Computational modeling and optimization of peptide-based CAR-T cell receptors targeting CD19 for enhanced efficacy and minimized toxicity

Chimeric Antigen Receptor T-cell (CAR-T) therapy has revolutionized the treatment of B-cell malignancies, with CD19 being a primary target due to its stable expression in lymphomas. However, current CAR-T therapies face challenges related to antigen escape, treatment resistance, and toxicity. In this study, we employed a computational approach to design and optimize peptide-based CAR-T cell receptors with improved specificity and reduced toxicity. We utilized in silico techniques, including PSI-BLAST sequence validation, molecular docking, machine learning-based toxicity prediction, and molecular dynamics simulations, to refine CAR-T receptor design. Our structural modeling and docking studies identified an optimized single-chain variable fragment (scFv) antibody (H8_L1) that demonstrated high binding affinity and stability with both wild-type and mutated CD19 variants. Toxicity assessments confirmed minimal off-target effects, ensuring safety in therapeutic applications. Additionally, computational mutation docking studies revealed that the optimized receptor maintained stable interactions despite antigenic variations, addressing a critical limitation of current CAR-T therapies. These findings provide a robust framework for designing next-generation CAR-T therapies with enhanced efficacy, reduced toxicity, and resilience against antigenic drift, paving the way for further experimental validation and clinical applications.