Cytokeratin 19 Immunolocalization in Cell Block Preparation of Thyroid Aspirates

Transcription

1 Cytokeratin 19 Immunolocalization in Cell Block Preparation of Thyroid Aspirates An Adjunct to Fine-Needle Aspiration Diagnosis of Papillary Thyroid Carcinoma Kamal K. Khurana, MD; Luan D. Truong, MD; Virginia A. LiVolsi, MD; Zubair W. Baloch, MD, PhD Context. Immunohistochemical staining for cytokeratin 19 (CK-19) is a useful ancillary technique for diagnosing papillary thyroid carcinoma (papillary carcinoma) in histologic specimens. Although similar results have been obtained on aspirate smears, to our knowledge the utility of CK-19 immunolocalization in cell block preparations as an adjunct to fine-needle aspiration diagnosis of papillary carcinoma has not been examined. Objective. The purpose of this study was to determine whether CK-19 immunostaining of cell block preparations of thyroid aspirates is a useful ancillary technique for diagnosing papillary carcinoma. Materials and Methods. Using a monoclonal antibody to CK-19 and a standard avidin-biotin complex technique, immunostaining was performed on paraffin-embedded cell blocks of 57 cases with the following cytologic diagnoses: (a) papillary carcinoma (20 cases); (b) atypical cytology, cannot exclude papillary carcinoma (19 cases); and (c) nonneoplastic thyroid (18 cases). The staining reaction in each case was graded on the basis of percentage of epithelial cells stained (1, 10%; 2, 10% 50%; 3, 50). Tissue follow-up was available in all cases. Fine-needle aspiration biopsy (FNAB) represents a major advance in the diagnosis of thyroid nodules. This method is now considered to be most effective test available for distinguishing between benign and malignant thyroid nodules, with accuracy approaching 95%. 1 Papillary thyroid carcinoma is the most commonly detected malignant thyroid neoplasm. 2 4 Although many cytologic features of papillary carcinoma are shared by other neoplasms or nonneoplastic lesions of thyroid, the presence of classic nuclear, cytoplasmic, and architectural features is virtually diagnostic of papillary carcinoma. However, in the absence of 1 or more of the classic features, especially Accepted for publication November 19, From the Departments of Pathology, State University of New York Upstate Medical University Hospital, Syracuse (Dr Khurana); Baylor College of Medicine, Houston, Tex (Dr Truong); and University of Pennsylvania, Philadelphia (Drs LiVolsi and Baloch). Presented in part at the 91st meeting of United States and Canadian Academy of Pathology, Chicago, Ill, February 23 March 1, Reprints: Kamal K. Khurana, MD, Department of Pathology, SUNY Upstate Medical University Hospital, 750 E Adams St, Syracuse, NY ( Khuranak@upstate.edu). Results. Nineteen (95%) of 20 cases with an unequivocal diagnosis of papillary carcinoma were positive for CK- 19 (3 ). Tissue follow-up confirmed papillary carcinoma in all 20 cases. Of the 19 cases with a diagnosis of atypical cytology, cannot exclude papillary carcinoma, 7 (37%) cases displayed 3 immunostaining and subsequent excision confirmed papillary carcinoma in all 7 cases. The remaining 12 cases with 1 immunostaining included surgically confirmed goiter (6 cases), adenoma (2 cases), lymphocytic thyroiditis (3 cases), and papillary carcinoma (1 case). The follicular cells in 18 cases with a cytologic diagnosis of nonneoplastic thyroid showed 1 immunostaining. Histologic follow-up of these cases confirmed the nonneoplastic cytologic diagnoses. Conclusions. Cytokeratin 19 immunostaining of cell block preparations of thyroid aspirates serves as a useful tool for the diagnosis of papillary carcinoma. Strong immunostaining (3 ) for CK-19 aids in accurate diagnosis of malignancy in cytomorphologically equivocal cases of papillary carcinoma. (Arch Pathol Lab Med. 2003;127: ) lack of