Evaluating the effectiveness of whole blood plasma versus protein precipitates in ovarian cancer detection through infrared spectroscopy

Early diagnosis of ovarian cancer remains challenging due to the absence of effective screening tests. The success of treatment and 5 year survival rates are significantly reliant on identifying the disease at a non-advanced stage, which highlights the urgent need for novel early detection and diagnostic approaches. Blood-based spectroscopic techniques, combined with chemometrics, have the potential to be used as tools for screening and diagnostic purposes in this context. In this study, we utilised attenuated total reflection Fourier-transform infrared (ATR-FTIR) spectroscopy to analyse blood plasma samples from benign (n = 15) and ovarian cancer (n = 15) cases. We conducted multivariate discrimination models to compare the results in terms of sensitivity, specificity, and diagnostic accuracy when using either plasmatic protein precipitates or whole plasma to distinguish between benign and ovarian cancer. Notably, diagnostic accuracy values of 96% (sensitivity and specificity of 96%) and 92% (sensitivity and specificity of 88% and 96%, respectively) were achieved for the protein precipitates and whole plasma datasets respectively using genetic algorithms with linear and quadratic discriminant analysis. Furthermore, this methodology demonstrated its capability to categorise samples within the ovarian cancer class, distinguishing between early stage (FIGO I) and advanced stage (FIGO II–III), with excellent accuracy exceeding 97% for protein precipitate dataset. These findings highlight the utilisation of a specific class of biomolecules in a proteomic-like approach based on infrared spectroscopy and chemometrics for detecting ovarian cancer using blood plasma samples.