OUP user menu

Combined Loss of E-cadherin and Aberrant β-Catenin Protein Expression Correlates With a Poor Prognosis for Small Intestinal Adenocarcinomas

Hee Jin Lee MDPhD, Ok-Jun Lee MDPhD, Kee-Taek Jang MDPhD, Young Kyung Bae MDPhD, Joon-Yong Chung PhD, Dae Woon Eom MDPhD, Joon Mee Kim MDPhD, Eunsil Yu MDPhD, Seung-Mo Hong MDPhD
DOI: http://dx.doi.org/10.1309/AJCPS54RTFCTHGWX 167-176 First published online: 1 February 2013

Abstract

Small intestinal adenocarcinomas (SIACs) are rare, and their molecular pathogenesis is largely unknown. To define the roles of E-cadherin and β-catenin, we performed immunohistochemistry for E-cadherin and β-catenin in 194 surgically resected SIACs with tissue microarrays and compared the data with clinicopathologic factors, including survival rates of patients with SIAC. Loss of E-cadherin expression and aberrant β-catenin expression were observed in 41.8% (81/194 cases) and 40.7% (79/194 cases) of SIACs, respectively. Combined loss of E-cadherin and aberrant β-catenin expression was observed in 24.2% (47/194 cases) of SIACs, and this feature was most frequently observed in mucinous adenocarcinomas and signet ring cell carcinomas (P < .001), poorly differentiated and undifferentiated carcinomas (P < .001), and tumors with advanced pT classification (P = .03). Survival times for patients with SIAC with both loss of E-cadherin and aberrant β-catenin expression (median, 13.9 months) were significantly shorter than those for patients without aberrant expression of both proteins (49.9 months), as determined by univariate (P < .001) and multivariate (P = .01) analyses. In conclusion, loss of E-cadherin and aberrant β-catenin expression correlate with poorly differentiated tumors, advanced T classification, and decreased patient survival time; therefore, it could be a prognostic factor in patients with SIAC.

Key Words
  • Small intestine
  • Adenocarcinoma
  • E-cadherin
  • β-catenin
  • Immunohistochemistry
  • Prognosis

The small intestine is the longest organ of the gastrointestinal (GI) tract, extending from the duodenum to the ileum, with a mucosal surface that covers about 90% of the absorptive surface area of the entire GI tract. Despite its length and large mucosal surface area, only 5% of malignant neoplasms of the GI tract occur in the small intestine. It is estimated that 8,070 Americans will be diagnosed with, and 1,150 people will die from, small intestinal cancers in 2012.1 The global age-adjusted incidence of small intestinal carcinoma is generally less than 1.0 per 100,000, ranging from 0.3 to 2.0 per 100,000.2 Adenocarcinomas are predominantly found in the duodenum and proximal jejunum, whereas neuroendocrine tumors (previously called carcinoid tumors) and lymphomas most commonly occur in the distal jejunum and ileum.2

Recent advances in imaging techniques and endoscopic modalities have improved the detection of small intestinal neoplasms.3 Small intestinal adenocarcinomas (SIACs) are diagnosed at an advanced disease state, and the 5-year survival rate is 41.2%.4 Lymph node metastasis and distal location of these tumors (in the jejunum and ileum) are reported to be the most important independent prognostic factors.4 Several molecular alterations have been reported to be correlated with carcinogenesis of SIACs. These include K-ras mutations, overexpression or mutation of p53, overexpression of cyclin D1 and p27, mutation of deleted pancreatic carcinoma 4 (DPC4), and microsatellite instability.59 However, compared with our knowledge of cancers of other GI tract organs, such as stomach and colorectal cancers, our knowledge of the molecular pathogenesis of SIACs is limited. Therefore, identification of new biomarkers for early detection and/or development of new therapeutic regimens based on a better understanding of the biological mechanisms are essential for this rare but aggressive disease.