intranuclear inclusions, accurate FNAB diagnosis of papillary carcinoma may be challenging. Cytokeratin 19 (CK-19) immunostaining as an ancillary technique in the diagnosis of papillary carcinoma is well recognized in the surgical pathology literature. On histologic sections, CK-19 is expressed strongly in tumor cells of papillary carcinoma and is absent or only focally present in follicular neoplasms and adenomatous goiters Although similar results have been reported recently in cytology smears, 14,15 immunostaining of cytology smears may be hampered by technical difficulties, including control availability and procedure standardization. In our laboratory, we have been able to obtain more consistent results with paraffin-embedded, formalin-fixed tissue. Hence, we chose to retrospectively determine the effectiveness of CK-19 immunolocalization in cell-block preparations in diagnosing papillary carcinoma. To the best of our knowledge, this approach has not been assessed previously. MATERIALS AND METHODS Samples were obtained from the archives of the Department of Pathology, University Hospital, Syracuse, NY, and the Department of Pathology, University of Pennsylvania, Philadelphia. Arch Pathol Lab Med Vol 127, May 2003 CK-19 Immunolocalization in Cell Block Preparations Khurana et al 579

2 Only cases with adequate cell blocks and available histologic follow-up were included. The availability of a cell block was required for inclusion in this retrospective study because in a preliminary evaluation we found that CK-19 immunostaining is optimal in tissue sections from the cell blocks, but is unsatisfactory in previously stained smears. The cytologic diagnoses rendered on the cell blocks in conjunction with Papanicolaou and Diff- Quik stained cytology smears were categorized into 3 broad categories as follows: (a) positive for malignancy, consistent with papillary carcinoma (20 cases); (b) atypical cytology, papillary carcinoma cannot be excluded (18 cases); and (c) nonneoplastic thyroid (19 cases). For each case, immunostaining was performed on formalinfixed, paraffin-embedded tissue sections with a monoclonal antibody against CK-19 (Dako Corporation, Carpinteria, Calif; 1:400 dilution) and an automatic immunostainer, which incorporates standard capillary gap (Microprobe, Fischer Scientific, Pittsburgh, Pa) and streptavidin-biotin-peroxidase techniques. Staining was independently reviewed by the 4 authors (K.K.K., L.D.T., Z.W.B., and V.A.L.). Cytokeratin 19 immunoreactivity was scored according to the percentage of stained cells as follows: negative or weakly positive (1 ), less than 10% of cells; moderate (2 ), 10% to 50%; and strong (3 ), more than 50%. We did not encounter any case with moderate staining. Hence, we divided our results into positive (strong staining) and negative (weak to no staining). Positive controls consisted of a histologic block of papillary carcinoma that had been previously demonstrated to be CK-19 positive. Negative controls included tissue sections from each case with replacement of the primary antibody by nonimmune mouse serum. Statistical Analysis The following parameters were calculated for the value of CK- 19 staining in predicting papillary carcinoma: (a) sensitivity, the percentage of malignant cases showing strong or moderate positivity (positive staining), and (b) specificity, percentage of benign cases with weak or no staining (negative staining) with CK-19. Statistical analysis was performed with a Fisher exact test (2- sided). The percentage of papillary carcinoma and benign/nonneoplastic thyroid lesion exhibiting strong immunopositivity for CK-19 were calculated, and statistically significant differences were noted. A P value of less than.05 was considered significant. RESULTS Cytologic Findings Smears from 20 cases with cytologic diagnoses of papillary carcinoma