Expression of Matrix Metalloproteinase-2, but not Caspase-3, Facilitates Distinction between Benign and Malignant Thyroid Follicular Neoplasms

Thyroid tumors with a follicular growth pattern include a broad range of lesions from benign hyperplastic nodules to malignant neoplasms (Carling & Udelsman, 2005). Follicular thyroid adenoma (FTA), a benign encapsulated tumor with evidence of follicular cell differentiation, is the most common thyroid neoplasm (Rosai, 2011). True follicular thyroid carcinoma (FTC) (not including follicular variant of papillary carcinoma) is the malignant thyroid tumor exhibiting evidence of follicular cell differentiation (Rosai, 2011). One of the most challenging tasks in pathologic evaluation of thyroid lesions, is the differentiation between FTC and FTA. Diagnosis of FTC is dependent on the presence of capsular or vascular invasion (Rosai, 2011; Bryson et al., 2008) .Fine needle aspiration biopsy is not capable to discriminate between FTA and FTC based on the mentioned criteria (Goldstein et al., 2002). Methods for distinguishing between FTA and FTC would obviate many unnecessary thyroid surgeries. However, to date no definitely reliable and feasible discriminatory marker has been introduced. Matrix metalloproteinase-2 (MMP-2) is a zinc-


Introduction
Thyroid tumors with a follicular growth pattern include a broad range of lesions from benign hyperplastic nodules to malignant neoplasms (Carling & Udelsman, 2005). Follicular thyroid adenoma (FTA), a benign encapsulated tumor with evidence of follicular cell differentiation, is the most common thyroid neoplasm (Rosai, 2011). True follicular thyroid carcinoma (FTC) (not including follicular variant of papillary carcinoma) is the malignant thyroid tumor exhibiting evidence of follicular cell differentiation (Rosai, 2011). One of the most challenging tasks in pathologic evaluation of thyroid lesions, is the differentiation between FTC and FTA. Diagnosis of FTC is dependent on the presence of capsular or vascular invasion (Rosai, 2011;Bryson et al., 2008) .Fine needle aspiration biopsy is not capable to discriminate between FTA and FTC based on the mentioned criteria (Goldstein et al., 2002).
Methods for distinguishing between FTA and FTC would obviate many unnecessary thyroid surgeries. However, to date no definitely reliable and feasible discriminatory marker has been introduced.
In this study, expression of MMP-2 and caspase-3 was examined by immunohistochemistry (IHC) in sections of FTAs and FTCs. The aim was to examine expression of MMP-2 and Caspase-3 in thyroid follicular neoplasms and to determine their usefulness for differential diagnosis between adenoma and carcinoma.

Materials and Methods
Paraffin blocks of formalin-fixed samples representing 60 cases of FTA and 41 cases of FTC diagnosed during the years 1988 -2010 in the Department of Pathology, Shariati Hospital, Tehran, Iran, were examined to determine the expression of MMP-2 and Caspase-3. Cases of FTC were diagnosed based on capsular and/ or vascular invasion. The study was conducted according to the protocol of the Ethics Committee of Endocrinology & Metabolism Research Center.
Two 3-micrometer sections from representative blocks of every tumor were deparaffinized, rehydrated, placed in 0.5% hydrogen peroxide for 10 minutes. Antigen retrieval and unmasking was performed according to manufacturer's instruction. Sections were separately incubated with primary antibodies, washed with TBS buffer, and then incubated with the appropriate biotinylated secondary antibody. Washing in TBS buffer, incubation in ABC reagent, washing in TBS buffer and incubation in suitable peroxidase substrate were then consecutively performed. Finally the slides were counterstained with hematoxyline, dehydrated and mounted. Expected staining pattern was cytoplasmic for both markers. For negative controls, the primary antibodies were substituted with normal serum at appropriate concentration. Tonsil and inflamed large bowel were used as positive controls for caspase-3 and MMP-2, respectively, as recommended by the manufacturer.
Immunoreactivity of tumor cells for each of the two markers were semi-quantitatively scored as to extent (Table 1).
After collection of data, the analysis of qualitative and  quantitative data was done using Statistical Package for Social Sciences (SPSS) version 15.0 (Chicago, Illinois, USA). P value < 0.05 was considered significant.

