Home About us Editorial board Ahead of print Current issue Search Archives Submit article Instructions Subscribe Contacts Login 

 Table of Contents  
Year : 2013  |  Volume : 9  |  Issue : 4  |  Page : 613-617

Role of Cyclooxygenase-2, Ezrin and Matrix metalloproteinase-9 as predictive markers for recurrence of basal cell carcinoma

1 Department of Dermatology, Al-Azhar University, Cairo, Egypt
2 Department of Pathology, National Cancer Institute, Cairo University, Cairo, Egypt

Date of Web Publication11-Feb-2014

Correspondence Address:
Mohamed A El-Khalawany
Department of Dermatology, Al-Azhar University, Box: 32515, Al Darasah, Cairo
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/0973-1482.126456

Rights and Permissions
 > Abstract 

Context: Recurrence of basal cell carcinoma (BCC) may form a prognostic problem that couldn't be fully predicted. Although there are different clinical and histologic risk factors for BCC recurrence, few reports are available for the role of biologic markers.
Aim: The aim of this study was to assess the value of Cyclooxygenase-2 (COX-2), Ezrin and Matrix metalloproteinase-9 (MMP-9) in recurrence of BCC.
Settings and Design: A retrospective controlled study.
Materials and Methods: Primary tumors of 22 patients who had recurrent basal cell carcinoma (R-BCC) and 22 matched controls that showed non-recurrent basal cell carcinoma (NR-BCC) were collected. Clinical, histopathological, and immunohistochemical results were recorded and analyzed. Statistical analysis used: SPSS software version 13 and Pearson χ2 test.
Results: R-BCC showed COX-2 expression in 20 (90.9%) cases compared to 13 (59.1%) in NR-BCC with a significant difference (P = 0.04). Moderate to strong intensity was recorded in 13 recurrent and two non-recurrent tumors. Higher frequency for Ezrin immunopositivity was noted in R-BCC (72.7%) than NR-BCC (40.9%), but the difference did not reach the level of significance (P = 0.07). Twelve R-BCC and three NR-BCC revealed moderate to strong staining. For MMP-9, there was no statistically significant difference (P = 1) between recurrent cases (63.6%) and controls (68.2%). No correlation was found between marker expressions and clinical or histologic features of R-BCC.
Conclusions: Biologic markers may have a promising role in assessment of BCC prognosis and early detection of recurrence. High COX-2 expression could be considered as a risk factor of BCC recurrence that can be added to other clinical and histologic factors.

Keywords: Basal cell carcinoma, recurrence, Ezrin, Cyclooxygenase-2, Matrix metalloproteinase-9

How to cite this article:
El-Khalawany MA, Abou-Bakr AA. Role of Cyclooxygenase-2, Ezrin and Matrix metalloproteinase-9 as predictive markers for recurrence of basal cell carcinoma. J Can Res Ther 2013;9:613-7

How to cite this URL:
El-Khalawany MA, Abou-Bakr AA. Role of Cyclooxygenase-2, Ezrin and Matrix metalloproteinase-9 as predictive markers for recurrence of basal cell carcinoma. J Can Res Ther [serial online] 2013 [cited 2020 Nov 29];9:613-7. Available from: https://www.cancerjournal.net/text.asp?2013/9/4/613/126456

 > Introduction Top

Basal cell carcinoma (BCC) is the most common skin cancer that constitutes up to 80% of all malignant skin tumors. Although there are some aggressive forms of the tumor, it usually shows a low metastatic tendency. [1] Recurrence even after proper management may form a serious prognostic problem that necessitates a careful monitoring and re-examination of the patient for at least 3 years or even a lifetime regular follow-up. [2]

There are several risk factors that can predict increased incidence of BCC recurrence, but none of them should be assessed individually and overall assessment must be considered in all cases. Clinical risk factors that were suggested in previous studies include male sex, topographic localization (central region of the face involving the inner canthus, nostrils, and peri-auricular area), tumor size (recurrence increases by approximately 7% for each millimeter of tumor diameter), sun-sensitive skin, and age over 60 years at first presentation. [2],[3]

