug discovery.J Comput Aided Mol Des2022;36:355-62 PMCID:PMC9325859

[49] Sakiyama Y.The use of machine learning and nonlinear statistical tools for ADME prediction.Expert Opin Drug Metab Toxicol2009;5:149-69

[50] Mansouri K,Judson RS.OPERA models for predicting physicochemical properties and environmental fate endpoints.J Cheminform2018;10:10 PMCID:PMC5843579

[51] Venkataraman M,Madavareddi JK.Leveraging machine learning models in evaluating ADMET properties for drug discovery and development.ADMET DMPK2025;13:2772 PMCID:PMC12205928

[52] Pantic I,Cumic J,Petroianu GA.Artificial neural networks in contemporary toxicology research.Chem Biol Interact2023;369:110269

[53] Gilmer J,Riley PF,Dahl GE. Neural message passing for quantum chemistry. arXiv 2017, arXiv:1704.01212. Available online: https://doi.org/10.48550/arXiv.1704.01212. (accessed 2026-09-21)

[54] Zhang Z,Lin Z.Development of machine learning and chemical language model-based QSAR models for predicting drug residue depletion half-lives in plasma and tissues of cattle across various administration routes.J Vet Pharmacol Ther2026;49:150-71

[55] Umer MS,Athar U.Large language models meet molecules: a systematic review of advances and challenges in AI-driven cheminformatics.Arch Computat Methods Eng2026;33:4867-908

[56] Limbu S,Dakshanamurthy S.Predicting dose-range chemical toxicity using novel hybrid deep machine-learning method.Toxics2022;10:706 PMCID:PMC9692315

[57] Zhang J,Zhang Y.Computational toxicology in drug discovery: applications of artificial intelligence in ADMET and toxicity prediction.Brief Bioinform2025;26:bbaf533 PMCID:PMC12499773

[58] Tropsha A.Best practices for QSAR model development, validation, and exploitation.Mol Inform2010;29:476-88

[59] Wallach I.Most ligand-based classification benchmarks reward memorization rather than generalization.J Chem Inf Model2018;58:916-32

[60] Sheridan RP.Time-split cross-validation as a method for estimating the goodness of prospective prediction.J Chem Inf Model2013;53:783-90

[61] OECD. Guidance document on the validation of (quantitative) structure-activity relationship [(Q)SAR] models. 2014. https://www.oecd.org/en/publications/guidance-document-on-the-validation-of-quantitative-structure-activity-relationship-q-sar-models_9789264085442-en.html. (accessed 2026-09-21)

[62] Kar S,Leszczynski J.Applicability domain: a step toward confident predictions and decidability for QSAR modeling.Methods Mol Biol2018;1800:141-69

[63] Alqahtani S.Improving on in-silico prediction of oral drug bioavailability.Expert Opin Drug Metab Toxicol2023;19:665-70

[64] Kumar R,Siddiqui MH.Prediction of human intestinal absorption of compounds using artificial intelligence techniques.Curr Drug Discov Technol2017;14:244-54

[65] Kamiya Y,Hayasaka R.Prediction of permeability across intestinal cell monolayers for 219 disparate chemicals using in vitro experimental coefficients in a pH gradient system and in silico analyses by trivariate linear regressions and machine learning.Biochem Pharmacol2021;192:114749

[66] Wang D,Shi G.ADMET evaluation in drug discovery: 21. Application and industrial validation of machine learning algorithms for Caco-2 permeability prediction.J Cheminform2025;17:3 PMCID:PMC11724520

[67] Rácz A,Volk B.Extending the limitations in the prediction of PAMPA permeability with machine learning algorithms.Eur J Pharm Sci2023;188:106514

[68] Narita I,Fujiwara C.In silico model to predict dermal absorption of chemicals in finite dose conditions.J Toxicol Sci2025;50:171-86

[69] Sarti D,Palma F.Interpretable machine learning unveils key predictors and default values in an expanded database of human in vitro dermal absorption studies with pesticides.Regul Toxicol Pharmacol2025;159:105801

[70] Chiu YW,Wang CC.Multitask learning for predicting pulmonary absorption of chemicals.Food Chem Toxicol2024;185:114453

[71] Simon L.Advancing exposure science through artificial intelligence: neural ordinary differential equations for predicting blood concentrations of volatile organic compounds.Ecotoxicol Environ Saf2025;292:117928

[72] Feschuk AM,Maibach HI.Comparative efficacy of reactive skin decontamination lotion (RSDL): a systematic review.Toxicol Lett2021;349:109-14

