Immunohistochemical levels of cyclo-oxygenase-2,matrix metalloproteinase-9 and vascular endothelial growth factor in papillary thyroid carcinoma and their clinicopathological correlations

Abstract

Objective

To measure the levels of cyclo-oxygenase (COX)-2, matrix metalloproteinase (MMP)-9 and vascular endothelial growth factor (VEGF) immunostaining in papillary thyroid carcinoma (PTC) and benign thyroid tumours, and to investigate potential correlations between their levels and clinicopathological characteristics.

Methods

The levels of immunohistochemical staining of COX-2, MMP-9 and VEGF proteins were measured in tumours from patients with PTC and compared with specimens from patients with benign thyroid tumours. The association between the levels of COX-2, MMP-9 and VEGF proteins and clinicopathological characteristics in patients with PTC was also analysed.

Results

A total of 66 patients with PTC and 40 patients with benign thyroid tumours participated in the study. The rates of positive immunostaining for COX-2, MMP-9 and VEGF in PTC tumours were significantly higher than those in benign thyroid tumours. There were significant positive associations between positive immunostaining for COX-2, MMP-9 and VEGF proteins and age (≥45 years), clinical stage (III–IV) and tumour diameter (≥2 cm).

Conclusion

Combined immunohistochemical evaluation of the levels of COX-2, MMP-9 and VEGF in PTC might be a useful marker for the diagnosis of PTC.

Keywords

Papillary thyroid carcinoma cyclo-oxygenase-2 matrix metalloproteinase-9 vascular endothelial growth factor

Introduction

Thyroid cancer is the most common malignancy of the endocrine system.¹ Papillary thyroid carcinoma (PTC) accounts for approximately 75–85% of thyroid cancers.² PTCs originate from the thyroid follicle cells and they are the most frequently seen malignancies in patients whose neck regions have been exposed to radiation.² PTCs are more prevalent in those <40 years old and are two to four times more common in women than men.³ In addition, PTC is the eighth most common cancer in women.⁴ Hence, it is important to understand the molecular mechanisms responsible for PTC development, progression and metastasis in order to be able to develop novel strategies for the early detection, prevention and treatment of this common thyroid cancer.

Angiogenesis is crucial for tumour growth and metastasis.⁵ It can be stimulated by several endogenous biological regulators including vascular endothelial growth factor (VEGF), transforming growth factor-β, platelet-derived growth factor, basic fibroblast growth factor, interleukin-8 and angiogenin.⁶ The most important of these angiogenic molecules is VEGF and it has considerable potential for research in the field of tumour biology.⁶

Cyclo-oxygenase (COX)-2 is a COX isoform that plays an important role in carcinogenesis.^7,8 COX-2 activity decreases intracellular adhesion and apoptosis, and increases angiogenesis and cell proliferation.^9,10 Matrix metalloproteinases (MMPs) are a diverse family of enzymes capable of degrading various components of the extracellular matrix (ECM).¹¹ MMP production has been found to be upregulated in several human tumours and correlates with advanced stage, invasion, angiogenesis, metastatic properties and poor prognosis.¹¹ Among the secreted MMPs, MMP-2 and MMP-9 are known to play a key role in tumour invasion and metastasis development.¹²

A novel correlation between COX-2 and VEGF has been reported in human lung adenocarcinoma and oesophageal squamous cell carcinoma, with COX-2 upregulating VEGF production.¹³ Some studies have indicated that there is a close relationship between VEGF and MMP expression.^12,14 On the basis of these studies, it is conceivable that VEGF expression may correlate with COX-2 and MMP expression. However, the relationships between VEGF expression and COX-2 and MMP expression in PTC have not been reported previously.

The aim of the present study was to compare the immunohistochemical levels of COX-2, MMP-9 and VEGF proteins in specimens of PTC with those in specimens of benign thyroid tumours. The study also investigated the correlations between the immunohistochemical levels of COX-2, MMP-9 and VEGF and several clinicopathological characteristics, including age, sex, tumour size and lymph node metastasis, in patients with PTC.

