Metabolic reprogramming-based characterization of circulating tumor cells in prostate cancer

BACKGROUND: Circulating tumor cells (CTCs), an advantageous target of liquid biopsy, is an important biomarker for the prognosis and monitoring of cancer. Currently, detection techniques for CTCs are mainly based on the physical and/or epithelial characteristics of tumor cells. However, biofunctional activity markers that can indicate the high metastatic capacity of CTCs are lacking. METHODS: Functional microarray, quantitative real-time polymerase chain reaction, and Western blot were used on five prostate cancer cell lines with different metastatic capacities to identify the metastasis-related metabolic genes. The identified genes were detected in the CTCs of 64 clinical samples using the RNA in situ hybridization. A multi-criteria weighted model was used to determine the optimal metabolic markers for the CTCs test. Based on five fluorescent signals targeting DAPI, CD45, metabolic, epithelial (EpCAM/CKs), and mesenchymal (Vimentin/Twist) markers, the filtration-enriched CTCs were classified as GM(+)CTCs/GM(−)CTCs (metabolic types) or E-CTCs/H-CTCs/M-CTCs (EMT types). Correlation analysis and ROC curve were conducted on 54 prostate cancer samples to evaluate the clinical significance of CTCs subtypes. RESULTS: Eight metastasis-related metabolic genes were identified, including HK2, PDP2, G6PD, PGK1, PHKA1, PYGL, PDK1, and PKM2. Among them, PGK1 and G6PD were determined as optimal glucose metabolic (GM) markers for CTCs. GM(+)CTCs (marked by PGK1/G6PD) were detectable in 64.8% (35/54) of prostate cancer patients, accounting for 46.5% (134/288) of total CTCs. An increased GM(+)CTCs level was associated with advanced tumor stage and metastasis (P < 0.05). In the discrimination of cancer metastasis from non-metastasis, GM(+)CTCs presented a higher AUC of the ROC curve (0.780) compared with the EMT CTCs subtypes (E-CTCs 0.729, H-CTCs 0.741, and M-CTCs 0.648). A triple tPSA–Gleason–GM(+)CTCs marker increased the AUC to 0.904, which was better than that of the tPSA–Gleason–H-CTCs marker (0.874). CONCLUSIONS: The metabolic marker (PGK1/G6PD) is determined as the indicator for the biofunctional activity analysis of CTCs, compared with the existing morphological (EMT) classification on CTCs. The metabolic characterization of CTCs demonstrates that hypermetabolic GM(+)CTCs are promising biomarkers for prostate cancer metastasis. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (10.1186/s13046-018-0789-0) contains supplementary material, which is available to authorized users.