Indian Journal of Pathology and Microbiology

ORIGINAL ARTICLE
Year
: 2017  |  Volume : 60  |  Issue : 2  |  Page : 161--166

The absence of CD56 expression can differentiate papillary thyroid carcinoma from other thyroid lesions


Ioana Golu1, Mihaela Maria Vlad1, Alis Dema2, Lavinia Cristina Moleriu3, Anca Tudor3, Mihaela Iacob4, Oana Popa5, Marioara Cornianu2,  
1 Department of Endocrinology, “Victor Babes” University of Medicine and Pharmacy, Timisoara, Romania
2 Department of Pathology, “Victor Babes” University of Medicine and Pharmacy, Timisoara, Romania
3 Department of Medical Informatics, “Victor Babes” University of Medicine and Pharmacy, Timisoara, Romania
4 Department of Pathology, “Pius Brinzeu” County Emergency Hospital, Timisoara, Romania
5 Department of Endocrinology, “Pius Brinzeu” County Emergency Hospital, Timisoara, Romania

Correspondence Address:
Mihaela Maria Vlad
Piata Eftimie Murgu 2, Timisoara
Romania

Abstract

Context: The neural cell adhesion molecule CD56 is an antigen important for the differentiation of the follicular epithelium. Recent studies have reported low or absent expression of CD56 in papillary thyroid carcinoma (PTC) and its presence in normal thyroid tissue, benign thyroid lesions, and most follicular non-PTC tumors. Aim: We wish to estimate the value of CD56 in the differentiation of PTC (including follicular variant-PTC [FV-PTC]) from other nontumoral lesions and follicular thyroid neoplasias. Settings and Design: This was a retrospective, case–control study. Subjects and Methods: We analyzed the expression of CD56 in normal thyroid follicular tissue, 15 nonneoplastic thyroid lesions (nodular hyperplasia, Graves' disease, and chronic lymphocytic thyroiditis/Hashimoto), and 38 thyroid follicular cell neoplasms (25 cases of PTC). The immunohistochemical reactions were performed on sections stained with anti-CD56 antibody. Statistical Analysis Used: We used the Chi-square test, values of P< 0.05 being considered statistically significant. Risk analysis was applied on these studied groups, by calculating the odds ratio (OR) value. Results: Our results indicated that CD56 immunoexpression had differentiated PTC from benign nonneoplastic lesions (P = 0.002), as well as from follicular neoplasias (P = 0.046). There were no significant differences regarding CD56 expression between FV-PTC and classical PTC (P = 0.436). The immunoexpression of CD56 has differentiated PTC from other thyroid non-PTC lesions (P < 0.001), with 26.4 OR value. Conclusions: CD56 has been proved to be a useful marker in the diagnosis of PTC, including FV-PTC. Its absence can help differentiate FV-PTC from other thyroid nodules with follicular patterns.



How to cite this article:
Golu I, Vlad MM, Dema A, Moleriu LC, Tudor A, Iacob M, Popa O, Cornianu M. The absence of CD56 expression can differentiate papillary thyroid carcinoma from other thyroid lesions.Indian J Pathol Microbiol 2017;60:161-166


How to cite this URL:
Golu I, Vlad MM, Dema A, Moleriu LC, Tudor A, Iacob M, Popa O, Cornianu M. The absence of CD56 expression can differentiate papillary thyroid carcinoma from other thyroid lesions. Indian J Pathol Microbiol [serial online] 2017 [cited 2019 Oct 15 ];60:161-166
Available from: http://www.ijpmonline.org/text.asp?2017/60/2/161/208378


Full Text

 Introduction



The diagnosis of thyroid tumors derived from follicular cells, especially the follicular variant of papillary thyroid carcinoma (FV-PTC), is always a controversial topic in human pathology. Neural cells adhesion molecules represent a glycoprotein family with an important role in intercellular adhesion, cell migration, and differentiation.[1] CD56, a member of this family, has been detected in follicular epithelial cells of the normal thyroid and in benign thyroid lesions, its expression becoming low or absent after malignant transformation.[2],[3],[4]

The aims of this study were to evaluate the expression of CD56 in different thyroid lesions and to estimate the value of this marker in the differentiation of PTC (including FV-PTC) from other nontumoral lesions and follicular thyroid neoplasias.

