Changes in immunoexpression of E-cadherin and
β-catenin in oral
squamous cell carcinoma with and without nodal metastasis
Fernanda Ferreira Lopes
a, Márcia Cristina da Costa Miguel
b, Antônio Luiz Amaral Pereira
a,
Maria Carmen Fontoura Nogueira da Cruz
a, Roseana de Almeida Freitas
b,
Leão Pereira Pinto
b, Lélia Batista de Souza
b,⁎
a
Dentistry School, Federal University of Maranhão, São Luiz 65085-580, Maranhão, Brazil
b
Department of Oral Pathology, Dentistry School, Federal University of Rio Grande do Norte, Rio Grande do Norte, Natal 59056-000, Brazil
Abstract The aim of this study is to analyze the immunohistochemical expression of E-cadherin andβ-catenin in oral squamous cell carcinoma to better understand the biological behavior of this lesion. The sample consisted of 15 cases of the tongue and 15 of the lower lip. The pattern and intensity of the labeling and the analysis of the percentage of tumor cells immunopositive in membrane for E-cadherin andβ-catenin were related to the anatomic location of the lesion, the presence or absence of nodal metastasis, and the histological gradation of malignancy in the tumor invasion front. The presence or absence of cytoplasmic and nuclear labeling was also recorded. The membrane expression for E-cadherin andβ-catenin predominately displayed a heterogeneous pattern in the carcinomas studied. No significant difference was observed between the expression pattern and the quantity of cells immunopositive for E-cadherin andβ-catenin and the anatomic location of the lesion or the presence or absence of nodal metastasis. However, a statistically significant difference was found between the reduced expressio\n of these proteins and the high malignancy score. The reduced immunoexpression of these proteins in the membrane may be related to the high degree of cell indifferentiation in cases of oral squamous cell carcinoma with high scores.
© 2009 Elsevier Inc. All rights reserved.
Keywords: Squamous cell carcinoma; Oral cancer; E-cadherin;β-catenin; Immunohistochemistry
1. Introduction
Cancer represents a global public health problem, which is currently the third cause of death resulting from disease
[1]. Among the malignant neoplasias that affect the oral cavity, squamous cell carcinoma is the most frequently reported, accounting for more than 90% of cases[2]. In spite of new and advanced therapeutic strategies, oral squamous cell carcinoma (OSCC) shows poor survival rates[3].
The anatomical areas most affected by OSCC are the tongue and lower lip[2], with tongue tumors showing a high
propensity for regional metastasis, poor prognosis, and worse survival rates, whereas lower lip neoplasias show better biological behavior than tumors in other anatomical areas[4,5].
In OSCC, the presence or absence of cervical lymph node involvement and distant metastases are important in patient prognosis. It therefore becomes relevant for the pathologist to search for characteristics in the histopathological analysis to determine the stage of neoplasias invasion. The histolo-gical characteristics of this carcinoma may differ from one area to another within the same tumor owing to its heterogeneous nature. Consequently, data collected in the tumor invasion front may be more significant[6,7].
Molecular events important to tumor growth, such as the loss and gain of adhesion molecules, occur in tumor-host interaction, that is, in the front of the invasion front[7,8].
Annals of Diagnostic Pathology 13 (2009) 22–29
⁎ Corresponding author. Av. Senador Salgado Filho, 1787, Lagoa Nova, Natal-Rio Grande do Norte, Brasil. CEP-59056-000. Tel.: +55 84 3215 4132; fax: +55 84 3215 4138.
E-mail address:[email protected](L.B. de Souza).
1092-9134/$– see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.anndiagpath.2008.07.006
E-cadherin is a transmembrane glycoprotein dependent on the calcium present in most epithelial cells that play an important role in intercellular adhesion and maintenance, cellular polarity, and tissue architecture. It is composed of an extracellular, transmembrane, and intercellular domain, which links to the catenins and which is involved in signal transduction mediated by these catenins. The cytoplasmic domain of E-cadherin is directly linked to β-catenin or γ-catenin, whereas the α-catenin connects the E-cadherin– β-catenin and γ-catenin to the actin cytoskeleton. Recent studies show thatβ- and γ-catenin act in cell adhesion and in the Wnt signaling pathway, with the stability of both proteins regulated by adenomatous polyposis coli[9].
