Background and aim: Thyroid gland carcinoma is a very prevalent neoplasia worldwide. The majority of malignant and benign lesions of the thyroid can be cytologically established. However, to distinguish between malignant and benign lesions histological demonstration is often required for a precise diagnosis. The aim of this study is to detect the ability of thyroid peroxidase (TPO) and to differentiate between thyroid malignant and non-malignant thyroid nodules. Methods: 50 formalin-fixed, paraffin embedded tissue blocks of different thyroid lesions were previously diagnosed and were sectioned at 3-4 μ thick and stained with hematoxylin and eosin (H&E) stain to confirm the histopathologic diagnosis, assess the subtypes of the lesions and assess the histopathologic grading and then were subjected to galectin-3 (Gal-3) and thyroid peroxidase (TPO) immunostaining. Results: In this retrospective histologic study, galectin-3 had a sensitivity of 93.3% for papillary thyroid carcinoma and 50% for follicular thyroid carcinoma. Thyroid peroxidase (TPO) had a sensitivity of 60% for papillary and 12.5% for follicular carcinoma. The combination of galectin-3 and TPO had a sensitivity of 100% for papillary and 50% for follicular carcinoma. Conclusion: This study confirms previous observations that galectin-3 alone is highly sensitive for papillary carcinoma but not adequately sensitive for follicular carcinoma. The combination of galectin-3 and TPO is complementary as a diagnostic and prognostic tool for patients with papillary carcinoma.