 Subjects and Methods



The study group was represented by 53 cases that were submitted to surgery for thyroid pathology between January 2010 and June 2012. They were represented by 38 thyroid neoplasias, divided into two subgroups (subgroup 1, consisting in 25 cases with PTC: Classic – 7 cases, FV-PTC – 11 cases, tall-cell variant – 2 cases, microcarcinoma – 5 cases; subgroup 2, represented by 13 cases of follicular neoplasias: Follicular adenoma [FA] – 5 cases, Hürthle cell FA – 2 cases, follicular thyroid carcinoma [FTC] – 2 cases, Hürthle cell carcinoma – 1 case, and well-differentiated thyroid tumors of uncertain malignant potential [WDT-UMP] – 3 cases), and 15 nonneoplastic thyroid lesions (nodular hyperplasia [NH] – 5 cases, Graves' disease (GD) – 5 cases, and Hashimoto chronic autoimmune thyroiditis [HT] – 5 cases).

The initial diagnosis in all these cases was based on WHO criteria from 2004.[5] Normal thyroid tissue sections obtained from thyroidectomy pieces were stained, too. The examined material followed routine processing procedures and was included into paraffin. The sections were colored by immunohistochemical (IHC) staining using the anti-CD56 antibody.

The IHC reactions were performed on sections having the thickness of 4 μ, applied on special slides, by the use of the visualization system EnVision + Dual Link System-HRP. For the antigen recovery, the de-paraffined sections were pretreated by a 20 min boiling process in retrieval solution (pH 6) and then incubated at room temperature for 60 min with primary anti-CD56 antibody, 1B6 clone, in 1:100 dilution.

For the positive control of the reaction, we used a small cell lung carcinoma CD56-positive tissue, and for the negative one, we replaced the primary antibody with a tampon solution.

The evaluation of IHC staining was performed by an optical microscope (with ×10 magnification, and selectively, ×20 and ×40). A positive CD56 immune reaction was considered the membrane immunostaining, with or without cytoplasmic staining, in ≥10% of the follicular epithelial cells, while the positive reaction encountered in <10% of the follicular epithelium was interpreted as negative CD56 immunostaining.

The Ethical Committee of the County Emergency Hospital, Timisoara, approved the protocol (Approval number 2/January 5, 2010).

The statistical analysis was performed by the aid of the SPSS version 17 (SPSS Inc., Chicago, U.S.A.) program. The Chi-square test was used to assess the differences between groups. The threshold for statistical significance was set at P< 0.05. Risk analysis was applied on these control and studied groups, by calculating the odds ratio (OR) value. Sensitivity, specificity, positive predictive value, and negative predictive value of CD56 were assessed in PTC-positive as compared to non-PTC thyroid lesions.

 Results



The results of IHC staining of the analyzed thyroid lesions are presented in [Table 1] and [Table 2].{Table 1}{Table 2}

In normal thyrocytes, nonneoplastic thyroid lesions (including diffuse colloid goiter, NH, GD, and HT), adenoma, and non-PTC tumors, we detected a strong, diffuse CD56 expression, with a pattern of membrane staining, the immunostaining often reaching 100% of the follicular cells.

In the cases with diffuse colloid goiter, NH, and GD, we observed a distinctive CD56 reaction limited to the cellular membrane of the follicular epithelial cells, with a diffuse pattern. In HT, we noted a CD56 expression with focal pattern, limited to the thyroid follicles, which was more pronounced in the proximity of the lymphoid infiltrate. The reaction was stronger in GD than in HT [Figure 1].{Figure 1}

Five of the total seven cases with FA presented a CD56-positive immune reaction [Figure 2], the intensity of CD56 expression being significantly higher than that of the normal adjacent thyroid tissue. In FTC and Hürthle cell carcinoma, a similar CD56 immune reaction has been found.{Figure 2}

The expression of CD56 protein was constant, being intense and diffuse in normal thyrocytes, in the follicular epithelium of the aforementioned nonneoplastic lesions and in some tumors, with the exception of PTC cases (including the classic subtype and the encapsulated FV-PTC) and papillary microcarcinomas, where the CD56 expression was low or even absent [Figure 3].{Figure 3}

In the majority of the cases with PTC (classical, encapsulated, tall-cell, and the FV), microcarcinoma, and PTC developed on HT, the CD56 immune reaction was absent. These aspects have been reported in all of the cases, independent on the size of the tumor or the stage of the disease. In the lymph node metastasis of PTC, a negative CD56 immune reaction was detected, similar to the primary lesions.