Consequently, immunohistochemical studies using anti– E-cadherin and anti–β-catenin antibodies, molecules that participate in the cell-cell adhesion process in epithelial tissues, have been performed to understand the significance of the expression of these proteins in human cancers[10-12]. Alterations in E-cadherin and catenin expression are associated with the loss of differentiation, acquisition of an invasive phenotype, and poor clinical course of many types of cancer. The role of these proteins in invasion and neoplastic metastasis has attracted great interest owing to its promise as a possible prognostic marker. This article aimed to analyze the immunohistochemical expression of E-cad-herin andβ-catenin in squamous cell carcinoma of the lower lip and tongue, correlating this expression to the anatomic location of the lesion, the history of nodal metastasis, and the histological gradation of malignancy proposed by Bryne in 1998[7] to obtain a better understanding of the biological behavior of this lesion.
2. Materials and methods
The sample is composed of 30 diagnosed cases of OSCC, 15 cases of the lower lip and 15 cases of the tongue, from the files of the Instituto Maranhense de Oncologia Aldenora Bello, São Luís/MA. Cases resulting from surgical treatment, without prior chemotherapy or radiotherapy, were included in the sample, whereas those whose paraffin-embedded material was insufficient were excluded.
The clinical data collected were age and sex of the patients and the presence or absence of nodal metastasis at diagnosis. Histological cuts 5μm thick were obtained from the paraffin-embedded material and stained using the hematoxilin/eosin technique to evaluate the areas of deepest tumor invasion based on the malignancy criteria recommended[7]. Specimens with a score of 4 to 8 points were classified as having a low degree of malignancy, whereas those with more than 8 points were considered to have a high degree of malignancy, as proposed by Miranda (2002) [13], which is an adaptation of Bryne's classifica-tion (1998) [7].
Histological cuts 3 μm thick were obtained and submitted to the immunohistochemical method, using the streptoavidin-biotin technique. Following antigenic recov-ery, using the citric acid (pH 6.0) steamer method, the primary anti–E-cadherin (clone NCH-38, 1:50, DakoCyto-mation, Carpinteria, CA) and anti–β-catenin antibodies were incubated (clone Ab-1, 1:500, Labvision/Neomarkers, Fremont, CA).
The pattern and intensity of the marking in the cytoplasmic membrane for E-cadherin and β-catenin were classified into 4 scores: 1 = absent, 2 = focal hetero-geneous, 3= reduced homohetero-geneous, and 4 = strong homogeneous. The semiquantitative analysis of the percentage of tumor cells positive for E-cadherin and β-catenin with membrane expression was classified into 4 scores: 1 = less than 25%, 2 = 26% to 50%, 3 = 51% to 75%, and 4 = greater than 75% of positive cells[14]. For the analysis of β-catenin, the presence of cytoplasmatic and nuclear marking in the tumor cells at the invasion front was also analyzed [15].
As a negative control, the primary antibody was substituted for 1% bovine serum albumin in a buffer solution, and the cuts were subsequently submitted to all the steps described in the technique. As a positive control, normal oral epithelium present in the areas circumjacent to the lesions was used[14].
The Mann-Whitney test, Spearman coefficient correla-tion, and Fisher exact test were used for the statistical analysis, considering a significance level of 5%. This study
Table 1
Expression pattern and amount of cells immunopositive for E-cadherin andβ-catenin in relation to lesion location
Location n Median Q25-Q75 Mean of ranks Sum of ranks U Pa
Expression pattern
E-cadherin Lower lip 15 2 1-3 15.27 229.00 109.00 .879 Tongue 15 2 1-3 15.73 236.00
β-catenin Lower lip 15 2 2-3 16.17 242.50 102.50 .650 Tongue 15 2 2-3 14.83 222.50
Immunopositive cells
E-cadherin Lower lip 15 2 1-3 15.53 233.00 112.00 .982 Tongue 15 1 1-3 15.47 232.00
β-catenin Lower lip 15 2 1-4 16.03 240.50 104.50 .730 Tongue 15 2 1-3 14.97 224.50
was approved by the Research Ethics Committee of the University Hospital of the Universidade Federal do Maranhão.
