A study published in Biosensors and Bioelectronics presents an advanced antigen-independent method for detecting circulating tumor cells (CTCs) using a deep-learning-assisted biolaser platform. The approach, which combines single-cell laser technology and deep learning, achieves high sensitivity and specificity in identifying CTCs.
The study used blood samples from seven healthy donors, two pancreatic cancer patients, and six lung cancer patients.
Key findings:
Sensitivity and specificity. The Deep Cell-Laser Classifier (DCLC) achieved 94.3 percent sensitivity and 99.9 percent specificity in distinguishing CTCs from white blood cells (WBCs).
Zero-shot generalization. The DCLC identified CTCs from previously unseen pancreatic and lung cancer cell lines without retraining.
Clinical validation. Results from patient blood samples aligned with traditional immunofluorescence techniques.
In terms of workflow, sample preparation included depletion of red blood cells (RBCs) ahead of CTC enrichment through microfluidic devices. Remaining cells were then stained with nucleic acid dyes before single-cell laser emission analysis in Fabry-Pérot cavities. Unique lasing mode patterns were analyzed with the DCLC to distinguish CTCs from WBCs.
By eliminating reliance on specific biomarkers, this antigen-independent approach could help address CTC heterogeneity that can limit traditional methods. However, the study’s small sample size and focus on pancreatic and lung cancers point to the need for larger studies to confirm generalizability across additional cancer types.
Newsletters
Receive the latest pathology news, personalities, education, and career development – weekly to your inbox.