In the majority of PTC cases that were included in this study, CD56 expression was completely absent both in the center and at the periphery of the tumor. In poorly delineated PTC cases, with infiltrative margins into the surrounding parenchyma, CD56-positive cells have been identified, in the majority of the cases these being located at the interface between the tumor and the nontumor thyroid tissue or interpenetrating the CD56-negative tumor cells [Figure 4].{Figure 4}

The expression of CD56, occasionally recorded in PTC, has not exceeded more than 10%–15% of the tumor cells located at the periphery of the tumor. Such CD56-positive cells were identified in two out of seven classical PTC, in two FV-PTC with multinodular growth, in one out of five microcarcinoma cases, and in one out of two tall-cell variants of PTC. They could be nontumoral residual cells, present at the interface between the infiltrative part of the tumor and the adjacent nontumoral thyroid tissue. We have noticed a focal CD56 expression less intense than the one in the surrounding thyroid tissue in three PTC cases (two classical and one tall-cell variant). In two cases of FV-PTC, we observed a diffuse CD56 immune reaction, with distinctive membrane pattern [Figure 5].{Figure 5}

CD56 stained some structures from the adjacent soft tissue of the invasive tumors, as well as the nervous fibers and the skeletal muscles. In the solid cell nests identified in chronic thyroiditis and in the lymph node metastasis of PTC [Figure 6], a CD56-negative immune reaction, similar to primary PTC, was obtained.{Figure 6}

All of the three cases of WDT-UMP included in this study presented a membrane CD56 expression in ≥10% of the tumor cells. When compared with CD56 expression in benign thyroid lesions/tumors, no statistically significant differences were detected.

The immunoexpression of CD56 allowed a clear differentiation (P = 0.002) between PTC and benign nonneoplastic lesions (NH, GD, HT), as well as between PTC and the follicular neoplasia (FTC, Hürthle cell carcinoma) (P = 0.046). The differences were significant when we compared the expression of CD56 in PTC with that of FA, NH, and HT (P = 0.012), as well as FV-PTC with the nonmalignant lesions (FA + NH + HT) (P = 0.008) or with the nontumor thyroid lesions (NH + HT + GD) (P < 0.001). We found no statistically significant differences between the CD56 expression in FV-PTC and the one in classical PTC [Table 2].

The expression of CD56 was positive in 11 follicular tumors (84.6%) and in six PTC (24%) out of the 38 cases of neoplasia included in this study.

The immunoexpression of CD56 has differentiated PTC from other thyroid non-PTC lesions, the results being statistically significant (P < 0.001), with 76% sensitivity, 89% specificity, a positive predictive value of 86%, a negative predictive value of 81%, and 26.4 OR value [Table 3].{Table 3}

 Discussion



CD56 is a homophilic protein from the Ig family, a neural cell adhesion molecule, present on follicular epithelial cells of normal thyroid tissue.[6],[7] Its expression can affect the migration of tumor cells.[8],[9] The anti-CD56 antibody recognizes an isoform of the neural cellular adhesion molecule. Usually, CD56 is expressed in NK cells, activated T cells, lymphocytes, endocrine, and cerebral tissue.[10] It is considered that CD56 protein is involved in the regulation of cellular motility, in the homophilic connection between neurons and in the stimulation of neural growth. Because cellular migration occurs only during the growth phase, the wound healing process, and the tumoral invasion, previous studies have shown that gene expression reduces the specific tumoral invasion.[11],[12]

The differentiation of FV-PTC from FA (essential for the long-term management and treatment of tumors) is one of the most difficult and controversial issues of thyroid pathology, considering the high variability between observers. Literature data have reported a category of atypical thyroid tumors,[13] suggesting the term of WDT-UMP for an encapsulated tumor with well-differentiated cells and uncertain nuclear appearances PTC-like, with absent or uncertain invasion of the capsule and absence of vascular invasion. This term does not solve the variability between observers regarding the diagnosis of FV-PTC but instead allows expressing the differences in opinion and serves to avoid excessive treatment in controversial cases that cannot be included with certainty in a benign or a malignant category.

