This article is mentioned in:

Introduction

Human epidermal growth factor receptor 2 (HER-2) is a 185-kDa transmembrane glycoprotein with tyrosine kinase activity. HER-2 is encoded by the c-erbB-2 proto-oncogene (1). HER-2 overexpression was first detected in human breast and ovarian cancers (2) and was also demonstrated in a wide range of human adenocarcinomas originating in different organs, such as the stomach, colon, lung and pancreas (3–5). HER-2 is generally an important factor for the growth and progression of malignant tumors and its overexpression in lung and breast cancer appears to be associated with an increased metastatic potential of the cells (6). The greatest value of HER-2 status as a predictive marker is in the prediction of response to therapies that specifically target HER-2, such as trastuzumab and lapatinib, in breast and gastric cancer (7–9).

Differentiated thyroid cancer (DTC), including papillary and follicular subtypes, accounts for 90% of all thyroid malignancies (10). DTC is generally associated with good prognosis and long-term survival rates >90% (11). However, despite their favorable prognosis, patients with DTC are at risk of tumor recurrence for several decades from the time of diagnosis. The therapeutic management of recurrent and metastatic DTC depends on the type of the initial treatment, the location and extent of the disease. Radioactive iodine (131I) administration and/or surgical excision of the lesions remain the first and main approach (12). A number of advanced thyroid cancers eventually develop lack of iodine avidity. Until recently, chemotherapy was the only option for systemic treatment, but only achieved low response rates and some metastatic patients were referred for best supportive care.

A element common to thyroid cancers is their vascularity, with elevated levels of vascular endothelial growth factor compared to those in normal thyroid tissue, which is the rationale behind the use of antiangiogenic drugs, such as axitinib and sorafenib, in the treatment of advanced DTC. These two drugs were found to achieve response rates of 30 and 23% respectively (13,14).

There is no consensus in the currently available literature regarding the expression of HER-2 and its prognostic and predictive value in thyroid cancer, although expression rates of 0–50% were reported by previous studies (15,16). These contradictory findings prompted us to investigate the HER-2 expression status in a retrospective series of DTC, including follicular and papillary subtypes.

Our second aim was to determine the association between HER-2 expression and the clinicopathological characteristics of DTC and determine whether HER-2 may be used as a prognostic factor and/or as a target for treatment with biological-targeted therapy with trastuzumab or other drugs in DTC patients.

Patients and methods

Patients and tumors

In this retrospective study, biopsies were collected from surgically removed DTCs from 69 patients treated at the Ziv Medical Center between 1999 and 2007. The patients exhibited no distant (visceral) metastases at the time of diagnosis. Histological classification was performed at our hospital according to the World Health Organization classification (17). Patient data at all the stages of the disease were retrospectively collected and summarized. The hospital records provided clinical and demographic data that included ethnicity, histology, age, surgical procedure, physical findings and disease stage according to the International Union Against Cancer (18). The pathological data included primary tumor diameter, histological type, tumor size, number of lymph nodes involved and tumor grade. HER-2 expression was evaluated using immunohistochemistry (IHC) or chromogenic in situ hybridization (CISH) if HER-2 expression was classed as +1 and +2 in IHC. Formalin-fixed, paraffin-embedded (FFPE) tissue was used for both tests. HER-2 was considered to be overexpressed if it was classed as +3 by IHC, using the scoring system described below, or if positive amplification was shown in the CISH test.

IHC

IHC was performed using the HercepTest (DakoCytomation, Carpinteria, CA, USA), following the manufacturer’s instructions. Briefly, this procedure includes deparaffinization and rehydration steps, followed by an epitope retrieval step, in which the tissue sample is incubated in EDTA buffer solution at >95°C for 20 min. The slide is then subjected to a series of alternating washes in phosphate-buffered saline (PBS) and incubation steps, beginning with a peroxidase-blocking reagent for 10 min and followed by a polyclonal antibody to HER-2 (A0485; DakoCytomation) applied to the sections at a 1:200 dilution in PBS and incubated for 60 min at room temperature, followed by a visualization reagent (dextran polymer) conjugated with horseradish peroxidase for 30 min and finally with a 3,3′-diaminobenzidine chromogen solution for 10 min. After a final wash, the slide is counterstained with hematoxylin.