E-cadherin is a transmembrane glycoprotein that serves as the prime mediator of epithelial cell adhesion10 The cytoplasmic domains of E-cadherin molecules are tethered to the actin fibers of the cytoskeleton via a complex that comprises α-catenin, β-catenin, axin, and glycogen synthase kinase 3β (GSK3β).10 Loss of E-cadherin from the plasma membrane releases β-catenin, which then migrates to the nucleus, associates with Tcf/Lef transcription factors, and induces expression of several genes orchestrating the epithelial-mesenchymal transition11 In addition, inactivation of the degradation complex protein APC stabilizes nuclear β-catenin, which activates transcription by binding to Tcf/Lef proteins. Reduced β-catenin expression without associated APC abnormality, as well as large deletions and insertions in the β-catenin gene (CTNNB1), has both been identified in SIACs.6,12 Previously, a few case studies reported aberrant E-cadherin and β-catenin protein expression in SIACs with small numbers of cases.6,1215 However, to the best of our knowledge, prognostic implications of E-cadherin and β-catenin expression have never been analyzed in the context of SIACs. In the current study, we analyzed E-cadherin and β-catenin expression in a large number of SIAC cases and demonstrated that E-cadherin and β-catenin expression in SIACs constitute prognostic factors in patients with SIAC.

Materials and Methods

Case Selection

This study focused on primary solitary adenocarcinomas originating in the mucosa of the small intestine. Carcinomas extending into the small intestines from surrounding organs of the GI tract, including the ampulla of Vater, appendix, cecum, pancreas, or stomach, were excluded. Tumors were regarded as having arisen by metastatic cancer to the small intestine when the epicenter of the tumor was located in the subserosa, multiple small intestinal tumors, or serosa of the intestine without an involvement of mucosa by histologic examination. Metastatic tumors were excluded from the study. Instead, the tumor was considered a primary SIAC when the tumor was solitary or predominantly involved the mucosa inconsiderate extension into the serosa, regardless of the presence of peritumoral dysplasia as 1 recent study reported.16

Appropriate measures were taken to protect the rights of all human subjects, and the necessary approvals were received from the institutional review boards of each hospital participating in the Korean Small Intestinal Cancer Study Group. Overall, 194 cases of surgically resected SIACs were collected from the surgical pathology archives of 22 South Korean institutions, as previously described.17 Data collected by reviewing the medical records of patients with SIAC included the age and sex of patients; diagnoses of prior or current malignancies; additional prior or current treatment modalities, including chemotherapy and/or radiation therapy; most recent follow-up dates; and survival status.

Data obtained from the gross examination included the growth pattern, location, and size of the tumor, as well as the operation date. Microscopic characteristics that were evaluated included the depth of invasion, degree of differentiation, and histologic subtype. Other features noted included lymphatic invasion, metastasis to the regional lymph node, metastasis to the pancreas and other intestinal loops, perineural invasion, peritoneal seeding, tumor size, and vascular invasion.

Tissue Microarray Construction

Tissue microarrays were constructed from archival formalin-fixed, paraffin-embedded tissue blocks, as previously described.17 Briefly, a representative tumor area was carefully selected for each tumor from an H&E-stained section of a donor block. Each case was represented by four 1-mm diameter cores, including matched normal small intestine.

