Search

Epithelial-mesenchymal transition (EMT) involving down-regulation of E-cadherin is known to play an important role in tumour progression. The aim of our study was to investigate the mRNA expression of two EMT regulators-Slug and E12/E47-in primary human gastric carcinomas and to compare this with the expression of E-cadherin and other EMT regulators (Snail, Twist, and SIP1). We studied a series of 59 gastric carcinomas by real-time quantitative RT-PCR in formalin-fixed and paraffin-embedded tissues. Thirty-four cases (58%) showed Slug up-regulation in the tumour; reduced or negative expression of E-cadherin was present in 24 of these (71%, p<0.0001). Twenty-one cases (36%) showed E12/E47 up-regulation that was not significantly associated with E-cadherin down-regulation (p=0.5734). Slug up-regulation accompanied by E-cadherin down-regulation correlated with the presence of distant metastases (p=0.0029) and with advanced pTNM stages (p=0.0424). A statistically significant association was found between Slug up-regulation and the expression of SIP1 in intestinal (p=0.0014) and Snail in diffuse (p=0.0067) carcinomas. We present the first study integrating the analysis of several EMT regulators in primary gastric carcinomas and conclude that Slug up-regulation is associated with E-cadherin down-regulation in diffuse and intestinal-type gastric carcinoma, and that this effect could be complemented by the presence of other EMT regulators., (Copyright (c) 2007 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd.)

Alterations of the cell adhesion molecule E-cadherin have been demonstrated in sporadic and hereditary gastric carcinomas. A cell adhesion molecule with functional similarity to E-cadherin is desmoglein 2 (Dsg2), a major component of the desmosomes. In this study, we investigated whether alterations of Dsg2 are involved in gastric carcinogenesis and whether germline mutations contribute to a genetic predisposition in familial gastric cancer patients with no germline mutations in the E-cadherin gene. Seventy-five formalin-fixed, paraffin-embedded tissues from 37 familial and 38 sporadic gastric carcinomas were analysed for Dsg2 expression by immunohistochemistry. DNA from 31 familial gastric cancer patients was analysed for germline mutations and five sporadic tumours were analysed for somatic mutations by DHPLC. Of the 75 tumours, 25 (33%) demonstrated abnormal (reduced and/or non-membrane-associated) Dsg2 expression. There was a trend towards more frequent abnormal expression in diffuse type (42%) than in intestinal type tumours (18%) (p = 0.066). One germline missense variant leading to a non-conservative amino acid change (c. 2810 C > A, Thr 937 Asn) was found in a familial gastric cancer patient with a diffuse type tumour. No somatic mutations were identified. The observed abnormal expression of Dsg2 protein suggests that this molecule is involved in the carcinogenesis of a subset of gastric carcinomas, in particular of the diffuse type. Somatic mutations in the gene do not seem to be a very frequent inactivation event and the finding of no clear pathogenic germline mutation rules out Dsg2 as a major gastric cancer predisposition gene., (Copyright (c) 2005 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd.)

Aberrant expression and mutation of E-cadherin is frequent in gastric carcinoma (GC) especially of the diffuse type. The frequency of CDH1 (gene encoding E-cadherin) mutation in populations with high incidence of diffuse GC and its prognostic significance is unknown. One hundred seventy-seven gastrectomies from Mexican mestizo patients with intestinal (53), mixed (55), or diffuse (69) GC were included. In addition, 101 endoscopic biopsies from patients with GC not subjected to surgery were analyzed. Immunohistochemistry against wild-type E-cadherin (clone 36) and against 2 mutation-specific antibodies (MSA) recognizing mutant CDH1 lacking exon-8 (del 8) or exon-9 (del 9) were performed. Staining was correlated with histotype, tumor node metastasis stage, and follow-up. Abnormal or absent E-cadherin expression (clone 36) was identified in 84% GC, predominantly in diffuse or mixed tumors (P = 0.004) in advanced stages (P = 0.003). No survival differences at 1 and 2 years were observed among patients showing normal, abnormal, or absent wild type E-cadherin expression. Overall reactivity with the MSA was observed in 10 (5.6%) patients who were treated with surgery. In 140 patients, dead from the disease or alive with the disease, the survival at 1 and 2 years was 37% versus 17% and 14% versus 0 for patients without and with del 8/9 positivity, respectively (log rank P = 0.01). Biopsies from patients with inoperable-GC (101) rendered 5 (4.95%) with del 8 or 9 immunoreactivity. Abnormal E-cadherin expression is frequent in GC. However, exon 8 or 9 deletions were observed in only 5.3% tumors in this series from Mexico, at a lower rate than previously published, but associated with a worse prognosis.