Tumor cells with circumferential membranous positivity were considered as overexpressing the HER-2 protein and scoring was performed according to the HercepTest manufacturer’s recommendations by pathologists experienced in breast pathology and immunohistochemical interpretation of HER-2 testing. According to the HercepTest criteria, immunoreaction was scored as follows: +3, >10% of tumor cells exhibiting strong and complete membranous staining; +2, moderate and complete membrane positivity in >10% cells; +1, weak and incomplete membrane positivity in >10% cells; and 0, membrane staining absent or present in <10% cells. Tumors scored as +3 were considered to be HER-2-positive, whereas tumors scored as 0/+1 were designated as HER-2-negative. In +2 tumors, as determined by IHC, CISH analysis was performed.

CISH

CISH allows detection of gene amplification, chromosome translocation and chromosome number using conventional peroxidase reactions under a bright-field microscope on FFPE tissues cut in 4–5-mm sections. The essence of CISH is the ability of labeled nucleic acid probes to hybridize, in situ, to specific sections of complementary nucleic acid in the sample. The probe hybridization results may be visualized within the context of the surrounding tissue morphology. Therefore, pathologists may view tissue morphology and gene aberrations simultaneously (18). The 84-0146 SPoT-Light1 HER-2 CISH™ kit (Zymed Laboratories, Invitrogen immunodetection, San Francisco, CA, USA) was used. The CISH staining results may be clearly visualized under a standard bright-field microscope and a 406 dry lens. Tumors with HER-2 gene amplification typically appear as a large peroxidase-positive intranuclear gene copy clusters, as numerous individual peroxidase-positive small signals, or as a mixture of clusters and individual gene copies. Tumors with a normal HER-2 gene status typically exhibit two signals or dots/nucleus.

Statistical analysis

The data were analyzed in association with histological type and other clinicopathological parameters. For evaluating clinical and pathological parameters, and the association with HER-2 levels, the χ2 test was used. P<0.05 was considered to indicate a statistically significant difference.

Results

Patient characteristics

A total of 69 cases were investigated and the patient characteristics are presented in Table I. The mean age at diagnosis was 44.6 years and 76.8% of the patients were female. Papillary carcinoma was more common compared to follicular carcinoma (84.1 and 15.9%, respectively). The majority of the patients were subjected to total or subtotal thyroidectomy. Of all the patients, 94.2% were treated with 131I ablation after surgery. In 63.8% of the patients the tumor was sized <2 cm and 15.9% had lymph node metastases (Table II). One patient had lung metastases diagnosed 4 months after the operation.

Table I Patients characteristics.

Table I

Patients characteristics.

Characteristics	No. (%)
Age at diagnosis (years)
Mean	44.6±14.4
Range	20–74
Gender
Female	53 (76.8)
Male	16 (23.2)
Histology
Papillary carcinoma	58 (84.1)
Follicular carcinoma	11 (15.9)
Type of operation
Total thyroidectomy	30 (43.5)
Subtotal thyroidectomy	24 (34.8)
Near-total thyroidectomy	13 (18.8)
Hemithyroidectomy	2 (2.9)
131I ablation after surgery
Yes	65 (94.2)
No	4 (5.8)

[i] ¹³¹I, radioactive iodine.

Table II Stage of the disease at diagnosis.

Table II

Stage of the disease at diagnosis.

Characteristics	No. (%)
Size of primary tumor
T1	44 (63.8)
T2	13 (18.8)
T3	10 (14.5)
T4	2 (2.9)
Lymph node status
N0	58 (84.0)
N1	11 (16.0)
Distant metastases
Lungs	1 (1.5)
Capsular invasion
No	50 (72.5)
Yes	19 (27.5)

HER-2 expression and correlation with clinicopathological characteristics

In 4 (6.9%) of the 58 patients with papillary carcinoma, HER-2 was overexpressed (+3) but the proportion of patients was not considered significant. One of these 4 patients was male and another patient had lymph node metastases. There was no HER-2 overexpression in any of the 11 patients with follicular carcinoma (Table III).

Table III HER-2 expression in thyroid cancer tissue.

Table III

HER-2 expression in thyroid cancer tissue.