Results
Results of IHC study of FTAs and FTCs are summarized in Figure 1.
Statistically significant relationship was observed between MMP-2 expression and malignant tumor behavior (i.e. diagnosis of FTC) (p=0.025).

Discussion
Differential diagnosis between FTA and FTC is based on presence of capsular or vascular invasion in carcinoma (Rosai, 2011;Bryson et al., 2008) which cannot be evaluated in material obtained from thyroid fine needle aspiration biopsy (Goldstein et al., 2002). Suspicious nodules must therefore be surgically removed in subtotal or total thyroidectomy. With the diagnosis of follicular carcinoma confirmed in formal pathologic evaluation, a second operation may be necessary (Stolf et al., 2006).
Regarding the fact that 5-10% of the population will develop a clinically significant thyroid nodule during their lifetime (Mazzaferi, 1993;Barden et al., 2003) and that by fine needle aspiration biopsy approximately 20% of thyroid nodules are follicular neoplasms (Carling & Udelsman, 2005), methods for distinguishing between adenoma and carcinoma would obviate many unnecessary thyroid surgeries.
Furthermore, such methods would be of great value in cases of minimally invasive (or encapsulated) follicular carcinoma, where discrimination from adenoma based on vascular or capsular invasion suffers inter-observer variation among pathologists (Hirokava et al., 2002;France et al., 2003) and also, in pathology specimens in which capsular detachment, tissue fragmentation or distortion, or subtleness of vascular or capsular invasion makes the distinction between carcinoma and adenoma difficult.
It is suggested that in molecular analysis, follicular thyroid adenoma and carcinoma have different characteristic microarray expression profiles (Barden et al., 2003;Cerutti et al., 2004, Finley et al., 2004Griffith et al., 2006). However, DNA microarray gene profiling is expensive, time consuming, and operator dependent and therefore not practical for clinical use (Bryson et al., 2008).
To overcome the difficulty in distinguishing between FTA and FTC by IHC study as a less costly and more available method, various markers have been evaluated. Despite some overlapping results, Dipeptidyl aminopeptidase IV (CD26) was reported to be useful for distinction between FTA and FTC, primarily in cytological specimens (Kotani et al., 1992;Maruta et al., 2004). Survivin (Hagh et al., 2006), HBME-1, CD15, thyroid peroxidase, CD10 (a zinc-dependent metalloproteinase, that may share certain functions with some MMPs) and cell proliferation markers have also been reported to be useful ( Cho et al., 2006). In a previous experience, we failed to show any discriminating role for nm-23 in this regard (Moradi et al., 2009).
In a study by Chen et al, EMMRPIN has been implicated in the metastatic potential of FTCs (Chen et al., 2001), Barden et al. (Barden et al., 2003) confirmed the differential expression of EMMRPIN by means of immunoblot analysis; however, no evaluation of its diagnostic potential was analyzed. A multicenter study demonstrated that IHC detection of galectin-3 (an enzyme regulating cell-cell communication and implicated in the initiation and regulation of cell growth and malignant transformation) in FNA biopsy samples from thyroid nodules served as a highly specific and sensitive means of identifying malignant lesions of the thyroid ( Bartolazzi et al., 2001). Other studies disprove the utility of galectin-3 as a single protein marker, especially in the case of follicular lesions (Bryson et al., 2008). Prasad et al. reported that a panel consisting of galectin-3, fibronectin-1 and HBME-1 may enhance the sensitivity and specificity for diagnosis of FTC versus FTA ( Prasad ML, et al 2005). Some studies have suggested trefoil factor 3 (TFF3) as a useful biomarker at the RNA level in thyroid cancer, and this gene was listed as 1 of 12 (of a total of 1795 genes) to have the most clinical promise (based on results of the array analysis) for the preoperative thyroid diagnosis (Griffith et al., 2006). IHC results of another study did not substantiate the anticipated promise of TFF3 as a clinical marker (Bryson et al., 2008).