[73] Dawson DE,Phillips KA,Wambaugh JF.Designing QSARs for parameters of high-throughput toxicokinetic models using open-source descriptors.Environ Sci Technol2021;55:6505-17 PMCID:PMC8548983

[74] Mi K,Chen Q.Predicting tissue distribution and tumor delivery of nanoparticles in mice using machine learning models.J Control Release2024;374:219-29 PMCID:PMC11886896

[75] Liu L,Feng H.Prediction of the blood-brain barrier (BBB) permeability of chemicals based on machine-learning and ensemble methods.Chem Res Toxicol2021;34:1456-67

[76] Nguyen T,Mukta FT,Nguyen DD.Geometric multi-color message passing graph neural networks for blood–brain barrier permeability prediction.Mol Syst Des Eng2026;11:436-46

[77] Huang ETC,Liao KYK.Predicting blood-brain barrier permeability of molecules with a large language model and machine learning.Sci Rep2024;14:15844 PMCID:PMC11233737

[78] Li J,Xu J,Zeng EY.Transplacental transfer of environmental chemicals: roles of molecular descriptors and placental transporters.Environ Sci Technol2021;55:519-28

[79] Chen X,Ma Y.Rapid screening of chemicals with placental transfer risk using interpretable machine learning.Environ Sci Technol Lett2024;11:798-804

[80] Guan R,Guo B,Zhao C.A data-driven computational framework for assessing the risk of placental exposure to environmental chemicals.Environ Sci Technol2024;58:7770-81

[81] Edbert Duru, C. Forever chemicals could expose the human fetus to xenobiotics by binding to placental enzymes: prescience from molecular docking, DFT, and machine learning.Comput Toxicol2023;26:100274

[82] Huang X,Liu P.Assessing chemical exposure risk in breastfeeding infants: an explainable machine learning model for human milk transfer prediction.Ecotoxicol Environ Saf2025;289:117707

[83] Maeshima T,Watanabe M.Prediction model for milk transfer of drugs by primarily evaluating the area under the curve using QSAR/QSPR.Pharm Res2023;40:711-9 PMCID:PMC10036427

[84] Zhao C,Zhang X.Prediction of milk/plasma drug concentration (M/P) ratio using support vector machine (SVM) method.Pharm Res2006;23:41-8

[85] Grant N,Yang Z,Wang C.Blood brain barrier permeability prediction with artificial intelligence and machine learning: a meta-review and future directions.Discov Artif Intell2025;5:494

[86] Nabi AE,Liu L,Ayubcha C.Machine learning in drug development for neurological diseases: a review of blood brain barrier permeability prediction models.Mol Inform2025;44:e202400325 PMCID:PMC11949286

[87] Agudelo-Pérez S,Rodríguez-Alvarado L,Raigoso-Díaz L.Artificial intelligence applied to the study of human milk and breastfeeding: a scoping review.Int Breastfeed J2024;19:79 PMCID:PMC11622664

[88] Lai Y,Di L.Recent advances in the translation of drug metabolism and pharmacokinetics science for drug discovery and development.Acta Pharm Sin B2022;12:2751-77 PMCID:PMC9214059

[89] Sun H,Xia M,Tice RR.Prediction of cytochrome P450 profiles of environmental chemicals with QSAR models built from drug-like molecules.Mol Inform2012;31:783-92 PMCID:PMC3583379

[90] Tyzack JD,Williamson MJ,Glen RC.Cytochrome P450 site of metabolism prediction from 2D topological fingerprints using GPU accelerated probabilistic classifiers.J Cheminform2014;6:29 PMCID:PMC4047555

[91] Yang H,Chen K.D-CyPre: a machine learning-based tool for accurate prediction of human CYP450 enzyme metabolic sites.PeerJ Comput Sci2024;10:e2040 PMCID:PMC11157575

[92] Ryu JY,Lee BH,Ahn S.PredMS: a random forest model for predicting metabolic stability of drug candidates in human liver microsomes.Bioinformatics2022;38:364-8

[93] Sasahara K,Sasabe H.Predicting drug metabolism and pharmacokinetics features of in-house compounds by a hybrid machine-learning model.Drug Metab Pharmacokinet2021;39:100395

[94] Kamiya Y,Miura T.Machine learning prediction of the three main input parameters of a simplified physiologically based pharmacokinetic model subsequently used to generate time-dependent plasma concentration data in humans after oral doses of 212 disparate chemicals.Biol Pharm Bull2022;45:124-8

