Simultaneous comparison of multiple molecules using tissue array analysis in the thyroid neoplasm

Aim: Differential diagnosis, including the respective distinctions between benign and malignant tumors, follicular and papillary neoplasms, and follicular adenoma and follicular carcinoma, are always required in clinical practice, because therapeutic strategy largely depends on the diagnosis made. Methods: The present study describes a novel approach to obtaining clinically useful markers by means of the simultaneous comparison of multiple molecules using tissue array analysis. The markers examined in this study include galectin‐3, CD44v6, p53, HBME‐1, maspin and S100A4, which were reportedly useful for making these distinctions in association with metastasis and invasion. A total of 45 cases of thyroid tumors (seven adenomatous goiters, 16 follicular adenomas, 12 follicular carcinomas and 10 papillary carcinomas) were analyzed. Results: The results demonstrated the following suggestive phenotypes: galectin‐3^‐, HBME‐1^‐ and maspin⁺ as benign lesions, galection‐3^‐, HBME‐1⁺ and maspin^‐ as follicular carcinoma, and galectin‐3⁺, HBME‐1⁺ and maspin⁺ as papillary carcinoma. Conclusions: The expression of the molecules was assessed in each case and the expression profiles were compared. Useful multiple molecules were selected for each distinction and were correlated with each other. To understand the complex relationship, a logistic regression model was constructed. These results suggested that combined analysis of multiple molecules enhanced the differential diagnostic accuracy.