Endocrine Abstracts

Introduction: Immunohistochemistry (ihc) in thyroid pathology is often recruited for differentiating between challenging benign and malignant tumors. There is no single antibody that can render definitive diagnosis. Most of the recently identified ihc markers have never been assimilated into daily clinical practice due to their low reproducibility and other associated limitations. Trop-2 ihc was recently reported as a useful tool for identifying papillary thyroid carcinoma (ptc) in fine-needle aspirate and tissue microarray samples. We aimed to validate trop-2 utility in the differential diagnosis of thyroid tumors on whole slide sections using automated digital image analysis.

Materials and methods: Trop-2 immunostaining was performed on a wide range of thyroid tumors (n=226) and controls (n=207).

Results: Normal thyroid, nodular goiter, hashimoto’s thyroiditis, graves’ disease, follicular adenoma, follicular carcinoma, and medullary carcinoma samples were negative for trop-2 ihc. The majority of ptc specimens (94/114, 82.5%) were positive for trop-2; however, the pattern of staining differed significantly between the histopathological variants. All papillary microcarcinomas (mptc), ptc classic variant (ptc cv), and tall cell variant (ptc tcv) were trop-2-positive with mainly diffuse staining. In contrast, less than half of ptc follicular variant specimens were positive for trop-2, with only focal immunoreactivity. Trop-2 could identify ptc cv with high diagnostic performance (sensitivity 98.1%, specificity 97.5%, area under curve 98%). Roc curve analysis found that >10% of trop-2-positive cells in a tumor support a diagnosis of ptc. Trop-2 h-score was significantly associated with ptc variant and presence of capsular invasion in encapsulated ptc fv (P<0.001). None of the baseline (age, gender) and clinical (tumor size, nodal disease, tnm stage) parameters were associated with trop-2 expression.

Conclusion: Trop-2 membranous staining is a very sensitive and specific marker for ptc cv, ptc tcv, and mptc – with high overall specificity for ptc.