E-selectin expression and BRAF status in papillary thyroid carcinomas: Correlation with clinicopathologic features Paolo Miccoli, MD,a Liborio Torregrossa, MD, PhD,a Nicla Borrelli, PhD,a Gabriele Materazzi, MD,a Andrea Cacciato Insilla, MD,a Mario Miccoli, PhD,b and Fulvio Basolo, MD,a Pisa, Italy
Background. Cell adhesion molecules, represented by the immunoglobulin family and selectins, play an important role in the progression of cancer. A correlation between selectins and tumor aggressiveness has been demonstrated in several reports. Methods. Eighty-eight patients (mean age, 41.0 ± 14 years) with papillary thyroid carcinoma (conventional variant and sized approximately 20 mm) were divided in 2 groups: 41 with encapsulated tumors and 47 with tumors with extrathyroidal extension. E-selectin expression was evaluated by immunohistochemical staining and semiquantitative real-time reverse-transcription polymerase chain reaction and normalized by calculating the z-score (positive: value above the population mean; negative: below the mean). Results. Lymph node metastasis (LNM) was found in 2 of 41 encapsulated tumors (4.8%) and in 19 of 47 tumors (40.4%) with extrathyroidal extension. BRAF mutation was present in 21 encapsulated tumors (51.2%) and in 31 tumors with extrathyroidal extension (65.9%). The mean E-selectin z-score was 0.32 for encapsulated tumors and 0.28 for tumors with extrathyroidal extension. E-selectin expression correlates with neoplastic infiltration (P = .04), the American Joint Commission on Cancer stage (P = .02), and BRAF mutation (P = .03). Conclusion. E-selectin overexpression in association with BRAF mutation status could promote a more aggressive phenotype in papillary thyroid carcinoma. (Surgery 2014;156:1550-8.) From the Department of Surgical, Medical, Molecular and Critical Area Pathology,a and Department of Translational Research and of New Surgical and Medical Technologies,b Universit a degli Studi di Pisa, Pisa, Italy
PAPILLARY THYROID CARCINOMA (PTC) is the most common type of thyroid cancer, accounting for more than 80% of all thyroid malignancies.1 Despite the excellent prognosis, more aggressive forms of the disease develop in some patients, including local and distant metastasis or tumor recurrence.2 One of the most important independent risk factors for an unfavorable outcome is extrathyroidal extension, defined as tumor penetration through the thyroid capsule into the adjacent fibroadipose tissue or skeletal muscle.3,4 The presence of tumor capsule, however, seems Accepted for publication August 19, 2014. Reprint requests: Paolo Miccoli, MD, Professor of Surgery, Department of Surgical, Medical, Molecular and Critical Area Pathology, Universita di Pisa, Via Paradisa 2, 56126 Pisa, Italy. E-mail: [email protected]. 0039-6060/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.surg.2014.08.049
1550 SURGERY
to be correlated to a less aggressive behavior of the neoplasm in lymph node metastases and tumor recurrence.5 A strong association between cell adhesion molecules and the mechanism of neoplastic progression and metastasis has been suggested.6-8 These molecules were first reported as receptors modulating leukocyte traffic.9,10 They are subdivided mainly into 2 groups: the immunoglobulin family, including intercellular adhesion molecule and vascular cell adhesion molecule, and selectins. The selectins comprise a family of 3 cell adhesion molecules: E-selectin, present in endothelial cells; P-selectin, in platelets and endothelial cells; and L-selectin, in leukocytes. Their structure is similar: the transmembrane domain anchors selectins to the membrane, and the cytoplasmic tail supports their action as signaling agents.8-10 Selectins are expressed when certain mediators are present, such as interleukins, tumor necrosis factor, or toxins. Selectins play a critical role in leukocyte
Surgery Volume 156, Number 6
recruitment, mediating leukocyte arrest in the vasculature, extravasation, and migration to inflamed tissue.10 Knowledge about leukocyte locomotion has been weaned from the understanding of tumor cell extravasation and metastasis.11 Numerous studies have reported the implication of some selectins (ie E-selectin) in neoplastic progression from primary to metastatic disease in different types of carcinomas, such as head and neck squamous cell carcinoma,6 colorectal carcinoma,13,14 and gastric carcinoma.15 We conducted this study to verify whether the overexpression of E-selectin is linked to a more aggressive tumor behavior in patients presenting a PTC with the same pathologic features. METHODS Formalin-fixed and paraffin-embedded (FFPE) thyroid specimens from 88 patients (20 male, 68 female) with a conventional variant of PTC who underwent operation at the Department of Surgical Pathology from January 2006 to December 2012 were retrospectively reviewed. Patients were a mean age of 41.0 ± 14 years (range, 12–78 years). The mean tumor size was 18.6 ± 1.4 mm (range, 15.0–20.0 mm). This study was approved by the Institutional Review Board, and informed consent was obtained