Machine Learning-Driven Prediction, Preparation, and Evaluation of Functional Nanomedicines Via Drug-Drug...

Area of Science:

• Nanomedicine
• Artificial Intelligence
• Drug Delivery

Background:

• Small molecules offer advantages as nanomedicine carriers, but selection for stable nanomedicines is challenging.
• Developing effective nanomedicines requires efficient screening of drug combinations and carrier molecules.

Purpose of the Study:

• To utilize artificial intelligence (AI) and machine learning to screen for small molecules capable of forming stable, self-assembling nanomedicines.
• To identify effective carrier molecules for antineoplastic drugs with high drug loading capacity and targeted delivery capabilities.

Main Methods:

• Employed AI and machine learning algorithms to predict nanomedicine formation based on physiochemical parameters.
• Screened Non-Steroidal Anti-Inflammatory Drugs (NSAIDs) as potential carriers for antineoplastic drugs.
• Developed and characterized PEG-coated (PiPTX) and laminarin-modified (LiDOX) indomethacin/paclitaxel nanomedicines.

Main Results:

• NSAIDs were identified as effective carriers for antineoplastic drugs, enabling high drug loading.
• Developed nanomedicines (PiPTX, LiDOX) exhibited extended circulation and active targeting functions.
• The indomethacin/paclitaxel nanomedicine (iDOX) demonstrated pH-responsive drug release in the tumor microenvironment.

Conclusions:

• AI-driven screening accelerates the development of novel nanomedicines.
• NSAID-based nanomedicines show potential for enhanced anti-tumor efficacy and reduced systemic toxicity.
• This approach offers a rapid pathway for the clinical translation of nanomedicine therapies.