Vimentin immunoexpression is associated with higher tumor grade, metastasis, and shorter survival in colorectal cancer

Abstract

Background/Aim: The association between vimentin immunoexpression and poor prognosis has been described in many human cancers. The objective of this study was to evaluate the relationship between vimentin immunostaining and colorectal carcinoma (CRC) clinicopathologic parameters. Materials and methods: Samples included 202 primary CRC tissues, 41 adenomas and 37 normal colonic mucosae. Anti-Vimentin (V9) monoclonal antibody was used for immunohistochemical staining. Vimentin expression was evaluated based on the percentage of cytoplasmic expression in epithelial cells. Results: Vimentin expression was identified in 35 (17.3%) of CRC samples. All normal mucosa and adenoma samples were vimentin negative. There was an association between positive vimentin immunostaining and high tumor grade, distant metastasis, and short overall (Log rank 5.112, P=0.024), as well as disease-free survival probabilities (Log rank 6.173, P=0.013). There was no association between vimentin expression and age, gender, tumor location, tumor size, tumor stage, nodal involvement, lymphovascular invasion, margin status, or tumor recurrence. Conclusion: Vimentin immunoexpression is associated with worse prognosis in CRC patients. Vimentin can be considered a potentially important disease biomarker and could be a target for CRC therapy.

Introduction

Colorectal carcinoma (CRC) is among the most common neoplasms in the world and is a leading cause of cancer-related death worldwide [1-3]. CRC is the most common cancer among Saudi men and the third-most common cancer among Saudi women [4]. The prognosis of CRC is determined by several factors, including tumor stage, pathological grade, molecular markers, DNA content, and markers that control proliferation and angiogenesis [5-7]. However, other prognostic factors, including currently-unknown molecular markers, may play a prognostic role as well. Epithelial-mesenchymal transition is a biologic phenomenon in which the epithelial cells lose epithelial features and acquire mesenchymal characteristics [8,9]. Vimentin is considered a hallmark of the epithelial-mesenchymal transition in epithelial cells [8,9]. Increased vimentin expression has previously been associated with tumor progression and metastasis in many human cancers [10,11]. However, studies on the role of vimentin are limited in CRC and need further evaluation. The objective of this study was to evaluate the relationship between vimentin immunoexpression and several clinicopathologic measures in a subset of CRC patients.

Materials and methods

Patients

The patients’ data and histopathologic material were collected from the Department of Pathology at King Abdulaziz University Hospital, Jeddah, Saudi Arabia. Tumor grades were reviewed and reclassified according to the World Health Organization’s classification [12]. Tumor stages were reviewed and reclassified according the cancer staging atlas of the American Joint Committee on Cancer [13]. The pathology tissue blocks of 202 primary tumors, 41 adenomas, and 37 normal colonic mucosae were used in this study.

The clinicopathologic findings are shown in Table 1. The study was approved by the Research Committee of the Biomedical Ethics Unit at our institution. The procedures followed were in accordance with the Declaration of Helsinki of 1975, as revised in 2000. Informed written consent was obtained from each patient to obtain permission to utilize their pathologic tissue specimens for laboratory studies.

Table 1.

Clinicopathologic findings of cases (n=202)

Finding		Number (%)
Age	<60 years	108 (53.5%)
	≥60 years	94 (46.5%)
Sex	Male	110 (55%)
	Female	92 (45%)
Tumor location	Right colon	52 (25.7%)
	Left colon	127 (62.9%)
	Rectum	23 (11.4%)
Tumor size	<5 cm	91 (45%)
	≥5 cm	111 (55%)
Grade	Well-differentiated	43 (21.3%)
	Moderately-differentiated	133 (65.8%)
	Poorly-differentiated	26 (12.9%)
Primary tumor	T1	3 (1.5%)
	T2	32 (15.8%)
	T3	149 (73.8%)
	T4	18 (8.9%)
Nodal metastasis	Negative	111 (54.9 %)
	Positive	84 (41.6%)
	Cannot be assessed	7 (3.5%)
Distant metastasis	Positive	57 (28.2%)
	Negative	145 (71.8%)
Lymphovascular invasion	Positive	31 (15.3%)
	Negative	171 (84.7%)
Margin status	Involved	10 (5%)
	Free	192 (95%)
Recurrence	Recurrence	63 (31.2%)
	No recurrence	139 (68.8%)

Tissue microarray

The tissue microarray was constructed as previously described [14,15]. Pathology slides (haematoxylin and eosin-stained) of primary CRC, metastatic tumor, adenomas and normal colonic mucosa tissue were evaluated and selected areas were marked. Areas that showed extensive necrosis, poor cellular preservation, crush artefacts, dominant stromal tissue, or autolytic changes were avoided. Donor paraffin blocks that matched the chosen sections were utilized to get two cores of the selected tissue and then transferred to recipient blocks via a tissue microarray machine (TMA Master 1.14 SP3 from 3D Histech Ltd., Budapest, Hungary). Unstained 4-µm-thick sections were cut from the TMA blocks and utilized for immunohistochemistry studies.

