Galectin-3 does not reliably distinguish benign from malignant thyroid neoplasms

Aims: To determine whether galectin-3 is a sensitive indicator of thyroid malignancy. It has been suggested as a potential marker for differentiating thyroid carcinoma from benign or non-neoplastic lesions in preoperative fine-needle aspirates (FNAs). Methods: Galectin-3 protein expression was assessed by immunohistochemistry in formalin-fixed thyroid tissues from 124 patients with histological diagnoses of papillary carcinoma (n = 38), follicular carcinoma (n = 19), follicular adenoma (n = 32) and dominant nodules of multinodular goitre (n = 35). Expression of galectin-3 was also assessed by Western blotting in 24 fresh thyroid tissues. Results: Galectin-3 expression was observed in the majority of carcinomas (papillary 92%; follicular 74%). However, a large proportion of follicular adenomas (72%) and multinodular goitres (57%) also expressed galectin-3. In addition, galectin-3 expression was observed in epithelial cells of normal thyroid tissue and Hashimoto's thyroiditis. Galectin-3 immunopositivity was significantly greater in papillary carcinomas than in dominant nodules or follicular adenomas (P < 0.0001, P = 0.0005, respectively). However, galectin-3 expression was no greater in follicular carcinomas than in follicular adenomas (P = 0.8735). Western blotting analysis confirmed both the specificity of the antiserum and expression of galectin-3 in multinodular goitres, follicular adenomas/carcinomas and papillary carcinomas. Conclusion: The data demonstrate that galectin-3 is not a reliable immunohistochemical marker to distinguish benign from malignant thyroid follicular lesions.