Predicting Mechanosensitive T Cell Expansion from Cell Spreading.

Area of Science:

• Biotechnology
• Immunology
• Cellular Engineering

Background:

• T cell performance variability challenges adoptive cellular immunotherapy (ACT) efficacy.
• Failure in T cell expansion can occur due to individual differences and disease states.
• Modulating substrate stiffness can improve T cell expansion, but optimal stiffness varies individually.

Purpose of the Study:

• To develop a predictive model for long-term T cell expansion based on short-term assays.
• To assess the utility of cell spreading as a predictor of mechanosensitive T cell expansion.
• To differentiate between healthy and Chronic Lymphocytic Leukemia (CLL) T cells using cell morphology.

Main Methods:

• Measuring short-term T cell spreading on substrates of varying mechanical stiffness.
• Utilizing a deep learning (DL) model for classification and prediction tasks.
• Correlating short-term cell spreading with long-term T cell expansion potential.

Main Results:

• Short-term cell spreading effectively predicts subsequent, mechanosensitive T cell expansion.
• A DL model accurately classified T cells from healthy donors versus CLL patients.
• The system successfully predicted long-term T cell expansion based on cell source and substrate stiffness.

Conclusions:

• Short-term cell spreading serves as a reliable predictor of long-term T cell function in ACT.
• Deep learning-based analysis of cell spreading can improve T cell production reliability.
• This approach enhances the efficacy of immunotherapy by enabling prediction from small diagnostic samples.