showed classic nuclear, cytoplasmic, and architectural features characteristic of papillary carcinoma. These features included nuclei with irregular membranes, grooved nuclei, pale chromatin, and intranuclear cytoplasmic invaginations; dense squamoid cytoplasm; and evidence of papillary architecture. Cell blocks in most of these cases revealed sheets, clusters, or papillae with similar cytologic features (Figure 1). Smears from the 19 FNABs with a cytologic diagnosis of atypical cytology, cannot exclude papillary carcinoma, lacked 1 or more features of classic papillary carcinoma, especially intranuclear cytoplasmic invagination. This category also included 4 FNAs with features suspicious for the follicular variant of papillary carcinoma (cellular smears with predominantly follicular pattern, occasional nuclear grooves, and no colloid). Cell blocks revealed groups and sheets of follicular cells with occasional nuclear grooves and nuclear clearing. Prominent follicular arrangement was also seen. However, no intranuclear inclusions or colloid were identified (Figure 2). Smears from 18 FNABs of nonneoplastic lesions revealed cytologic features of goiter (12 cases, moderately Figure 1. Papillary carcinoma. Cell block preparation showing papillary fronds lined by tumor cells with intranuclear inclusions and nuclear grooves (hematoxylin-eosin, original magnification 200). Figure 2. Atypical cytology. Cell block preparation showing prominent follicular arrangement. Occasional nuclei with nuclear grooves and nuclear clearing are also seen. No intranuclear inclusions are identified (hematoxylin-eosin, original magnification 200). Figure 3. Nonneoplastic thyroid lesion. Cell block preparation showing follicles of variable size filled with colloid (hematoxylin-eosin, original magnification 200). cellular smears with bland follicular cells and histiocytes admixed with abundant colloid), lymphocytic thyroiditis (3 cases, follicular cells and Hürthle cells admixed with polymorphous population of lymphocytes and scattered fragment of colloid), or colloid nodules (3 cases, abundant 580 Arch Pathol Lab Med Vol 127, May 2003 CK-19 Immunolocalization in Cell Block Preparations Khurana et al

3 Immunoreactivity for Cytokeratin 19 (CK-19) in 57 Cases of Thyroid Lesion With Follow-up Histologic (Final) Diagnoses Cytologic Diagnosis Papillary carcinoma (n 20) 19/3 (95) 1/1 (5) Atypical cytology, cannot exclude papillary carcinoma (n 19) 7/3 (37) 12/1 (63) CK-19 Immunostaining, No. of Cases/Staining Score (%)* Final Diagnosis Papillary carcinoma (n 19) Papillary carcinoma (n 1) Papillary carcinoma (n 7) Benign/nonneoplastic lesions (n 11) Papillary carcinoma (n 1) Nonneoplastic (n 18) 18/1 (100) Nonneoplastic lesions (n 18) *1 indicates less than 10% of cells stained; 3, more than 50% of cell stained. See text for description of atypical cytology, cannot exclude papillary carcinoma cells. watery or inspissated colloid admixed with few groups of bland follicular cells). Cell blocks revealed sheets or clusters of follicular cells and variably sized follicles filled with colloid (Figure 3). Prominent lymphoid component and histiocytes were identified in cases with a cytologic diagnosis of lymphocytic thyroiditis and goiter, respectively. Immunohistochemistry and Correlation With Tissue Diagnoses Both positive and negative controls for CK-19 immunostaining showed appropriate results. The immunostaining results are summarized in the Table. Among the 20 cases with an unequivocal cytologic diagnosis of papillary carcinoma, strongly positive staining for CK-19 (Figure 4) was noted in 19 (95%) and weak staining was seen in 1 (5%). Total thyroidectomy performed in all these 20 cases confirmed the diagnosis in each case (see Table). Among the 19 FNABs with the diagnosis atypical cytology, papillary carcinoma cannot be excluded, the