The aim of the study was to determine epidermal growth factor receptor (EGFR) expression in gastric adenocarcinoma by standardized immunohistochemistry and to correlate EGFR expression with clinical features and patient survival. EGFR expression was investigated in paraffin sections of resection specimens of 89 gastric carcinomas from Mexican Mestizo patients using standardized immunohistochemistry with antigen retrieval (Dako EGFRpharmDx assay detection system). Membrane staining of EGFR was evaluated in the neoplastic cells and graded using a semiquantitative score (0-3+). Of the 89 carcinomas examined, staining of neoplastic cells was weak in 17 (19.1%, score 1+), moderate in 16 (18.0%, score 2+), and strong in nine cases (10.1%, score 3+). EGFR reactivity was heterogeneous, frequently showing completely negative up to 3+ positive areas within an individual tumor. EGFR reactivity score correlated with distant metastases (P=0.002) and clinical stage (P=0.033). EGFR score 0/1+ was significantly associated with an increase in patient survival when compared to score 2+/3+ (P=0.0006). In a multivariate analysis, EGFR positive cells in muscularis or subserosa (P=0.004), distant metastases (P=0.016) and residual disease (P=0.039) were significantly correlated with decreased survival. The prognosis was associated with the EGFR reactivity score (P=0.003), distant metastases (P=0.0001) and residual disease (P=0.012) in a univariate analysis. EGFR reactivity in neoplastic cells is an independent prognostic factor in gastric adenocarcinoma. The relevance of the heterogeneity in EGFR expression with regard to tumor progression, metastasis and anti-EGFR therapy needs to be studied.

Background: The mandible is a very uncommon place for a metastasis of a gastric carcinoma. Normally the area of the temporomandibular joint (TMJ) remains unaffected. The separate vascularization is discussed as one reason among others. Primary reconstruction after resection of the condyle is often problematic because an early onset of adjuvant systemic therapy is required. In this case, the insertion of a Quinn joint prosthesis is presented after resection of a TMJ metastasis., Case: We report a hematogenic metastatic gastric adenocarcinoma in a 51-year old male who initially presented with increasing disclusion in the left molar region. Suspecting a metastatic adenocarcinoma of the TMJ, a condylectomy with immediate replacement by a total joint prosthesis was performed via a preauricular approach. Corresponding to the clinically and radiologically suspected diagnosis, the decalcified histological specimen presented as a metastatic gastric adenocarcinoma within the intracapsular region., Results: The healing period of the implanted modified Quinn prosthesis was fast and uncomplicated after resection of this, to our knowledge, first documented metastatic gastric adenocarcinoma of the intracapsular region. After early restoration of joint function and patient satisfaction, the required radiochemotherapy of further unresectable bony metastases could be started in time., Discussion: This example of an extremely rare case of a metastasis shows that such a total joint prosthesis appears to be a very good alternative to extended autogenous reconstruction or an unsatisfactory primary resection. Due to the mating of the spherical condylar head and glenoid fossa, the modified Quinn prosthesis is very suitable for total joint replacement after extended resection or in multiply preoperated cases.

