Abstract
BACKGROUND: Endometrial cancer (EC) is the 6th most common cancer among women worldwide. No effective non-invasive screening methods or approved blood biomarkers for EC exist. Previous research explored Attenuated Total Reflection-Fourier Transform Infrared (ATR-FtIR) and Raman spectroscopies, using dried blood plasma. Fresh, 'wet', blood samples, that might provide faster results, have not been investigated. This study compared ATR-FtIR and Raman spectroscopies on 'wet' and dry blood plasma samples for EC detection. It also conducted a preliminary exploration into their diagnostic potential for EC in high-risk individuals with polycystic ovary syndrome (PCOS). METHODS: 'Wet' and dry blood plasma samples from participants with EC, PCOS and healthy controls were analysed using ATR-FtIR and Raman spectroscopies. Machine learning algorithms and multivariate statistical analyses assessed spectral variance across datasets to evaluate the techniques' diagnostic performance. RESULTS: Raman analysis of 'wet' plasma achieved 82% accuracy in detecting EC, while ATR-FtIR spectroscopy reached 78%. When combined, diagnostic accuracy reached 86%. In comparison, dry plasma analysis with ATR-FtIR detected EC with 83% accuracy. Spectral similarities were found between EC and PCOS. CONCLUSIONS: Our study suggests that ATR-FtIR and Raman spectroscopies could revolutionise early diagnosis of EC. More research is required to validate these promising findings.