Immunohistochemical Staining and Scoring

Immunohistochemistry was performed on 4-μm-thick tissue microarray sections as previously described.17 Briefly, tissue sections were deparaffinized in xylene and hydrated in serially diluted ethanol. Endogenous peroxidase was blocked by incubation in 3% H2O2 for 10 minutes. Antigen retrieval was performed in a steam pressure cooker with preheated antigen retrieval buffer, pH 6 (DAKO, Glostrup, Denmark), at 95°C for 10 minutes. Nonspecific binding of antibodies was minimized by incubating sections with Protein Block (DAKO) for 15 minutes. Microarrays were incubated at room temperature for 36 minutes with antibodies against E-cadherin (4A2C7; 1:200 dilution; Invitrogen, Carlsbad, CA) and β-catenin (CAT-5H10, 1:200 dilution; Invitrogen). Sections were labeled using an automated immunostaining system with an I-View detection kit (Benchmark XT; Ventana Medical Systems, Tucson, AZ). Immunostained sections were lightly counterstained with hematoxylin, dehydrated in ethanol, and cleared in xylene. Normal intestinal epithelial cells included in tissue microarrays were used as positive controls for both E-cadherin and β-catenin. For validation of immunohistochemical staining on tissue microarray slides, we selected 10 conventional slides from 10 included cases in this study, performed immunohisto-chemical staining for E-cadherin and β-catenin, and compared the staining pattern of tissue microarray slides and matched conventional slides from the same case. Both tissue microarray slides and matched conventional slides showed similar staining patterns for E-cadherin and β-catenin staining. To be included for analysis, each cancer had to have sufficient numbers of E-cadherin–labeled cells to permit quantification of the percentage of cells with E-cadherin labeling (>100 cancer cells). E-cadherin expression was evaluated primarily according to the percentage of cells that labeled, although we also evaluated if labeling intensity was an important variable. The results of membranous immunohistochemical staining for E-cadherin and β-catenin were scored with a previously described histologic score (also known as “histoscore”) scheme.1820 The intensity of staining was categorized as 0 (negative), 1 (weak), 2 (moderate), or 3 (strong). In addition, the results of abnormal cytoplasmic catenin were scored the same way. We counted the proportion of labeled cancer cells of approximately 100 cancer cells in each tissue microarray core. The percentage of positive epithelial cells was scored as 0 (<5%), 1 (6%–25%), 2 (26%–50%), 3 (51%–75%), or 4 (>76%). A histoscore was generated as the product of the intensity and the area of staining. The histoscore was then dichotomized into loss of expression (histoscore, 0–6) and intact expression (histoscore, 8–12). We arbitrarily selected a histoscore of 8 as the cutoff point for intact expression because histoscores of 8 or more matched with cases with diffuse (>51%) and strong or moderate intensity. We did not compare other cutoff points in this study. For β-catenin, nuclear staining was also evaluated as present or absent, separately from evaluation of cytoplasmic expression of each specimen. Aberrant β-catenin expression of defined tumor cells showed either cytoplasmic or nuclear expression of β-catenin.

Statistical Analysis

Statistical analyses were performed using SPSS version 17 (SPSS, Chicago, IL). Associations between categorical variables were examined using the Pearson χ2 and Fisher exact tests. Survival curves were calculated by the KaplanMeier method, and statistical significance was evaluated using the log-rank test and the Cox proportional hazards regression model. A P value less than .05 was considered statistically significant.

Results

Clinicopathologic Characteristics of Cases

Clinicopathologic characteristics of the cases in this study are summarized in Table 1. Patient ages ranged from 23 to 86 years (mean, 59.0 years). Of the 194 patients, 121 were men and 73 were women. Tumor sizes ranged from 0.8 to 16 cm (mean, 4.4 cm). Chemotherapy and radiation therapy were performed in 74 (38.2%) and 25 (12.9%) cases, respectively. The length of patient follow-up time ranged from 1 to 158 months, and median survival time at last follow-up was 28 months.

View this table:
Table 1

E-Cadherin and β-Catenin Expression

In normal small intestinal epithelia, moderate to strong signals related to E-cadherin and β-catenin were identified in the cytoplasmic membranes in all 194 cases Image 1A and Image 1B. Representative expression of E-cadherin and β-catenin in SIAC tumor specimens is also depicted Image 1C and Image 1D. Loss of E-cadherin expression, resulting in either complete elimination of or simply a decrease in E-cadherin expression, was observed in 41.8% (81/194 cases) of SIACs. As shown in Table 2, mucinous adenocarcinomas (7/9 cases, 77.8%), signet ring cell carcinomas (4/4 cases, 100.0%), and undifferentiated carcinomas (4/5 cases, 80.0%) were all characterized by a significantly greater reduction in levels of E-cadherin expression than tubular adenocarcinoma (66/176 cases, 37.5%; P = .001, χ2 test). Poorly differentiated cancers (26/42 cases, 61.9%) demonstrated more loss of E-cadherin expression than well-differentiated (17/42 cases, 40.5%) or moderately differentiated (34/105 cases, 32.4%) tumors (P = .002).

