Abstract:Current microparticle (MP) development still strongly relies on the laborious trial-and-error approach. Herein, we developed a systemic method to evaluate the significance of MP formulation factors and predict drug loading efficiency (DLE) using design of experiment (DoE) and machine learning. A first-in-class 3D printing concept was initially employed to fabricate polymeric MPs by a 3D printer. Sprayed Multi Adsorbed-droplet Reposing Technology (SMART) was developed to combine extrusion-based printing with multi-materials emulsion evaporation technique to fabricate a small molecule drug i.e., 6-thioguanine (6-TG) loaded poly (lactide-co-glycolide) (PLGA) MPs. Compared to conventional emulsion evaporation method, SMART employs the shear force exerted by the printing nozzle rather than the sonication energy to generate micro-sized emulsion droplets in a single step. Furthermore, the applied shear force in the 3D printing process reported herein is controllable and consistent since the emulsion is extruded through the nozzle under preset printing conditions. The formulated MPs exhibited spherical structure with size distribution ∼1-3 μ m in diameter and reached ∼100% drug release at 10h. Also, the papain-like protease (PLpro) inhibition efficacy of 6-TG in formulated MPs was maintained even after the printing process under different printing parameters. Furthermore, the formulation factor importance was assessed by DoE statistical analysis and further validated by machine learning modeling. Among the four process parameters (drug amount, printing speed, printing pressure, and nozzle size), drug amount was the most influential formulation factor. Moreover, it is interesting that nearly all the machine learning models, especially decision tree (DT), demonstrated superior performance in predicting DLE compared to DoE regression models. Overall, incorporating DoE and machine learning modeling shows great promises in the prediction and optimization of MP formulations factors with precise control over particle size by means of a novel SMART technology. Moreover, this systemic approach helps streamline the development of MP with programmable pharmaceutical attributes, representing a new paradigm for digital pharmaceutical science.

Gaussian processes modeling for the prediction of polymeric nanoparticle formulation design to enhance encapsulation efficiency and therapeutic efficacy

Gaussian Process Regression Approach for Robust Design and Yield Enhancement of Self-Assembled Nanostructures

Machine learning models to accelerate the design of polymeric long-acting injectables

Computationally guided high-throughput design of self-assembling drug nanoparticles

Optimising the production of PLGA nanoparticles by combining design of experiment and machine learning

Toward the Integration of Machine Learning and Molecular Modeling for Designing Drug Delivery Nanocarriers

A Machine Learning Approach for PLGA Nanoparticles in Antiviral Drug Delivery

Towards rational design of API-poly(D, L-lactide-co-glycolide) based micro- and nanoparticles: The role of API-polymer compatibility prediction

Machine learning assisted exploration of the influential parameters on the PLGA nanoparticles

Computationally guided design and synthesis of dual‐drug loaded polymeric nanoparticles for combination therapy

Modeling and Simulation of Process Technology for Nanoparticulate Drug Formulations—A Particle Technology Perspective

Nanodelivery of a Self-Assembling Prodrug with Exceptionally High Drug Loading Potentiates Chemotherapy Efficacy.

3D bioprinted microparticles: Optimizing loading efficiency using advanced DoE technique and machine learning modeling

Therapeutic peptides are preferentially solubilized in specific microenvironments within PEG-PLGA polymer nanoparticles

Coarse‐Grained Model‐Assisted Design of Polymer Prodrug Nanoparticles with Enhanced Cytotoxicity: A Combined Theoretical and Experimental Study

Dimensionality reduction, and function approximation of poly(lactic-co-glycolic acid) micro- and nanoparticle dissolution rate

Empirical search for factors affecting mean particle size of PLGA microspheres containing macromolecular drugs

Drug Properties Prediction Based on Deep Learning

Machine learning of performance space mapping for the DPD simulation of drug delivery to endothelial cells

Machine learning framework to predict the performance of lipid nanoparticles for nucleic acid delivery

Assembling Biocompatible Polymers on Gold Nanoparticles: Toward a Rational Design of Particle Shape by Molecular Dynamics