The current standard in the diagnosis of thyroid lesions is by histologic examination of routine H&E stained sections. However, it is widely known that the interpretation of follicular patterned lesions can be quite difficult [1
]. A somewhat common dilemma is encountered with encapsulated tumors showing follicular growth pattern. Presence or absence of capsular and/or vascular invasion distinguishes benign from malignant follicular tumors, but identification of this finding can be challenging due to incomplete capsular penetration, equivocal vascular invasion or technical difficulties due to processing or sectioning artifacts. Another challenging situation is encountered when some but not all of the diagnostic nuclear features of papillary carcinoma are present. Recent study by Elsheikh et al [25
] and editorial review by J. Rosai [4
] pointed clearly to this issue.
Thyroid nodules are fairly common clinical findings affecting approximately 40% of the population between 30 and 60 years old in the U.S.A., and thyroid cancer is the most common endocrine malignancy [8
]. Fortunately, most of these nodules are benign tumors or hyperplastic lesions; however, it is important to identify these benign lesions for proper management and to realize maximum benefit for the patients [28
]. Accurate diagnosis then is very critical for post-operative management of patients with thyroid nodules, and incorrect interpretation can lead to significant psychological and social problems, and unnecessary increase in healthcare cost [15
]. Additionally, since FNA cytology in itself is not a reliable method to differentiate between benign and malignant follicular tumors or lesions, these patients usually undergo surgical resection, although only about 10% will actually have malignant tumors.
For all of the aforementioned reasons, investigators have focused during the last several years on finding molecular or IHC markers that can help in the distinction between benign and malignant lesions of the thyroid [1
]. Identifying markers that can separate hyperplastic/adenomatous nodules from follicular tumors can be of tremendous benefits to the patients and the healthcare system[23
]. As a result, many surgeries for benign lesions can be avoided and patients can be managed medically as needed [34
Galectin-3 is a beta-galactoside binding polypeptide with a 31 kDa molecular weight. It is a member of the lectin family, and seems to play a significant role in a number of biological processes. It has a role in regulating cell-cell and cell-matrix interaction, adhesion, migration and damaged cell repair. It also has a role in inflammation and neoplastic transformation. Normally, it is expressed in various tissue types and tumors and appears to have a role in the invasive and metastatic potential of various tumors [15
]. Kovacs et al found that IHC expression of galectin-3 may help in the differential diagnosis of solitary encapsulated follicular lesions, especially the minimally invasive follicular carcinoma [15
]. Several other investigators showed that galectin-3 is very useful in distinguishing benign from malignant tumors, especially PTC, with high sensitivity and specificity [10
]. Galectin-3 can aid in identifying FVPC, and distinguishing minimally invasive FC from FA. In our study, galectin-3 showed 85.2% sensitivity for immunoexpression distinction between carcinomas and benign nodules (positive in 27.5% of benign vs. 85.1% of malignant nodules). However, the specificity was lower at 72.4. In addition, we found that galectin-3 was somewhat more strongly and diffusely positive in PTC than in FC and FVPC, and expression in FA/HA was more focal and less intense than in malignant tumors. Also, galectin-3 expression was also detected in 8 of 52 benign non-neoplastic lesions (HN), but this was quite focal and weak. Other authors have also noted similar findings [3
]. The positivity of galectin-3 in HN may be explained by the fact that follicular cells normally contain endogenous biotin that can cause false positivity. We and others also found that false positivity can be seen in cystic and inflammatory lesions. Similarly, Kovacs et al also indicated that there might be some interpretation problems caused by the observation of focal positivity in inflammatory and cystic lesions. They postulated that expression of non-neoplastic follicular cells in inflamed areas may result from cytokines secreted by the inflammatory cells or simple permeation of galectin-3 abundantly shed by lymphocytes in the neighboring follicular cells. Therefore, they suggested that IHC staining has to be evaluated in conjunction with the histological features and use of biotin-free detection system.
Some authors consider true galectin-3-positive follicular adenoma as an indication of potentially early or incipient carcinoma, in which the capsular and/or vascular invasion can not be histologically observed yet [15
]. Also, some authors believe that galectin-3 immunoexpression in PTC may promote the release of tumor cells resulting in metastasis [15
HBME-1 is a monoclonal antibody directed against an antigen on the mesothelial cell membrane. Several studies have demonstrated its preferential reactivity in malignant thyroid tumors [3
]. It has been found to be reactive mostly in papillary thyroid carcinoma and some follicular carcinomas, but usually negative in follicular adenomas. Papotti et al in a study of well-differentiated thyroid tumors of uncertain malignant potential found that a diffuse and strong expression of HBME-1, and to a lesser extent galectin-3, is preferentially observed in the tumors with nuclear changes suggestive of papillary carcinoma [2
]. However, they concluded that the diagnosis of these tumors should also depend on previously defined morphologic criteria. In our study, HBME-1 was expressed in 47 of total 54 thyroid carcinomas with a diagnostic sensitivity of 87%. However, it was also expressed in 9/52 (17.3%) of benign non-neoplastic lesions and in 26/46 (56.5%) of adenomas. Thus our study shows that HBME-1 is not a very good marker to distinguish adenomas from thyroid carcinomas with over half of the adenomas expressing this marker.