E-cadherin mutations are found in 50% of diffuse-type gastric carcinoma, but not in intestinal gastric carcinoma. Because cell-cell adhesion mediated by E-cadherin plays an important role in epithelial cell survival, E-cadherin mutations could alter the apoptotic behavior of tumor cells. p53 and Bcl-2 family members are also important regulators of cellular apoptosis. This is the first study that investigates the relationship between E-cadherin gene mutation and p53 gene mutation, p53 accumulation, Bcl-2 expression, and Ki-67 expression in diffuse-type gastric carcinoma (24 cases, E-cadherin mutation status: wild-type in 8 patients and mutant in 16 patients). The mutation status of exons 5-8 of p53 was analyzed by denaturing high pressure liquid chromatography (DHPLC) in formalin-fixed, paraffin-embedded tumor sections, followed by direct sequencing of cases with aberrant chromatographic patterns. p53 mutations were found in 1 of 8 tumors without E-cadherin mutation (12.5%) and in 1 of 16 tumors with E-cadherin mutation (6.3%), a difference that was not statistically significant (p = 1.00). p53 accumulation was found in 8 of 24 tumors (33.3%) by immunohistochemical staining. p53 accumulation was significantly more frequent in tumors without E-cadherin mutations (5 of 8 tumors, 62.5%) than in gastric carcinoma tissues with E-cadherin mutations (3 of 16 tumors, 18.8%, p = 0.03). Bcl-2 staining was not observed in gastric carcinoma cells without E-cadherin mutations, but was detectable in 5 of 16 tumors with E-cadherin mutations (31.3%), a difference that was not statistically significant (p = 0.13). No relationship was observed between Ki-67 staining and the E-cadherin mutation status (p = 1.00). These data suggest that the presence of E-cadherin mutations can significantly alter the accumulation of the apoptosis-regulating p53 protein, whereas no correlation with the p53 mutation status or with Ki-67 staining was observed., (Copyright 2002 Wiley-Liss, Inc.)

Epithelial-mesenchymal transition (EMT) involving down-regulation of E-cadherin is thought to play a fundamental role during early steps of invasion and metastasis of carcinoma cells. The aim of our study was to elucidate the role of EMT regulators Snail, SIP1 (both are direct repressors of E-cadherin), and Twist (an activator of N-cadherin during Drosophila embryogenesis), in primary human gastric cancers. Expression of Snail, SIP1, and Twist was analyzed in 48 gastric carcinomas by real-time quantitative RT-PCR in paraffin-embedded and formalin-fixed tissues. The changes of expression levels of these genes in malignant tissues compared to matched non-tumorous tissues were correlated with the expression of E- and N-cadherin. From 28 diffuse-type gastric carcinomas analyzed reduced E-cadherin expression was detected in 11 (39%) cases compared to non-tumorous tissues. Up-regulated Snail could be found in 6 cases with reduced or negative E-cadherin expression. However, there was no correlation to increased SIP1 expression. Interestingly, we could detect abnormal expression of N-cadherin mRNA in 6 cases, which was correlated with Twist overexpression in 4 cases. From 20 intestinal-type gastric cancer samples reduced E-cadherin expression was found in 12 (60%) cases, which was correlated to up-regulation of SIP1, since 10 of these 12 cases showed elevated mRNA levels, whereas Snail, Twist, and N-cadherin were not up-regulated. We present the first study investigating the role of EMT regulators in human gastric cancer and provide evidence that an increase in Snail mRNA expression is associated with down-regulation of E-cadherin in diffuse-type gastric cancer. We detected abnormally positive or increased N-cadherin mRNA levels in the same tumors, probably due to overexpression of Twist. SIP1 overexpression could not be linked to down-regulated E-cadherin in diffuse-type tumors, but was found to be involved in the pathogenesis of intestinal-type gastric carcinoma. We conclude that EMT regulators play different roles in gastric carcinogenesis depending on the histological subtype.

