Surface-enhanced Raman spectroscopy of serum accurately detects prostate cancer in patients with prostate-specific antigen levels of 4–10 ng/mL
Chen N,Rong M,Shao X,Zhang H,Liu S,Dong B,Xue W,Wang T,Li T,Pan J
IF: 7.033
2017-01-01
International Journal of Nanomedicine
Abstract:Na Chen,1 Ming Rong,1 Xiaoguang Shao,2 Heng Zhang,1 Shupeng Liu,1,3 Baijun Dong,2 Wei Xue,2 Tingyun Wang,1 Taihao Li,3 Jiahua Pan2 1Key Laboratory of Specialty Fiber Optics and Optical Access Networks, School of Communication and Information Engineering, Shanghai University, 2Department of Urology, Ren Ji Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, 3Beijing Advanced Innovation Center for Imaging Technology, Capital Normal University, Beijing, People’s Republic of China Abstract: The surface-enhanced Raman spectroscopy (SERS) of blood serum was investigated to differentiate between prostate cancer (PCa) and benign prostatic hyperplasia (BPH) in males with a prostate-specific antigen level of 4–10 ng/mL, so as to reduce unnecessary biopsies. A total of 240 SERS spectra from blood serum were acquired from 40 PCa subjects and 40 BPH subjects who had all received prostate biopsies and were given a pathological diagnosis. Multivariate statistical techniques, including principal component analysis (PCA) and linear discriminant analysis (LDA) diagnostic algorithms, were used to analyze the spectra data of serum from patients in control (CTR), PCa and BPH groups; results offered a sensitivity of 97.5%, a specificity of 100.0%, a precision of 100.0% and an accuracy of 99.2% for CTR; a sensitivity of 90.0%, a specificity of 97.5%, a precision of 94.7% and an accuracy of 98.3% for BPH; a sensitivity of 95.0%, a specificity of 93.8%, a precision of 88.4% and an accuracy of 94.2% for PCa. Similarly, this technique can significantly differentiate low- and high-risk PCa with an accuracy of 92.3%, a specificity of 95% and a sensitivity of 89.5%. The results suggest that analyzing blood serum using SERS combined with PCA–LDA diagnostic algorithms is a promising clinical tool for PCa diagnosis and assessment. Keywords: Ag nanoparticles, linear discriminant analysis, gray zone, principle component analysis, benign prostatic hyperplasia, spectral classification