Although IHC and molecular methods are used for the diagnosis of tumors derived from the follicular thyroid epithelium, the histopathological examination remains the gold standard.[14] IHC studies are useful, but until now, there is no marker that could certainly differentiate PTC from other thyroid lesions, with a 100% consistency.[8]

Due to the lack of specificity and sensitivity of the known markers for differentiation of follicular cell-derived lesions/neoplasia (often with a superposable immune profile), it is recommended to use the IHC methods with precaution for the diagnosis of FV-PTC.[15],[16]

Previous studies reported that CD56 is an antigen involved in follicular epithelium differentiation.[17] Normal thyroid tissue, benign thyroid disease, and follicular tumors show membrane CD56 immunostaining, whereas PTC presents a low or absent expression of this protein.[18] Furthermore, it has been shown that CD56 decreases tumor invasion through the suppression of vascular endothelial growth factor D.[19]

In our study, we evaluated the diagnostic value of CD56 protein expression in normal follicular thyroid tissue, 15 nontumoral thyroid lesions, and 38 thyroid tumors with follicular origin, in order to assess the expression of CD56 as a marker for differentiating PTC, including FV-PTC, from other follicular lesions and thyroid follicular neoplasia. In this study, three cases of WDT-UMP have been included, very rare and extremely controversial tumors, whose diagnosis needed to be re-evaluated.

Following the staining of the sections with anti-CD56 antibody (1B6 clone), we have obtained a diffuse CD56 expression, constantly present, in normal follicular epithelium, in nonneoplastic and neoplastic epithelium (the reaction was limited to the cellular membrane of follicular cells), except PTC, including FV-PTC. Our results notice a positive CD56 immune-expression in nontumoral thyroid lesions and thyroid tumors, except classic PTC, which has revealed a negative CD56 immune reaction or an occasionally focal one, but of very low intensity (in 3 cases). However, two of the FV-PTC cases (18.2%) have presented a high CD56 expression in >50% of the tumor cells. Similar results were reported by Abd El Atti and Shash,[20] who observed a high CD56 expression in three (18.7%) cases of FV-PTC. In five cases of CD56-negative PTC, nonneoplastic follicular cells adjacent to the tumor have presented a positive CD56 immune profile with a diffuse membrane expression in <10% of the cells.

CD56 immunoexpression has differentiated PTC from FA, NH, and HT (P = 0.012) and from FTC (P = 0.046), the differences being statistically significant. Similar results were also obtained after comparing CD56 expression between FV-PTC and the nonmalignant thyroid lesions (FA + NH + HT) (P = 0.008). We have not noticed statistically significant differences between the CD56 expression of WDT-UMP and benign thyroid lesions/tumors.

Our results are comparable to the observations reported by Scarpino et al.,[19] who noticed a low or absent CD56 expression in 61 PTC cases and their lymph node metastasis. El Demellawy et al.[21] described a diffuse CD56 expression in normal and neoplastic follicular epithelium, but this aspect was absent in PTC, supporting, therefore, the contribution of CD56 in the differentiation of PTC (including FV-PTC) from other follicular thyroid lesions/neoplasia (HT, FA) and highlighted the diagnostic value of CD56 protein in PTC. In addition, the authors observed the absence of the CD56 expression in the nests of solid cells, similar to PTC, the same aspects being identified in our study, as well.

