Application of protein markers in combination with ThinPrep bronchial brush cytology in classification of lung cancer subtypes

2008 
Objective To evaluate the value of application of cellular protein markers stained by immunoeytoehemistry in combination with ThinPrep bronchial brush cytology in classification of lung cancer subtypes. Methods Remaining bronchial brush cytology samples from 206 lung cancer patients with positive cytological diagnosis and 45 fine needle aspiration samples of resected lung carcinomas were collected. The expressions of CK10/13, CK7, CK18, CD56 and SYN in those samples were detected by immunoeytochemistry (ICC) using corresponding antibodies. Results The sensitivity and specificity of CK10/13 were 94.7% and 72.0%, respectively, in diagnosis of squamous cell carcinoma. The sensitivity and specificity of CK7 were 98.6% and 61.5%, and those of CK18 were 98.6% and 37.5%, respectively, in diagnosis of adenoeareinoma. The sensitivity and specificity of CD56 were 86.3% and 82.9%, and those of SYN were 81.6% and 93.5%, respectively, in diagnosis of small cell lung cancer. No significant difference was found in the expressions of CK10/13, CK7 and CK18 protein markers among differently differentiated lung squamous cell carcinomas and adenocarcinomas ( P 〉 0. 05 ). The classification rate of cytology in combination with ICC in differential diagnosis for 44 cases of unclassified lung cancer reached 90.0% for squamous cell carcinoma, 96. 3% for adenocarcinoma, and 100. 0% for small cell lung carcinoma. Conclusion Application of cellular protein markers in combination with ThinPrep bronchial brush cytology is helpful to improve the differential diagnosis of lung cancer subtypes, and may become a supplementary diagnostic method in subclassification of lung cancer. Key words: Lung neoplasms;  Subclassify;  Protein markers;  Bronchial brushing;  Cytology
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    7
    Citations
    NaN
    KQI
    []