Distant metastases mainly influence the prognosis of patients with prostate cancer (PC), however, development of novel biomarkers for predicting metastatic PC and understanding of the molecular mechanisms remain essential. The objective of this study was to investigate the metabolic differences in the primary tumor tissues between localized PC and metastatic PC using Raman micro-spectroscopy and metabolomics analysis, and then explore potential biomarkers for predicting metastasis and the potential metabolic pathways during the progression from localized prostate cancer to metastasis. We used confocal Raman microscopy (CRM) and liquid chromatography-mass spectrometry (LC–MS) based metabolomics to analyze the primary prostate tumor tissues of localized PC and metastatic PC. Subsequently, we used a convolutional neural network (CNN) structure to develop a classification model to predict metastatic PC based on the tissue Raman spectra, and then explored potential metabolic pathways via Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis. We collected a total of 547 spontaneous Raman spectra and 2D Raman images of primary prostate tumor from 21 localized PC and 21 metastatic PC. Compared with localized PC, the Raman peak associated with unsaturated fatty acids was significantly higher in metastatic PC, while the Raman peaks associated with amino acids and proteins were significantly lower. Subsequently, we used a CNN structure to develop a classification model to predict metastatic PC based on the tissue Raman spectra and the model showed a testing accuracy of 81.3 ± 3.7%. The LC–MS based metabolomics results of tissues validated the CRM findings that the primary prostate tumor tissue of metastatic PC exhibited a similar changing trend in prenol lipids, linolenic acid, and multiple classes of amino acids. The CRM could be a potential tool for predicting metastases by analyzing prostate biopsy tissues at the time of diagnosis. Our study found that metabolic remodelling of primary tumors occurred during the metastasis process. The metabolic alterations in primary tumor tissue can help elucidate the underlying mechanisms of the metastasis process, leading to the development of new therapies.

Surface-enhanced Raman spectroscopy was employed to measure lettuce seeds at five different germination stages. The experimental results show that 85% of the spectra of normally germinating seeds exhibited consistency in both peak positions and quantities at the same germination stage, while spectra from seeds that failed to germinate normally demonstrated significant differences compared to the normal ones. These data indicate that the surface-enhanced Raman spectrum of seeds could serve as an effective method for detecting seed germination rates.

In the originally published version of this article