Young Eun Jeon, Seok Cheol Lee, Seung Sam Paik, Kyeong Geun Lee,So Young Jin, Hyo Rim Kim, Chong Woo Yoo, Hye Min Park, Sang Yun Hani, Dong Ho Choi, Hark Kyun Kim
Pathology International 2011, 61 (8), 449-455 To date, protein profiles for hepatocellular carcinomas and cholangiocarcinomas have not been systematically evaluated and compared with each other in an unbiased way. Thirty-six hepatocellular carcinomas and adjacent normal tissue samples were analyzed using histology-directed, matrix-assisted laser desorption/ionization (MALDI) mass spectrometry (MS). Four cholangiocarcinomas and adjacent normal tissue samples were also evaluated. Tissue samples were sectioned at 10 mm, with 1-3 sections thaw-mounted on a conductive indium tin oxide-coated glass slide. Sinapinic acid was manually deposited on areas of each tissue section enriched by epithelial cells, either tumor or normal, and mass spectra were acquired using a MALDI-time of flight instrument. According to class prediction analysis, average prediction accuracy in test sets (composed of 18 hepatocellular carcinoma-normal pairs) ranged from 93.0 to 95.8%. Cholangiocarcinomas and hepatocellular carcinomas had different proteins in profiles, as evidenced by average prediction accuracy of >95% in the test set for all classifiers. Permutation P-values for 0.632 + bootstrap cross validated misclassification rates (at feature selection P < 0.001) were less than 0.05 for predicting p53 immunostaining status. We conclude that MALDI MS profiles may be useful in assisting with the diagnosis and the differential diagnosis of primary liver cancers.
Categories:
Tags:
No responses yet