Patients and methods

Patients

This prospective study included analysis of surgical specimens excised from consecutive patients with PTC and patients with benign thyroid tumours between July 2007 and October 2012 in the Department of Thyroid Surgery, First Affiliated Hospital, Jilin University, Changchun, Jilin Province, China. Clinical staging was based on the tumour node metastasis (TNM) classification system of the International Union Against Cancer.¹⁵ All patients with PTC who received PTC radical surgery and central and/or lateral cervical lymph node clearance were enrolled in the study. All case information was retrieved from the patients’ medical records. Samples of benign thyroid tumours from patients with thyroid adenoma or nodular goitre were collected from the patients during surgical removal of the benign tumour mass.

The study procedures were approved by the Ethical Committee of the First Affiliated Hospital, Jilin University (no. J5247-23). Written informed consent was obtained from all of the study participants.

Immunohistochemistry

Primary tumour specimens were fixed in 10% buffered formalin, dehydrated in ethanol, embedded in paraffin wax and cut into 4-µm sections for use in immunohistochemical staining. The sections were dewaxed in xylene and rehydrated in a descending series of alcohol solutions. Antigen retrieval was achieved by boiling the sections for 10 min in 0.01 M citrate buffer (pH 6.0) (Wuhan Boster Biological Technology, Wuhan, Hubei, China) and endogenous peroxidase was blocked by incubation in 0.3% hydrogen peroxide (Wuhan Boster Biological Technology) in methanol for 30 min at room temperature. After blocking with 3% bovine serum albumin (Wuhan Boster Biological Technology) in 0.01 m M phosphate buffered saline (PBS; pH 7.2) for 2 h at 37℃, the sections were incubated with mouse monoclonal antibodies against human COX-2, MMP-9 and VEGF proteins (all from Santa Cruz Biotechnology, Santa Cruz, CA, USA) at a dilution of 1:500 overnight at 4℃. The samples were then washed three times in 0.01 mM PBS (pH 7.2) for 5 min. The sections were incubated with biotinylated rabbit antimouse antibody (1:100 dilution; ZSGB-BIO, Beijing, China) for 2 h at 37℃. The slides were then washed three times with 0.01 mM PBS (pH 7.2). The immunolabelling was visualized by incubation with 0.05% diaminobenzidine (Invitrogen, Carlsbad, CA, USA) in 0.01 mM PBS (pH 7.2) for 5 min at room temperature and the slides were then rinsed for 2 min under running tap water. The immunohistochemical staining was examined at ×100 and ×400 magnifications using an Olympus BX51 light microscope (Olympus Optical, Tokyo, Japan). The percentage of positive cells was calculated semiquantitatively by counting the number of labelled cells in eight randomly selected high-power fields for each specimen, at ×400 magnification.

The degree of immunoreactivity was assessed similarly to the system described previously.¹⁶ The intensity of the staining was graded on a 4-point scale as 0, 1+, 2+ or 3+. The extent of immunostaining was considered as: 0 if there were no immunoreactive cells; 1+ if the proportion of immunoreactive cells was <25%; 2+ if there were 25–75% immunoreactive cells; and 3+ if there were >75% immunoreactive cells. In cases with variable staining intensities, the most common pattern was recorded. Specimens with 0 or 1+ staining were classified as negative, and specimens with 2+ or 3+ staining were classified as positive. All slides were scored independently by two observers (Q.L. and S.L.). Five specimens of PTC with discordant results were re-evaluated to obtain consensus between the two observers.

Statistical analyses

All statistical analyses were performed using the SPSS® statistical package, version 13.0 (SPSS Inc., Chicago, IL, USA) for Windows®. The data are presented as n (%). Pearson’s χ²-test was used to compare the data between groups. Spearman’s rank correlation coefficient analysis was used to investigate the potential correlation between the levels of COX-2, MMP-9 and VEGF protein immunostaining. A P-value <0.05 was considered statistically significant.