staining was strong in 7 (37%) (Figure 5) and weak in the remaining 12 (73%). Total thyroidectomy performed in all 19 cases revealed papillary carcinoma in 8 (42%) and benign/nonneoplastic thyroid lesions in 11 (58%; 6 goiter, 2 follicular adenoma, and 3 lymphocytic thyroiditis). The CK-19 staining was strongly positive in 7 of 8 cases of papillary carcinoma, but was negative in the remaining papillary carcinoma and in 11 cases of nonneoplastic/benign thyroid lesions. All 18 FNABs with the cytologic diagnosis of nonneoplastic thyroid were negative or weakly positive for CK- 19 (Figure 6). Total thyroidectomy performed in all cases confirmed nonneoplastic lesions in each (nodular goiter [15 cases] and lymphocytic thyroiditis [3 cases]). In our evaluation of all cases for CK-19 immunostaining, we did not encounter any case with moderate staining. Statistical Analysis Strong immunoreactivity for CK-19 was identified in cell block sections of 26 of 28 cases of papillary carcinoma (sensitivity 93%). Weak or negative staining was noted in each of the 29 cases with final diagnoses of nonneoplastic/ benign thyroid lesions (specificity 100%). The differences in the frequency of CK-19 positive papillary carcinoma and nonneoplastic/benign thyroid lesions for CK-19 (93% vs 0%, respectively) were statistically significant (P.05). COMMENT The diagnosis of papillary carcinoma in FNA specimens is usually straightforward when classic cytologic features are present, including nuclei with irregular membranes, grooved nuclei, pale chromatin, and intranuclear cytoplasmic invaginations; dense squamoid cytoplasm; and evidence of papillary architecture At times, however, it is difficult to make an unequivocal cytologic diagnosis of papillary carcinoma in an aspirate smear. The diagnostic dilemma is related to the observation that some typical cytologic features of papillary carcinoma (nuclear grooves, giant cells, psammoma bodies, papillary fragments) are also seen in nonneoplastic lesions and follicular neoplasms of thyroid Benign lesions can be misdiagnosed as papillary cancer. Follicular adenomas, adenomatous nodules, and hot autonomous nodules can show papillary fragments, nuclear enlargement, and nuclear grooves. Optical clearing and nuclear atypia and enlargement can be found in lymphocytic thyroiditis. Dystrophic calcifications in goiter may be overinterpreted as psammoma bodies. Thus, a marker that would differentiate papillary carcinoma from nonneoplastic thyroid lesions and follicular neoplasms would be of immense assistance in resolving these common diagnostic dilemmas. Cytokeratin 19 is a low-molecular-weight keratin widely present in simple epithelial cells and is a minor component of stratified epithelium, such as basal cell layers. 8,9,25 In histologic specimens, CK-19 has been found to be strongly and diffusely expressed in papillary carcinoma, whereas it is usually absent or focally expressed in adenomatous goiter and follicular neoplasms Cytokeratin 19 immunostaining has been reported to be useful on cytology smears of thyroid nodules. 14,15 However, immunostaining of smears is well known to be technically difficult and capricious. This is especially true for destained smears, which may be the only material available. In contrast, immunostaining of tissue sections prepared from cell blocks in our laboratories, including that for CK-19, have provided excellent results. In our study, 45% of cases were strongly positive for CK-19, and 55% of cases were negative or weakly expressed CK-19. Since we did not encounter any cases that exhibited moderate staining, we refer to all cases with strong expression for CK-19 as positive and all cases with less than 10% cells expressing CK-19 as negative. Similar definitions for negative CK-19 staining have been used by Nasser et al. 14 The lack of cases with moderate staining made the demarcation between positive and negative staining very striking and the results easily reproducible and concordant among the pathologists (K.K.K., L.D.T., V.A.L., and Z.W.B.). We found that CK-19 immunohistochemistry in cell block preparations is of considerable help in the diagnosis Arch Pathol Lab Med Vol 127, May 2003 CK-19 Immunolocalization in Cell Block Preparations Khurana et al 581