3. Results
At diagnosis, the age of the patients varied between 34 and 91 years (mean age, 64.2 years). There were more men than women. Nodal metastasis occurred in 11 (73.33%) of the 15 cases of tongue carcinoma and in 3 (20%) of the 15 lower lip cases, a statistically significant association (P = .009), with the tongue more often affected.
Following the morphological analysis, we observed that 11 (73.33%) of the 15 (73.33%) cases of squamous cell carcinoma of tongue and 9 (60%) of the 15 lower lip cases were among the group of high-score malignancy, with no
statistically significant association found between lesion location and malignancy scores (P = .700).
The presence of nodal metastasis was observed at diagnosis in 14 cases of OSCC and absence in the remaining 16 cases. However, there was also no statistically significant association (P = .709) found between nodal metastasis and histological gradation of malignancy.
The cytoplasmic membrane expression for E-cadherin was predominant, with a heterogeneous pattern (score 2) for both lower lip and tongue carcinomas, as well as membrane expression forβ-catenin (Table 1) (Figs. 1-6). No significant difference was found between the means for expression pattern and the amount of immunopositive cells in the antibodies analyzed.
In relation to nodal metastasis, no significant difference was found between the means of expression pattern and the amount of immunopositive cells for either E-cadherin or
Fig. 1. Immunohistochemical expression of E-cadherin in OSCC. Cytoplasmic membrane expression with less than 25% of immunopositive cells (arrow) (streptavidin-biotin, original magnification ×400).
Fig. 2. Immunopositivity for E-cadherin in 26% to 50% of cells (arrow) (streptavidin-biotin, original magnification ×400).
Fig. 3. Immunoreactivity in cytoplasmic membrane for E-cadherin in greater than 75% of tumoral cells (arrows) (streptavidin-biotin, original magnifica-tion ×100).
Fig. 4. Staining of β-catenin in OSCC. Reduced homogeneous pattern (streptavidin-biotin, original magnification ×400).
β-catenin (Table 2). A statistically significant difference was observed in the membrane expression of the E-cadherin and β-catenin proteins in relation to malignancy scores. Cases with a low-malignancy score had significantly higher means in both the expression pattern and amount of cells immunopositive for E-cadherinβ-catenin (Table 3).
There was no significant association between the nuclear expression of β-catenin with the anatomic location of the lesion, nodal metastasis, and histological gradation of malignancy. There was also no significant association betweenβ-catenin expression in cytoplasm and the location of the lesion and nodal metastasis. A significant association was found between β-catenin expression in the cytoplasm and histological gradation of malignancy, with β-catenin present in the cytoplasm more often in cases of high malignancy scores (Table 4).
4. Discussion
Among the squamous cell carcinomas that occur in the oral cavity, those of the tongue are the neoplasias that most frequently develop nodal metastasis in initial tumors and are associated with poor prognosis and low survival rates[3]. In this study, we observed a high frequency of nodal metastasis in the tongue cases (more than 70% of the sample), which is in agreement with previous studies [16]; however, less alarming values were presented [10], who detected the presence in 41.18% of his sample.
In the lower lip, the presence of nodal metastasis was significantly less frequent than in the tongue, corroborating others studies, which found 79% of patients with squamous cell carcinoma of lip without nodal metastasis [4], and reported that lip neoplasias have better biological behavior than that of tumors in other anatomical sites, possibly due to the low rate of nodal metastasis and early detection because of their location[5].