Other studies [9] confirm the utility of CD56 IHC-expression in the thyroid gland, revealing its value in the selection of PTC cases (including FV-PTC) from the group of follicular cell-derived thyroid tumors, with a 100% sensitivity and specificity. The authors describe the absence of CD56 expression in PTC and microcarcinoma, including therefore CD56 in the group of IHC markers effective for the diagnosis of PTC, especially FV-PTC. Even so, Etem et al.[22] have described a negative CD56 immune reaction in 35% of the follicular tumors and PTC, without any statistically significant differences between the two groups of tumors. The authors do not consider CD56 as a reliable marker in the differentiation of PTC from other follicular tumors and underlie the need of further similar studies on a higher number of cases. In addition, the authors appreciate the limited value of Hector Battifora mesothelial cell-1 (HBME-1) as an individual marker for differentiating FV-PTC from other follicular tumors and consider adequate the use of an antibody panel that includes HBME-1, cytokeratin-19, p63, and galectin-3 for this purpose.[22]

Although the differential diagnosis of follicular thyroid tumors relies on histological and cytomorphological criteria, Shin et al.[23] appreciate the utility of CD56 and high molecular weight cytokeratin for the diagnosis of PTC and consider that negative CD56 expression is a valuable marker in differentiating PTC from other benign thyroid lesions with papillary architecture.

By analyzing the diagnostic value of CD56 and claudin-1 in PTC and solitary thyroid nodules with follicular pattern, Abd El Atti and Shash [20] described a negative CD56 expression in 82.8% of PTC cases and a positive CD56 expression in 89.4% of solitary thyroid nodules with follicular pattern. In their study, CD56 differentiated FV-PTC from other thyroid nodules with follicular pattern, such as FA, hyperplastic nodules and follicular tumors with UMP, the results obtained being statistically significant (P < 0.001). However, a statistically significant difference between the CD56 expression in PTC and the one in FV-PTC was not reported (the results being similar to our study).

The weaknesses of our study are represented by the reduced number of cases from each subgroup and the use of a single IHC marker.

The strengths of our data are the great variability of the thyroid lesions analyzed by IHC using CD56 expression, as well as the interpretation of the IHC staining for all the samples by a single person with a long-standing experience in thyroid cytology and pathology.

In our opinion, the question raised by our data is in relation with WDT-UMP, which presented a CD56 positivity similar to benign thyroid lesions. Extended studies for these borderline thyroid lesions are necessary to find the utility of those markers for an accurate diagnosis that could prevent unnecessary therapy.

 Conclusions



To conclude, the expression of CD56 protein was constant, intense and diffuses in normal thyrocytes, in follicular epithelium of the nonmalignant lesions and some of the tumors, except PTC (absent or low focal immune reaction). From a total of 38 thyroid tumors that were analyzed, we identified CD56 expression in 84.6% of follicular malignant tumor cases and 24% of PTC. Based on our results, we consider that (a) CD56 proves to be an useful marker for the diagnosis of PTC, including the FV and microcarcinoma; (b) the absence of CD56 expression in FV-PTC can be useful in differentiating it from other thyroid nodules with follicular pattern; (c) CD56 expression allows the differentiation of PTC from nonneoplastic lesions and other thyroid tumors derived from follicular cells with a very good sensitivity, specificity, and positive and negative predictive values. WDT-UMP had a similar CD56 profile to benign thyroid tumors/lesions, suggesting that, most likely, they have a benign biological behavior rather than a malignant one. Long-term studies including CD56 in the panel of IHC markers used for the diagnosis of PTC could predict the evolutional potential of these borderline lesions.

Acknowledgment

The authors thank Associate Prof. Dr. Adrian Vlad and Dr. Andreea Tanase for their valuable help and suggestions in preparing the manuscript.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

