Abstract OBJECTIVES: The diagnosis of cholangiocarcinoma is often difficult, making management approaches problematic. A reliable serum tumor marker for cholangiocarcinoma would be a useful additional diagnostic test. Previous studies have demonstrated that elevated serum concentrations of CA 19-9, a tumor-associated antigen, have good sensitivity and specificity for cholangiocarcinoma in patients with primary sclerosing cholangitis. However, the value of this tumor marker for cholangiocarcinoma unassociated with primary sclerosing cholangitis is unclear. Thus, the aims of this study were to determine the usefulness of a serum CA 19-9 determination in the diagnosis of de novo cholangiocarcinoma. METHODS: We prospectively measured serum CA 19-9 concentrations in patients with cholangiocarcinoma (n = 36), nonmalignant liver disease (n = 41), and benign bile duct strictures (n = 26). Serum CA 19-9 concentrations were measured by an immunoradiometric assay (CIS Bio International) without knowledge of the clinical diagnosis. RESULTS: The sensitivity of a CA 19-9 value >100 U/ml in diagnosing cholangiocarcinoma was 53%. When compared with the nonmalignant liver disease and the benign bile duct stricture groups, the true negative rates were 76% and 92%, respectively. Patients with unresectable cholangiocarcinoma had significantly greater mean CA 19-9 concentrations compared to patients with resectable cholangiocarcinoma. CONCLUSIONS: These data suggest that the serum CA 19-9 determination is a useful addition to the available tests for the differential diagnosis of cholangiocarcinoma.