Immunohistochemical analysis has been used to show that expression of the homophilic cell-to-cell adhesion molecule, E-cadherin, is frequently altered in human cancers, including gastric and breast carcinoma. Besides genetic down-regulation, structural mutations such as in-frame deletions of exon 8 and exon 9 were frequently found; these may affect the binding of monoclonal antibodies used for immunohistochemical analysis. In this study it was found that antibodies HECD-1 and E9, two monoclonal antibodies often used in E-cadherin immunoanalysis, react with epitopes present at least in part in exon 8 and exon 9, respectively. This study generated and characterized a mutation-specific monoclonal antibody, E-cad delta 8-1, reacting with the mutant protein lacking exon 8 but not with the wild-type molecule. By using E-cad delta 8-1 and HECD-1, it was possible separately to analyse the immunoreactivity of mutant and normal E-cadherin proteins, respectively, in an allele-specific manner in archival material. A similar analysis was performed using E9 and the previously characterized mutation-specific antibody E-cad delta 9-1. Typically, in gastric and breast cancer harbouring E-cadherin splice site gene mutations, the mutant proteins were expressed but the wild-type protein was not detected in malignant tissues. These results indicate that variant-specific monoclonal antibodies can be used to identify differentially expressed E-cadherin proteins. For immunohistochemical analysis of E-cadherin, at least two different monoclonal antibodies should be used to exclude alterations of the epitopes resulting in failure to detect a mutant protein., (Copyright 2002 John Wiley & Sons, Ltd.)

The influence of proliferation and proliferation kinetics on prognosis in gastric cancer after complete resection are controversial. In a prospective study we investigated the tumour specimens of 111 patients after resection of gastric cancer, who received 200 mg intravenous (i.v.) bromodeoxyuridine (BrdU) pre-operatively. The following biological parameters were analysed in the tumour tissue using flow-cytometry: DNA ploidy, proportion of S-phase cells, BrdU labelling index (LI), DNA synthesis time (T(s)), potential tumour doubling time (T(pot)), proliferating cell nuclear antigen (PCNA) and Ki-67 LI. The median follow-up time was 40 months (range 19-62 months). Besides the established pathohistological prognostic factors, univariate analysis revealed a prognostic influence on survival for BrdU LI, T(pot) and the proportion of S-phase cells. By multivariate Cox analysis of the completely resected cases, only tumour stage and T(pot) had a significant, independent influence on survival. By classification and regression trees (CART) analysis, resection status, tumour stage and T(pot) defined risk groups with significantly different outcomes. A short T(pot) was a predictor of better survival in stage I, II and IIIA tumours. Ploidy and the other investigated proliferation-related parameters failed to demonstrate any influence on prognosis after resection of gastric cancer.

A monoclonal antibody (E-cadherin delta 9-1) directed against a characteristic E-cadherin mutation (in-frame deletion of exon 9), found in diffuse-type gastric cancer but not in any normal tissue, was conjugated with the high linear energy transfer alpha-emitter 213Bi and tested for its binding specificity in s.c. and i.p. nude mice tumor models. After intratumoral application in s.c. tumors expressing mutant E-cadherin, the 213Bi-labeled antibody was specifically retained at the injection site as shown by autoradiography. After injection into the peritoneal cavity, uptake in small i.p. tumor nodules expressing mutant E-cadherin was 17-fold higher than in tumor nodules expressing wild-type E-cadherin (62% injected dose/g versus 3.7% injected dose/g). 78% of the total activity in the ascites fluid was bound to free tumor cells expressing mutant E-cadherin, whereas in control cells, binding was only 18%. The selective binding of the 213Bi-labeled, mutation-specific monoclonal antibody E-cadherin delta 9-1 suggests that it will be successful for alpha-radioimmunotherapy of disseminated tumors after locoregional application.

In-frame deletions from the E-cadherin mRNA, coding for a homophilic cell adhesion molecule, are characteristic for diffuse-type gastric carcinomas. Using immunohistochemical analysis the mutant form cannot be distinguished from normal E-cadherin, making results difficult to interpret. In this study, a rat monoclonal antibody, designated E-cad delta 9-1, was generated against a peptide spanning the fusion junction region between exons 8 and 10. This new epitope is present in an E-cadherin variant that lacks exon 9 from the mRNA due to different splice-site gene mutations. Using Western blotting and immunohistochemistry of E-cadherin-transfected cells, we demonstrate that E-cad delta 9-1 specifically reacts with E-cadherin lacking exon 9 but not with the wild-type protein. No immunoreactivity was observed in 31 nontumorous and embryonal tissues analyzed. In gastric carcinoma specimens known to express mutant E-cadherin mRNA lacking exon 9, E-cad delta 9-1 targets exclusively tumor cells in routine formalin-fixed and paraffin-embedded material from biopsies, primary tumors, and lymph node metastases. In a retrospective series of 172 diffuse-type gastric carcinomas expressing E-cadherin, E-cad delta 9-1 reacted with 22 tumors (13%). This new tumor marker-monoclonal antibody system could open novel avenues for selective diagnosis and specific therapy of a subgroup of diffuse-type gastric cancer patients.