from all patients. All patients had undergone a total thyroidectomy and a full exploration of the central compartment on both sides. A lymphadenectomy was performed when enlarged lymph nodes were present preoperatively or intraoperatively. Histopathologic review was conducted by 2 pathologists (L.T. and A.C.I). The diagnoses were reassessed according to the World Health Organization’s classification of thyroid malignancy.16 Hematoxylin and eosin–stained sections were re-examined, taking into account the known risk factors for thyroid cancer, such as tumor size, tumor extension, multifocality, vascular invasion, and the presence or absence of lymph node metastasis. Tumors were staged according to the tumor, node, metastasis (ie, TNM)-based staging system recommended by the American Joint Commission on Cancer (AJCC; 7th Edition) and the Union Internationale Contre le Cancer.17 During the histopathologic review, particular attention was paid to the definition of the degree of tumor infiltration, subdividing the PTCs into 2 groups: (1) ‘‘totally encapsulated’’ when the tumor was completely surrounded by a well-defined fibrous capsule separating it from the surrounding non-neoplastic thyroid tissue and (2) ‘‘extrathyroidal extension,’’ when the tumor showed evident features of
Miccoli et al 1551
neoplastic infiltration beyond the thyroid capsule into the adjacent fibroadipose tissue or skeletal muscle (Fig 1). The immunohistochemical analyses were performed automatically using the Ventana Benchmark immunostaining system (Ventana Medical Systems, Tucson, AZ). Embedded sections (3–5 mm) were deparaffinized in xylene, dehydrated through a graded series of alcohols, and processed using a diaminobenzidine detection system (Ventana), following the manufacturer’s instructions. E-selectin immunostaining was performed using a rabbit polyclonal antibody generated from an immunogen represented by a synthetic peptide surrounding amino acid 584 of human CD62E (ab 18981; Abcam, Cambridge, UK) at a dilution of 1:100. The specificity of the E-selectin antibody was assessed using a specific blocking peptide (ab 113900; Abcam), following the manufacturer’s instructions. Diaminobenzidine was used as a chromogen, and commercial hematoxylin was used for counterstaining. Positive controls were prepared using samples of small intestine, following the manufacturer’s instruction. For the negative control, all of the described reagents were used without the inclusion of the primary antibody. At least 10 representative fields under highpower magnification (original magnification, 340) were chosen. The cellular localization of the immunostaining was recorded. E-selectin immunoreactivity was independently evaluated by 2 pathologists (L.T. and A.C.I.) who were blinded to the clinicopathologic data. Different scores between the 2 investigators were observed in less than 10% of the samples, and a consensus was achieved in all cases after discussion. Semiquantitative scores were given as the score of the percentage of positive cells plus the score of the staining intensity, as previously described.18 The E-selectin expression of each PTC sample was normalized by calculating the z-score for each final staining score. The z-score normalization functions to equalize the variability of the immunohistochemical staining among the different samples. Every final staining score was converted to a z-score by subtracting the tumor cell population mean and dividing by the standard deviation. Thus, a positive z-score indicates a value above the population mean, and a negative score indicates a value below the mean. RNA purification and complementary DNA synthesis. After standard deparaffinization, FFPE thyroid tissue sections (5 mm) were enriched by manual microdissection. Then, RNA was
1552 Miccoli et al
Surgery December 2014
Fig 1. Representative hematoxylin and eosin–stained sections and E-selectin immunohistochemical staining are shown of 2 papillary thyroid carcinomas with a different degree of neoplastic infiltration. (A–C) Totally encapsulated papillary thyroid carcinoma with a final staining score of 0. (A) The tumor appears completely surrounded by a well-defined fibrous capsule separating it from the surrounding non-neoplastic thyroid tissue (arrow indicates fibrous capsule; original magnification 31.6). (B, C) E-selectin immunostaining shows absence of immunoreactivity of the neoplastic cells at original magnification 31.6 (B) and original magnification 320 (C). (D–F) Papillary thyroid carcinoma with extrathyroidal extension and with a final staining score of 6. (D) The tumor shows evident features of neoplastic infiltration beyond the thyroid capsule into the adjacent fibroadipose tissue (asterisk indicates soft perithyroidal tissues; original magnification 31.6). (E, F) E-selectin immunostaining shows immunoreactivity of moderate intensity in most of the neoplastic cells (E) original magnification 31.6; (F) original magnification 320. Each threesome of hematoxylin and eosin–stained slides and immunohistochemical stains is from the same sample.