4 Figure 4. Papillary carcinoma. A, A case of papillary carcinoma with malignant cells showing strong immunopositivity for cytokeratin 19 (CK- 19) (same case as shown in Figure 1). B, Another case of papillary carcinoma with malignant cells expressing CK-19 (CK-19 immunohistochemistry, original magnification 200). Figure 5. Atypical cytology. A, Neoplastic follicular cells arranged in follicular pattern showing strong positivity for cytokeratin 19 (CK-19). Subsequent tissue follow-up confirmed papillary carcinoma (same case as shown in Figure 2). B, Papillae with slight nuclear atypia and nuclear grooves showing strong positivity for CK-19. Subsequent tissue follow-up confirmed papillary carcinoma (CK-19 immunohistochemistry, original magnification 200). Figure 6. Nonneoplastic thyroid lesion. Follicular cells show focal and weak expression of cytokeratin 19 ( 10% of cells stained; same case as in Figure 3) (CK-19 immunohistochemistry, original magnification 200). of papillary carcinoma and its distinction from nonneoplastic/benign thyroid lesions. Hirokawa et al 15 reported similar observations in touch imprints prepared from resected thyroid specimen. Nasser et al 14 also demonstrated the usefulness of CK-19 immunolocalization in thyroid aspirate smears as an adjunct to conventional morphologic features in the diagnosis of papillary carcinoma. In our study, CK-19 immunostaining was detected not only in 582 Arch Pathol Lab Med Vol 127, May 2003 CK-19 Immunolocalization in Cell Block Preparations Khurana et al

5 95% of cases for which the cytologic diagnosis was papillary carcinoma, but also in 37% of cases (confirmed cases of papillary carcinoma by tissue follow-up) with equivocal cytologic diagnoses. Although none of the benign thyroid lesions (confirmed by tissue follow-up) stained positively for CK-19, lack of CK-19 expression in papillary carcinoma was observed in one case with obvious cytologic features of papillary carcinoma and in another case with equivocal cytology. Hence, lack of CK-19 staining could not exclude malignancy in 2 (7%) of 28 cases. In our series, the sensitivity and specificity values for CK-19 were 93% and 100%, respectively, and are comparable to the sensitivity and specificity values of 92% and 97%, respectively, reported by Nasser et al. 14 Conventional immunohistochemistry performed on cell block preparations as used in our study is easy to use in contrast to immunostaining of smears. The latter is hampered by the difficulty in standardization of appropriate controls and by technical difficulties related to immunostaining of destained slides. False-positive CK-19 staining for nonneoplastic and benign thyroid lesions was not found in our study. Nasser et al 14 reported an isolated CK-19 positive case that was cytologically considered suspicious for papillary carcinoma but was diagnosed histologically as follicular adenoma with focal papillary hyperplasia. In conclusion, CK-19 immunostaining of cell blocks from thyroid aspirates with features suspicious but not diagnostic for papillary carcinoma serves as a useful adjunct. Positive CK-19 staining in this context is highly specific and sensitive for papillary thyroid carcinoma. References 1. Gharib H. Fine needle aspiration biopsy of thyroid nodules: advantages, limitations, and effect. Mayo Clin Proc. 1994;69: LiVolsi VA. Surgical pathology of the thyroid. In: Bennington JL, ed. Major Problems in Pathology. Vol 22. Philadelphia, Pa: WB Saunders; 1990: Carcangiu ML, Zampi G, Pupi A, Castagnoli A, Rosai J. Papillary carcinoma of the thyroid: a clinicopathologic study of 241 cases treated at the University of Florence, Italy. Cancer. 