Immunohistochemistry

Immunocytochemistry was performed by utilizing CONFIRM anti-Vimentin (V9) primary antibody, which is a mouse monoclonal antibody (IgG1) directed against vimentin (Ventana Medical Systems, Inc. 1910 E. Innovation Park Drive Tucson, Arizona 85755 USA). An automated immunostainer (Ventana Bench Mark XT, Ventana Inc., Tucson, AZ) was used to perform the immunohistochemistry procedure. The positive control was a leiomyoma tissue that is known to be vimentin-positive. Negative controls were processed without adding the primary antibody.

Evaluation of vimentin immunostaining

The vimentin immunohistochemistry results were evaluated by examining cytoplasmic staining. Vimentin cytoplasmic in >5% of tumor cells considered positive and <5% of tumor cells considered negative.

Statistical analysis

The chi-squared test was used to test the differences between two groups of variables. The overall survival (OS) and disease-free survival (DFS) values were measured by the Kaplan-Meier method with the log-rank (Mantel-Cox) comparison test. DFS was calculated as the time from diagnosis to the appearance of recurrent disease (or date of the last seen disease-free appearance). Statistical analyses were performed using the SPSS® (IMB NY, USA) software package, version 20. P<0.05 was considered significant.

Results

The clinicopathologic features of the cases are summarized in Table 1. Cytoplasmic staining of vimentin was found in 35 (17.3%) of the CRC cases and was not detected in any normal mucosae or adenomas (Figure 1). The expression pattern of vimentin in CRC, normal mucosae, and adenomas is summarized in Table 2. There was an association between positive vimentin immunostaining and high tumor grade (P: 0.004). Vimentin positive staining was identified in 10%, 19% and 32% of well, moderately, and poorly-differentiated tumors, respectively. Vimentin immunoexpression was also associated with distant metastasis (P: 0.042). There was no association between vimentin expression and age, gender, tumor location, tumor size, nodal metastasis, lymphovascular invasion, margin status and tumor recurrence (Table 3). Regarding survival analysis, patients with vimentin-positive tumors tended to have short OS (log rank 5.112, P=0.024) and DFS probabilities (log rank 6.173, P=0.013) (Figures 2 and 3).

Figure 1.

Vimentin immunostaining. A. Sections from colorectal carcinoma show strong positive immunostaining in poorly differentiated area (upper arrow), while a well-differentiated area of the same tumor shows negative staining (lower arrow). B. Sections from colorectal carcinoma with scattered positive staining cells (>5%), considered positive. C. Sections from colorectal carcinoma with rare positive staining cells (<5%), considered negative. D. Sections from colorectal carcinoma with completely negative immunostaining. E. A section from a colonic adenoma shows negative staining (200×). F. A section from a normal colonic mucosa shows negative staining (200×).

Table 2.

Categories of vimentin immunostaining in CRC

Table 3.

Distribution of positive vimentin immunoexpression in relation to clinicopathologic findings

Figure 2.

Overall survival curve (Kaplan-Meier) in relation to vimentin immunoexpression in CRC patients. There is association between OS and vimentin immunostaining (Log rank 5.112, P=0.024).

Figure 3.

Disease-free survival curve (Kaplan-Meier) in relation to vimentin immunoexpression in CRC patients. There is association between high vimentin immunostaining and high DFS probability (Log rank 6.173, P=0.013).

Discussion

Epithelial-mesenchymal transition (EMT) is a biologic phenomenon in which epithelial cells lose their epithelial features, due to processes like the down regulation of E-cadherin, and acquire mesenchymal characteristics, such as the acquisition of vimentin expression [8,9]. The EMT process plays an important role in many human cancers and is associated with cancer progression and poor prognosis [16]. Recent literature revealed that EMT in many human cancers is associated with increased tumor invasiveness and increased metastatic ability [17]. Vimentin is considered a hallmark of EMT epithelial cells [8,9]. Vimentin immune-expression was used to evaluate an immunophenotype that reflected EMT in epithelial neoplasm [18]. Increased vimentin expression has been associated with tumor progression and metastasis in human cancers [10,11].