The vascular and lymphatic systems vary between distinct anatomical sites, a fact that may influence tumor develop-ment[17]. The lower lymphatic vascularization in the lower lip area appears to contribute to the low nodal metastasis rate in cases of squamous cell carcinomas in this region at diagnosis, as observed in this study. However, despite the lower lip carcinoma's being negative at diagnosis, it may develop nodal metastasis at a later stage, in most cases within a 2-year period[18].
In the morphological analysis of the sample in this study, we observed a predominance of lesions classified as high-score malignancy, both for cases of the tongue (73.33%) and of the lower lip (60%), with no statistically significant association found between anatomical location and malig-nancy scores.
In relation to the tongue lesions, our results were similar to others reported by the literature [19], which observed a predominance of squamous cell carcinomas of the tongue with high malignancy scores. In other study [13], the 12 cases of tongue carcinoma (100%) had high malignancy scores, whereas among the 12 cases of lower lip carcinoma, only 8.33% were classified with a high score compared to 91.67% with a low malignancy score, differing from our findings for the lower lip.
The distinct biological behavior of OSCC, when the anatomical location of the primary tumor is analyzed, is defended by some authors [4,5]. We found an association between anatomical location and the formation of nodal metastasis, which reflects the differences in biological behavior between neoplasias and distinct anatomical regions
[2]. However, no association was found between either the location of the tumor and the histological gradation of malignancy, or location and nodal metastasis. This corrobo-rates the previous findings of Okamoto et al [20] and diverges from those of Odell et al[6], who found a relation between nodal metastasis and the histological gradation of malignancy.
Fig. 5. Immunoreactivity forβ-catenin with a focal heterogeneous pattern (arrow) (streptavidin-biotin, original magnification ×100).
Fig. 6. Cytoplasmic membrane immunoexpression for β-catenin with a strong homogeneous pattern (streptavidin-biotin, original magnification ×400).
E-cadherin links directly with theβ-catenin protein, forming the E-cadherin/β-catenin complex. β-Catenin may also act in the Wnt signaling pathway. In the absence of Wnt signaling, β-catenin has rapid phosphorylation and degradation; however, with the activation of this signaling method, the concentration ofβ-catenin in the cytoplasm rises, facilitating its translocation to the nucleus. In the nucleus, theβ-catenin links with members of the TCF/LEF family of transition factors and activates the expression of TCF/LEF target genes, which include the protooncogenes c-Myc and cyclin D1[8,21].
The results of this study, in relation to the immunohis-tochemical expression of E-cadherin and β-catenin mole-cules, proved what was described previously[22,23]. These studies revealed that cells positive for E-cadherin and β-ca-tenin had a cuboidal format, whereas the negative cells were fusiforms with greater loss of differentiation. In our research, the tumor cells were predominantly negative when found in small infiltrated or isolated islands, whereas the central cells of the tumor masses were generally immunopositive, albeit with a variable immunomarking pattern[24].
Also correlated in this study was the expression of E-cadherin and β-catenin with the anatomic location of the lesion, the presence of nodal metastasis and the histological gradation of malignancy proposed by Bryne[7]. The results showed that the heterogeneous focal pattern for E-cadherin was predominant in both the lower lip and the tongue. These results corroborate those of previous study [25], who
reported that in malignant tumors, the E-cadherin immu-noexpression is heterogeneous, with quantitative and qualitative marking alterations. They also observed that in normal oral mucous adjacent to the tumors, the E-cadherin molecule displayed a strong pericellular immunopositivity in the basal, suprabasal, and spinosum layers, with varying intensity, confirming others' findings [5], who analyzed 85 epidermoid carcinomas of the tongue.
There was no significant difference in the pattern of immunoexpression and amount of immunopositive cells for the antibodies analyzed in this study, between the anatomical locations of lesions in the lower lip and tongue. This result corroborates with that of Bowie et al[26], who found no association between E-cadherin expression and the anato-mical location of tumors in 28 cases of squamous cell carcinoma of the head and neck. There is, however, a scarcity of studies comparing the variables of E-cadherin and β-ca-tenin expression with tumor location. Most studies of these proteins are correlated to histological gradation and the appearance of metastasis.