1985;55: Rosai J, Zampi G, Carcangiu ML. Papillary carcinoma of the thyroid: a discussion of its several morphologic expressions with particular emphasis on the follicular variant. Am J Surg Pathol. 1983;8: Permanetter W, Nathrath WBL, Lohrs U. Immunohistochemical analysis of thyroglobulin and keratin in benign and malignant thyroid tumors. Virchows Arch A Pathol Anat Histopathol. 1982;398: Buley ID, Gatter KC, Heryet A, Mason DY. Expression of intermediate filament proteins in normal and diseased thyroid glands. J Clin Pathol. 1987;40: Dockhorn-Dvornickzak B, Frank WW, Schroder S, et al. Patterns of cytoskeletal proteins in human thyroid gland and thyroid carcinoma. Differentiation. 1987;35: Schelfhout LJDM, Van Muijen GNP, Fleuren GJ. Expression of keratin 19 distinguishes papillary thyroid carcinoma from follicular carcinoma and follicular thyroid adenoma. Am J Clin Pathol. 1989;92: Rapheal SJ, McKeown-Eyssen G, Asa SL. High-molecular-weight cytokeratin and cytokeratin-19 in the diagnosis of thyroid tumors. Mod Pathol. 1994;7: Miettinen M, Franssila K, Lehto VP, Paasivuo R, Virtanen I. Expression of intermediate filaments in thyroid and thyroid tumors. Lab Invest. 1984;50: Miettinen M, Kovatich AJ, Karkkainen P. Keratin subsets in papillary and follicular thyroid lesions: a paraffin section analysis with diagnostic implications. Virchows Arch. 1997;431: Fonseca E, Nesland JM, Hoie J, Sobrinho-Simoes M. Patterns of expression of intermediate cytokeratin filaments in the thyroid gland: an immunohistochemical study of simple and stratified epithelial-type cytokeratins. Virchows Arch. 1997;430: Baloch ZW, Abraham S, Roberts S, LiVolsi VA. Differential expression of cytokeratins in follicular variant of papillary carcinoma: an immunohistochemical study and its diagnostic utility. Hum Pathol. 1999;30: Nasser SM, Pitman MB, Pilch BZ, Faquin W. Fine-needle aspiration biopsy of papillary thyroid carcinoma: diagnostic utility of cytokeratin 19 immunostaining. Cancer. 2000;90: Hirokawa M, Inagaki A, Sonoo H. Expression of cytokeratin 19 in cytologic specimens of thyroid. Diagn Cytopathol. 2000;22: Kini SR, Miller JM, Hamburger JI, Smith MJ. Cytopathology of papillary carcinoma of the thyroid by fine needle aspiration. Acta Cytol. 1980;24: Miller TR, Bottles K, Holly EA, Friends NF, Abele JS. A stepwise logistical regression analysis of papillary carcinoma of the thyroid. Acta Cytol. 1986;30: Kaur A, Jayaram G. Thyroid tumors: cytomorphology of papillary carcinoma. Diagn Cytopathol. 1991;7: Basu D, Jayaram G. A logistical model for thyroid lesions. Diagn Cytopathol. 1992;8: Baloch ZW, Sacks MJ, Yu GH, LiVolsi VA, Gupta PK. Fine needle aspiration of thyroid: an institutional experience. Thyroid. 1998;8: Mai KT, Yazdi HM, Common AS, Perkins DG, MacDonald L. Neoplastic non-papillary thyroid carcinoma lesions with fine chromatin pattern. Pathol Int. 1999;49: Martinez-Parra D, Campos Fernandez J, Hierro-Guolmain CC, Sola Perez J, Perez-Guillermo M. Follicular variant of papillary carcinoma: cytologic and histologic correlations. Diagn Cytopathol. 1996;15: Baloch ZW, Gupta PK, Yu GH, Sacks MJ, LiVolsi VA. Follicular variant of papillary carcinoma: cytologic and histologic correlation. Am J Clin Pathol. 1999; 111: Khurana KK, Baloch ZW, LiVolsi VA. Aspiration cytology of pediatric solitary papillary hyperplastic nodule: potential pitfall. Arch Pathol Lab Med. 2001; 125: Sun T-T, Shih C, Green H. Keratin cytoskeleton in epithelial cells of internal organs. Proc Natl Acad Sci U S A. 1979;22: Arch Pathol Lab Med Vol 127, May 2003 CK-19 Immunolocalization in Cell Block Preparations Khurana et al 583