Image 1

Representative images of normal and cancer cells following E-cadherin and β-catenin immunohistochemical staining. Both E-cadherin (A, ×100) and β-catenin (B, ×100) were stained in the cytoplasmic membrane in normal small intestinal epithelia. Cancer cells show a loss of E-cadherin labeling (C, ×200) and aberrant nuclear and cytoplasmic staining of β-catenin

View this table:
Table 2

Changes in patterns of β-catenin expression were also observed at the subcellular level, with decreased expression in membranes, increased expression in the cytoplasm, and abnormal expression in the nucleus. Combined loss of β-catenin from membranes and accumulation of β-catenin in the cytoplasm was observed in 39.2% (76/194 cases) of SIACs. Abnormal nuclear β-catenin expression was found in 9.3% (18/194 cases) of SIACs. Three cases were characterized by aberrant accumulation of nuclear β-catenin only without accompanying increased cytoplasmic labeling. Therefore, aberrant β-catenin expression, characterized by a combination of loss of membranous β-catenin and aberrant accumulation of nuclear β-catenin, was present in 40.7% (79/194 cases) of SIACs. The frequency of aberrant β-catenin expression was higher in tumors with infiltrative growth patterns (65/139 cases, 46.8%) than in those with polypoid (8/35 cases, 22.9%) or nodular patterns (3/12 cases, 25.0%; P = .02). Poorly differentiated (24/42 cases, 57.1%) and undifferentiated (4/5 cases, 80.0%) cancers showed significantly more abnormal β-catenin expression than well differentiated (10/42 cases, 23.8%) and moderately differentiated (41/105 cases, 39.0%) cancers (P = .04). The frequency of aberrant β-catenin expression was higher in more deeply invasive cancers (pT3 cancers [26/63 cases, 41.3%] and pT4 cancers [53/111 cases, 47.7%]) than in superficially invasive tumors (pTis-T2 cancers [0/20 cases, 0%]; P < .001). Aberrant β-catenin expression was observed more frequently in SIACs with other loop invasion (5/5 cases, 100%) than those without other loop invasion (74/189 cases, 39.2%; P = .01). Abnormal β-catenin expression was also observed more frequently in SIACs with lymphovascular invasion (49/100 cases, 49.0%) than those without lymphovascular invasion (30/94 cases, 31.9%; P = .019).

A combined loss of E-cadherin and aberrant β-catenin expression was observed in 24.2% (47/194 cases) of SIACs. Simultaneous loss of E-cadherin and aberrant β-catenin expression was more frequently observed in mucinous adenocarcinomas (7/9 cases, 77.8%), signet ring cell carcinomas (4/4 cases, 100%), and undifferentiated carcinomas (3/5 cases, 60%) than in tubular adenocarcinomas (33/176 cases, 18.8%; P < .001). Moreover, SIACs with combined decreased E-cadherin and aberrant β-catenin expression were more associated with poorly differentiated (21/42 cases, 50.0%) or undifferentiated (3/5 cases, 60.0%) cancers than well-differentiated (6/42 cases, 18.8%) or moderately differentiated (17/105 cases, 16.2%) cancers (P < .001). Concomitant loss of E-cadherin and aberrant β-catenin expression in SIACs were mainly observed in more deeply invaded tumors (pT3 tumors [18/63 cases, 28.6%] and pT4 tumors [29/111 cases, 26.1%]) than less deeply invaded cancers (pTis-pT2 cancers, 0/20 cases, 0%; P = .03).

Survival Analysis

Median survival time in patients with SIAC with either no or reduced E-cadherin expression (median survival, 22.6 months) trended shorter than that in patients with intact expression (median survival, 48.4 months; P = .054, log-rank test) Figure 1A.

