Machine Learning-Enhanced Analysis of Exosomal Surface Sialic Acid Using Surface-Enhanced Raman Spectroscopy for...

Area of Science:

• Biomedical Engineering
• Oncology
• Nanotechnology

Background:

• The lack of ovarian cancer (OC)-specific biomarkers hinders precise noninvasive diagnostic and monitoring strategies.
• Exosomal surface sialic acid (SA) is a potential biomarker, but its role in OC and detection methods require further investigation.
• Current exosome isolation and detection techniques have limited clinical applicability.

Purpose of the Study:

• To develop a sensitive and specific method for detecting exosomal SA in OC patients.
• To evaluate the potential of exosomal SA as a biomarker for OC diagnosis and treatment monitoring.
• To integrate nanotechnology and machine learning for improved OC management.

Main Methods:

• Development of a CD63 aptamer-functionalized gold array chip for exosome isolation.
• Integration of a surface-enhanced Raman scattering (SERS) nanosensor for sensitive SA detection.
• Application of machine learning algorithms to analyze SERS spectra for OC diagnosis.

Main Results:

• The developed chip efficiently isolated exosomes from clinical serum samples.
• The SERS nanosensor selectively detected exosomal SA, enabling sensitive analysis.
• Machine learning achieved 93% accuracy in diagnosing OC based on SA signatures.
• Longitudinal monitoring of SA correlated with treatment response, as indicated by clinical markers like CA125 and HE4.

Conclusions:

• Exosomal SA, detected via an aptamer-functionalized SERS chip and analyzed by machine learning, shows significant potential for OC diagnosis.
• This approach offers a powerful tool for sensitive, noninvasive diagnosis and longitudinal monitoring of OC treatment response.
• The findings support the role of exosomal SA in precision medicine for ovarian cancer management.