FORMULATION-AWARE MACHINE LEARNING OF TIME-DEPENDENT DRUG RELEASE FROM PLGA MICROPARTICLES
DOI:
https://doi.org/10.65327/bp.v12i1.2545Keywords:
PLGA microparticles, drug release kinetics, machine learning, formulation descriptors, external validationAbstract
Models that rely only on sampling time may overlook formulation properties used to achieve a drug release that varies over time in PLGA microparticles. In this study, the researchers checked if the cumulative in vitro release could be better predicted by the drug or formulation descriptors. A literature-curated database of 321 release experiments and 4,913 measurements were analyzed. Several models using time as well as ten drug and formulation descriptors (histogram-based gradient boosting and random forest) were compared to a time-only boosted base model. Data from source publications were held out for separate validation split, formulation held out complete release curves. The random forest model had a mean absolute error (MAE) of 0.1216, an R² of 0.741, and a 31.7% decrease in MAE compared to the time-only baseline. Among the formulation-related variables, time was the most important parameter while the polymer M.W. and drug encapsulation efficiency were less important. On unseen studies, forest performance declined to an MAE of 0.1909 and R² of 0.361. Formulation data was more effective at predicting measured release; however, this was less consistent across independent studies. These models should be standardized in terms of release conditions and subsequently be validated before use in formulation selection.
Downloads
References
Bao, Z., Kim, J., Kwok, C., Le Devedec, F., & Allen, C. (2024). A dataset on formulation parameters and characteristics of drug-loaded PLGA microparticles (Version 2) [Data set]. Mendeley Data. https://doi.org/10.17632/zzvtdrcy76.2
Bruk, L. A., Fan, X., Resnick, J. L., & DiLeo, M. V. (2023). Controlled Release of Mesenchymal Stem Cell‐Conditioned Media from a Microsphere/Gel‐Based Drug Delivery System for Wound Healing of Tympanic Membrane Perforations. Journal of Tissue Engineering and Regenerative Medicine, 2023(1), 6039254.
Caputo, T. M., Cusano, A. M., Principe, S., Cicatiello, P., Celetti, G., Aliberti, A., ... & Cusano, A. (2023). Sorafenib-loaded PLGA carriers for enhanced drug delivery and cellular uptake in liver cancer cells. International Journal of Nanomedicine, 4121-4142.
Choi, S., Kang, B. S., Choi, G., Kang, M., Park, H., Kim, N., ... & Jung, H. S. (2023). Multichamber PLGA Microparticles with Enhanced Monodispersity and Encapsulation Efficiency Fabricated by a Batch‐Microfluidic Hybrid Approach. Advanced NanoBiomed Research, 3(10), 2300044.
Costa, M. S., Ramos, A. M., & Cardoso, M. M. (2025). Drug release kinetics of PLGA-PEG microspheres encapsulating aclacinomycin A: the influence of PEG content. Processes, 13(1), 112.
Erdoğan, H., Bacanlı, M. G., Karayavuz, B., Eşim, Ö., Sarper, M., Erdem, O., & Özkan, Y. (2024). Plasmonic photothermal therapy based synergistic drug release application through gold nanoparticles coated and embedded PLGA nanoparticles. Journal of Drug Delivery Science and Technology, 97, 105794.
Guerriero, I., Pesce, C., Spanò, R., Sganga, S., Tirelli, N., Palange, A. L., ... & Decuzzi, P. (2026). Programming Degradation and Drug Release Through Micropatterning of PLGA Films. ACS Applied Materials & Interfaces, 18(20), 28464.
Kim, M., Kim, J. H., Kim, S., Maharjan, R., Kim, N. A., & Jeong, S. H. (2022). New long-acting injectable microspheres prepared by IVL-DrugFluidic™ system: 1-month and 3-month in vivo drug delivery of leuprolide. International journal of pharmaceutics, 622, 121875.
Kregar, M. L., Krtalić, I., & Šagud, I. (2026). An Overview of In Vitro Release Methods for Long-Acting Injectable Products Based on PLGA. Methods and protocols, 9(3), 87.
Lee, J., Bae, J., Kwak, D., Kim, H., Kim, J., Hlaing, S. P., ... & Yoo, J. W. (2023). 5-Fluorouracil crystal-incorporated, pH-responsive, and release-modulating PLGA/Eudragit FS hybrid microparticles for local colorectal cancer-targeted chemotherapy. International Journal of Pharmaceutics, 630, 122443.
