Abstract
Aim: To investigate the potential involvement of claudin-1 (CL-1) in the tumorigenesis of rectal cancer by analyzing the correlation between CL-1 expression, clinicopathological factors and prognosis. Patients and Methods: Rectal cancer tissue specimens from 306 patients that had undergone surgical treatment were evaluated using immunohistochemical analysis for expression of CL-1 and correlated with clinicopathological factors. Results: A reduced expression of CL-1 (less than 30% of tumor cells strongly, positively stained) correlated significantly with poor prognosis in stage II and III rectal cancer. Moreover, the expression levels of CL-1 correlated significantly with tumor differentiation and perineural invasion (p=0.037 and 0.009, respectively). However, no significant differences were detected between the expression levels of CL-1 and other clinicopathological factors. Conclusion: Loss of claudin-1 expression is a strong predictor of disease recurrence and poor patient survival in stage II and III rectal cancer.
Claudin-1 (CL-1) and claudin-2 (Cl-2) are transmembrane proteins which are an integral component of tight junction strands (1). The altered expressions of some claudins (CLs) have been found in many types of human cancer, such as breast, ovary, prostate, liver, stomach and colon (2, 3). CLs have also been identified as potential targets in the development of molecular-based strategies for diagnosis (4), prediction of progression (5), disease recurrence (6), cell invasion and metastasis (7-11), as well as for the improvement of therapeutic strategies (12). Recent studies have implicated CLs in tumorigenesis, cancer cell invasion and migration (13).
The up-regulated expression of CL-1 has been seen in colorectal cancer (CRC) (14, 15). The loss of function of the tight junction proteins (TJPs) in CRC (16) is believed to increase the access of tumor cells to nutrients and signaling peptides (17), down-regulate cell-to-cell adhesion and increase motility and metastasis (18, 19). It has also been seen that β-catenin and CL-1 co-localize in the nucleus of many metastatic colonic adenocarcinomas (20). The differential expression of genes encoding TJPs, which increases the metastatic potential of tumor cell lines (22), has also been reported in CRC (6, 21).
CL-1 expression has been reported to have a prognostic value in CRC (14, 15, 21). Resnick et al. (23) reported that CL-1 expression was a positive prognostic indicator and correlated with lower tumor grade, absence of lymphovascular invasion and increased patient survival in stage II colonic cancer. But the relationship between CL-1 expression and prognosis in rectal cancer has not yet been examined.
The aim of this study was to analyze the correlation between the expression of CL-1 in rectal cancer tissues and clinicopathological factors and to investigate the effectiveness of CL-1 as a possible prognostic marker in rectal cancer patients.
Patients and Methods
Patients and tissue samples. A total of 306 patients with rectal cancer who had undergone surgical treatment at Kurume University Hospital in Fukuoka, between January 1999 and December 2007 were included in the study. Informed consent was obtained from all the patients prior to surgical resection. The study was approved by the Institutional Review Committee for Research on Human Subjects, at Kurume University Hospital. Clinical follow-up was performed for the duration of patient survival. Patients were classified as having no evidence of disease or having evidence of relapse. Relapse was defined as local recurrence or the development of distant metastasis.
Histopathology. After surgery, tumor specimens, the resection margin and lymph nodes were fixed in formalin and embedded in paraffin. Hematoxylin and eosin-stained sections were reviewed to establish the pathological diagnosis. Tumor differentiation and the degree of invasion were examined by pathologists and histopathological classification was performed according to the General Rules for Colorectal Cancer Study (24) and/or TNM. The patient clinical information is summarized in Table I.
Patient demographics.
Immunohistochemistry. Immunohistochemistry was performed as described in our previous studies (14, 25). The tissue sections were stained with monoclonal antibodies (diluted at 1:400) against CL-1 (Zymed Laboratories Inc., San Francisco, CA, USA). The sections were stained using a BenchMark XT device IHC automated system (Ventana medical systems, Tucson, Arizona, USA).
Quantification of immunostaining. To evaluate the cytoplasmic staining intensity, an intensity score (IS) was used as follows: zero staining intensity was scored as 0, marginal intensity as +1, medium intensity as +2 and strong intensity as +3 (Figure 1) (26). At the same time, the immunoreactivity of the membranous CL-1 expression was also scored by estimating the percentage of strongly positive tumor cells (PS), termed CL-1 PS (27) (Figure 2). All the immunohistochemical studies were evaluated by two experienced observers who were blinded to the condition of the patients.
Statistical analysis. Statistical analysis was performed using JMP version 8.0 (SAS Institute, Cary, NC, USA). Statistical comparisons were made using Fisher's exact test, the chi-square test or the Wilcoxon rank-sum test, depending on the type of data. Values of p<0.05 were considered to indicate statistical significance.
The relationships between CL-1 expression and overall (OS) and disease-free survival (DFS), as well as between other clinicopathological findings and molecular markers, were examined using the Kaplan-Meier method and the log-rank test. Hazard ratios were estimated by Cox regressions.
Results
Patient clinical course. For the rectal cancer patients, the median duration of follow-up after surgery was 38 months, with a range of 1-123 months over all stages. Relapse was diagnosed in 21.1% of the patients and the five-year survival rate was 86.2% in both stage II and III patients.
Expression of CL-1 protein in human rectal cancer tissue specimens. The CL-1 expression was correlated with the CL-1 PS as shown in Figure 3. Figure 4A shows the CL-1 expression in terms of the IS at all cancer stages. No significant correlation was detected. Surprisingly, positive cases (1+, 2+ and 3+) were not as common in stage IV as in the other stages. On the other hand, in the group where PS was <30%, CL-1 PS showed increased correlation with disease stage in the same cases, while correlation decreased in the 30%<PS<60% group (Figure 4B).
