46 chapter 2 accurately differentiated follicular adenomas from FTC with 92% sensitivity and 94% specificity if a cytoplasmic stain in ≥10% of the cells was considered positive [212]. The prospective multicentre clinical validation study by Bartolazzi et al. demonstrated 78% sensitivity and 93% specificity in Thy3 nodules if a cytoplasmic galectin-3 stain in >5% of the cells was considered positive. Nineteen of the 22 false-positive nodules were follicular adenoma. However, a group of 33 difficult-to-diagnose (follicular) tumour of unknown malignant potential lesions was disregarded, 22 of which were galectin-3 negative. If these neoplasms were considered malignant, sensitivity dropped to 69% [202]. HBME-1 HBME-1 is a monoclonal antibody targeting an unknown antigen on the microvilli of mesothelial cells. It is usually negative in normal thyroid follicular cells. Abnormal expression of HBME-1 shows cytoplasmic location with membrane accentuation. It is associated with, but does not necessarily indicate PTC [203, 207, 216, 217]. Its low detection limit enables assessment in liquid-based cytology [218]. Reported sensitivity and specificity of HBME-1 in indeterminate nodules ranged from 61% to 100% and from 75% to 96%, respectively [212, 214, 218, 219]. Approximately two out of five nodules showed positive staining. If only non-oncocytic follicular neoplasms were selected, Saggiorato et al. demonstrated that HBME-1 had excellent 93% sensitivity and 98% specificity in indeterminate thyroid nodules [212]. Cytokeratin 19 Cytokeratin 19 (CK-19) is a type I keratin. It belongs to the group of intermediate filament proteins, which arrange the cell cytoskeleton and structural integrity. CK-19 is widely present in epithelial cells, but also found in basal cells layers of stratified epithelium [212, 220]. Strong and diffuse abnormal expression of CK-19 indicates PTC, including FVPTC. Expression in FTC is less intense and more variable, warranting nuanced interpretation of CK-19 staining intensity. CK-19 usually shows no or only focal expression in follicular neoplasms, hyperplastic nodules and adenomatous goitre [212, 216, 220, 221]. The reported sensitivities and specificities for CK-19 staining in indeterminate cytology ranged from 76% to 88% and 80% to 100%, respectively [212, 218, 220]. Lacoste-Collin et al. demonstrated the importance of an accurate threshold. CK-19 staining in 31 Bethesda IV nodules accurately diagnosed five out of six malignancies, including a PTC, two FTC and two out of three FVPTC. At a threshold of ≥30% stained cells, 5 of 25 benign lesions tested false-positive; at a more sensitive threshold of ≥10% stained cells, 12 of 25 tested false positive [218]. Other immunocytochemistry markers Immunohistochemistry studies identified more potential ICC markers. Some, like CD44v6, have not yet been investigated in indeterminate cytology [155]. Other markers were sporadically investigated in preclinical studies, including Ki-67, TROP-2, emerin, keratan sulphate, thyroperoxidase, CD57 and GLUT-1.
RkJQdWJsaXNoZXIy MTk4NDMw