The AOX testing of CCC is a novel method with excellent prognostic and predictive properties enabling the monitoring of therapies, e.g. effects of RP and endocrine therapy. We speculate that the continuing elevated expression of AOX with special emphasis on GPX1 acts as survival and defence mechanism in CCC required in an atypical environment prone to escape from immune surveillance.

What's known on the subject? and What does the study add?The role of circulating cancer cells in metastogenesis is generally accepted. Two forms of these cells have been reported in a number of studies, cancer cell clusters (CCCs) and individual epithelial cancer cells. Clusters appear at higher frequencies in the blood. CCCs have been reported to be rich in vimentin and poor in E‐cadherin expression The resulting epithelial to mesenchymal transition, a prerequisite for metastasis formation, occurs in CCCs. We have developed a new set of biomarkers, namely the antioxidant genes GPX1, SOD2 and TXNRD1, specific to cell trafficking in the blood.Firstly, the study shows that diagnosis of distant metastases is feasible by applying molecular phenotyping with a five gene test that has 94% sensitivity and 81% accuracy. Again SOD2 and GPX1 showed the highest sensitivities. Secondly, the study shows the efficacy of palliative chemotherapy in clearing the blood of CCCs overexpressing diagnostic genes. Clinically the overall lifespan ranged from 5 to 99 months under taxotere. We aimed to investigate the molecular reasons and found that MDR1 overexpression worsened survival by 31%. To the best of our knowledge, this is the first study to show the clinical impact of drug targeting and the counter‐effect of drug resistance in CCCs on overall survival. The findings may, therefore, add a novel tool for clinicians in tailoring therapies individually.OBJECTIVES To find the molecular phenotype in circulating cancer cells from patients with prostate cancer (PCa) in order to predict distant metastases. To determine genes affecting the study endpoints of overall survival and time to progression. PATIENTS AND METHODS Twenty‐five urologists in several clinics participated in the study, with 51 patients with metastatic and 77 with non‐metastatic PCa. Molecular analysis was carried out in two forms of circulating cancer cells, cancer cell clusters (CCCs) and individual epithelial cancer cells (CECs). Gene expression was studied using real‐time reverse‐transcriptase PCR. Cycle threshold values were normalized with glyceraldehyde 3‐phosphate dehydrogenase in cancer cells and mononucleated cells, yielding comparative specific expression values from the relative quantification method with the help of the standard curve method for each patient and each gene locus. RESULTS Preclinical validation was performed using aggregated and non‐aggregated SW480 cells showing the independence of CCCs and CECs. Prediction of metastases was achieved with five genes showing the highest sensitivity, SOD2, GPX1, AR, cyclin B and bFGF. The following results were obtained: 94% sensitivity, 65% specificity, 76% positive predictive value and 89% negative predictive value. The prevalence was 63%. Test accuracy was 81% with an odds ratio of 32 (P < 0.001). Overall survival was worsened by preceding chemotherapies when leaving insufficient GPX1 clearance in blood. Drug resistance genes were found to worsen the endpoints, among them MDR1 (P = 0.003; hazard ratio: 1.31; 95% CI: 1.09–1.58). CONCLUSIONS SOD2, GPX1 and AR represent a novel biomarker set for circulating cancer cells (clusters and scattered individual cells) in PCa. The clinical usefulness of these biomarkers ranges from the prediction of clinical tumours to disease prognostication, therapy monitoring and therapy outcome prediction (hormonal therapies, chemotherapies). The presence of CCCs and CECs after batch isolation allows the addition of genes for intensive studies, e.g. drug resistance.