[95] Litsa EE,Kavraki LE.Machine learning models in the prediction of drug metabolism: challenges and future perspectives.Expert Opin Drug Metab Toxicol2021;17:1245-7 PMCID:PMC8759823

[96] Tyzack JD.Computational methods and tools to predict cytochrome P450 metabolism for drug discovery.Chem Biol Drug Des2019;93:377-86 PMCID:PMC6590657

[97] Wang D,Shen Z.Deep learning based drug metabolites prediction.Front Pharmacol2019;10:1586 PMCID:PMC7003989

[98] Chao P,Cheng KC.Use of intrinsic clearance for prediction of human hepatic clearance.Expert Opin Drug Metab Toxicol2010;6:189-98

[99] Tran TTV,Chong KT.Artificial intelligence in drug metabolism and excretion prediction: recent advances, challenges, and future perspectives.Pharmaceutics2023;15:1260 PMCID:PMC10143484

[100] Bois FY,Clewell HJ.PBPK modelling of inter-individual variability in the pharmacokinetics of environmental chemicals.Toxicology2010;278:256-67

[101] Tucker GT.Measurement of the renal clearance of drugs.Br J Clin Pharmacol1981;12:761-70 PMCID:PMC1401922

[102] Lee W.Transporters and renal drug elimination.Annu Rev Pharmacol Toxicol2004;44:137-66

[103] Wang ZJ,Tomlinson B.OCT2 polymorphisms and in-vivo renal functional consequence: studies with metformin and cimetidine.Pharmacogenet Genomics2008;18:637-45

[104] Ryu S,Sadegh Modaresi SM.Evaluation of 14 PFAS for permeability and organic anion transporter interactions: implications for renal clearance in humans.Chemosphere2024;361:142390 PMCID:PMC11774580

[105] Sharifi M.Estimation of biliary excretion of foreign compounds using properties of molecular structure.AAPS J2014;16:65-78 PMCID:PMC3889537

[106] Tátrai P,Krajcsi P.Role of hepatocyte transporters in drug-induced liver injury (DILI)-in vitro testing.Pharmaceutics2022;15:29 PMCID:PMC9866820

[107] Nakanishi T.Interaction of drug or food with drug transporters in intestine and liver.Curr Drug Metab2015;16:753-64

[108] Baker M.Kinetic determinants of hepatic clearance: plasma protein binding and hepatic uptake.Xenobiotica2007;37:1110-34

[109] Paine SW,Bird J.A rapid computational filter for predicting the rate of human renal clearance.J Mol Graph Model2010;29:529-37

[110] Watanabe R,Esaki T.Development of an in silico prediction system of human renal excretion and clearance from chemical structure information incorporating fraction unbound in plasma as a descriptor.Sci Rep2019;9:18782 PMCID:PMC6906481

[111] Hosey CM,Benet LZ.Predicting when biliary excretion of parent drug is a major route of elimination in humans.AAPS J2014;16:1085-96 PMCID:PMC4147063

[112] Chou WC,Yuan L.An artificial intelligence-assisted physiologically-based pharmacokinetic model to predict nanoparticle delivery to tumors in mice.J Control Release2023;361:53-63 PMCID:PMC11008607

[113] Obrezanova O,Whitehead T.Prediction of in vivo pharmacokinetic parameters and time-exposure curves in rats using machine learning from the chemical structure.Mol Pharm2022;19:1488-504

[114] Jia X,Zhu H.Advancing computational toxicology by interpretable machine learning.Environ Sci Technol2023;57:17690-706 PMCID:PMC10666545

[115] Tjoa E.A survey on explainable artificial intelligence (XAI): toward medical XAI.IEEE Trans Neural Netw Learn Syst2021;32:4793-813

[116] Hassija V,Mahapatra A.Interpreting black-box models: a review on explainable artificial intelligence.Cogn Comput2024;16:45-74

[117] Wu K,Zhou Z.Predicting pharmacodynamic effects through early drug discovery with artificial intelligence-physiologically based pharmacokinetic (AI-PBPK) modelling.Front Pharmacol2024;15:1330855 PMCID:PMC10904617

[118] Wang W,Wu Y.An integrated AI-PBPK platform for predicting drug in vivo fate and tissue distribution in human and inter-species extrapolation.Clin Pharmacol Ther2025;118:865-75

[119] Kamiya Y,Miura T.Plasma and hepatic concentrations of chemicals after virtual oral administrations extrapolated using rat plasma data and simple physiologically based pharmacokinetic models.Chem Res Toxicol2019;32:211-8