extracted and purified from the FFPE tissue sections using the QIAGEN RNeasy FFPE kit (QIAGEN GmbH, Hilden, Germany) in accordance with the manufacturer’s instructions. RNA integrity was assessed spectrophotometrically using a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE), with the A260/A230 ratio greater than 1.7, and the A260/A280 ratio greater than 1.8. Total RNA was converted into complementary (c)DNA using the RT2 First Strand Kit (QIAGEN GmbH) for each sample. Semiquantitative real-time reverse-transcription polymerase chain reaction. The quantification of E-selectin messenger RNA (SELE) was performed using the 2-step real-time quantitative reverse-transcription polymerase chain reaction (PCR) Rotor Gene SYBR Green PCR Kit (QIAGEN) on a Rotor Gene 6000 (QIAGEN) instrument. Quanti Tect Primer Assay (QIAGEN) was used for SELE (NM_000450), and b-actin (NM_001101) was used as an internal control. In addition, 2.5 mL of cDNA was used in a final
volume of 25 mL. Amplification conditions were in accordance with the manufacturer’s instructions. The comparative threshold cycle (Ct) method, defined as 2-DDCt, was used for the calculation of fold amplification. Each experiment was evaluated with 2 PCRs. Data are presented as the mean value ± standard deviation. The detection of BRAF mutations was performed according to standard procedures.19 Statistical analysis. Statistical analyses were performed using IBM SPSS Statistics software (version 17.0.1; IBM, Armonk, NY). A ShapiroWilk test was performed to verify the normality of distributions. A Student t test was used to compare z-score, and a Mann-Whitney U test was performed to compare final staining score. The v2 test was used in presence of categorical variables. The power tests (ex post) were performed to estimate the sample sizes. RESULTS We analyzed 88 tumors histologically diagnosed as conventional variant of PTC and whose size was
Miccoli et al 1553
Surgery Volume 156, Number 6
Table I. Clinicopathologic features of 88 patients with conventional variant of papillary thyroid carcinoma of approximately 20 mm Age at diagnosis, y Patients
Background. Cell adhesion molecules, represented by the immunoglobulin family and selectins, play an important role in the progression of cancer. A correlation between selectins and tumor aggressiveness has been demonstrated in several reports. Methods. Eighty-eight patients (mean age, 41.0 ± 14 years) with papillary thyroid carcinoma (conventional variant and sized approximately 20 mm) were divided in 2 groups: 41 with encapsulated tumors and 47 with tumors with extrathyroidal extension. E-selectin expression was evaluated by immunohistochemical staining and semiquantitative real-time reverse-transcription polymerase chain reaction and normalized by calculating the z-score (positive: value above the population mean; negative: below the mean). Results. Lymph node metastasis (LNM) was found in 2 of 41 encapsulated tumors (4.8%) and in 19 of 47 tumors (40.4%) with extrathyroidal extension. BRAF mutation was present in 21 encapsulated tumors (51.2%) and in 31 tumors with extrathyroidal extension (65.9%). The mean E-selectin z-score was 0.32 for encapsulated tumors and 0.28 for tumors with extrathyroidal extension. E-selectin expression correlates with neoplastic infiltration (P = .04), the American Joint Commission on Cancer stage (P = .02), and BRAF mutation (P = .03). Conclusion. E-selectin overexpression in association with BRAF mutation status could promote a more aggressive phenotype in papillary thyroid carcinoma. (Surgery 2014;156:1550-8.) From the Department of Surgical, Medical, Molecular and Critical Area Pathology,a and Department of Translational Research and of New Surgical and Medical Technologies,b Universit a degli Studi di Pisa, Pisa, Italy
PAPILLARY THYROID CARCINOMA (PTC) is the most common type of thyroid cancer, accounting for more than 80% of all thyroid malignancies.1 Despite the excellent prognosis, more aggressive forms of the disease develop in some patients, including local and distant metastasis or tumor recurrence.2 One of the most important independent risk factors for an unfavorable outcome is extrathyroidal extension, defined as tumor penetration through the thyroid capsule into the adjacent fibroadipose tissue or skeletal muscle.3,4 The presence of tumor capsule, however, seems Accepted for publication August 19, 2014. Reprint requests: Paolo Miccoli, MD, Professor of Surgery, Department of Surgical, Medical, Molecular and Critical Area Pathology, Universita di Pisa, Via Paradisa 2, 56126 Pisa, Italy. E-mail: [email protected]. 0039-6060/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.surg.2014.08.049
1550 SURGERY
