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

Area of Science:

• Biomaterials Science
• Drug Delivery Systems
• Computational Chemistry

Background:

• Long-acting injectables offer improved efficacy, safety, and patient compliance for chronic disease management.
• Polymer materials provide diverse properties for advanced drug formulation strategies.
• Predicting the performance of these systems is challenging due to complex drug-polymer interactions.

Purpose of the Study:

• To demonstrate the application of machine learning in predicting drug release from long-acting injectables.
• To show how trained machine learning models can guide the design of new long-acting injectable formulations.
• To highlight the potential of a data-driven approach to streamline drug formulation development.

Main Methods:

• Utilizing machine learning algorithms to analyze experimental data.
• Training predictive models based on physicochemical properties of drugs and polymers.
• Validating model performance in predicting in vitro drug release profiles.

Main Results:

• Machine learning models accurately predict experimental drug release from long-acting injectable systems.
• Trained models successfully guided the design of novel long-acting injectable formulations.
• Demonstrated feasibility of a data-driven approach for formulation development.

Conclusions:

• Machine learning offers a powerful tool for predicting drug release in long-acting injectables.
• Data-driven methods can significantly accelerate the design and optimization of drug delivery systems.
• This approach has the potential to reduce the time and cost associated with developing new therapeutics.