CIAdex: Single-Cell FTIR Spectral Fingerprinting for Cell Identity Verification and Aging Quantification in Therapeutic Cell Manufacturing

Abstract

Ensuring the identity and optimal aging state of cell products is critical for the efficacy and safety of cell therapies. Despite rapid iterations, there remains an urgent need for robust and easy-to-implement tests to characterize cell products. Here, we present CIAdex (Cell Identification and Aging Index), an analytical framework that utilizes single-cell Fourier-transform infrared (FTIR) spectral fingerprints and machine learning to achieve precise label-free cell identity assessment and aging quantification. CIAdex employs a Feature Extraction Processor to automatically extract FTIR spectral variables corresponding to distinct cellular biomolecular features, enabling reliable distinction of lineage-, donor-, and batch-specific cell populations using linear discriminant analysis. Through application of the XGBoost algorithm, a quantitative aging index (trPDL/trPN) was further generated for tracking cellular aging dynamics along culture expansion. Notably, trPDL/trPN quantitatively represent subtle age-related shifts among different batches and drug-induced senescence or rejuvenation effects which are unmeasurable by existing methods. Together, our work demonstrates that CIAdex, by simultaneous label-free identity verification and aging quantification of cell populations, offers a transformative approach to interpret single-cell FTIR spectral fingerprints and provides novel metrics for quality control in cell manufacturing with significant potential for optimization and assurance of cell therapies safety and efficacy.