Results

A total of 75 consecutive patients with PTC and 40 patients with benign thyroid tumours (20 with thyroid adenoma and 20 with nodular goitres) were enrolled in the study. The appropriate clinical information and follow-up data were available for 69 (92%) patients with PTC; of these, three were excluded from the analysis because two had cystic small papillary carcinoma, and one had histology indicative of the tall cell variant of PTC and anaplastic carcinoma. Therefore, 66 patients (31 male and 35 female) with PTC and 40 patients (13 male and 27 female) with benign thyroid tumours were included in this immunohistochemical study. The mean age of the 66 patients with PTC was 45.8 years (range 19–68 years). The mean age of the 40 patients with benign thyroid tumours was 42.1 years (range 21–72 years). Demographic and clinicopathological characteristics of the patients with PTC are presented in Table 1.

Table 1.

The levels of immunohistochemical staining of cyclo-oxygenase (COX)-2, matrix metalloproteinase (MMP)-9 and vascular endothelial growth factor (VEGF) proteins in patients with benign thyroid tumours (n = 40) and papillary thyroid carcinomas (PTC; n = 66) and the associations between the immunohistochemical staining levels and clinicopathological characteristics in patients with PTC.

Parameter	n	COX-2		Statistical significance^a	MMP-9		Statistical significance^a	VEGF		Statistical significance^a
Pathological tumour type	–	+	–	+	–	+
Papillary thyroid carcinoma (group 1)	66	7	59	1 vs 2 P < 0.001	5	61	1 vs 2 P < 0.001	8	58	1 vs 2 P < 0.001
Thyroid adenoma (group 2)	20	16	4	1 vs 3 P < 0.001	14	6	1 vs 3 P < 0.001	14	6	1 vs 3 P < 0.001
Nodular goitre (group 3)	20	14	6	2 vs 3 P = 0.012	18	2	2 vs 3 P = 0.05	18	2	2 vs 3 NS
Clinicopathological characteristics in patients with PTC (n = 66)
Lymph node metastasis
Negative	32	5	27	NS	2	30	NS	5	27	NS
Positive	34	2	32	3	31	3	31
Tumour diameter, cm
<2.0	27	5	22	P < 0.001	3	24	P < 0.001	2	25	P = 0.031
≥2.0	39	2	37	2	37	6	33
Clinical TNM stage
I–II	40	6	34	P = 0.003	5	35	P = 0.006	8	32	P < 0.001
III–IV	26	1	25	0	26	0	26
Sex
Male	31	4	27	NS	1	30	NS	5	26	NS
Female	35	3	32	4	31	3	32
Age, years
<45	37	6	31	P = 0.004	5	32	P = 0.003	8	29	P < 0.001
≥45	29	1	28	0	29	0	29

Data presented as number of patients.

Pearson’s χ²-test; between-group comparisons for pathological tumour type shown as group 1 versus group 2 (1 vs 2), group 1 versus group 3 (1 vs 3) and group 2 versus group 3 (2 vs 3).

TNM, tumour node metastasis classification system; NS, no significant between-group differences (P ≥ 0.05).

Using immunohistochemistry, very little positive COX-2 immunostaining was observed in benign thyroid tissues (Figure 1A). Positive COX-2 immunostaining was detected in 59 of 66 (89.4%) PTC tumour specimens (Figure 1B and Table 1). COX-2 immunostaining was diffusely positive in all 59 PTC tumour specimens, but was negative in seven PTC tumour specimens. Positive COX-2 immunostaining was detected in six of 20 (30.0%) nodular goitres and in four of 20 (20.0%) thyroid adenomas. In terms of MMP-9 immunostaining, eight of 40 (20.0%) benign thyroid tumours were positively stained (thyroid adenoma: six of 20 [30.0%]; nodular goitre: two of 20 [10.0%]; Figure 1C) compared with 61 of 66 (92.4%) PTC tumour specimens (Figure 1D). Positive VEGF immunostaining was evident in eight of 40 (20.0%) benign thyroid tumours (thyroid adenoma: six of 20 [30.0%]; nodular goitre: two of 20 [10.0%]; Figure 1E) compared with 58 of 66 (87.9%) PTC tumour specimens (Figure 1F). These data showed that the rates of positive immunostaining for COX-2, MMP-9 and VEGF in PTC tumour specimens were significantly higher than those of benign thyroid tumours (P < 0.001 for all comparisons; Table 1).