There was also no significant difference between expression pattern and the amount of immunopositive cells for either E-cadherin orβ-catenin and the nodal metastasis variable. Studies with uterine carcinoma also found no relation between the E-cadherin and β-catenin expression and clinical-pathological data, such as recurrence, metasta-sis, and survival[27]. They were also similar to the results of Chow et al [10], who found no correlation between
Table 2
Expression pattern and amount of cells immunopositive for E-cadherin andβ-catenin in relation to nodal metastasis
Nodal metastasis n Median Q25-Q75 Mean of ranks Sum of ranks U Pa
Expression pattern
E-cadherin Without metastasis 16 2 1.25-2.75 15.56 249.00 111.00 .965 With metastasis 14 2 1-3 15.43 216.00
β-catenin Without metastasis 16 2 2-3 15.28 244.50 108.50 .874 With metastasis 14 2 2-3.25 15.75 220.50
Immunopositive cells
E-cadherin Without metastasis 16 2 1-2.75 15.72 251.50 108.50 .875 With metastasis 14 1 1-3.25 15.25 213.50
β-catenin Without metastasis 16 2 1-4 15.91 254.50 105.50 .779 With metastasis 14 2 1-3.25 15.04 210.50
a Mann-Whitney test.
Table 3
Expression pattern and amount of cells immunopositive for E-cadherin andβ-catenin in relation to the histological gradation of malignancy
Score of malignancy n Median Q25-Q75 Mean of ranks Sum of ranks U Pa
Expression pattern E-cadherin Low 10 3 2-3.25 21.50 215.00 40.00 .005 High 20 2 1-2 12.50 250.00 β-catenin Low 10 3 2-4 21.05 210.50 44.50 .008 High 20 2 2-2 12.73 254.50 Immunopositive cells E-cadherin Low 10 3 2-4 21.80 218.00 37.00 .003 High 20 1 1-2 12.35 247.00 β-catenin Low 10 4 2.75-4 22.80 228.00 27.00 .001 High 20 2 1-2 11.85 237.00 a Mann-Whitney test.
E-cadherin andβ-catenin expression and nodal metastasis in 85 squamous cell carcinomas of the tongue.
Diverging from our results, studies on oral carcinomas
[12] and on tongue carcinomas [19] showed significant association between E-cadherin expression and lymph node metastasis, considering the low expression of this protein as an indicative value for cervical metastasis. However, Takes et al [11], despite finding a significant correlation between the presence of nodal metastasis and low E-cadherin expression in OSCC, did not observe this relationship in cases located in the pharynx.
In relation toβ-catenin expression in the membrane, the results of our study confirm those of Kurtz et al[28], who found no significant association between β-catenin expres-sion in the membrane and nodal metastasis in head and neck carcinomas. However, our results diverge those reported by others studies [9,12], which found that low β-catenin expression was significantly related to nodal metastasis in OSCCs and, according to the latter, also related to poor patient prognosis.
This may be justified by the fact that the process of tumor development and metastasis is complex, and it is unlikely that a single molecular parameter could predict the behavior of metastatic tumor cells [11]. The existence of other factors involved in the formation of nodal metastasis is still unclear[28].
The complexity of molecular mechanisms involved in the formation of nodal metastasis was shown previously in a study about head and neck squamous cell carcinomas[28], which found no correlation between the presence of nodal metastasis and low E-cadherin expression despite their finding a significant association of the latter variable with vascular invasion and lower survival of patients. Perhaps the loss of cellular adhesion is merely an initial event in tumor invasion, with other molecules necessary, such as matrix metalloproteinases and integrins that function in extracel-lular cell-matrix interaction, to consolidate the invasion and promote the formation of nodal metastasis.
