Precise Control of Drug Release in Machine Learning-Designed Antibody-Eluting Implants for Postoperative Scarring...

Area of Science:

• Biomaterials Science
• Drug Delivery Systems
• Ophthalmology

Background:

• Postoperative fibrosis and scarring are significant clinical challenges.
• Sustained delivery of anti-fibrotic agents is crucial for effective treatment.
• Current drug delivery methods often lack precise control over release kinetics.

Purpose of the Study:

• To develop a smart, scalable subconjunctival implant for sustained delivery of basic fibroblast growth factor monoclonal antibody (FGFb mAb).
• To utilize machine learning for precise prediction and control of drug release profiles.
• To evaluate the efficacy and biocompatibility of the developed implant system in preclinical models.

Main Methods:

• Fabrication of polycaprolactone (PCL) and polyethylene glycol (PEG) micro-cylindrical implants.
• Machine learning (LightGBM) analysis to predict drug release kinetics based on fabrication parameters.
• In vitro evaluation of fibroblast-mediated collagen contraction and fibrotic gene expression.
• In vivo studies in a rat glaucoma filtration surgery (GFS) model with histological analysis.

Main Results:

• LightGBM accurately predicted drug release kinetics (R² = 0.9000 ± 0.0058).
• Optimized PCL-PEG implants with PLGA bio-coating showed sustained antibody release.
• In vitro and in vivo studies demonstrated significant inhibition of fibrosis markers and collagen contraction.
• The implant system exhibited no significant toxicity and good biocompatibility.

Conclusions:

• The developed smart implant system offers precise, sustained delivery of FGFb mAb for anti-fibrotic therapy.
• Machine learning integration enables accurate prediction and optimization of drug release.
• The PCL-PEG/FGFb mAb@PLGA implant shows significant potential for modulating wound healing and preventing postoperative fibrosis.