The use of adjusted preoperative CA 19-9 to predict the recurrence of resectable pancreatic cancer

Background. Despite the usefulness of CA 19-9 in the diagnosis and prognosis of pancreatic cancer, cholestasis can falsely elevate CA 19-9 levels, which contributes to limited clinical utility in patients with biliary obstruction. This study was designed to evaluate the usefulness of adjusted preoperative CA 19-9 levels in predicting a prognosis of pancreatic cancer. Methods. The available medical records of patients with resected pancreatic cancer from January 1990 to June 2005 were retrospectively viewed at Yonsei Medical Center, Seoul, Korea. The adjusted CA 19-9 value was obtained by dividing the serum CA 19-9 level by the values of serum bilirubin in case of bilirubin 2 mg/dL. Disease-free survival was evaluated according to the adjusted preoperative CA 19-9 value. Results. Sixty-one patients were investigated. Their adjusted preoperative CA 19-9 values were significantly different from the actual baseline CA 19-9 value (129.4 +/- 225.2 U/mL, versus 442.1 +/- 645.5 U/mL, P < 0.0001). On univariate analysis, peripancreatic microscopic invasion (P = 0.0142), lymphovascular invasion (P = 0.0038), and adjusted preoperative CA 19-9 >= 50 U/mL (P = 0.0049) were predictive factors for cancer recurrence after curative resection. Adjusted preoperative CA 19-9 >= 50 U/mL (Exp, (B) = 2.097, P = 0.027) was an independent predictive factor in multivariate analysis. Conclusions. The adjusted preoperative CA 19-9 value can predict the risk of recurrence after curative resection of pancreatic cancer. Interpreting the preoperative CA 19-9 value adjusted to the serum biliru- bin values seems to be more reasonable in evaluating prognosis of pancreatic cancer. (D) 2007 Elsevier Inc. All rights reserved.