HER-2 status	No. (%)
Papillary carcinoma (n=58)
Negative	54 (93.1)
Positive	4 (6.9)
Follicular carcinoma (n=11)
Negative	11 (100.0)
Positive	0 (0.0)
Total no. (n=69)
Negative	65 (94.2)
Positive	4 (5.8)

Five tumors were scored as +2 and 10 as +1 by IHC for HER-2. All these tumors were evaluated by CISH. No gene amplification was found in either of these tumors. There was no correlation between HER-2 expression and tumor size, lymph node metastases, age and gender.

Discussion

The role of the HER-2 proto-oncogene in thyroid cancer remains controversial, particularly regarding the prognostic significance of its amplification and overexpression. Haugen et al (16) investigated the c-erbB-2 protein expression in frozen thyroid cancer specimens and detected immunoreactivity at the cell membrane as well as the cytoplasm in papillary thyroid carcinomas. Additionally, in those specimens, the HER-2/neu mRNA levels were analyzed and found to be significantly elevated. Kremser et al (19) observed a highly significant correlation between the HER-2/neu expression in papillary and follicular thyroid carcinomas and the clinical course regarding metastatic disease. Both tumor entities without distant metastases exhibited non-significant cytoplasmic immunostaining compared to patients with metastases. There was no association identified between patient gender and HER-2 expression, although a significant correlation with age was identified. Advanced tumor stage presented with a tendency towards higher HER-2 expression. Cervical lymph node metastases did not correlate with HER-2 levels. Furthermore, there was no correlation between HER-2 expression and any site of metastasis, histological subtype or variant (19).

Similar to our findings, previous studies reported no HER-2 overexpression in thyroid tumors. Lemoine et al (15) investigated a large series of benign and malignant thyroid tumors by means of Southern blot analysis, polymerase chain reaction and IHC, without detection of abnormalities in structure or expression of c-erbB-2, and concluded that activation of this oncogene is not necessarily involved in neoplastic transformation or the prognosis of thyroid tumors. However, in our cases, 4 of the 58 (6.9%) patients with papillary carcinoma overexpressed HER-2, as determined by IHC. No correlation was found in these 4 cases between HER-2 and tumor size, lymph node metastasis, age and gender. In addition, there was no HER-2 overexpression in the 11 patients with follicular carcinoma.

Utrilla et al (20) investigated 67 thyroid tumors (10 follicular adenomas, 11 follicular carcinomas, 3 anaplastic carcinomas, 25 papillary carcinomas, 18 medullary carcinomas and 16 metastases) for HER-2 expression by IHC using an antigen aretrieval method for paraffin-embedded material and a specific polyclonal antibody against the intracytoplasmic part of the HER-2 oncoprotein. No expression of this oncoprotein was detected in follicular adenomas, follicular carcinomas and anaplastic carcinomas; however, 52% of the papillary carcinomas (some membranous staining, but mostly diffuse cytoplasmic staining patterns) and all the medullary carcinomas (granular cytoplasmic staining pattern) were immunopositive. Sugg et al (21) reported no DNA amplifications of the c-erbB-2 gene in 47 human thyroid tumors, including papillary and follicular carcinoma. There was also no membrane staining with NCL-CBE1, a monoclonal antibody specific for the external domain of HER-2. Consistent with previous studies, they found that the cytoplasmic immunoreactivity of HER-2 correlates with differentiation of DTCs and may be a prognostic factor in this disease. The occurrence of cytoplasmic immunoreactivity of HER-2 has not been fully elucidated. De Potter et al (22) discovered a 155-kDa (p155) protein reacting with the antibody. p155 may be localized in the mitochondrial cristae, explaining the partly granular aspect of the staining. In bladder cancer, c-erbB-2 amplification, high mRNA levels and strong cytoplasmic immunoreactivity were observed, correlating with the high levels of p155. This may be attributed to reactions with a homologue of the c-erbB-2 protein, which may be due to an alteration or instability of the oncoprotein itself or the mRNA (23).

In conclusion, according to the present study and other available publications, HER-2 is not considered suitable to be used routinely as a prognostic or predictive factor in DTC. We detected HER-2 overexpression in only a few cases of papillary thyroid carcinoma, but its levels do not appear to correlate with other parameters of aggressive behavior. There were no HER-2-positive cases of follicular carcinoma. Therefore, only a limited number of patients with papillary DTC refractory to 131I may have HER-2-positive tumors and may benefit from biological-targeted therapy. We are planning a new study to evaluate HER-2 expression in metastatic DTC. However, the profile of other epidermal growth factor receptors in different thyroid cancer types requires further investigation.

References