Histological risk factors are usually associated with the aggressive-growth variants such as sclerosing, mixed (nodular and sclerosing), and rarely superficial (multifocal) type. [4],[5] It was also observed that deep margin involvement, the presence of sclerodermiform or metatypic basaloid squamous cells as well as pleomorphous morphology and/or peritumoral inflammatory infiltrates were all related to an increase in the probability of recurrence. [6] In this study, we assessed the expression of different biologic markers in BCC aiming to evaluate their possible role in recurrence.

 > Materials and Methods Top

The primary tumors of 22 cases of BCC that showed recurrent basal cell carcinoma (R-BCC) were collected through the period from 2002 to 2008. Clinical data were retrieved and all lesions were proved to be totally surgically excised with histological negative surgical margins. Histopathologic examination and immunohistochemistry for Cyclooxygenase-2 (COX-2), Ezrin and Matrix metalloproteinase-9 (MMP-9) were performed. Twenty-two matched non-recurrent basal cell carcinoma (NR-BCC) who followed-up for 5 years were used as a control group.

Histological sub-typing

The criteria of sub-typing were based on the histological features mentioned by Weedon; nodular (solid) type is composed of islands of cells with peripheral palisading and a haphazard arrangement of the more central cells; micronodular type resembles the solid type, but the nests are much smaller and the peripheral palisading is not always as well-developed; superficial (multifocal) type is composed of multiple small islands of basaloid cells attached to the undersurface of the epidermis, and usually confined to the papillary dermis; infiltrating type shows elongated strands of basaloid cells infiltrating between collagen bundles; and metatypical type composed of nests and strands of cells maturing into larger and paler cells with common loss of peripheral palisading. [7]

Immunohistochemical staining

Immunohistochemical staining was performed using avidin-biotin complex method, [8] on formalin-fixed, paraffin-embedded tissue sections according to manufacturer's protocols. The primary antibodies included COX-2 (monoclonal mouse anti-human, clone CX-294, dilution 1:100, Dako, Denmark), MMP-9 (polyclonal rabbit anti-human, dilution 1:100, Dako, Denmark), and Ezrin (monoclonal mouse anti-human, clone 3C12, dilution 1:100, Abcam, UK).

Evaluation and analysis of immunostaining

Cytoplasmic positivity for COX-2 and MMP-9, and cytoplasmic positivity with or without membranous staining for Ezrin were evaluated. For each case, six fields (×400) with the highest density of stained cells were selected and the percentage of positive cells were scored in relation to the total tumor cell population, 0 (≤10% +ve cells), 1+ (11-25%), 2+ (26-50%), 3+ (>50% +ve cells). The intensity of staining was evaluated as 0 (no staining), 1+ (weak staining), 2+ (moderate staining), 3+ (strong staining). The percentage of positive tumor cells and staining intensity were multiplied to produce a weighted score for each case. Tumors with a weighted score = 0 were designated as negative; all others were considered positive.

Statistical analysis

The SPSS software version 13 was used. The statistical significance of the results was analyzed using the Pearson χ2 test. A level of P < 0.05 was considered significant.

 > Results Top

Cases with R-BCC included 16 males (72.7%) and 6 females (27.3%) with age ranged from 39 years to 78 years (mean 56 ± 10.2). All lesions were localized to the head and neck area; 12 (54.5%) on the face, 9 (40.9%) on the scalp, and one lesion (4.5%) on the ear. The size of primary lesions ranged from 7 mm to 24 mm with a mean of 13.4 ± 4.9 mm. Ten cases (45.5%) were presented with pigmented nodule while other clinical forms included ulcerative nodule (22.7%), atrophic plaque (27.3%), and pigmented patch (4.5%). The relapse time ranged from 30 months to 91 months with a mean of 57 ± 18.5 months. Histological examination showed nodular type in 9 cases (40.9%), infiltrating in 6 (27.3%), micronodular in 4 (18.2%), metatypical in 2 (9.1%) and superficial in 1 (4.5%). Data are summarized in [Table 1].
Table 1: Clinical and histologic features of primary lesions in 22 cases of recurrent BCC