References

1Panicker AK, Buhusi M, Thelen K, Maness PF. Cellular signalling mechanisms of neural cell adhesion molecules. Front Biosci 2003;8:d900-11.
2Crnic I, Strittmatter K, Cavallaro U, Kopfstein L, Jussila L, Alitalo K, et al. Loss of neural cell adhesion molecule induces tumor metastasis by up-regulating lymphangiogenesis. Cancer Res 2004;64:8630-8.
3Satoh F, Umemura S, Yasuda M, Osamura RY. Neuroendocrine marker expression in thyroid epithelial tumors. Endocr Pathol 2001;12:291-9.
4Kim JH, Kim YH, Han JH, Lee KB, Sheen SS, Lee J, et al. Silencing of homeobox B9 is associated with down-regulation of CD56 and extrathyroidal extension of tumor in papillary thyroid carcinoma. Hum Pathol 2012;43:1221-8.
5DeLellis RA, Williams ED. Pathology of the thyroid and parathyroid. In: DeLellis RA, Lloyd RV, Heitz PU, Eng C, editors. Pathology and Genetics of Tumours of Endocrine Organs. World Health Organization Classification of Tumours. Lyon: IARC Press; 2004. p. 57-66.
6Zeromski J, Dworacki G, Jenek J, Niemir Z, Jezewska E, Jenek R, et al. Protein and mRNA expression of CD56/N-CAM on follicular epithelial cells of the human thyroid. Int J Immunopathol Pharmacol 1999;12:23-30.
7Zeromski J, Lawniczak M, Galbas K, Jenek R, Golusinski P. Expression of CD56/N-CAM antigen and some other adhesion molecules in various human endocrine glands. Folia Histochem Cytobiol 1998;36:119-25.
8El Demellawy D, Nasr A, Alowami S. Application of CD56, P63 and CK19 immunohistochemistry in the diagnosis of papillary carcinoma of the thyroid. Diagn Pathol 2008;3:5.
9El Demellawy D, Nasr A, Alowami S. An updated review on the clinicopathologic aspects of arrhythmogenic right ventricular cardiomyopathy. Am J Forensic Med Pathol 2009;30:78-83.
10Hoos A, Stojadinovic A, Singh B, Dudas ME, Leung DH, Shaha AR, et al. Clinical significance of molecular expression profiles of Hürthle cell tumors of the thyroid gland analyzed via tissue microarrays. Am J Pathol 2002;160:175-83.
11Prag S, Lepekhin EA, Kolkova K, Hartmann-Petersen R, Kawa A, Walmod PS, et al. NCAM regulates cell motility. J Cell Sci 2002;115(Pt 2):283-92.
12Owens GC, Orr EA, DeMasters BK, Muschel RJ, Berens ME, Kruse CA. Overexpression of a transmembrane isoform of neural cell adhesion molecule alters the invasiveness of rat CNS-1 glioma. Cancer Res 1998;58:2020-8.
13Williams ED. Two proposals regarding the terminology of thyroid tumors. Int J Surg Pathol 2000;8:181-3.
14Ozolins A, Narbuts Z, Strumfa I, Volanska G, Gardovskis J. Diagnostic utility of immunohistochemical panel in various thyroid pathologies. Langenbecks Arch Surg 2010;395:885-91.
15Saleh HA, Jin B, Barnwell J, Alzohaili O. Utility of immunohistochemical markers in differentiating benign from malignant follicular-derived thyroid nodules. Diagn Pathol 2010;5:9.
16Serra S, Asa SL. Controversies in thyroid pathology: The diagnosis of follicular neoplasms. Endocr Pathol 2008;19:156-65.
17Migita K, Eguchi K, Kawakami A, Ida H, Fukuda T, Kurata A, et al. Detection of Leu-19 (CD56) antigen on human thyroid epithelial cells by an immunohistochemical method. Immunology 1991;72:246-9.
18Zeromski J, Biczysko M, Stajgis P, Lawniczak M, Biczysko W. CD56 (NCAM) antigen in glandular epithelium of human thyroid: Light microscopic and ultrastructural study. Folia Histochem Cytobiol 1999;37:11-7.
19Scarpino S, Di Napoli A, Melotti F, Talerico C, Cancrini A, Ruco L. Papillary carcinoma of the thyroid: Low expression of NCAM (CD56) is associated with downregulation of VEGF-D production by tumour cells. J Pathol 2007;212:411-9.
20Abd El Atti RM, Shash LS. Potential diagnostic utility of CD56 and claudin-1 in papillary thyroid carcinoma and solitary follicular thyroid nodules. J Egypt Natl Canc Inst 2012;24:175-84.
21El Demellawy D, Nasr AL, Babay S, Alowami S. Diagnostic utility of CD56 immunohistochemistry in papillary carcinoma of the thyroid. Pathol Res Pract 2009;205:303-9.
22Etem H, Ozekinci S, Mizrak B, Şentürk S. The role of CD56, HBME-1, and p63 in follicular neoplasms of the thyroid. Turk J Pathol 2010;26:238-42.
23Shin MK, Kim JW, Ju YS. CD56 and high molecular weight cytokeratin as diagnostic markers of papillary thyroid carcinoma. Korean J Pathol 2011;45:477-84.