Surface-enhanced Raman spectroscopy before radical prostatectomy predicts biochemical recurrence better than CAPRA-S.

Abstract

ObjectiveThe objective of this study was to evaluate the performance of surface-enhanced Raman spectroscopy (SERS) in the prediction of early biochemical recurrence after radical prostatectomy (RP).Patients and methodsWe synthesized monodisperse gold nanoparticles as SERS-enhanced substrates and analyzed preoperative plasma samples of patients who underwent RP. The roles of clinical risk model (Cancer of the Prostate Risk Assessment [CAPRA] score) and distinctive SERS spectra on prediction of early biochemical recurrence were evaluated. The principal component analysis and linear discriminant analysis (PCA-LDA) were used to manage the spectral data and develop diagnostic algorithm.ResultsA total of 306 preoperative plasma Raman spectra from 102 patients were collected. SERS spectrum from those who developed early biochemical recurrence were compared to those who remained biochemical recurrence-free. The SERS detected more abundant circulating free nucleic acid bases in biochemical recurrence population, presenting significant stronger intensities at SERS spectral bands 725 and 1,328 cm−1. The addition of Raman spectral peak 1,328 cm−1 to CAPRA postsurgical (CAPRA-S) score significantly improved the predictive power of logistic regression model compared to simple CAPRA score (P<0.001). Meanwhile, the leave-one-out cross-validation method was used to validate the PCA-LDA model and revealed the sensitivity, specificity, and accuracy of 65.8%, 87.5%, and 79.4%, respectively. The receiver operating characteristic (ROC) curve was used to evaluate the performance of different models. Area under the ROC curve of the CAPRA-S score model alone was 0.77, however, when combined with Raman spectral peak 1,328 cm−1, it improved to 0.81.ConclusionOur primary results suggested that SERS could be a meaningful technique for prediction of early biochemical recurrence in prostate cancer.