to be correlated to a less aggressive behavior of the neoplasm in lymph node metastases and tumor recurrence.5 A strong association between cell adhesion molecules and the mechanism of neoplastic progression and metastasis has been suggested.6-8 These molecules were first reported as receptors modulating leukocyte traffic.9,10 They are subdivided mainly into 2 groups: the immunoglobulin family, including intercellular adhesion molecule and vascular cell adhesion molecule, and selectins. The selectins comprise a family of 3 cell adhesion molecules: E-selectin, present in endothelial cells; P-selectin, in platelets and endothelial cells; and L-selectin, in leukocytes. Their structure is similar: the transmembrane domain anchors selectins to the membrane, and the cytoplasmic tail supports their action as signaling agents.8-10 Selectins are expressed when certain mediators are present, such as interleukins, tumor necrosis factor, or toxins. Selectins play a critical role in leukocyte
Surgery Volume 156, Number 6
recruitment, mediating leukocyte arrest in the vasculature, extravasation, and migration to inflamed tissue.10 Knowledge about leukocyte locomotion has been weaned from the understanding of tumor cell extravasation and metastasis.11 Numerous studies have reported the implication of some selectins (ie E-selectin) in neoplastic progression from primary to metastatic disease in different types of carcinomas, such as head and neck squamous cell carcinoma,6 colorectal carcinoma,13,14 and gastric carcinoma.15 We conducted this study to verify whether the overexpression of E-selectin is linked to a more aggressive tumor behavior in patients presenting a PTC with the same pathologic features. METHODS Formalin-fixed and paraffin-embedded (FFPE) thyroid specimens from 88 patients (20 male, 68 female) with a conventional variant of PTC who underwent operation at the Department of Surgical Pathology from January 2006 to December 2012 were retrospectively reviewed. Patients were a mean age of 41.0 ± 14 years (range, 12–78 years). The mean tumor size was 18.6 ± 1.4 mm (range, 15.0–20.0 mm). This study was approved by the Institutional Review Board, and informed consent was obtained from all patients. All patients had undergone a total thyroidectomy and a full exploration of the central compartment on both sides. A lymphadenectomy was performed when enlarged lymph nodes were present preoperatively or intraoperatively. Histopathologic review was conducted by 2 pathologists (L.T. and A.C.I). The diagnoses were reassessed according to the World Health Organization’s classification of thyroid malignancy.16 Hematoxylin and eosin–stained sections were re-examined, taking into account the known risk factors for thyroid cancer, such as tumor size, tumor extension, multifocality, vascular invasion, and the presence or absence of lymph node metastasis. Tumors were staged according to the tumor, node, metastasis (ie, TNM)-based staging system recommended by the American Joint Commission on Cancer (AJCC; 7th Edition) and the Union Internationale Contre le Cancer.17 During the histopathologic review, particular attention was paid to the definition of the degree of tumor infiltration, subdividing the PTCs into 2 groups: (1) ‘‘totally encapsulated’’ when the tumor was completely surrounded by a well-defined fibrous capsule separating it from the surrounding non-neoplastic thyroid tissue and (2) ‘‘extrathyroidal extension,’’ when the tumor showed evident features of
Miccoli et al 1551
neoplastic infiltration beyond the thyroid capsule into the adjacent fibroadipose tissue or skeletal muscle (Fig 1). The immunohistochemical analyses were performed automatically using the Ventana Benchmark immunostaining system (Ventana Medical Systems, Tucson, AZ). Embedded sections (3–5 mm) were deparaffinized in xylene, dehydrated through a graded series of alcohols, and processed using a diaminobenzidine detection system (Ventana), following the manufacturer’s instructions. E-selectin immunostaining was performed using a rabbit polyclonal antibody generated from an immunogen represented by a synthetic peptide surrounding amino acid 584 of human CD62E (ab 18981; Abcam, Cambridge, UK) at a dilution of 1:100. The specificity of the E-selectin antibody was assessed using a specific blocking peptide (ab 113900; Abcam), following the manufacturer’s instructions. Diaminobenzidine was used as a chromogen, and commercial hematoxylin was used for counterstaining. Positive controls were prepared using samples of small intestine, following the manufacturer’s instruction. For the negative control, all of the described reagents were used without the inclusion of the primary antibody. At