Figure 1

Survival of small intestinal adenocarcinomas (SIAC) patients based on E-cadherin and β-catenin labeling. A, Median survival in patients with SIAC with loss of E-cadherin expression (median survival, 22.6 months) trended lower than that in patients with intact E-cadherin expression (median survival, 48.4 months; P = .054, log-rank test). B, Median survival in patients with SIAC with aberrant β-catenin expression (20.1 months) was significantly worse than that in patients with intact β-catenin expression pattern (50.1 months; P = .003). C, Survival of patients with SIAC according to combined E-cadherin and β-catenin expression. The median survival times of patients with SIAC with intact expression of both E-cadherin and β-catenin (n = 81), loss of E-cadherin expression only (n = 34), aberrant β-catenin expression only (n = 32), and combined loss of E-cadherin and aberrant β-catenin expression (n = 47) were 48.5 months, 74.1 months, 41.6 months, and 13.9 months, respectively. There was a significant survival difference among the 4 groups by overall comparison (P < .001). When compared in a pairwise manner, patients with SIAC with combined loss of E-cadherin and aberrant β-catenin expression had significantly shorter survival times than those with intact expression of both E-cadherin and β-catenin expression (P < .001), those with loss of E-cadherin expression only (P = .004), and those with aberrant β-catenin only (P = .04). However, no differences in survival were apparent from other group comparisons. D, Median survival in patients with SIAC with combined loss of E-cadherin and aberrant β-catenin expression (13.9 months) was significantly lower than those without combined decreased E-cadherin and aberrant β-catenin expression (49.9 months; P < .001).

Median survival in patients with aberrant β-catenin expression (20.1 months) was significantly worse than that in patients with an intact β-catenin expression pattern (50.1 months; P = .003, log-rank test) Figure 1B. The 1-, 3-, and 5-year survival rates in the aberrant β-catenin expression group were 62%, 37%, and 29%, respectively, whereas the corresponding rates in the retained membranous β-catenin expression group were 79%, 59%, and 46%.

The survival differences were compared after combining E-cadherin and β-catenin expression patterns. The median survival times for patients with SIAC with both intact E-cadherin and β-catenin expression (n = 81), reduced E-cadherin expression only (n = 34), aberrant β-catenin only (n = 32), and both reduced E-cadherin and aberrant β-catenin (n = 47) were 48.5 months, 74.1 months, 41.6 months, and 13.9 months, respectively. There was a significant survival difference among the 4 groups (P < .001, log-rank test, overall comparison) Figure 1C. When compared in a pairwise manner, patients with SIAC with both reduced E-cadherin and aberrant β-catenin had significantly shorter survival than those with intact expression of both E-cadherin and β-catenin (P < .001), those with reduced E-cadherin expression only (P = .004), and those with aberrant β-catenin expression only (P = .04). However, there was no difference in survival between other group comparisons.

Median survival in patients with SIAC with combined reduced E-cadherin and aberrant β-catenin expression (13.9 months) was significantly lower than in those without combined decreased E-cadherin and aberrant β-catenin expression (49.9 months; P < .001) Figure 1D.

Univariate Analysis of Other Clinicopathologic Factors

The relationships between other clinicopathologic variables and survival are summarized in Table 3. The clinicopathologic factors associated with shorter patient survival by univariate survival analysis were tumor location (P = .03), pT classification (P = .02), lymph node metastasis (P < .001), other loop invasion (P = .025), retroperitoneal seeding (P < .001), perineural invasion (P = .008), lymphovascular invasion (P < .001), and radiation therapy (P = .002). In contrast, survival was not associated with sex, growth pattern, histologic subtype, differentiation, pancreatic invasion, or chemotherapy.

View this table:
Table 3

Multivariate Analysis of Clinicopathologic Factors

The independent prognostic significance of a combined reduced E-cadherin and aberrant β-catenin expression as well as other clinicopathologic parameters were determined by applying the Cox proportional hazards model. Although β-catenin expression, pT classification, and lymphovascular invasion were significant by univariate analyses, they were not included for the Cox regression model because β-catenin expression was associated with other factors, such as combined reduced E-cadherin and aberrant β-catenin expression. Similarly, loop invasion and retroperitoneal seeding, which were components of pT classification as well as lymphovascular invasion, were closely linked with pN classification.