E-Cadherin alterations have been reported frequently in sporadic diffuse type gastric and lobular breast carcinomas. Germline mutations of this gene have been identified recently in several gastric cancer families. We analyzed seven patients with a family history of the disease who had diffuse type gastric cancer diagnosed before the age of 45 for germline mutations in CDH1, the gene encoding the E-cadherin protein. We identified a frameshift mutation in exon 3 in one patient with a strong family history of gastric cancer. The same germline mutation was found in the patient's mother, who had metachronous development of lobular breast and diffuse type gastric carcinomas. Immunohistochemistry for E-cadherin protein expression revealed an abnormal staining pattern in both of these tumors, suggesting complete inactivation of the cell adhesion molecule. Thus, our finding suggests that besides diffuse type gastric cancer, lobular breast carcinomas may be associated with germline CDH1 mutations.

Somatic deletion mutations in the cell adhesion molecule E-cadherin are present in almost 50% of diffuse type gastric cancer. We recently generated monoclonal antibodies against an in-frame deletion of exon 9. The aim of this study was to generate and characterize monoclonal antibodies against the second mutational hot spot, in-frame deletions of exon 8. Lou/C rats were immunized using a KLH-coupled peptide that represents a unique sequence generated by fusion of exon 7 and exon 9 from an E-cadherin deletion mutation lacking exon 8. Hybridoma supernatants were tested in a solid-phase immunoassay using BSA-coupled peptide. Positive reacting hybridomas were confirmed by Western Blots, FACS analysis, and immunohistochemistry of E-cadherin negative carcinoma cells that had been transfected with mutant and wild-type E-cadherin cDNA, respectively. In addition, routine formalin fixed and paraffin embedded tissues from gastric cancer patients were analyzed using both mutation-specific and commercial monoclonal antibodies against E-cadherin, including HECD-1 and AEC. Two hybridoma supernatants, termed E-cad delta 8-1, were selected that reacted with the mutant peptide used for immunization and gave strong signals in Western Blot and FACS analysis with cells expressing mutant E-cadherin lacking exon 8. Wild-type protein expressing cells only reacted with the commercial antibodies but not with the two selected hybridoma supernatants. In contrast to AEC, monoclonal antibody HECD-1 did not react with exon 8 deleted E-cadherin, suggesting that the previously unknown epitope for this often used monoclonal antibody is located at least in part within exon 8. Four gastric cancer specimens known to express mutated E-cadherin mRNA strongly reacted with both mutation-specific supernatants and with AEC monoclonal antibody but not with HECD-1. Taken together, we succeeded in generating monoclonal antibodies reacting with mutant E-cadherin protein lacking exon 8. Furthermore, using both HECD-1 and the new mutation-specific antibodies E-cadherin immunoreactivity can for the first time be evaluated in an allele-specific manner in archival tissues.

The cell adhesion molecule E-cadherin (CDH1; MIM# 192090) has been implicated in numerous cellular functions, ranging from controlling morphogenesis to suppressing tumor invasion. We describe 11 previously unreported somatic E-cadherin mutations in two subgroups of gastric and breast cancer showing markedly reduced homophilic cell-to-cell interactions. Using reverse transcription-polymerase chain reaction (RT-PCR) and direct sequencing of the entire coding region 5 mutations were detected in diffuse-type gastric cancer specimens. The sequence alterations include 3 missense mutations affecting exons 3, 10, and 12. Furthermore, two in-frame deletions were identified removing 63 and 9 base pairs from exon 4 and 5, respectively. In invasive Lobular breast cancer 6 E-cadherin mutations were detected after RT-PCR amplification and direct sequencing or using single strand conformation polymorphism (SSCP) analysis followed by sequencing. In addition to two nonsense mutations affecting exon 2, four out-of-frame deletions removing 115 base pairs (entire exon 2), 224 base pairs (entire exon 3), 8 base pairs from exon 12 or 1 base pair from exon 13 were seen. Our report confirms the general principle that in diffuse-type gastric cancer E-cadherin mutations result in structurally altered proteins with possible reduced adhesive functions whereas in invasive lobular breast carcinomas complete loss-of-function mutations are characteristic.