In this study, a statistically significant difference was found in membrane expression of the E-cadherin and β-catenin proteins in relation to malignancy scores. A moderately negative correlation can be observed between the expression
pattern, the amount of immunopositive cells for E-cadherin andβ-catenin, and the total malignancy score. These results are similar to some studies[19,23,24]that who showed that E-cadherin expression was inversely proportional to the level of tumor differentiation; however, they differ from the studies
[10,26]that did not find this correlation.
The significant correlation between the reducedβ-catenin expression in the membrane and the high malignancy score observed in this study differ from those of Castro et al[29], who found no such relation in esophageal carcinomas. However, our results and those of Bànkfalvi et al[8]support the theory that E-cadherin and β-catenin are involved in tissue polarity and architecture and are significantly related to histological gradation of the tumor invasion front in OSCC. The immunohistochemical expression of β-catenin, not only in the membrane but also in the cytoplasm and the nucleus, showed that this expression in the cytoplasm was more frequent than in the nucleus because, among 30 cases of OSCC in this study, 22 (73.33%) showed cytoplasmatic expression and 6 (20%) showed expression in the nucleus, confirming that nuclear β-catenin appears not to be a common finding in OSCC[30].
With regard toβ-catenin expression in the cytoplasm, our results were similar to those of Lo Muzio et al [22], who found this marking in 18 of 30 cases of OSCC. In this study, a significant association was observed between the cyto-plasmic marking forβ-catenin and the histological gradation of malignancy, with the latter being more frequent in case of high malignancy scores.
On the contrary, a study with ovarian carcinomas [21]
found no association between β-catenin expression in the membrane and in cytoplasm with histological gradation of malignancy. This difference in results may be explained by the type of tumor, diversity of histological subtypes of ovarian carcinomas, and differences in data analysis.
It is worth emphasizing that Lyakhovitsky et al [14]
observed the presence ofβ-catenin in the cytoplasm only in cutaneous squamous cell carcinomas and not in precursor lesions, suggesting that cytoplasmic β-catenin may reflect the invasive potential of cells. The location of β-catenin is considered important because the higher frequency of cytoplasmicβ-catenin was, the predetermining factor in the
Table 4
Correlation of theβ-catenin expression in the cytoplasm and the nucleus and the anatomical location of the lesion, nodal metastasis, and histological gradation of malignancy
β-catenin expression
Cytoplasm Nucleus
Negative n (%) Positive n (%) Pa Negative n (%) Positive n (%) Pa Location Lower lip 05 (33.3) 10 (66.7) .682 14 (93.3) 01 (6.7) .169
Tongue 03 (20.0) 12 (80.0) 10 (66.7) 05 (33.3)
Nodal metastasis Without metastasis 05 (31.3) 11 (68.7) .689 14 (87.5) 02 (12.5) .378 With metastasis 03 (21.4) 11 (78.6) 10 (71.4) 04 (28.6)
Score of malignancy Low 06 (60.0) 04 (40.0) .007 09 (90.0) 01 (10.0) .633 High 02 (10.0) 18 (90.0) 15 (75.0) 05 (25.0)
a
transference of this molecule to the nucleus of colorectal carcinomas[15].
In this study, we found 6 (20%) cases with β-catenin expression in the nucleus, and no association was found between lesion location, nodal metastasis, and histological gradation. In a sample of 24 OSCCs, only 2 (8.33%) cases revealed nuclear expression located only in invasive cells
[31], whereas no nuclear marking was observed in 18 cases of metastatic OSCCs[30].
The 20% frequency in our study of cases with ofβ-catenin nuclear expression is similar to that detected by Pukkila et al
[32], who confirmed this expression in 23% (27 cases) of the 138 cases of squamous cell carcinoma of the pharynx. They observed a correlation betweenβ-catenin nuclear expression and patient survival, whereas in our study, no such correlation was found. In addition, patients with squamous cell carcinoma of the esophagus that showed β-catenin nuclear expression had lower mean survival in months than those who did not show this expression[29], confirming the need for further studies onβ-catenin in carcinomas.