Click here to view

NR-BCC included 15 males and 7 females with a mean age of 53 ± 11.7 years. Fourteen lesions were located on the face while 8 lesions were located on the scalp. Histological analysis showed nodular type in 13 (59.1%), infiltrating in 4 (18.2%), micronodular in 3 (13.6%), and superficial in 2 (9.1%).

Immunohistochemical study of R-BCC showed COX-2 expression in 20 (90.9%) cases compared to 13 (59.1%) in the control group with a significant difference (P = 0.04). Moderate to strong intensity [Figure 1] was recorded in 13 R-BCC (five nodular, four infiltrating, two micronodular, one metatypical, one superficial) and in two NR-BCC.
Figure 1: Cyclooxygenase-2 immunostaining showing strong expression in variable types of basal cell carcinoma including nodular (a ×400 and b ×1000), infiltrating (c ×400 and d ×1000), superficial (e ×1000) and metatypical (f ×1000) types

Click here to view

Immunopositivity for Ezrin showed a higher frequency in recurrent cases (72.7%) than NR controls (40.9%), but without a significant difference (P = 0.07). Moderate to strong staining [Figure 2] was observed in 12 R-BCC (five nodular, two infiltrating, three micronodular, one superficial, one metatypical), and in three NR-BCC.
Figure 2: Strong Ezrin immunopositivity in nodular (a ×1000) and superficial (b ×1000) basal cell carcinomas

Click here to view

For MMP-9, there was no statistically significant difference between R-BCC and NR-BCC (63.6%. 68.2%, P = 1). Six cases [Figure 3] and five controls revealed moderate to strong intensity. The immunohistochemical profile of the studied markers is presented in [Table 2].
Figure 3: Matrix metalloproteinase-9 showing strong staining in nodular (a ×1000) and superficial (b ×1000) basal cell carcinomas

Click here to view
Table 2: Immunohistochemical profile of COX-2, Ezrin and MMP-9 in both groups of BCC

Click here to view

In R-BCC, there were no significant correlations between markers expression and clinical parameters (age, sex, site, size, and duration of the lesions) or histologic type [Table 3].
Table 3: Correlation between biologic markers and clinicohistologic features of R-BCC

Click here to view

 > Discussion Top

The pathogenic mechanism of BCC recurrence still unclear and up till now there is no definitive mechanism that explains it. Clinical and histological risk factors were studied in previous reports but their value remains limited. Therefore, attempts to identify additional predictive biologic markers were needed. [9],[10] In this study, we evaluated the expression of COX-2, Ezrin and MMP-9 and assessed their role in BCC recurrence.

COX-2 is an enzyme that enhances the synthesis of prostaglandins in inflamed and neoplastic tissues. It was suggested that cellular over-expression of COX-2 has a significant role in the development of cancer through different mechanisms including enhancement of cellular proliferation, promotion of angiogenesis, inhibition of apoptosis, stimulation of invasion and suppression of immune response. [11] Moreover, COX-2 inhibitors have been successful at preventing UV-induced skin cancers in mouse models. [12],[13] Recently, Johannesdottir et al.[14] observed that non-steroidal anti-inflammatory drugs and the COX-2 inhibitors (diclofenac, etodolac, and meloxicam) were associated with a decreased risk of cutaneous cancer. The risk reduction was greatest among long-term and high-intensity users, suggesting a cumulative and dose-dependent protective effect.

