Identification of clinically relevant T cell receptors for personalized T cell therapy using combinatorial algorithms

Area of Science:

• Immunology
• Bioinformatics
• Cancer Research

Background:

• Personalized cancer immunotherapy relies on identifying T cell receptors (TCRs) that target tumors.
• Distinguishing tumor-reactive T cells from bystander cells is a significant challenge.

Purpose of the Study:

• To develop and validate an antigen-agnostic computational tool, TRTpred, for predicting tumor-reactive T cell receptors (TCRs).
• To integrate TRTpred with an avidity predictor to create a combinatorial algorithm for selecting clinically relevant TCRs for personalized T cell therapy.

Main Methods:

• TRTpred was built by leveraging the distinct transcriptomic profiles of tumor-reactive versus bystander T cells.
• The performance of TRTpred was benchmarked in silico.
• TRTpred was integrated with an avidity predictor to form a combinatorial algorithm.

Main Results:

• TRTpred demonstrated effectiveness as an antigen-agnostic predictor of tumor-reactive TCRs.
• The combinatorial algorithm, integrating TRTpred and an avidity predictor, was benchmarked in patient-derived xenografts.

Conclusions:

• TRTpred represents a novel computational approach to identify tumor-reactive TCRs.
• The developed combinatorial algorithm shows promise for selecting TCRs for personalized cancer immunotherapy and requires further clinical validation.