Background: The value of serum tissue polypeptide specific antigen (TPS) as a complement to CA 19-9 in the detection of pancreatic carcinoma was determined prospectively. TPS and CA 19-9 levels obtained at the time of diagnosis in patients suspected of having chronic pancreatitis or pancreatic carcinoma were evaluated in receiver operating characteristic (ROC) curve analysis.
Methods: Serum TPS and CA 19-9 levels were measured by immunoassays in 122 subjects, 48 with pancreatic carcinoma and 74 with chronic pancreatitis.
Results: Elevated levels of CA 19-9 were detected preoperatively in 70% of pancreatic carcinoma patients and in 19% of chronic pancreatitis patients. Elevated levels of TPS were detected in 100% of patients with pancreatic carcinoma and in 22% of patients with chronic pancreatitis. The median levels of TPS and CA 19-9 for pancreatic carcinoma were significantly higher than those for chronic pancreatitis (P < 0.0001). Increasing the upper reference value of TPS allowed for better discrimination between chronic pancreatitis and pancreatic carcinoma. ROC curve analysis showed that the introduction of 200 U/L as a decision criterion for TPS did not reduce its sensitivity but significantly improved its specificity. At a specificity of 98% for TPS, discrimination between pancreatic carcinoma and chronic pancreatitis was found to be 97%. Increasing the upper reference level for CA 19-9 to attain a specificity of 98% decreased its sensitivity from 70% to 33%.
Conclusions: At an elevated cut-off level for TPS (200 U/L), almost complete discrimination between pancreatic carcinoma and chronic pancreatitis was obtained. TPS will be more useful than CA 19-9 in the differential diagnosis of pancreatic carcinoma and chronic pancreatitis.
Copyright 2000 American Cancer Society.