Vimentin is a major component of the intermediate filament family of proteins. Its expression has been found in many epithelial cancer cells, such as gastric [19], breast [20], esophageal [21], prostate [22], cervical [23], pancreatic [24], and non-small lung carcinomas [25] and found to be associated with aggressive tumor types and poor prognoses. On the other hand, some authors demonstrated that high levels of vimentin were associated with an improved overall survival rate of patients with malignant ovarian tumors [26].

To date, limited studies have evaluated the association between vimentin expression and prognosis in cases of CRC. In such cases, the progression from normal mucosa to invasive colon cancer usually takes years and is associated with many genetic and epigenetic changes [27,28]. The detection of biomarkers can help identify the potential for progression, which is of clinical and possibly therapeutic significance.

Based on our existing knowledge, there is no other study reported from Saudi Arabia that evaluates the expression of vimentin in CRC. In the current study, cytoplasmic staining of vimentin was seen in 17.3% of patients with CRC and was not detected in any normal mucosae or adenomas. Vimentin staining was observed in 56% of the CRC by Xiao et al. [29], 29.9% by Gao et al. [17], 24% by Rashed et al. [30], 14% by Choi et al. [31], and only in 2% by Jaca et al. [32]. The wide variation in the detection rate of vimentin immune-expression could be due to several factors, including different tumor phenotypes, different scoring systems, different antibody clones, and (possibly) different ethnic groups of patients. For example, the cut-off point used to identify vimentin positive tumors was variable between authors and ranged between 5% [18,31,32] and 10% [17,29,33].

In the current study, there was an association between positive vimentin immunostaining and high tumor grade. Vimentin positive staining was identified in 10%, 19%, and 32% of well, moderately, and poorly differentiated tumors, respectively. Vimentin immune-expression was also associated with distant metastasis (P: 0.042). There was no association between vimentin expression and age, gender, tumor location, tumor size, nodal involvement, lymphovascular invasion, margin status, or tumor recurrence. The current study also reveals that CRC patients with vimentin immune-expression tend to experience disease-free survival. Previous studies showed that vimentin expression is associated with tumor grade [17,30,31], tumor stage [17,31], tumor size [18], nodal metastasis [17,18,30,31], distant metastasis [30], lymphovascular invasion [18,31], serosal invasion [18], and disease-free survival [33]. In a meta-analysis by Du et al. [34], it was found that positive vimentin expression is a predictor of poor overall survival (OS) and disease-free survival (DFS) in CRC patients. This analysis also indicated that vimentin expression is related to lymph node metastasis, TNM stage, and N stage. The weak point in this meta-analysis is that all the populations included in that study were in Southeast Asia, thus more studies from all over the world would result in better general dconclusions and confirm the current conclusions.

Historically, management planning for CRC was primarily based on tumor stage. However, the rapid development of molecular biology may open more channels for a therapeutic approach to managing CRC. It is widely accepted that in cases of CRC, EMT is associated with invasive and metastatic phenotype. Vimentin immune-expression could be utilized as a valuable biomarker that is indicative of EMT. Recent publications highlighted the significance of paracrine signaling between tumor cells and the surrounding microenvironment in determining EMT and the acquisition of invasive and metastatic features [35]. There are many molecules interacting with vimentin, which all have different functions [10]. Vimentin gene methylation has been found to occur often in cases of advanced CRC, and can be utilized as a high-sensitivity technique for detecting CRC by testing serum and stool samples [10,36].

In our study, as well as in most of the other studies on CRC, vimentin immune-expression was evaluated in tumor epithelial cells. However, some authors studied vimentin expression in tumor stroma and found that the vimentin expression is also a predictive marker for disease recurrence in CRC patients [37]. Liu et al. found that stromal vimentin immune-expression is an encouraging indicator for survival prediction and therapy response in a subset of CRC patients [38]. Recent in vitro studies revealed that vimentin is a crucial target for anti-metastasis therapy in some cancers [39,40]. The research in this area needs to be further scrutinized.

In summary, there is encouraging evidence for the clinical significance of vimentin immune expression in cases of CRC. Vimentin immune expression is associated with worse prognosis in CRC patients, thus vimentin may be a potentially important disease biomarker and target for CRC therapy.

Disclosure of conflict of interest

None.