Bladder cancer is the second most common malignancy of the genitourinary system in European countries. Despite the use of the current multimodality approach to treatment of bladder cancer, the patients with the disease have a very high recurrence rate ; therefore, the early detection of new or recurrent bladder tumor lesions is crucial for successful treatment and favorable prognosis. Currently, diagnosis and monitoring of bladder cancer is based on cystoscopy, combining the additional information provided by urinary cytology. Despite their great clinical utility, these diagnostic method still have some limitation mainly related to their sensitivity. Accordingly, alternative methods are urgently required. An objective marker detection that is not based only on morphologic criteria should have been established. An ideal cancer marker should be objective, noninvasive, easy to detect and interpret, and posses high sensitivity and specificity.Objective: Investigate the predictive value of serum fibronectin, serum cytokeratin 20, urine fibronectin, urine fibronectin per creatinine ratio, in the diagnosis of bladder cancer patients, and study their correlation with grade and stage of the tumor. Subjects and methods: The study was conducted during the period from September 2012 till August 2013 in the department of chemistry and biochemistry/college of medicine/ Al-Nahrain University.Thirty five bladder cancer patients, 33 cases diagnosed as transitional cell carcinoma and 2 cases diagnosed as sequamous cell carcinoma, and 35 normal volunteers (control group) were enrolled throughout this study . They all attended from Al-kadhmia Teaching hospital and Gazi Al-Hariry for specialized surgery , they were diagnosed by specialist urologists.Serum fibronectin, serum cytokeratin 20 and urine fibronectin levels measured by commercial Enzyme Linked Immuno Sorbent Assay (ELISA) kits, urine creatinine level was determined by a spectrophotometric method. Results: The results of the study showed that the mean of serum Fibronectin level was significantly higher in the group of bladder cancer compared to control group (P< 0.001). The Receiver Operator Characteristic (ROC) curve demonstrated a significant discriminatory ability of increase serum FN level for the diagnosis of bladder cancer. When using a serum FN concentration of 25.65µg/ml as a cut-off value for the diagnosis of bladder cancer from control group, the sensitivity was 71.4%, specificity 100%, the positive predictive value was 100% and the negative predictive value 77.78%. The mean serum cytokeratin 20 level was significantly higher in bladder cancer compared with normal subjects (P<0.001). The Receiver Operator Characteristic (ROC) curve demonstrated a significant discriminatory ability of increased serum CK20 level for the diagnosis of bladder cancer. When using serum CK20 concentration of 20.15ng/ml as a cut-off value for the diagnosis of bladder cancer from control group, sensitivity was 71.4%, specificity 91.4%, the positive predictive value was 89.28% and the negative predictive value 76.19%. Urine fibronectin level was significantly higher in bladder cancer subjects compared with control group (P<0.001). The cut-off value was determined by the ROC curve which was 20.00ng/ml, the sensitivity and specificity were 94.3% and 97.1% espectively, the positive predictive value was 97.05% and the negative predictive value was 94.44% while accuracy was 95.71%. Urine fibronectin/Creatinine ratio was significantly higher in bladder cancer compared with control group (P< 0.001), the sensitivity and specificity were 74.3% and 88.6% respectively. Conclusion: Serum FN levels is a specific test for early diagnosis of bladder cancer with 100% specificity, 71.4% sensitivity, measurement of serum CK20 levels can be used in the differentiation of bladder cancer from normal subjects. Urine FN level is a better test for diagnosis of bladder cancer with an accuracy of 95.71%. Serum CK20 correlate positively and significantly with grade of tumor (P<0.05)