The clinical significance of β-catenin expression in the cytoplasm and/or nucleus is controversial, where the intense nuclear superexpression of this protein has been considered a significant prognostic parameter for the low survival rate in colorectal carcinoma. On the contrary, at other times, the discovery of this high nuclear expression has been shown to be a significant independent marker for a more favorable prognosis in non–small cell lung carcinomas[9].
When studyingβ-catenin, we found that its cytoplasmatic and nuclear immunoexpression findings are omitted in various studies cited in the literature. More research on the immunohistochemical expression of E-cadherin and β-catenin is needed. Studies should investigate the presence or absence of nodal metastasis in addition to patient survival times to reach more significant conclusions.
Finally, we found that alterations in the immunohisto-chemical expression of E-cadherin and β-catenin in the membrane and the presence of β-catenin in the cytoplasm and nucleus of OSCC cells observed in this study suggest that these proteins may be involved in the aggressive biological behavior of this neoplasia in invasive tumor areas, with a greater loss of differentiation where this association is evident.
Acknowledgments
The authors would like to thank Doctor Raimunda Ribeiro da Silva (Instituto Maranhense de Oncologia Aldenora Bello) for his kind help with the samples used in this study, and CAPES-PQI/UFMA/UFRN-Accord 00099/ 03-1 and Project PQI0047031.
References
[1] Wunsch-Filho V. The epidemiology of oral pharynx cancer in Brazil. Oral Oncol 2002;38:737-46.
[2] Gervásio OL, Dutra RA, Tartaglia SM, et al. Oral squamous cell carcinoma: a retrospective study of 740 cases in a Brazilian population. Braz Dent J 2001;12:57-61.
[3] Al-Rajhi N, Khafaga Y, El-Husseiny J, et al. Early stage carcinoma of oral tongue: prognostic factors for local control and survival. Oral Oncol 2000;36:508-14.
[4] Vartanian JG, Carvalho AL, Araújo Filho MJ, et al. Predictive factors and distribution of lymph node metastasis in lip cancer patients and their implications on the treatment of the neck. Oral Oncol 2004;40: 223-7.
[5] Choi KK, Kim MJ, Yun PY, et al. Independent prognostic factors of 861 cases of oral squamous cell carcinoma in Korean adults. Oral Oncol 2006;42:208-17.
[6] Odell EW, Jani P, Scherriff M, et al. The prognostic value of individual histologic grading parameters in small lingual squamous cell carcinomas. The importance of the pattern of invasion. Cancer 1994; 74:789-94.
[7] Bryne M. Is the invasive front of an oral carcinoma the most important area for prognostication? Oral Dis 1998;4:70-7.
[8] Bànkfalvi A, Kraβort M, Végh A, et al. Deranged expression of the E-cadherin/β-catenin complex and the epidermal growth factor receptor in the clinical evolution and progression of oral squamous cell carcinomas. J Oral Pathol Med 2002;31:450-7.
[9] Ueda G, Sunakawa H, Nakamori K, et al. Aberrant expression of beta-and gamma-catenin is an independent prognostic marker in oral squamous cell carcinoma. Int J Oral Maxillofac Surg 2006;35:356-61. [10] Chow V, Yuen AP, Lam KY, et al. A comparative study of the clinicopathological significance of E-cadherin and catenins (α,β,γ) expression in the surgical management of oral tongue carcinomas. J Cancer Res Clin Oncol 2001;12:59-63.
[11] Takes RP, Baatenburg De Jong RJ, Alles MJ, et al. Markers for nodal metastasis in head and neck squamous cell cancer. Arch Otolaryngol Head Neck Surg 2002;128:512-8.
[12] Tanaka N, Odajima T, Ogi K, et al. Expression of E-caderin,α-catenin eβ-cetenin in the process of lymph node metastasis in oral squamous cell carcinoma. Br J Cancer 2003;89:557-63.
[13] . Miranda JL. Expressão de proteínas de matriz extracelular em carcinoma epidermóide de lábio inferior e língua. Tese (Doutorado em Patologia Oral)– Departamento de Odontologia da UFRN, Natal-RN. 119p, 2002.