Li, X., Li, K., Liu, S., Yang, C., Yang, S., Guo, J., & Zhang, J. (2026). Ultralong-Acting PLGA Microspheres Engineered to Sustain Release across the Entire Bone-Regeneration Timeline. ACS Biomaterials Science & Engineering, 12(4), 2254-2269.
Liu, G., & McEnnis, K. (2022). Glass transition temperature of PLGA particles and the influence on drug delivery applications. Polymers, 14(5), 993.
Mittapelly, N., Djehizian, A., Telaprolu, K. C., McNally, K., Puttrevu, S. K., Arjmandi-Tash, O., ... & Bois, F. Y. (2024). Mechanistic model for drug release from PLGA-based biodegradable implants for in vitro release testing: development and validation. ACS Applied Bio Materials, 7(11), 7453-7465.
Nwazojie, C. C., Obayemi, J. D., Salifu, A. A., Borbor-Sawyer, S. M., Uzonwanne, V. O., Onyekanne, C. E., ... & Soboyejo, W. O. (2023). Targeted drug-loaded PLGA-PCL microspheres for specific and localized treatment of triple negative breast cancer. Journal of Materials Science: Materials in Medicine, 34(8), 41.
Petrov, I., Ivanova, O., & Smirnov, D. (2026). PREDICTING LONG-ACTING INJECTABLE RELEASE USING POLYMER DEGRADATION, DRUG LOADING, AND MICROSPHERE MORPHOLOGY. Pharmacophore, 17(1), 33.
Rahmani, F., Naderpour, S., Nejad, B. G., Rahimzadegan, M., Ebrahimi, Z. N., Kamali, H., & Nosrati, R. (2023). The recent insight in the release of anticancer drug loaded into PLGA microspheres. Medical Oncology, 40(8), 229.
Ran, X., Wang, Q., Sun, Y., Pan, Q., Chen, H., Ren, W., ... & Wang, X. (2024). Dual microparticles programmed delivery system regulating stem cell-based cartilage regeneration by cartilage-specific matrix hydrogels. Composites Part B: Engineering, 272, 111221.
Robles, K. N., & Samad, M. D. (2026). Predicting early and complete drug release from long-acting injectables using explainable machine learning. International journal of pharmaceutics, 126922.
Sagoe, P. N. K., Velázquez, E. J. M., Espiritusanto, Y. M., Gilbert, A., Orado, T., Wang, Q., & Jain, E. (2023). Fabrication of PEG-PLGA microparticles with tunable sizes for controlled drug release application. Molecules, 28(18), 6679.
Shafiee, K., Bazraei, S., Mashak, A., & Mobedi, H. (2024). The impact of temperature on the formation, release mechanism, and degradation of PLGA-based in-situ forming implants. Journal of Polymers and the Environment, 32(8), 3591-3608.
Sharifi, F., Otte, A., & Park, K. (2022). Initial formation of the skin layer of PLGA microparticles. Advanced healthcare materials, 11(7), 2101427.
Shinde, U. A., De, A., Godse, R. D., Rathod, M., Thorat, R., Sharma, N., ... & Singh, K. (2024). Efficacy evaluation of carboplatin-etoposide Co-loaded PLGA microparticles for subconjunctival chemotherapy in retinoblastoma. Journal of Drug Delivery Science and Technology, 92, 105333.
Song, Y., Bashyal, S., Cho, H. J., Woo, M. R., Kim, J. O., & Hwang, D. (2026). Interpretable Two-Stage Machine Learning for Early and Full Drug Release Prediction in PLGA Microspheres. Pharmaceuticals, 19(5), 767.
Sun, Y., Li, Y., & Liu, J. (2023, June). Mathematical modeling and numerical simulations for drug release from PLGA particles. In International Conference on Computational Science (pp. 347-360). Cham: Springer Nature Switzerland.
Tang, Z. A., Li, H., Li, T., & Xue, F. (2026). MC-NODE: A Mechanism-Decomposed Neural Differential Model for PLGA Microsphere Drug Release Prediction and Attribution. Pharmaceuticals, 19(8), 1227.
Wang, Y. (2025). Soft biomaterials for applicatons as local drug delivery systems for use in the brain (Doctoral dissertation, Cardiff University).
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 NNANGUMA Kingsley Ahamefula
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
