Quantifying drug tissue biodistribution by integrating high content screening with deep-learning analysis.

Area of Science:

• Pharmacology and Drug Discovery
• Biotechnology
• Computational Biology

Background:

• Accurate quantification of in vivo drug concentrations in target organs is crucial for drug discovery and translation.
• Current bioassay technologies struggle to differentiate drug binding mechanisms, hindering target engagement confirmation.
• Lack of high-throughput, sensitive methods limits the assessment of drug distribution and binding kinetics in preclinical models.

Purpose of the Study:

• To develop and validate a novel multiplexed, high-throughput method for quantifying drug distribution in tissues.
• To enable objective determination of drug binding kinetics at the cellular level in animal models.
• To overcome the limitations of existing bioassays in assessing drug target engagement.

Main Methods:

• Integration of high content screening (HCS) for multiplexed imaging of tissues.
• Application of U-Net based deep learning (DL) models for image analysis and quantification.
• Development of a workflow for direct visualization and quantification of biologic drug binding in targeted tissues.

Main Results:

• Successful development of a high-throughput method for quantifying drug distribution in tissues.
• Demonstrated ability to visualize and quantify drug binding with cellular resolution.
• Enabled objective assessment of drug binding kinetics in targeted organs of animal models.

Conclusions:

• The integrated HCS and DL approach provides a powerful tool for quantitative drug distribution analysis.
• This technology significantly advances the ability to confirm target engagement and understand drug binding kinetics.
• Facilitates more efficient and objective drug discovery and translation processes.