least 10 representative fields under highpower magnification (original magnification, 340) were chosen. The cellular localization of the immunostaining was recorded. E-selectin immunoreactivity was independently evaluated by 2 pathologists (L.T. and A.C.I.) who were blinded to the clinicopathologic data. Different scores between the 2 investigators were observed in less than 10% of the samples, and a consensus was achieved in all cases after discussion. Semiquantitative scores were given as the score of the percentage of positive cells plus the score of the staining intensity, as previously described.18 The E-selectin expression of each PTC sample was normalized by calculating the z-score for each final staining score. The z-score normalization functions to equalize the variability of the immunohistochemical staining among the different samples. Every final staining score was converted to a z-score by subtracting the tumor cell population mean and dividing by the standard deviation. Thus, a positive z-score indicates a value above the population mean, and a negative score indicates a value below the mean. RNA purification and complementary DNA synthesis. After standard deparaffinization, FFPE thyroid tissue sections (5 mm) were enriched by manual microdissection. Then, RNA was
1552 Miccoli et al
Surgery December 2014
Fig 1. Representative hematoxylin and eosin–stained sections and E-selectin immunohistochemical staining are shown of 2 papillary thyroid carcinomas with a different degree of neoplastic infiltration. (A–C) Totally encapsulated papillary thyroid carcinoma with a final staining score of 0. (A) The tumor appears completely surrounded by a well-defined fibrous capsule separating it from the surrounding non-neoplastic thyroid tissue (arrow indicates fibrous capsule; original magnification 31.6). (B, C) E-selectin immunostaining shows absence of immunoreactivity of the neoplastic cells at original magnification 31.6 (B) and original magnification 320 (C). (D–F) Papillary thyroid carcinoma with extrathyroidal extension and with a final staining score of 6. (D) The tumor shows evident features of neoplastic infiltration beyond the thyroid capsule into the adjacent fibroadipose tissue (asterisk indicates soft perithyroidal tissues; original magnification 31.6). (E, F) E-selectin immunostaining shows immunoreactivity of moderate intensity in most of the neoplastic cells (E) original magnification 31.6; (F) original magnification 320. Each threesome of hematoxylin and eosin–stained slides and immunohistochemical stains is from the same sample.
extracted and purified from the FFPE tissue sections using the QIAGEN RNeasy FFPE kit (QIAGEN GmbH, Hilden, Germany) in accordance with the manufacturer’s instructions. RNA integrity was assessed spectrophotometrically using a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE), with the A260/A230 ratio greater than 1.7, and the A260/A280 ratio greater than 1.8. Total RNA was converted into complementary (c)DNA using the RT2 First Strand Kit (QIAGEN GmbH) for each sample. Semiquantitative real-time reverse-transcription polymerase chain reaction. The quantification of E-selectin messenger RNA (SELE) was performed using the 2-step real-time quantitative reverse-transcription polymerase chain reaction (PCR) Rotor Gene SYBR Green PCR Kit (QIAGEN) on a Rotor Gene 6000 (QIAGEN) instrument. Quanti Tect Primer Assay (QIAGEN) was used for SELE (NM_000450), and b-actin (NM_001101) was used as an internal control. In addition, 2.5 mL of cDNA was used in a final
volume of 25 mL. Amplification conditions were in accordance with the manufacturer’s instructions. The comparative threshold cycle (Ct) method, defined as 2-DDCt, was used for the calculation of fold amplification. Each experiment was evaluated with 2 PCRs. Data are presented as the mean value ± standard deviation. The detection of BRAF mutations was performed according to standard procedures.19 Statistical analysis. Statistical analyses were performed using IBM SPSS Statistics software (version 17.0.1; IBM, Armonk, NY). A ShapiroWilk test was performed to verify the normality of distributions. A Student t test was used to compare z-score, and a Mann-Whitney U test was performed to compare final staining score. The v2 test was used in presence of categorical variables. The power tests (ex post) were performed to estimate the sample sizes. RESULTS We analyzed 88 tumors histologically diagnosed as conventional variant of PTC and whose size was
Miccoli et al 1553
Surgery Volume 156, Number 6
Table I. Clinicopathologic features of 88 patients with conventional variant of papillary thyroid carcinoma of approximately 20 mm Age at diagnosis, y Patients
