Point-of-Care Disease Screening in Primary Care Using Saliva: A Biospectroscopy Approach for Lung Cancer and Prostate...

Area of Science:

• Biomedical diagnostics
• Spectroscopy
• Cancer screening

Background:

• Saliva is an accessible, non-invasive biofluid containing disease-associated biomarkers.
• Current diagnostic methods for cancers like lung and prostate can be invasive or costly.
• Developing rapid, point-of-care screening tools is crucial for early disease detection.

Purpose of the Study:

• To develop and validate a non-invasive saliva-based diagnostic test for lung and prostate cancers.
• To assess the efficacy of attenuated total reflection Fourier-transform infrared (ATR-FTIR) spectroscopy combined with machine learning for cancer detection in saliva.
• To evaluate the potential of this method for large-scale screening in primary care settings.

Main Methods:

• Collected 998 saliva samples from patients in a lung cancer screening program.
• Analyzed saliva samples using ATR-FTIR spectroscopy.
• Employed principal component analysis (PCA) and quadratic discriminant analysis (QDA) for spectral data classification, with data split into training and testing sets.

Main Results:

• The PCA-QDA model achieved 91% accuracy, 100% sensitivity, and 91% specificity for lung cancer detection in the test set.
• For prostate cancer detection in the male cohort, the PCA-QDA model achieved 93% accuracy, 100% sensitivity, and 92% specificity.
• The developed method demonstrated high performance in distinguishing cancer-positive samples from others.

Conclusions:

• A novel saliva-based ATR-FTIR spectroscopy test shows significant potential for non-invasive lung and prostate cancer screening.
• The PCA-QDA algorithm provides a robust classification method for spectral data.
• This approach could enable cost-effective, large-scale cancer screening in primary care settings.