Aberrant E-cadherin and β-catenin expression (P = .01), pN classification (P = .01), and retroperitoneal seeding (P = .02) were all independently prognostic in our model. The hazard ratio for SIAC with reduced E-cadherin and aberrant β-catenin expression was 1.82 (95% confidence interval, 1.15–2.87) compared with those of intact E-cadherin and β-catenin Table 4.

View this table:
Table 4

Because cases with pT3 and pT4 composed about 90% of the cases, additional multivariate analysis was performed with pT3 and pT4 cases. Again, aberrant E-cadherin and β-catenin expression (P = .008), pN classification (P = .02), and retroperitoneal seeding (P = .02) remained as independent prognostic factors after stratifying with pT3 and pT4 classifications.

Discussion

We observed abnormal expression of the E-cadherin and β-catenin proteins in a large number of SIACs. Loss of E-cadherin was observed in 41.8% (81/194 cases) of SIACs, and aberrant β-catenin was observed in 40.7% (79/194 cases) of SIACs. Combined loss of E-cadherin and aberrant β-catenin expression was observed in 24.2% (47/194 cases) of SIACs. Only a few previous studies analyzed abnormal β-catenin expression in SIACs, with frequencies of 19% to 81% reported with small numbers (fewer than 32 cases) of SIACs.6,1215 A study concerning aberrant E-cadherin expression reported that 38% of SIACs had decreased membranous expression of E-cadherin,6 which is similar to the percentage frequency we observed.

Although a few previous studies reported aberrant E-cadherin and β-catenin protein expression in SIACs, they did not compare rates of aberrant expression of E-cadherin and β-catenin protein owing to small sample sizes.6,1215 In the current study, we observed more common loss of E-cadherin expression in poorly differentiated or undifferentiated carcinomas than in moderately differentiated or well-differentiated carcinomas. Signet ring cell carcinomas, undifferentiated carcinomas, and mucinous carcinomas showed more frequent reduced or no E-cadherin expression than conventional tubular adenocarcinomas. It is reasonable to deduce that reduced E-cadherin expression was more frequent in poorly differentiated or undifferentiated carcinomas because the main function of E-cadherin involves adhesion of epithelial cells. Wheeler et al6 previously observed decreased membrane expression of E-cadherin in 38% (8/21 cases) of SIACs and did not find an association between reduced E-cadherin expression and differentiation of SIACs owing to the small number of examined cases. Loss of E-cadherin expression was associated with loss of intercellular junctional or cellular polarity of cancer cells, which are characteristics of poorly cohesive or noncohesive cancer cell phenotypes, such as undifferentiated carcinomas or signet ring cell carcinomas, which were previously described in stomach and pancreatic cancers.2123 Likewise, the poorly cohesive phenotype of poorly differentiated or undifferentiated SIACs and signet ring cell carcinomas in the present study was associated with complete loss or a reduced level of E-cadherin expression.

Combined loss of E-cadherin and aberrant β-catenin expression was more commonly seen in poorly differentiated or undifferentiated SIACs than in moderately differentiated or well-differentiated SIACs. The incidence of aberrant β-catenin expression was higher in signet ring cell carcinomas, undifferentiated carcinomas, and mucinous carcinomas than in conventional tubular adenocarcinomas. We also observed that SIACs with aberrant β-catenin expression were associated with increased depth of invasion and invasion of other intestinal loops and the lymphovascular system. To the best of our knowledge, ours is the first study to evaluate β-catenin expression in the context of clinicopathologic variables, including survival of patients with SIAC.