Previous reports found that high COX-2 expression may increase the risk of BCC, possibly through the increase in the expression of p53 protein, and suppression of apoptosis. [15],[16] Tjiu et al., [17] revealed that higher levels of COX-2 expression were correlated with tumor invasion depth. Furthermore, Karahan et al.,[18] showed a significant higher expression of COX-2 in primary infiltrating BCC compared to nodular and superficial cases. However, Karagece Yalçin and Seckin found no difference between aggressive and non-aggressive types. [19]

In the current series, we showed a significant difference for COX-2 expression between R-BCC and NR-BCC although no correlation was observed with either clinical or histological features. This suggests that the risk factors of BCC recurrence are variable and may not be correlated with each other. The pathogenic mechanism of COX-2 in the aggressive behavior of BCC still unclear; however, a relationship was previously mentioned between COX-2 over-expression and increased levels of vascular endothelial growth factor-A, cluster of differentiation (CD) 31 positive vessels, and apoptosis-regulating genes Mcl-1 and Bcl-2. [17]

Ezrin is a cytoplasmic peripheral membrane protein, which is belongs to the Ezrin, Radixin and Moesin protein family. It has a significant role in cell surface structure adhesion, migration, and organization. During the last few years, it was suggested that it plays a role in tumor development and progression and can be used as a biomarker for treatment planning in the head and neck squamous cell carcinoma (SCC). [20]

Regarding BCC, Abdou et al., [21] found that score values of Ezrin expression tended to be high in cases showing infiltrative edges, but the association was away from significance (P = 0.84). In addition, cytoplasmic pattern of Ezrin was related to young age while other parameters were non-significantly correlated. In our study, although there was no statistical significant difference for Ezrin expression between cases and controls, a higher frequency was noted among cases (72.7%. 40.9%, P = 0.07). This may add Ezrin as a potential marker of BCC recurrence and more studies are needed to evaluate its role.

MMPs are zinc-dependent endopeptidases that were reported to have an important role in the initiation, growth and invasion of malignant tumors. [22],[23],[24],[25] Hernαndez-Pérez et al. [26] found that MMP-14 may be used as a prognosticator in SCC while MMP-2 may be of relevance to tumor progression.

In BCC, it was suggested that membrane type-1 MMP can be a novel marker for high-risk BCC and it plays an important role in locally invasive and highly destructive growth behavior. [27] Boyd et al. [28] studied the stromal expression of different MMPs in BCC and found that MMP-1 and MMP-9 were expressed more frequently in stromal macrophages and MMP-9 positive neutrophils were more detected in superficial BCC compared with nodular type, which showed more stromal expression of MMP-10.

Although O'Grady et al., [29] stated that MMP-9 was associated with local aggressiveness in BCC, we could not observe a significant difference for MMP-9 expression in recurrent and NR tumors. In accordance Karahan et al., [18] did not detect a difference between the histological subtypes of primary BCCs and between primary and recurrent cases.

 > Conclusion Top

Recurrence of BCC is a prognostic problem that needs proper evaluation. Although the role of biologic markers is not fully studied, some markers can help in early detection of this behavior. We found that COX-2 is a promising marker in prediction of BCC recurrence; high expression can be added as a risk factor in combination with other clinical and histological factors.

