bs_bs_banner
Pathology International 2015; 65: 119–125
doi:10.1111/pin.12252
Original Article
HBME-1 and CD15 immunocytochemistry in the follicular variant of thyroid papillary carcinoma
Makoto Ohta,1,2 Tadakazu Ookoshi,2 Hironobu Naiki2 and Yoshiaki Imamura3 Department of Pathology, Fukui Red Cross Hospital, 2Division of Molecular Pathology, Department of Pathological Sciences, Faculty of Medical Sciences, University of Fukui and 3Division of Surgical Pathology, University of Fukui Hospital, Fukui, Japan
1
Papillary carcinoma is the most common thyroid malignancy. As the cytological diagnosis of papillary carcinoma is not difficult in patients with the usual type of lesion, fineneedle aspiration (FNA) cytology is an effective method for preoperative evaluation. However, this modality is often ineffective in identifying the follicular variant of papillary thyroid carcinoma (FVPTC) due to its similarity to other follicular lesions and the incompleteness of typical nuclear features. Therefore, we investigated the expression of immunocytochemical markers of papillary carcinoma in cytological specimens of FVPTC and evaluated their utilities. The immunoreactivity of HBME-1 and CD15 was investigated using 50 imprint smear cytological specimens obtained from thyroid lesions, including 13 FVPTC. The sensitivity and specificity of HBME-1 for FVPTC were 92% and 89%, respectively, while those of CD15 were 23% and 100%, respectively. In conclusion, HBME-1 is a sensitive marker of papillary carcinoma, including both usual type and FVPTC, in cytological specimens. Therefore, using HBME-1 immunocytochemistry in FNA cytology will lead to reduction of the incidence of falsenegative diagnoses of FVPTC. Although CD15 is apparently inferior in terms of sensitivity for FVPTC, its excellent specificity will support the definitive diagnosis of thyroid malignancies, including FVPTC, after screening with HBME-1. Key words: CD15, cytology, HBME-1, immunocytochemistry, papillary carcinoma, thyroid cancer
The follicular variant of papillary thyroid carcinoma (FVPTC) is one of the major variants of thyroid papillary carcinoma. Histologically, it is mostly composed of a follicular structure and does not exhibit a papillary growth pattern, which characterizes the usual type of papillary carcinoma (UPC). As
Correspondence: Yoshiaki Imamura, MD, PhD, Division of Surgical Pathology, University of Fukui Hospital, 23-3 Matsuoka-Shimoaizuki, Eiheiji, Yoshida, Fukui 910-1193, Japan. E-mail: [email protected] Received 29 October 2014. Accepted for publication 12 December 2014. © 2015 Japanese Society of Pathology and Wiley Publishing Asia Pty Ltd
the prognosis of FVPTC is generally similar to that of UPC1 except for some encapsulated cases,2–5 making an accurate preoperative diagnosis and providing appropriate treatment are required. Although fine-needle aspiration (FNA) cytology is an effective method for preoperatively evaluating nodular thyroid lesions,6 making the diagnosis of FVPTC is generally more difficult than that of UPC. This diagnostic difficulty in the setting of FVPTC is primarily due to the lack of a papillary architecture and weakness of typical nuclear features, such as grooving, pseudoinclusion and optical clearance. To make matters worse, it has been revealed that these nuclear features are not always specific to papillary carcinoma.7,8 In order to reduce the incidence of false-negative diagnoses and diagnostic uncertainty in the cytological diagnosis of thyroid malignancies, several immunocytochemical markers, for example, HBME-1,9–16 CD15,10 cytokeratin 19 (CK19),12–17 CA19-9,10 CD44,15,18,19 CD44v6,20 Galectin-3,11,13,15,20 cyclin D119 and RET11,13 have been proposed. Although some of these markers appear to be very effective for diagnosing UPC, their utility for making a diagnosis of FVPTC remains unclear. In this study, we investigated the expression of immunocytochemical markers of papillary carcinoma, HBME-1 and CD15, in cytological specimens of FVPTC and evaluated the utility of these markers for making the cytological differential diagnosis of FVPTC. We previously reported the utility of CD15 for the histopathological diagnosis of various thyroid lesions.21 Thus, we combined CD15 with HBME-1, which many investigators reported to be useful for the cytological diagnosis of UPC.
