1

Introduction

The abundant lymphatic network of the supraglottic larynx causes strong predilection of cervical lymph node metastases .The incidence of metastasis has been reported to vary from 12% to 52%, approximately 30% of them being occult . Therefore, many authors advocate elective treatment of the N0 neck. However, as a result of this strategy, 70% of patients are overtreated. On the other hand, there are still some limitations to assess the true status of the neck disease using recent diagnostic techniques.

In existing studies, tumor size, localizations, and differentiation were usually used to predict the neck metastases risk in supraglottic laryngeal cancer rather than immunohistochemical parameters. More sensitive diagnostic tools are required to detect or predict the risk of lymph node metastasis preoperatively to constitute an accurate treatment plan . Relevant tumor markers of different metastatic steps can be useful to predict the risk of nodal metastasis. If any immunochemical parameter that predicts neck metastasis can be found, patients would be treated therapeutically instead of electively.

CD44 molecules are a family of transmembrane cell adhesion glycoprotein that are involved in cell-cell and cell-matrix interactions and signal transduction. CD44 family consists of standard form (CD44s) and alternative splice variants (CD44v) . Isoform CD44s is the smallest and most abundant member of these glycoproteins. The CD44 molecule provides a cell surface docking receptor for matrix metalloproteinase (MMP)–9 and mediates functions of MMP-9 indirectly such as tumor invasion and angiogenesis . The relationship between CD44 expression and cervical lymph node metastasis remains controversial in current literature. Furthermore, data for supraglottic carcinoma of the larynx are extremely rare .

Matrix metalloproteinases are a group of zinc-dependent enzymes that include more than 25 different members. They are capable of degrading extracellular matrix components. MMP–9, also called gelatinase B , and MMP-2 are both involved in the digestion of type IV collagen that is known as the main component of the basement membrane, thereby facilitating tumor cell infiltration and metastasis . The relationship between gelatinases expression and lymph node metastases of head neck tumors has been widely reported in medical literature , but there has been only one study that focused on supraglottic carcinomas .

CD44 provides a cell surface docking receptor for active MMP-9, and this localization of MMP-9 to cell surface is required to promote tumor invasion and angiogenesis. Screening the relevant literature in English, there was no study about supraglottic laryngeal cancer with CD44 and MMP-9 immunostaining concurrently. The aim of this study is to evaluate the predictive value of these markers in patients with lymph node metastasis of supraglottic carcinomas.

2

Material and methods

Approval for this study was obtained from the institutional review board of Erciyes University School of Medicine in Kayseri. Two hundred ninety-four supraglottic laryngeal cancers were treated surgically in our head and neck surgery department from 1991 to 2005. Ninety-four of the 294 patients had pathologically metastatic lymph node (pN+). From these 94 patients, we selected 30 pN+ patients via random sampling. We also selected 30 sex-, T value–, and differentiation-matched pathologically negative neck (pN0) patients who had not developed any neck recurrence during the 3 years after surgery. None of the pN0 patients underwent any adjuvant therapy.

Tumors were classified according to the American Joint Committee on Cancer . In all patients, the surgical procedure began with ipsilateral or bilateral neck dissection according to the primary tumor site and was followed by various types of laryngectomies. T1 to T2 tumors mainly involving one side and not crossing the midline were treated with only ipsilateral neck dissection ( Table 1 ). Laryngectomy types are selected on the basis of the status of the primary tumor, the patient’s condition, and pulmonary function. Patients who received previous therapy such as surgery, irradiation, and/or chemotherapy were excluded.

CD44 and MMP-9 antibodies were applied to the tumor representative sections that were derived from paraffin sections by using the streptavidin biotin method . All specimens were examined by 2 blinded pathologists. The association between immunohistochemical results and histopathologic lymph node metastasis was analyzed statistically. We also analyzed association between immunostaining of CD44 and MMP-9.

2

Material and methods

Approval for this study was obtained from the institutional review board of Erciyes University School of Medicine in Kayseri. Two hundred ninety-four supraglottic laryngeal cancers were treated surgically in our head and neck surgery department from 1991 to 2005. Ninety-four of the 294 patients had pathologically metastatic lymph node (pN+). From these 94 patients, we selected 30 pN+ patients via random sampling. We also selected 30 sex-, T value–, and differentiation-matched pathologically negative neck (pN0) patients who had not developed any neck recurrence during the 3 years after surgery. None of the pN0 patients underwent any adjuvant therapy.

Tumors were classified according to the American Joint Committee on Cancer . In all patients, the surgical procedure began with ipsilateral or bilateral neck dissection according to the primary tumor site and was followed by various types of laryngectomies. T1 to T2 tumors mainly involving one side and not crossing the midline were treated with only ipsilateral neck dissection ( Table 1 ). Laryngectomy types are selected on the basis of the status of the primary tumor, the patient’s condition, and pulmonary function. Patients who received previous therapy such as surgery, irradiation, and/or chemotherapy were excluded.

CD44 and MMP-9 antibodies were applied to the tumor representative sections that were derived from paraffin sections by using the streptavidin biotin method . All specimens were examined by 2 blinded pathologists. The association between immunohistochemical results and histopathologic lymph node metastasis was analyzed statistically. We also analyzed association between immunostaining of CD44 and MMP-9.

3

Immunohistochemical study

Tumors were analyzed using full tissue sections (n:60). All cases were fixed in 10% buffered formalin, processed routinely, and embedded in paraffin. Tissue sections (3 or 4 μ m thick) were deparaffinized in xylene and rehydrated in a graded series of ethanols. Heat-induced epitope retrieval was performed by placing sections in plastic Coplin jars containing a preheated target retrieval solution (DAKO, Carpinteria, CA) boiled for 2 × 10 minutes in a microwave oven in a EDTA buffer. Sections then were allowed to cool at room temperature for at least 15 minutes. We used a mouse monoclonal antibody specific for CD44s (clone DF1485; Novocastra, Newcastle upon Tyne, United Kingdom) at a dilution of 1:40 and a rabbit polyclonal antibody MMP-9 (Lab Vision, Fremont, CA) at a dilution 1:70. Subsequent steps of the immunostaining procedure were performed in a humidity chamber as previously described. Formalin-fixed and paraffin-embedded samples of breast carcinoma and tonsil carcinoma were used as positive control for MMP-9 and CD44, respectively. The substitution of phosphate-buffered saline for the primary antibodies was used as negative control.