 > References Top

1.Elghissassi I, Mikou A, Inrhaoun H, Ennouhi A, Gamra L, Errihani H. Metastatic basal cell carcinoma to the bone and bone marrow. Int J Dermatol 2009;48:481-3.  Back to cited text no. 1
2.Bartoš V, Pokorný D, Zacharová O, Haluska P, Doboszová J, Kullová M, et al. Recurrent basal cell carcinoma: A clinicopathological study and evaluation of histomorphological findings in primary and recurrent lesions. Acta Dermatovenerol Alp Panonica Adriat 2011;20:67-75.  Back to cited text no. 2
3.Bøgelund FS, Philipsen PA, Gniadecki R. Factors affecting the recurrence rate of basal cell carcinoma. Acta Derm Venereol 2007;87:330-4.  Back to cited text no. 3
4.Friedman HI, Williams T, Zamora S, al-Assaad ZA. Recurrent basal cell carcinoma in margin-positive tumors. Ann Plast Surg 1997;38:232-5.  Back to cited text no. 4
5.Zagrodnik B, Kempf W, Seifert B, Müller B, Burg G, Urosevic M, et al. Superficial radiotherapy for patients with basal cell carcinoma: Recurrence rates, histologic subtypes, and expression of p53 and Bcl-2. Cancer 2003;98:2708-14.  Back to cited text no. 5
6.Sartore L, Lancerotto L, Salmaso M, Giatsidis G, Paccagnella O, Alaibac M, et al. Facial basal cell carcinoma: Analysis of recurrence and follow-up strategies. Oncol Rep 2011;26:1423-9.  Back to cited text no. 6
7.Weedon D. Tumors of the epidermis. In: Weedon D, editor. Weedon′s Skin Pathology. 3 rd ed. New York: Elsevier (Churchill Livingstone); 2010. p. 682-91.  Back to cited text no. 7
8.Hsu SM, Raine L, Fanger H. Use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase techniques: A comparison between ABC and unlabeled antibody (PAP) procedures. J Histochem Cytochem 1981;29:577-80.  Back to cited text no. 8
9.Bolshakov S, Walker CM, Strom SS, Selvan MS, Clayman GL, El-Naggar A, et al. p53 mutations in human aggressive and nonaggressive basal and squamous cell carcinomas. Clin Cancer Res 2003;9:228-34.  Back to cited text no. 9
10.Yerebakan O, Ciftçioglu MA, Akkaya BK, Yilmaz E. Prognostic value of Ki-67, CD31 and epidermal growth factor receptor expression in basal cell carcinoma. J Dermatol 2003;30:33-41.  Back to cited text no. 10
11.Telliez A, Furman C, Pommery N, Hénichart JP. Mechanisms leading to COX-2 expression and COX-2 induced tumorigenesis: Topical therapeutic strategies targeting COX-2 expression and activity. Anticancer Agents Med Chem 2006;6:187-208.  Back to cited text no. 11
12.An KP, Athar M, Tang X, Katiyar SK, Russo J, Beech J, et al. Cyclooxygenase-2 expression in murine and human nonmelanoma skin cancers: Implications for therapeutic approaches. Photochem Photobiol 2002;76:73-80.  Back to cited text no. 12
13.Rundhaug JE, Mikulec C, Pavone A, Fischer SM. A role for cyclooxygenase-2 in ultraviolet light-induced skin carcinogenesis. Mol Carcinog 2007;46:692-8.  Back to cited text no. 13
14.Johannesdottir SA, Chang ET, Mehnert F, Schmidt M, Olesen AB, Sørensen HT. Nonsteroidal anti-inflammatory drugs and the risk of skin cancer: A population-based case-control study. Cancer 2012;118:4768-76.  Back to cited text no. 14
15.Vogel U, Christensen J, Wallin H, Friis S, Nexø BA, Tjønneland A. Polymorphisms in COX-2, NSAID use and risk of basal cell carcinoma in a prospective study of Danes. Mutat Res 2007;617:138-46.  Back to cited text no. 15
16.Chen Z, Yang J, Huang Q. Correlation and expression of COX-2 and P53 protein in basal cell carcinoma of eyelid. J Huazhong Univ Sci Technolog Med Sci 2009;29:383-6.  Back to cited text no. 16