MATERIALS AND METHODS Clinical cases From the pathology files of the University of Fukui Hospital, cytological specimens were collected between January
120
M. Ohta et al.
2006 and December 2013. Alcohol-fixed, Papanicolaoustained imprint smears were obtained from pre-fixed 50 surgically resected thyroid lesions, including 19 usual type of papillary carcinomas (UPCs, male/female = 15/4, mean age ± SD = 61.5 ± 16.7 years, mean tumor size ± SD = 1.9 ± 1.0 cm, extrathyroidal extention in 13 cases), 13 follicular variant of papillary carcinomas (FVPTCs, male/ female = 8/5, mean age ± SD = 59.8 ± 13.9 years, mean tumor size ± SD = 1.6 ± 1.0 cm, extrathyroidal extention in four cases), seven follicular adenomas (FAs, male/female = 3/4, mean age ± SD = 64.2 ± 10.3 years, mean tumor size ± SD = 3.2 ± 0.9 cm) and 11 adenomatous goiters (AGs, male/female = 9/2, mean age ± SD = 61.3 ± 11.8 years, mean tumor size ± SD = 2.7 ± 1.5 cm). In all cases, the diagnoses were made based on histological examinations of the surgically resected specimens conducted by three pathologists (Ohta M., Ookoshi T. and Imamura Y.) using the World Health Organization (WHO) criteria.22 Follicular lesions for which the diagnosis was not in agreement among the three pathologists were excluded from this study. As these excluded cases showed an incomplete or limited distribution of typical papillary carcinoma nuclei, some of the lesions were classified as well-differentiated tumors of uncertain malignant potential, as proposed by Williams et al.23 Follicular carcinoma, medullary carcinoma, poorly differentiated carcinoma and undifferentiated carcinoma were excluded from this study. The study was approved by the ethical committee of University of Fukui and performed in adherence to the tenets of the Declaration of Helsinki. Informed consent to research use of biological specimens was obtained from all patients.
Immunohistochemical staining In order to compare the findings with those of the histological specimens, formalin-fixed, paraffin-embedded tissues obtained from 50 surgically resected specimens were immunohistochemically stained. All 50 histological specimens and 50 cytological specimens were obtained from the same lesions of 50 patients. Four-micrometer-thick tissue sections were stained with the same primary antibodies as those used for the cytological specimens. The primary antibody to HBME-1 was used at a dilution of 1:100 and that to CD15 at a dilution of 1:50. For both HBME-1 and CD15, antigen retrieval was achieved by steaming at 105°C in citrate buffer solution for 14 min, and the primary antibodies were incubated for 30 min at room temperature. The remaining steps and evaluation procedures followed those used for the cytological specimens.
Statistical analyses The sensitivity and specificity of HBME-1 and CD15 for UPC and FVPTC were calculated. The statistical analyses were performed with the Chi-square test and median test using the MedCalc software program (MedCalc Software, Mariakerke, Belgium). A P-value of
Pathology International 2015; 65: 119–125
doi:10.1111/pin.12252
Original Article
HBME-1 and CD15 immunocytochemistry in the follicular variant of thyroid papillary carcinoma
Makoto Ohta,1,2 Tadakazu Ookoshi,2 Hironobu Naiki2 and Yoshiaki Imamura3 Department of Pathology, Fukui Red Cross Hospital, 2Division of Molecular Pathology, Department of Pathological Sciences, Faculty of Medical Sciences, University of Fukui and 3Division of Surgical Pathology, University of Fukui Hospital, Fukui, Japan
1
Papillary carcinoma is the most common thyroid malignancy. As the cytological diagnosis of papillary carcinoma is not difficult in patients with the usual type of lesion, fineneedle aspiration (FNA) cytology is an effective method for preoperative evaluation. However, this modality is often ineffective in identifying the follicular variant of papillary thyroid carcinoma (FVPTC) due to its similarity to other follicular lesions and the incompleteness of typical nuclear features. Therefore, we investigated the expression of immunocytochemical markers of papillary carcinoma in cytological specimens of FVPTC and evaluated their utilities. The immunoreactivity of HBME-1 and CD15 was investigated using 50 imprint smear cytological specimens obtained from thyroid lesions, including 13 FVPTC. The sensitivity and specificity of HBME-1 for FVPTC were 92% and 89%, respectively, while those of CD15 were 23% and 100%, respectively. In conclusion, HBME-1 is a sensitive marker of papillary carcinoma, including both usual type and FVPTC, in cytological specimens. Therefore, using HBME-1 immunocytochemistry in FNA cytology will lead to reduction of the incidence of falsenegative diagnoses of FVPTC. Although CD15 is apparently inferior in terms of sensitivity for FVPTC, its excellent specificity will support the definitive diagnosis of thyroid malignancies, including FVPTC, after screening with HBME-1. Key words: CD15, cytology, HBME-1, immunocytochemistry, papillary carcinoma, thyroid cancer
The follicular variant of papillary thyroid carcinoma (FVPTC) is one of the major variants of thyroid papillary carcinoma. Histologically, it is mostly composed of a follicular structure and does not exhibit a papillary growth pattern, which characterizes the usual type of papillary carcinoma (UPC). As