Representative examples of histological and immunohistological staining. Left panel: HE staining. Right panel: CL-1. A: Well-differentiated adenocarcinoma, CL-1 IS=3. B: Moderately differentiated adenocarcinoma, CL-1 IS=2. C: Signet-ring cell adenocarcinoma, CL-1 IS=1. D: Moderately differentiated adenocarcinoma, CL-1 IS=0. Magnification: ×200. IS: Intensity score.
CL-1 PS scoring system. Red frames indicate strongly positive staining regions. Magnification: ×20.
Using these results, the cut-off value of the CL-1 PS was determined by means of a receiver operating characteristic (ROC) curve, fitted using JMP software. The cut-off value was determined to be at 30%.
Correlation between CL-1 expression and clinicopathological features. The patients were divided into two groups: those with a CL-1 PS of 30% or less and those with a CL-1 PS >30%. As shown in Table II, using multivariate analysis, the expression of CL-1 significantly correlated with tumor differentiation and perineural invasion (PNI) (p=0.037 and 0.009, respectively). Surprisingly, no significant differences were detected between the expression levels of CL-1 and other clinicopathological factors, such as age, sex, tumor location, T category, lymph node metastasis and preoperative serum carcinoembryonic antigen (CEA) level. Univariate analysis of recurrence revealed a significant association with tumor differentiation (p=0.007), lymphatic invasion (p=0.048), PNI (p=0.001), preoperative CEA level (p=0.01) and low levels of CL-1 expression (CL-1 PS) (p=0.002) in stage II and III patients (Table III). Multivariate Cox analysis revealed that tumor location (p=0.005), venous invasion (p=0.038), PNI (p=0.033), preoperative CEA level (p=0.027) and low level of CL-1 expression (CL-1 PS) (p=0.046) were associated with recurrent disease in the same group (Table III). Kaplan-Meier plots illustrating the association of CL-1 expression with recurrence and survival are shown in Figures 5A and 5B. A clearly reduced expression of CL-1 (less than 30% of CL-1 PS) correlated significantly with poor prognosis.
Correlation between the percentage of cells positive for membranous CL-1 (CL1-PS) and the intensity of cytoplasmic CL-1 expression in tumor tissue (p<.001, Spearman rank correlation analysis).
Analysis of patient characteristics and CL-1 expression.
CL-1 expression at each cancer stage. A: Staining intensity score (IS) of cytoplasm: no staining=0; marginal staining=1; medium staining=2 and strong staining=3. B: Percentage of strongly CL-1-positive membrane-stained tumor cells.
A: Overall survival rates at stages II and III. B: Disease-free survival rates at stages II and III.
Analysis of patient characteristics and disease-free survival.
Discussion
Surgical resection is the primary treatment modality for CRC. The most powerful tool for assessing prognosis following surgery is pathological analysis of the resected specimen. Although the parameters that determine the pathological stage are the strongest predictors of postoperative outcome, such as histological grade, lymphatic or vessel invasion and PNI which have prognostic significance independent of stage (28). While tumor extent, lymph node status, tumor grade and the assessment of lymphatic and venous invasion remain the most important additional histological parameters, they are not regarded as essential in prognosis.
While molecular analysis and gene expression profiling as a means of improving identification of patients likely to have a poor clinical outcome and therefore more likely to benefit from adjuvant treatment have been reported (29-31), reliable prognostic markers identified by immunohistochemical protein profiling are yet to be established.
We have previously reported that CL-1 plays a pivotal role in the regulation of cellular morphology and behavior in the colonic epithelium using a CL-1 overexpressing CRC cell line (14, 15). While it is unclear why reduced expression of CL-1 correlated significantly with poor prognosis, in the present study, this finding suggested that the loss of certain CLs has a greater impact on tumor aggression than others, and CL-1 loss may lead to tumor progression.
Interestingly, the present study showed that reduced expression of CL-1 was significantly associated with PNI. PNI is the process of neoplastic invasion of nerves and is an under-recognized route of metastatic spread. PNI has emerged as a key pathological feature of many other malignancies, including those of the pancreas, colorectal, prostate, biliary tract and stomach. For many of these malignancies, PNI is a marker of poor outcome and a harbinger of decreased survival (32-37). Shirouzu et al. reported that PNI was an important factor in influencing the prognosis of rectal cancer patients with stage III disease (38).
Taken together, these data suggest that CL-1 protein expression may have significant clinical relevance and it may therefore become a potentially useful immunohistochemical and prognostic marker in rectal cancer. This study was the first to examine comprehensively the expression of CL-1 in rectal cancer and to correlate CL-1 expression with disease progression. A reduced expression of CL-1 was found to be significantly associated with poorly differentiated tumors and PNI. It was concluded that the loss of CL-1 expression was a strong predictor of disease recurrence and poor survival in stage II and III rectal cancer patients. The use of CL-1 as a novel prognostic factor in rectal cancer is therefore proposed.
Acknowledgements
This research was funded by Grant-in-Aid for Scientific Research (No. 21591740), Japan for which the Authors are grateful. We would also like to thank Kenta Murotani of the Translational Research Informatics Center for statistical advice.
- Received April 14, 2011.
- Revision received June 18, 2011.
- Accepted June 20, 2011.
- Copyright© 2011 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.