Analysis of prostate cancer using serum surface-enhanced Raman spectroscopy and multivariate statistical algorithm

Abstract

Prostate cancer (PCa) is an epithelial malignant tumor occurring in the prostate gland and is the most common malignancy of the male genitourinary system. Prostate cancer is often asymptomatic in its early stage, and the best treatment time is usually missed when it is found. Therefore, early diagnosis and treatment is the key to reduce the mortality of prostate cancer patients. In this work, we developed a method to identify patients with prostate cancer and healthy volunteers. We collected serum samples from patients with prostate cancer and healthy volunteers, detected the SERS spectra of these serum samples, and analyzed the SERS spectrum of the obtained serum samples with a preliminary spectral peak assignment. Then principal component analysis (PCA) combined with linear discriminant analysis (LDA) was used to diagnose the SERS spectra of serum from patients with prostate cancer and healthy volunteers. Difference spectrum analysis showed that there are obvious differences in several characteristic peaks between the serum of patients with prostate cancer and the serum of healthy volunteers, which may be related to the special changes of nucleic acids, proteins, lipids and other biological molecules in the process of carcinogenesis. Using principal component analysis (PCA) combined with linear discriminant analysis (LDA) multivariate statistical method, the diagnostic sensitivity and specificity were 85% and 95%, respectively. The receiver operating characteristic (ROC) curve further proves the effectiveness of diagnosis algorithm based on PCA-LDA technology, and the area under the curve (AUC) is 0.998. These results show that the combination of SERS technology and PCA-LDA algorithm has great potential in screening prostate cancer patients.