Despite all the studies, of which only some were mentioned, no definitely reliable and feasible discriminatory marker has been introduced to date.
Matrix metalloproteinases (MMPs), including MMP-2, are a group of zinc-dependent endopeptidases with extracellular matrix-degrading properties that affect early events in carcinogenesis and tumor invasion and ⁄or metastasis (Nelson et al., 2000;Nabeshima et al., 2002;Cho et al., 2006).
Over expression of MMP-2 in tumor tissue compared to adjacent normal tissue has been documented in many types of solid tumors, including breast, colorectal and pancreatic carcinomas, as well as squamous carcinomas of the head and neck. Human thyroid carcinoma tissues have been also reported to express MMP-2 (Nakamura et al., 1999;Dubravka et al., 2000). Others proposed MMPs, could not be shown to definitely distinguish between widely invasive and minimally invasive follicular thyroid carcinomas or between the latter and adenomas (Cho et al., 2006).
However, in a study by Liang et al. 113 patient samples were analyzed for expression of several biomarkers including MMP2, which was determined as independent factor associated with the metastatic status in thyroid cancers. Finally, MMP2 was considered as a potential tumor marker to identify greater risk of metastasis (Liang et al., 2011). Moreover, MMPs are known to promote tumor metastasis and are determined as targets of 3,3'-di indolylmethane(DIM),which is believed to have antiestrogen like property and can be considered as a potential novel preventive/therapeutic dietary supplement in thyroid cancers (Rajoria et al., 2011).
Caspases, including caspase-3, are cystein-dependent aspartate-directed proteases and important mediators of apoptosis. Many studies confirm contribution of deregulated apoptotic pathways in cancer development and progression; particularly, an important feature in thyroid carcinogenesis is represented by aberrations in control of apoptosis (Veidinger et al., 2010). Inactivation of proapoptotic and/ or activation of antiapoptotic components, including decreased expression of caspase-3 and other caspases has been documented in many cancers shown to correlate with response to treatment and prognosis in some cancers, including those of breast, lung colon, uterine cervix and thyroid. (Kotani et al., 1992;Devarajan et al., 2002;Philchenkov et al., 2004;Fennel et al., 2005;Furuya, 2007;Moradi et al., 2009;) .
It could be inferred from the mentioned facts that expression of MMP-2 and caspase-3, could potentially be related to -respectively-aggressive and indolent tumor behavior. In this study, expression of MMP-2 and caspase-3 was examined by IHC in samples of FTA and FTC in order to determine their usefulness for preoperative differential diagnosis between FTA and FTC.
We have observed a statistically significant relationship between positive expression of MMP-2 and malignant behavior of thyroid follicular neoplasm. This means that positive expression of MMP-2 in a follicular thyroid neoplasm could be equal to the diagnosis of FTC (versus FTA) with 9.7% sensitivity, 100% specificity,100% Predictive value of a positive test (PPV) and 61.8% predictive value of a negative test (NPV) ( Table 2).
We failed to show any relationship between expression of caspase-3 and the probability of a thyroid follicular neoplasm, being diagnosed as FTA or FTC.
In this study, MMP-2 was differentially expressed by FTAs and FTCs. Alas, because of low sensitivity of positive expression of MMP-2 , this is not a suitable test for screening follicular thyroid neoplasms for diagnosis of FTC. However, in cases with equivocal criteria of malignancy (vascular or capsular invasion) or in a panel with potential sensitive marker, positive IHC result for MMP-2 could be of great value to confirm the diagnosis of FTC with 100% specificity and 100% PPV.