Figure 1.

Representative photomicrographs showing immunohistochemical staining for cyclo-oxygenase-2, matrix metalloproteinase-9 and vascular endothelial growth factor proteins in benign thyroid adenomas (A, C and E, respectively) and in papillary thyroid carcinomas (B, D and F, respectively). The colour version of this figure is available at: http://imr.sagepub.com.

This present study included analysis of the relationship between the COX-2, MMP-9 and VEGF protein immunostaining levels and clinicopathological variables, including sex, age, clinical TNM stage, tumour diameter and the presence of lymph node metastasis in patients with PTC using Pearson’s χ²-test. Positive immunostaining for COX-2, MMP-9 and VEGF proteins was significantly associated with age (≥45 years), clinical TNM stage (III–IV), and tumour diameter (≥2 cm) (P < 0.05 for all comparisons; Table 1). There were no significant associations between the COX-2, MMP-9 and VEGF protein immunostaining levels and sex or the presence of lymph node metastasis.

The potential correlation between the levels of COX-2, MMP-9 and VEGF protein immunostaining in all patients with PTC was assessed using Spearman’s rank correlation coefficient analysis. COX-2 immunostaining was positively correlated with VEGF (r = 0.578, P < 0.01) and MMP-9 immunostaining (r = 0.543, P < 0.01), while MMP-9 immunostaining was also positively correlated with VEGF immunostaining (r = 0.471, P < 0.01).

Discussion

This present study demonstrated that the rates of positive immunostaining for COX-2, MMP-9 and VEGF proteins in PTC tumour specimens were significantly higher than those of benign thyroid tumours. There were also significant positive associations between positive immunostaining for COX-2, MMP-9 and VEGF proteins and age (≥45 years), clinical TNM stage (III–IV) and tumour diameter (≥2 cm). Spearman’s rank correlation coefficient analysis demonstrated that COX-2 immunostaining was positively correlated with VEGF and MMP-9 immunostaining, and MMP-9 immunostaining was positively correlated with VEGF immunostaining.

VEGF, a potent mitogenic and angiogenic agent, is capable of inducing endothelial cell proliferation, increasing vascular permeability, modifying the status of the ECM and altering gene expression.^17,18 Several studies have reported VEGF overexpression in PTCs, but no correlations between VEGF immunostaining were observed with sex, age, tumour diameter and the presence of metastasis.^19–21 In contrast, the present study demonstrated a correlation between positive VEGF immunostaining and age, clinical TNM stage and tumour diameter. In a study of 19 cases, consisting of 15 papillary and four follicular thyroid carcinomas, VEGF immunostaining was more frequently seen in PTC and was closely correlated with tumour diameter, which was in agreement with the present results.²²

The presence of COX-2 protein has been demonstrated in various types of thyroid carcinomas but the levels were low in non-neoplastic thyroid tissue from tumour samples and it was not present in normal or benign thyroid tissue.²³ Subsequently, it has been found to be occasionally produced in benign follicular cells, including follicular adenoma and thyroiditis.^24,25 COX-2 was more frequently upregulated in PTC compared with other types of thyroid malignancy.^24,25 In addition, a previous study reported that COX-2 levels were higher in tumours from older patients with PTC.²⁶ In the present study, there was a correlation between the levels of COX-2 immunostaining and age, and with clinical TNM stage and tumour diameter. Several studies have reported that VEGF and COX-2 are coexpressed in tumour tissues, and that COX-2 modulates VEGF expression.^27,28 A further study demonstrated that VEGF stimulated the release of COX-2 in endothelial cells by increasing COX-2 transcription and prolonging the half life of COX-2 mRNA.²⁹ A novel correlation between COX-2 and VEGF-C was reported in patients with PTC³⁰; in addition, COX-2 was found to upregulate VEGF-C expression,³⁰ which was in agreement with the current findings that COX-2 immunostaining was positively correlated with VEGF immunostaining. In contrast, however, Erdem et al. found a correlation between COX-2 and VEGF expression only in negative staining groups (r = 0.3, P = 0.02).²² This might be because the PTC sample number and quality were different from those in our study.

