Antigen-independent single-cell circulating tumor cell detection using deep-learning-assisted biolasers.
Journal:
Biosensors & bioelectronics
PMID:
39615221
Abstract
Circulating tumor cells (CTCs) in the bloodstream are important biomarkers for clinical prognosis of cancers. Current CTC identification methods are based on immuno-labeling, which depends on the differential expression of specific antigens between the cancer cells and white blood cells. Here we present an antigen-independent CTC detection method utilizing a deep-learning-assisted single-cell biolaser. Single-cell lasers were measured from nucleic-acid-stained cells inside optical cavities. A Deep Cell-Laser Classifier (DCLC) was developed to detect tumor cells from a patient CTC-derived pancreatic cell line using their unique single-cell lasing mode patterns. We further showed that the knowledge learned from one type of pancreatic cancer cell line can be transferred to detect other pancreatic cancer cell lines by the DCLC in zero-shot. A sensitivity of 94.3% and a specificity of 99.9% were achieved. Finally, enumeration was performed on CTCs obtained from pancreatic cancer patients. We further demonstrated the DCLC's ability in zero-shot generalization of enumeration on lung cancer patients' CTCs. The counting trends were consistent with those observed using conventional immunofluorescence imaging techniques. Employing our DCLC model, single-cell lasers open new avenues for both future biological studies and clinical applications, including classification of cell types and identification of rare cells.