Formalin-fixed and paraffin-embedded tissues are a valuable resource for diagnosis and research. PCR is one of the most powerful methods of retrospective analysis of the DNA present in fixed tissues. One major problem with the molecular analysis of tissue samples, however, is cellular heterogeneity, ie, the large variety of cell types usually present in these specimens can mask cell-specific genetic alterations associated with disease. Herein we describe a procedure for obtaining and analyzing single cells recovered from stained histologic tissue sections without risking contamination from neighboring cells. An ultraviolet laser microbeam was used to physically destroy the tissue surrounding the single cells of interest. These cells, now freed from adjacent cells, were then easily retrieved with a motorized, computer-controlled micromanipulator and molecularly characterized through the use of PCR-based microanalysis. This accurate microdissection technique, followed by DNA amplification and direct sequencing, revealed a novel mutation in the gene coding for the cell adhesion molecule E-cadherin in single tumor cells that was absent in the adjacent single epithelial cells of a patient with early gastric cancer of the diffuse type. In this form of malignancy, tumor cells lose homophilic cell-to-cell interactions and invade the connective tissue as single cells. E-cadherin gene mutations have previously been detected in advanced diffuse-type gastric cancer and gastric carcinoma cell lines. The present study suggests that E-cadherin gene mutations may be an early event in gastric tumorigenesis. The laser-based isolation and subsequent molecular characterization of individual cells, as described herein, allows for micrometer-sized precision and should prove useful in detecting the nucleic acid abnormalities that underlie cancer, infection, and genetic disease.

The calcium-dependent homophilic cell adhesion molecule and candidate suppressor gene, E (epithelial)-cadherin, plays a major role in the organization and integrity of most epithelial tissues. Diffusely growing gastric carcinomas show markedly reduced homophilic cell-to-cell interactions. We speculated that mutations in the E-cadherin gene may be responsible for the scattered phenotype of this type of carcinoma. For that reason we have examined E-cadherin in 26 diffuse type, 20 intestinal type and 7 mixed gastric carcinomas (Laurén's classification) at the DNA, RNA, and protein levels. Reverse transcription polymerase chain reaction and direct sequencing of amplified E-cadherin complementary DNA fragments revealed inframe skipping of either exon 8 or exon 9 in 10 patients with diffuse tumors and an exon 9 deletion in one patient with a mixed carcinoma; both exons encode putative calcium binding domains. These alterations were not seen in nontumorous gastric tissues. Splice site mutations responsible for the exon deletions were identified in six of these patients, eliminating the possibility of alternative splicing mechanisms. Five of these splice site alterations were confirmed as somatic mutations. Non-splice site mutations were observed in three diffuse type tumors, namely a 69-base pair deletion of exon 10 and two point mutations, one of which destroys a putative calcium binding region. Immunohistochemical evaluation showed E-cadherin immunoreactivity in tumors and lymph node metastases of patients expressing abnormal mRNA. The allelic status of the E-cadherin gene was analyzed in one patient, revealing loss of heterozygosity with retention of a mutated E-cadherin allele. Overall, E-cadherin mutations were identified in 50% (13 of 26) of the diffuse type and in 14% (1 of 7) of the mixed carcinomas. In contrast, two silent E-cadherin mutations (not changing the amino acid sequence) were detected in two tumors of the intestinal type. Our study provides strong in vivo evidence that E-cadherin gene mutations may contribute to the development of diffusely growing gastric carcinomas and support a tumor/metastasis suppressor gene hypothesis.