Correspondence: Yoshiaki Imamura, MD, PhD, Division of Surgical Pathology, University of Fukui Hospital, 23-3 Matsuoka-Shimoaizuki, Eiheiji, Yoshida, Fukui 910-1193, Japan. E-mail: [email protected] Received 29 October 2014. Accepted for publication 12 December 2014. © 2015 Japanese Society of Pathology and Wiley Publishing Asia Pty Ltd
the prognosis of FVPTC is generally similar to that of UPC1 except for some encapsulated cases,2–5 making an accurate preoperative diagnosis and providing appropriate treatment are required. Although fine-needle aspiration (FNA) cytology is an effective method for preoperatively evaluating nodular thyroid lesions,6 making the diagnosis of FVPTC is generally more difficult than that of UPC. This diagnostic difficulty in the setting of FVPTC is primarily due to the lack of a papillary architecture and weakness of typical nuclear features, such as grooving, pseudoinclusion and optical clearance. To make matters worse, it has been revealed that these nuclear features are not always specific to papillary carcinoma.7,8 In order to reduce the incidence of false-negative diagnoses and diagnostic uncertainty in the cytological diagnosis of thyroid malignancies, several immunocytochemical markers, for example, HBME-1,9–16 CD15,10 cytokeratin 19 (CK19),12–17 CA19-9,10 CD44,15,18,19 CD44v6,20 Galectin-3,11,13,15,20 cyclin D119 and RET11,13 have been proposed. Although some of these markers appear to be very effective for diagnosing UPC, their utility for making a diagnosis of FVPTC remains unclear. In this study, we investigated the expression of immunocytochemical markers of papillary carcinoma, HBME-1 and CD15, in cytological specimens of FVPTC and evaluated the utility of these markers for making the cytological differential diagnosis of FVPTC. We previously reported the utility of CD15 for the histopathological diagnosis of various thyroid lesions.21 Thus, we combined CD15 with HBME-1, which many investigators reported to be useful for the cytological diagnosis of UPC.
MATERIALS AND METHODS Clinical cases From the pathology files of the University of Fukui Hospital, cytological specimens were collected between January
120
M. Ohta et al.
2006 and December 2013. Alcohol-fixed, Papanicolaoustained imprint smears were obtained from pre-fixed 50 surgically resected thyroid lesions, including 19 usual type of papillary carcinomas (UPCs, male/female = 15/4, mean age ± SD = 61.5 ± 16.7 years, mean tumor size ± SD = 1.9 ± 1.0 cm, extrathyroidal extention in 13 cases), 13 follicular variant of papillary carcinomas (FVPTCs, male/ female = 8/5, mean age ± SD = 59.8 ± 13.9 years, mean tumor size ± SD = 1.6 ± 1.0 cm, extrathyroidal extention in four cases), seven follicular adenomas (FAs, male/female = 3/4, mean age ± SD = 64.2 ± 10.3 years, mean tumor size ± SD = 3.2 ± 0.9 cm) and 11 adenomatous goiters (AGs, male/female = 9/2, mean age ± SD = 61.3 ± 11.8 years, mean tumor size ± SD = 2.7 ± 1.5 cm). In all cases, the diagnoses were made based on histological examinations of the surgically resected specimens conducted by three pathologists (Ohta M., Ookoshi T. and Imamura Y.) using the World Health Organization (WHO) criteria.22 Follicular lesions for which the diagnosis was not in agreement among the three pathologists were excluded from this study. As these excluded cases showed an incomplete or limited distribution of typical papillary carcinoma nuclei, some of the lesions were classified as well-differentiated tumors of uncertain malignant potential, as proposed by Williams et al.23 Follicular carcinoma, medullary carcinoma, poorly differentiated carcinoma and undifferentiated carcinoma were excluded from this study. The study was approved by the ethical committee of University of Fukui and performed in adherence to the tenets of the Declaration of Helsinki. Informed consent to research use of biological specimens was obtained from all patients.
Immunohistochemical staining In order to compare the findings with those of the histological specimens, formalin-fixed, paraffin-embedded tissues obtained from 50 surgically resected specimens were immunohistochemically stained. All 50 histological specimens and 50 cytological specimens were obtained from the same lesions of 50 patients. Four-micrometer-thick tissue sections were stained with the same primary antibodies as those used for the cytological specimens. The primary antibody to HBME-1 was used at a dilution of 1:100 and that to CD15 at a dilution of 1:50. For both HBME-1 and CD15, antigen retrieval was achieved by steaming at 105°C in citrate buffer solution for 14 min, and the primary antibodies were incubated for 30 min at room temperature. The remaining steps and evaluation procedures followed those used for the cytological specimens.
Statistical analyses The sensitivity and specificity of HBME-1 and CD15 for UPC and FVPTC were calculated. The statistical analyses were performed with the Chi-square test and median test using the MedCalc software program (MedCalc Software, Mariakerke, Belgium). A P-value of