[14] Lyakhovitsky A, Barzilai A, Fogel M, et al. Expression of E-caderin and beta-catenin in cutaneous squamous cell carcinoma and its precursors. Am J Dermatopathol 2004;26:372-8.
[15] El-Bahrawy MA, Poulsom R, Jeffery R, et al. The expression of E-cadherin and catenins insporadic colorectal carcinoma. Hum Pathol 2001;32:1216-24.
[16] Amorim Filho FS, Sobrinho JA, Rapoport A, et al. Estudo clínico-epidemiológico do carcinoma epidermóide da base de língua. Rev Bras Otorrinolaringol 2003;69:175-9.
[17] Massano J, Regateiro FS, Januário G, et al. Oral squamous cell carcinoma: review of prognostic and predictive factors. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2006;102:67-76.
[18] Bucur A, Stefanescu L. Management of patients with squamous cell carcinoma of the lower lip and N0-neck. J Craniomaxillofac Surg 2004;32:16-8.
[19] Lim SC, Zhang S, Ishii G, et al. Predictive markers for late cervical metastasis instage I and II invasive squamous cell carcinoma of the oral tongue. Clin Cancer Res 2004;10:166-72.
[20] Okamoto M, Nishimine M, Kishi M, et al. Prediction of delayed neck metastasis in patients with stage I/II squamous cell carcinoma of the tongue. J Oral Pathol Med 2002;31:227-33.
[21] Lee CM, Shvartsman H, Deavers MT, et al. β-Catenin nuclear localization is associated with grade in ovarian serous carcinoma. Gynecology Oncology 2003;88:363-8.
[22] Lo Muzio L, Staibano S, Pannone G, et al. Beta- and gama-catenin expression in oral squamous cell carcinomas. Anticancer Research 1999;19:3817-26.
[23] Yokoyama K, Kamata N, Hayashi E, et al. Reverse correlation of E-cadherin and snail expression in oral squamous cell carcinoma cells in vitro. Oral Oncol 2001;37:65-71.
[24] Yamada K, Jordan R, Mori M, et al. The relationship between E-cadherin expression, clinical stage and tumor differentiation in oral squamous cell carcinoma. Oral Dis 1997;3:82-5.
[25] Mareel M, Leroy A. Clinical, cellular, and molecular aspects of cancer invasion. Physiol Rev 2003;83:337-76.
[26] Bowie GL, Caslin AW, Roland NJ, et al. Expression of the cell-cell adhesion molecule E-cadherin in squamous cell carcinoma of the head and neck. Clin Otolaryngol 1993;18:196-201.
[27] Van de Putte G, Kristensen GB, Baekelandt M, et al. E-cadherin and catenins in early squamous cervical carcinoma. Gynecologic Oncology 2004;94:521-7.
[28] Kurtz KA, Hoffman HT, Zimmerman MB, et al. Decreased E-cadherin but notβ-catenin expression is associated with vascular invasion and
decreased survival in head and neck squamous carcinomas. Otolar-yngol Head Neck Surg 2006;134:142-6.
[29] Castro J, Gamallo C, Palacios J, et al.β-Catenin expression pattern in primary esophageal squamous cell carcinoma. Relationship with clinicopathologic features and clinical outcome. Virchows Arch 2000; 437:599-604.
[30] Kudo Y, Kitajima S, Ogawa I, et al. Invasion and metastasis of oral cancer cells require methylation of E-cadherin and/or degradation of membranousβ-catenin. Clin Cancer Res 2004;10: 5455-63.
[31] Gasparoni A, Chaves A, Fonzi L, et al. Subcellular localization of beta-catenin in malignant cell lines and squamous cell carcinomas of the oral cavity. J Oral Pathol Med 2002;31:385-94.
[32] Pukkila MJ, Virtaniem JA, Kumpulainen EJ, et al. Nuclearβ-catenin is related to unfavourable outcome in oropharyngeal and hypopharyngeal squamous cell carcinoma. J Clin Pathol 2001;54:42-7.
