Machine Learning-Assisted Optimization of Drug Combinations in Zeolite-Based Delivery Systems for Melanoma Therapy

Area of Science:

• Nanotechnology and Materials Science
• Computational Chemistry and Cheminformatics
• Pharmacology and Therapeutics

Background:

• Zeolite-based delivery systems (ZDS) offer potential for targeted drug delivery.
• Silver (Ag+) and 5-fluorouracil (5-FU) are key antimicrobial and antineoplastic agents.
• Optimizing drug combinations for enhanced therapeutic efficacy is crucial in cancer treatment.

Purpose of the Study:

• To determine the optimal drug combination of ZDS for cancer therapy using artificial neural network (ANN) models.
• To evaluate the anticancer and antimicrobial efficacy of silver and 5-FU incorporated into NaY zeolite.
• To develop a virtual cell viability assay protocol for predicting drug efficacy.

Main Methods:

• Preparation and characterization of NaY zeolite loaded with silver (Ag+) and 5-fluorouracil (5-FU).
• In vitro cell viability assays using A375 cancer cells with ZDS, 5-FU, and Ag+ solutions.
• Training two independent machine learning (ML) models using experimental cell viability data.

Main Results:

• Successful incorporation of Ag+ and 5-FU into NaY zeolite without structural changes.
• ANN models accurately predicted experimental cell viability, enabling virtual assay development.
• Zeolite-incorporated Ag+ and 5-FU demonstrated potentiated anticancer activity compared to liquid phase administration.
• Optimal AgY/5-FU@Y ratios were identified for maximum cell viability reduction.
• ZDS exhibited significant efficacy against Escherichia coli and Staphylococcus aureus.

Conclusions:

• The developed ZDS effectively delivers combined antimicrobial and antineoplastic agents.
• Machine learning models provide a reliable method for predicting drug efficacy and optimizing ZDS formulations.
• This approach holds promise for treating cancer and associated bacterial infections, particularly Staphylococcus aureus infections.