Murata et al24 first reported that SIACs occasionally carry mutations of CTNNB1, which encodes β-catenin. They observed somatic interstitial deletion of exon 3 in CTNNB1 and deduced that an activating mutation of CTNNB1 is involved in the carcinogenesis of SIACs.24 Wheeler and colleagues6 evaluated the roles of mutation of APC and used immunohistochemistry to characterize E-cadherin and β-catenin levels in 21 cases of SIAC. They did not find any APC mutation but observed frequent abnormal expression of E-cadherin and β-catenin, which led them to conclude that SIACs have a distinct genetic pathway with colorectal cancers. Likewise, Blaker et al14 examined 21 SIACs and identified the CTNNB1 mutation in 1 case and abnormal β-catenin expression in 5 cases. They insisted that accumulation of nuclear β-catenin following the activation of Wnt signaling is important in the carcinogenesis of SIACs and concluded that nuclear accumulation of β-catenin may not be caused by mutations that either inactivate APC or activate CTNNB1.14

We did not include any case of ampullary carcinoma in the present study because ampullary carcinomas may have combined features of carcinomas of the small intestine (especially duodenal cancers), distal bile duct, or pancreas. Interestingly, the prognostic significance of reductions in the levels of either E-cadherin or β-catenin was reported in 2 previous studies of ampullary carcinomas.25,26 Loss of E-cadherin expression was reported for 34% to 63% of the tested cases, and aberrant β-catenin expression was reported for 41% to 60%.25,26 The proportion of loss of E-cadherin and the extent of aberrant β-catenin expression were both similar to those seen in the current study. Loss of E-cadherin and aberrant β-catenin expression was associated with worse survival in patients with ampullary carcinomas.25,26 Similarities between these studies and ours in the extent of the loss of E-cadherin, the extent of aberrant β-catenin expression, and correlation of both parameters with worse survival suggest that common molecular profiles may exist between patients with SIAC and ampullary carcinoma.

In the present study, we observed that β-catenin expression and a combination of reduced E-cadherin and aberrant β-catenin expression were both associated with a worse prognosis in patients with SIAC. Combined abnormal E-cadherin and β-catenin expression has been reported as a prognostic indicator in cancers from other organs, including colorectal cancers,27,28 as well as head and neck squamous cancers.29

In summary, loss of E-cadherin and aberrant β-catenin expression are commonly observed in surgically resected SIACs. Loss of membranous E-cadherin and aberrant β-catenin expression are more frequently observed in tumors with poorly cohesive phenotypes, including poorly differentiated, undifferentiated, and signet ring cell carcinomas. Combined abnormalities in both E-cadherin and β-catenin expression are negatively correlated with patient survival, thus providing a prognostic factor for patients with SIAC.

Acknowledgments

We thank the members of the Korean Small Intestinal Cancer Study Group: Dr Hee-Kyung Chang, Kosin University College of Medicine, Pusan; Dr Eun Sun Jung, The Catholic University of Korea College of Medicine, Seoul; Drs Ghil Suk Yoon and Han-Ik Bae, Kyungpook National University, Dague; Dr Young-Ha Oh, Hanyang University College of Medicine, Seoul; Dr Gwangil Kim, Bundang CHA Medical Center, CHA University, Seongnam; Dr Soo Jin Jung, Inje University College of Medicine, Busan; Dr Mi Jin Gu, Fatima Hospital, Daegu; Dr Jung Yeon Kim, Inje University Sanggye Paik Hospital, Seoul; Dr Kyu Yun Jang, Chonbuk National University Medical School, Jeonju; Dr Kye Won Kwon, Bundang Jesaeng General Hospital, Seongnam; Dr Gyeong Hoon Kang, Seoul National University College of Medicine, Seoul; Dr Jae Bok Park, Catholic University of Daegu, Daegu; Dr Soon Won Hong, Yonsei University College of Medicine, Seoul; and Dr Ji Shin Lee, Chonnam National University Medical School, Gwangju, South Korea.

Footnotes

  • Dr H. J. Lee and Dr O.-J. Lee equally contributed to this work.

  • This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) and funded by the Ministry of Education, Science, and Technology (2010-0004807) and a grant (2013-554) from the Asan Institute for Life Sciences, Seoul, South Korea.

References

View Abstract