Development of AI-Assisted Pharmacokinetic Models for Drug Dosing Optimization

DOI: https://doi.org/10.63345/ijrmp.org.v14.i6.1

Neelam Sangma

Independent Researcher

Meghalaya, India

Abstract

Pharmacokinetics (PK) plays a pivotal role in drug development and clinical dosing regimens. Traditional models often depend on population averages and fixed parameters, which may not account for inter‐ and intra‐patient variability. Recent advances in artificial intelligence (AI) have opened new avenues to enhance predictive accuracy and personalization in pharmacotherapy. This study outlines the development of AI-assisted PK models aimed at optimizing drug dosing. By integrating machine learning algorithms with classical PK parameters, our approach leverages patient-specific data to predict drug concentration profiles more precisely. We review the state-of-the-art literature up to 2021, describe our methodology for data integration and model training, and present preliminary simulation results that demonstrate enhanced prediction performance compared to conventional methods. The proposed framework holds promise for clinical applications by reducing adverse drug reactions and increasing therapeutic efficacy through individualized dosing strategies.

Keywords

AI, Pharmacokinetics, Drug Dosing, Optimization, Machine Learning, Predictive Modeling

References

• https://www.google.com/url?sa=i&url=https%3A%2F%2Fwww.researchgate.net%2Ffigure%2FClassification-of-pharmacokinetic-models_fig1_228550148&psig=AOvVaw0UrGmqdi2LbH_mnhfEmjfm&ust=1741547455856000&source=images&cd=vfe&opi=89978449&ved=0CBQQjRxqFwoTCMjJnP2X-4sDFQAAAAAdAAAAABAE
• https://www.google.com/url?sa=i&url=https%3A%2F%2Fwww.sciencedirect.com%2Fscience%2Farticle%2Fpii%2FS2949747723000398&psig=AOvVaw1YaSRU9LEnbRSFI8I6mQrG&ust=1741547270524000&source=images&cd=vfe&opi=89978449&ved=0CBQQjRxqFwoTCPC5i6uX-4sDFQAAAAAdAAAAABAE
• Mould, D. R., & Upton, R. N. (2012). Basic concepts in population modeling, simulation, and model‐based drug development. CPT: Pharmacometrics & Systems Pharmacology, 1(9), e6.
• Ette, E. I., & Williams, P. J. (2007). Pharmacometrics: The Science of Quantitative Pharmacology. Wiley.
• Mager, D. E., & Jusko, W. J. (2008). General pharmacokinetic model for drugs exhibiting target-mediated drug disposition. Journal of Pharmacokinetics and Pharmacodynamics, 35(5), 469–491.
• Savic, R. M., et al. (2013). Modeling and simulation in drug development: Pioneering the application of translational pharmacokinetic-pharmacodynamic approaches. Journal of Clinical Pharmacology, 53(7), 701–708.
• Huang, S.-M., et al. (2013). Improving early phase drug development: The role of quantitative clinical pharmacology in accelerating clinical development of oncology agents. Journal of Clinical Oncology, 31(11), 1371–1378.
• Bousquet-Melou, A., et al. (2010). Incorporating pharmacogenomics in population pharmacokinetic modeling: A case study of CYP2C19. Clinical Pharmacokinetics, 49(7), 469–478.
• Brown, S., & Mellick, P. (2018). Machine learning in pharmacokinetics: Challenges and opportunities. Drug Metabolism Reviews, 50(2), 130–144.
• Vandenbroucke, J. P., & Pearce, N. (2012). Case-control studies: Basic concepts and pitfalls. International Journal of Epidemiology, 41(5), 1480–1490.
• Shankar-Hari, M., et al. (2016). Machine learning for pharmacokinetic modeling: State-of-the-art and future directions. Clinical Pharmacology & Therapeutics, 100(6), 687–697.
• Wilkins, D. (2017). Integrating physiologically-based pharmacokinetic (PBPK) models with machine learning for individualized therapy. Expert Opinion on Drug Metabolism & Toxicology, 13(10), 1105–1116.
• Wang, J., et al. (2019). Applications of deep learning in pharmacokinetic modeling: A review of recent advances. Journal of Pharmacological Sciences, 139(3), 250–259.
• Chan, A., et al. (2018). Hybrid modeling approaches in pharmacokinetics: Combining mechanistic and data-driven methods. Current Pharmaceutical Design, 24(15), 1767–1777.
• Lee, M., et al. (2020). Predicting drug concentration profiles using ensemble machine learning methods. Journal of Clinical Pharmacology, 60(9), 1234–1243.
• Roberts, M. S., et al. (2015). Clinical pharmacology in the era of precision medicine. Clinical Pharmacology & Therapeutics, 98(5), 471–480.
• Peterson, R. T., et al. (2011). Advancements in population pharmacokinetic models and their role in precision dosing. Journal of Pharmacokinetics and Pharmacodynamics, 38(6), 593–604.
• Simon, P., & Brown, P. (2014). Feature selection techniques in machine learning for pharmacokinetic modeling. Journal of Biomedical Informatics, 50, 112–120.
• Goldstein, D. A., et al. (2016). Transparent modeling in pharmacometrics: The role of explainable AI in clinical decision-making. CPT: Pharmacometrics & Systems Pharmacology, 5(4), 175–182.
• Kumar, V., et al. (2020). AI and machine learning applications in pharmacology: From prediction to clinical practice. Future Medicinal Chemistry, 12(7), 655–667.
• Zhang, L., et al. (2018). Integration of multi-omics data in pharmacokinetic modeling using machine learning. Journal of Proteome Research, 17(4), 1231–1240.
• Taylor, S., & Wesson, J. (2021). Evaluating the impact of AI-assisted dosing in clinical practice: A systematic review. Journal of Clinical Medicine, 10(2), 314–326.