Tissue Classification Using Optical Spectroscopy Accurately Differentiates Cancer and Chronic Pancreatitis

Area of Science:

• Biomedical optics
• Medical diagnostics
• Gastroenterology

Background:

• Current diagnostic methods for pancreatic cancer lack reliability in early detection and differentiation from chronic pancreatitis.
• Accurate differentiation is crucial for timely and effective treatment of pancreatic diseases.

Purpose of the Study:

• To evaluate the efficacy of multimodal optical spectroscopy in distinguishing between normal pancreatic tissue, chronic pancreatitis, and pancreatic adenocarcinoma.
• To develop and test a clinically compatible optical spectroscopy technology for pancreatic tissue analysis.

Main Methods:

• Utilized freshly excised human pancreatic tissue specimens (normal, chronic pancreatitis, adenocarcinoma).
• Employed multimodal optical spectroscopy to measure tissue reflectance and endogenous fluorescence.
• Applied a multinomial logistic model with principal component analysis and generalized estimating equations for tissue classification.

Main Results:

• Optical spectroscopy achieved high accuracy in differentiating pancreatic cancer from normal and chronic pancreatitis tissues (Sensitivity: 91%, Specificity: 82%, AUC: 0.89).
• Reflectance measurements alone provided comparable diagnostic accuracy to combined reflectance and fluorescence.
• Indicated potential for a simplified, reflectance-based optical device without compromising diagnostic performance.

Conclusions:

• Developed a novel, clinically compatible, label-free optical diagnostic technology for accurate pancreatic tissue characterization.
• Optical spectroscopy demonstrates significant potential to enhance the diagnostic capabilities for pancreatic cancer and chronic pancreatitis.
• These findings provide a foundation for improved diagnostic tools in pancreatic disease management.