BACKGROUND/AIM: Classical serum cancer biomarkers, such as carcinoembryonic antigen (CEA) and cancer antigen 19-9 (CA 19-9), remain important tools in colorectal cancer (CRC) management for disease follow up. However, their sensitivity and specificity are low for diagnostic and prognostic evaluation. The aim of this study was to evaluate the potential of biomarkers reflecting biological activity of tumors - tissue polypeptide specific antigen (TPS), cytokeratin fragment 19 (CYFRA 21-1), thymidine kinase (TK), insulin-like growth factor 1 (IGF-1) and insulin-like growth factor binding protein 3 (IGF-BP3) - together with the CEA and CA 19-9 in CRC diagnosis and prognosis. PATIENTS AND METHODS: This is a retrospective study including 148 CRC patients and 68 age-matched healthy subjects. Serum biomarkers were measured in pre-operative serum samples using immunoanalytical methods. The end-point for the diagnostic evaluation was the area under the receiving operating characteristic curve (AUC ROC) of the biomarkers. The end-point for the prognostic evaluation was overall survival. RESULTS: Serum levels of CEA, CA 19-9, TPS, and TK were significantly increased in CRC early-stage patients compared with healthy controls. Each of the studied biomarkers had AUC between 0.6 and 0.7. Analysis of survival demonstrated that the patients with CEA, CA 19-9, cytokeratin, and TK above optimal cut offs had significantly shorter survival. A multivariate analysis performed on all the study biomarkers resulted in the selection of CYFRA 21-1 as the best performing biomarker with hazard ratio 10.413. CONCLUSION: The combination of cytokeratins and thymidine kinase with classical cancer biomarkers enables the prediction of tumor aggressiveness and long-term prognosis.

Department of Surgery University Hospital Pilsen and Charles University Medical Faculty in Pilsen Pilsen Czech Republic

Division of Immunochemical Diagnostics University Hospital Pilsen and Charles University Medical Faculty in Pilsen Pilsen Czech Republic

References provided by Crossref.org