17.Tjiu JW, Liao YH, Lin SJ, Huang YL, Tsai WL, Chu CY, et al. Cyclooxygenase-2 overexpression in human basal cell carcinoma cell line increases antiapoptosis, angiogenesis, and tumorigenesis. J Invest Dermatol 2006;126:1143-51.  Back to cited text no. 17
18.Karahan N, Baspinar S, Bozkurt KK, Caloglu E, Ciris IM, Kapucuoglu N. Increased expression of COX-2 in recurrent basal cell carcinoma of the skin: A pilot study. Indian J Pathol Microbiol 2011;54:526-31.  Back to cited text no. 18
19.Karagece Yalçin U, Seçkýn S. The expression of p53 and COX-2 in basal cell carcinoma, squamous cell carcinoma and actinic keratosis cases. Turk Patoloji Derg 2012;28:119-27.  Back to cited text no. 19
20.Schlecht NF, Brandwein-Gensler M, Smith RV, Kawachi N, Broughel D, Lin J, et al. Cytoplasmic ezrin and moesin correlate with poor survival in head and neck squamous cell carcinoma. Head Neck Pathol 2012;6:232-43.  Back to cited text no. 20
21.Abdou AG, Maraee AH, El-Sayed EM, Elnaidany NF. Immunohistochemical expression of ezrin in cutaneous basal and squamous cell carcinomas. Ann Diagn Pathol 2011;15:394-401.  Back to cited text no. 21
22.Roh MR, Zheng Z, Kim HS, Kwon JE, Jeung HC, Rha SY, et al. Differential expression patterns of MMPs and their role in the invasion of epithelial premalignant tumors and invasive cutaneous squamous cell carcinoma. Exp Mol Pathol 2012;92:236-42.  Back to cited text no. 22
23.Piao S, Zhao S, Guo F, Xue J, Yao G, Wei Z, et al. Increased expression of CD147 and MMP-9 is correlated with poor prognosis of salivary duct carcinoma. J Cancer Res Clin Oncol 2012;138:627-35.  Back to cited text no. 23
24.Elahi M, Rakhshan V, Ghasemian NT, Moshref M. Prognostic value of transforming growth factor beta 1 TGF-â1 and matrix metalloproteinase 9 MMP-9 in oral squamous cell carcinoma. Biomarkers 2012;17:21-7.  Back to cited text no. 24
25.Sullu Y, Demirag GG, Yildirim A, Karagoz F, Kandemir B. Matrix metalloproteinase-2 (MMP-2) and MMP-9 expression in invasive ductal carcinoma of the breast. Pathol Res Pract 2011;207:747-53.  Back to cited text no. 25
26.Hernández-Pérez M, El-hajahmad M, Massaro J, Mahalingam M. Expression of gelatinases (MMP-2, MMP-9) and gelatinase activator (MMP-14) in actinic keratosis and in in situ and invasive squamous cell carcinoma. Am J Dermatopathol 2012;34:723-8.  Back to cited text no. 26
27.Oh ST, Kim HS, Yoo NJ, Lee WS, Cho BK, Reichrath J. Increased immunoreactivity of membrane type-1 matrix metalloproteinase (MT1-MMP) and â-catenin in high-risk basal cell carcinoma. Br J Dermatol 2011;165:1197-204.  Back to cited text no. 27
28.Boyd S, Tolvanen K, Virolainen S, Kuivanen T, Kyllönen L, Saarialho-Kere U. Differential expression of stromal MMP-1, MMP-9 and TIMP-1 in basal cell carcinomas of immunosuppressed patients and controls. Virchows Arch 2008;452:83-90.  Back to cited text no. 28
29.O′Grady A, Dunne C, O′Kelly P, Murphy GM, Leader M, Kay E. Differential expression of matrix metalloproteinase (MMP)-2, MMP-9 and tissue inhibitor of metalloproteinase (TIMP)-1 and TIMP-2 in non-melanoma skin cancer: Implications for tumour progression. Histopathology 2007;51:793-804.  Back to cited text no. 29


  [Figure 1], [Figure 2], [Figure 3]

  [Table 1], [Table 2], [Table 3]


Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

  >Abstract>Introduction>Materials and Me...>Results>Conclusion>Discussion>Article Figures>Article Tables
  In this article

 Article Access Statistics
    PDF Downloaded134    
    Comments [Add]    

Recommend this journal