Several cytokeratins have been evaluated for the differential diagnosis of thyroid nodules, of which CK19 has been found to be the most useful. Studies showed that CK19 is strongly and diffusely positive in malignant thyroid tumors including classic PTC, FVPC and FC [9
]. However, other studies showed variable results and yet others demonstrated that the CK19 expression is mostly focal and weak in FC, FA and benign hyperplastic nodules [3
]. Sahoo et al, reported that 25% of their follicular adenomas had extensive immunoreactivity for CK19, and Miettinen et al also reported that 59% of their follicular carcinomas showed CK19 reactivity in more than 10% of the lesions, suggesting that CK19 expression patterns are not reliable for the distinction between papillary carcinomas and follicular neoplasms [5
]. In our study, 46/54 (85.1%) of all the malignant tumors were positive for CK19 (diffusely and strongly), and 23/46 (50%) of the adenomas were also positive (but more focal and less intense). Our results showed a higher rate of CK19 reactivity in FAs than other studies, but the number of cases is relatively small for making strong conclusion. Also, we observed that positivity of CK19 in adenomas was more focal and weak than in carcinomas. In general, although most authors agree that CK19 reactivity is more frequent, diffuse and strong in papillary carcinoma, its reactivity in follicular neoplasms may limit its utility as a diagnostic marker [22
The Ret gene is located on chromosome 10 q and encodes a tyrosine kinase transmembrane receptor [12
]. It is typically absent in the normal thyroid follicular cells; however, gene rearrangement occurs in most PTCs. This oncogene is believed to be specific to PTC and encodes an oncoprotein product that contains the cytoplasmic portion of Ret gene. Therefore, some investigators believe that IHC expression of Ret oncoprotein is a reliable marker for PTC [12
]. Cheung et al showed immunoexpression of Ret in 78% of PTC, 63% of FVPC and 57% of Hurthle cell carcinoma, while all benign nodules were non-immunoreactive for Ret. In our study, Ret was positive in 18/20 cases of PTC, in 45/54 (83.3%) of all carcinomas, and in 30/98 (30.6%) benign lesions. Rossie et al found that Ret had focal or moderate immunoreactivity in benign lesions while it showed prevalent cytoplasmic expression in classic papillary cacrcinoma and its variants. The general conclusion among researchers is that diffuse immunoexpression represents a good supportive evidence for the diagnosis of papillary carcinomas; however, focal staining is often found in other lesions including benign nodules. Perhaps, a more important finding is that immunoreactivity of a panel that includes Ret, HBME-1 and CK19 is very specific for papillary carcinoma. Cheung et al concluded that HBME-1 positivity indicates malignancy, whereas diffuse CK19 and/or Ret positivity confirm papillary differentiation [12
Our study shows some different findings in staining reactions with other studies. We believe that some of the differences are due to various factors including antibodies used, dilution and antigen retrieval methods, type of tissue fixative used and time of fixation. Some authors have also investigated the method of IHC staining and found that galectin-3 reaction may be impacted by biotin-like activity produced by some thyroid lesions [14
]. They recommended performing galectin-3 IHC staining of thyroid lesions using biotin-free detection system. We believe there is a need to standardize fixation, antibody specifics, detection systems and IHC processes to be more able to compare results of various studies.
Many studies have evaluated the immunoexpression of a single marker such as galectin-3 or CK19 in the investigation of various thyroid lesions. In this study, we aimed at evaluating multiple markers to compare their sensitivity and usefulness, and to find out if a specific combination of the evaluated markers (galectin-3, HBME-1, CK-19 and Ret oncoprotein) can be of additional value in discriminating between benign and malignant thyroid lesions. We found that these markers combinations of galectin-3+HBME-1; galectin-3+CK19 or HBME-1+ CK19 do not improve the sensitivity or specificity for the distinction between benign and malignant thyroid lesions. Furthermore, using a combination panel of three markers (galectin-3+ HBME-1+ CK19) also did not increase the sensitivity or specificity for the distinction between benign and malignant thyroid lesions. On the other hand, other investigators have found that panels of various combinations of IHC markers can add to the value of diagnosing malignant thyroid tumors and discriminating them from benign tumors and non-neoplastic lesions [23
]. For example, De Matos et al found that a panel of galectin-3, HBME-1 and CK19 is useful in differentiating the follicular patterned lesions, and specifically distinguishing FVTC from FC or FA [35
]. Cheung et al recommended using a panel of CK19, HBME-1 and Ret as a useful means for diagnosing papillary carcinoma; whereas Rossie et al concluded that a panel of only HBME-1 and galectin-3 can correctly diagnose classic and variants of papillary carcinomas.
In summary, immunoexpression of galectin-3, CK19 and HBME-1 is an important supplementary test in the diagnosis of thyroid neoplasms, albeit it does not replace the conventional histomorphological examination. We found that these markers have somewhat similar sensitivity and specificity of immunoexpression in thyroid malignancy. We also found that combination panels of 2 or 3 of these markers (galectin-3+ HBME-1; galectin-3+ CK19; HBME-1+ CK19 or galectin-3+ HBME-1+ CK19) do not significantly improves the sensitivity or specificity of immunoexpression in malignant tumors. However, using a panel of two markers is advised to avoid instances of technical problems or processing issues. Therefore, we recommend using these panels as useful means to increases the likelihood of detecting malignant tumors. This practical low cost IHC test of commercially available markers can help to optimize the management of patients with thyroid nodules and reduce unnecessary surgical resection of benign nodules. Nonetheless, there are still questions to answer and additional studies are needed toward the quest of identifying useful markers for differentiating benign from malignant thyroid nodules.