The gelatinases MMP-2 and MMP-9, which are two important isoforms of the MMP family, are considered to be closely correlated with the pathological mechanisms involved in tumour invasion and metastasis.³¹ VEGF has been shown to increase the release of MMP-2 in brain and ovarian tumour cells.^11,12 In smooth muscle cells, VEGF promotes MMP-9 mRNA transcription and protein activity.³² This present study demonstrated that MMP-9 immunostaining was positively correlated with VEGF immunostaining: this supports the theory that VEGF overexpression might increase the production and activity of MMP-9.In addition, our results showed that COX-2 immunostaining was positively correlated with VEGF immunostaining.

Although our study showed that COX-2, MMP-9 and VEGF expression in PTC are related to age, clinical TNM stage and tumour diameter, the details of the roles of COX-2, MMP-9 and VEGF in PTC remain unclear. Further studies are need to clarify these roles.

In conclusion, the present study demonstrated that the levels of COX-2, MMP-9 and VEGF proteins were significantly upregulated in PTCs compared with benign thyroid tumours as demonstrated by immunohistochemistry. There was a significant relationship between age, clinical TNM stage and tumour diameter and increased levels of COX-2, MMP-9 and VEGF immunostaining. These results showed that the combined immunohistochemical evaluation of the levels of COX-2, MMP-9 and VEGF in PTC might be useful as a molecular marker for the diagnosis of PTC.

Footnotes

Declaration of conflicting interest

All authors declare that there are no conflicts of interest.

Funding

The authors gratefully acknowledge the financial support provided by the Health Bureau of Jilin (no. 2008Z017).

References

Nikiforova

Nikiforov

. Molecular genetics of thyroid cancer: implications for diagnosis, treatment and prognosis. Expert Rev Mol Diagn 2008; 8: 83–95.

Brown

de Souza

Cohen

. Thyroid cancer: burden of illness and management of disease. J Cancer 2011; 2: 193–199.

Weber

Schilling

Büchler

. Thyroid carcinoma. Curr Opin Oncol 2006; 18: 30–35.

Nikiforov

. Diagnostic Pathology and Molecular Genetics of the Thyroid: A Comprehensive Guide for Practicing Thyroid Pathology, 2nd edn. Philadelphia: Lippincott Williams & Wilkins, 2012.

Wehland

Bauer

Magnusson

. Biomarkers for anti-angiogenic therapy in cancer. Int J Mol Sci 2013; 14: 9338–9364.

Folkman

. Seminars in Medicine of the Beth Israel Hospital, Boston. Clinical applications of research on angiogenesis. N Engl J Med 1995; 333: 1757–1763.

Reddy

Rao

Seibert

. Evaluation of cyclooxygenase-2 inhibitor for potential chemopreventive properties in colon carcinogenesis. Cancer Res 1996; 56: 4566–4569.

Sano

Kawahito

Wilder

. Expression of cyclooxygenase-1 and -2 in human colorectal cancer. Cancer Res 1995; 55: 3785–3789.

Kakiuchi

Tsuji

Tsujii

. Cyclooxygenase-2 activity altered the cell-surface carbohydrate antigens on colon cancer cells and enhanced liver metastasis. Cancer Res 2002; 62: 1567–1572.

10.

Yoshimura

Matsuyama

Kawahito

. Study of cyclooxygenase-2 in renal cell carcinoma. Int J Mol Med 2004; 13: 229–233.

11.

Rundhaug

. Matrix metalloproteinases and angiogenesis. J Cell Mol Med 2005; 9: 267–285.

12.

Rooprai

Rucklidge

Panou

. The effects of exogenous growth factors on matrix metalloproteinase secretion by human brain tumor cells. Br J Cancer 2003; 82: 52–55.

13.

