Prediction of the Probability of Malignancy in Mucinous Cystic Neoplasm of the Pancreas With Ovarian-Type Stroma: A Nationwide Multicenter Study in Japan.

Abstract

The aim of the study was to develop a formula for predicting the probability of malignancy of mucinous cystic neoplasm (MCN) of the pancreas with ovarian-type stroma. A total of 364 patients were enrolled. A total score was calculated as the sum of the approximate integers of the odds ratios of the predictive factors identified by multivariate analysis. The relationship between the total score and pathological results was assessed. A total of 321 patients had benign MCN and 43 had malignant MCN. Five possible predictive factors were analyzed: 56 years or older, high serum carcinoembryonic antigen level, high carbohydrate antigen 19-9 level, tumor size of 51 mm or greater, and the presence of mural nodules. The total score was significantly higher in patients with malignant MCN (median, 24; range, 0-37) compared with benign MCN (median, 5; range, 0-33; P < 0.001). Receiver operating characteristic curve analysis demonstrated that the area under the curve was 0.86, and the sensitivity and specificity of the total score for discriminating malignant MCNs were 72% and 83%, respectively, using a cut-off value of 22. The current simple formula can predict the malignancy of MCN and may thus contribute to the adequate management of patients with MCN.