Shih

Yen

. Cyclooxygenase-2 induces EP1- and HER-2/Neu-dependent vascular endothelial growth factor-C up-regulation: a novel mechanism of lymphangiogenesis in lung adenocarcinoma. Cancer Res 2004; 64: 554–564.

14.

Zhang

Meng

Wang

. Enhanced in vitro invasiveness of ovarian cancer cells through up-regulation of VEGF and induction of MMP-2. Oncol Rep 2006; 15: 831–836.

15.

Sobin

Wittekind

. TNM: Classification of malignant tumours. International Union Against Cancer, 5th edn. New York, NY: Wiley-Liss, 1997.

16.

Campo

Merino

Liotta

. Distribution of the 72-kd type IV collagenase in nonneoplastic and neoplastic thyroid tissue. Hum Pathol 1992; 23: 1395–1401.

17.

Dvorak

Brown

Detmar

. Vascular permeability factor/vascular endothelial growth factor, microvascular hyperpermeability, and angiogenesis. Am J Pathol 1995; 146: 1029–1039.

18.

Fujio

Walsh

. Akt mediates cytoprotection of endothelial cells by vascular endothelial growth factor in an anchorage-dependent manner. J Biol Chem 1999; 274: 16349–16354.

19.

Bozbora

Erbil

Türe

. Role of vascular endothelial growth factor in the prognosis of papillary thyroid cancer. Endokrinol Diyalog Cilt 2004; 1: 18–23.

20.

Bunone

Vigneri

Mariani

. Expression of angiogenesis stimulators and inhibitors in human thyroid tumors and correlation with clinical pathological features. Am J Pathol 1999; 155: 1967–1976.

21.

Fenton

Patel

Dinauer

. The expression of vascular endothelial growth factor and the type 1 vascular endothelial growth factor receptor correlate with the size of papillary thyroid carcinoma in children and young adults. Thyroid 2000; 10: 349–357.

22.

Erdem

Gundogdu

Sipal

. Correlation of E-cadherin, VEGF, COX-2 expression to prognostic parameters in papillary thyroid carcinoma. Exp Mol Pathol 2011; 90: 312–317.

23.

Specht

Tucker

Hocever

. Cyclooxygenase-2 expression in thyroid nodules. J Clin Endocrinol Metab 2002; 87: 358–363.

24.

Ito

Yoshida

Nakano

. Cyclooxygenase-2 expression in thyroid neoplasms. Histopathology 2003; 42: 492–497.

25.

Cornetta

Russell

Cunnane

. Cyclooxygenase-2 expression in human thyroid carcinoma and Hashimoto’s thyroiditis. Laryngoscope 2002; 112: 238–242.

26.

Siironen

Ristimäki

Nordling

. Expression of COX-2 is increased with age in papillary thyroid cancer. Histopathology 2004; 44: 490–497.

27.

Huang

Shun

. Cyclooxygenase-2 increases hypoxia-inducible factor-1 and vascular endothelial growth factor to promote angiogenesis in gastric carcinoma. J Biomed Sci 2005; 12: 229–241.

28.

Dirim

Haberal

Goren

. VEGF, COX-2, and PCNA expression in renal cell carcinoma subtypes and their prognostic value. Int Urol Nephrol 2008; 40: 861–868.

29.

Mannam

. Hypoxia induces myocyte-dependent COX-2 regulation in endothelial cells: role of VEGF. Am J Physiol Heart Circ Physiol 2003; 285: H2420–H2429.

30.

Siironen

Ristimäki

Narko

. VEGF-C and COX-2 expression in papillary thyroid cancer. Endocr Relat Cancer 2006; 13: 465–473.

31.

Rocks

Paulissen

El Hour

. Emerging roles of ADAM and ADAMTS metalloproteinases in cancer. Biochimie 2008; 90: 369–379.

32.

Yao

Chen

Zhai

. Minocycline exerts multiple inhibitory effects on vascular endothelial growth factor-induced smooth muscle cell migration: the role of ERK1/2, PI3K, and matrix metalloproteinases. Circ Res 2004; 95: 364–371.