Abstract
Background: Angiopoietin-like protein-2 (ANGPTL2) mediates chronic inflammation. Tumor cell-derived ANGPTL2 promotes tumor invasion and angiogenesis. ANGPTL2 expression has not been fully investigated in gastric cancer (GC). Materials and Methods: ANGPTL2 expression in 354 patients with GC was assessed by immunohistochemistry (IHC). IHC scores were calculated, and the association of ANGPTL2 with clinicopathological factors and patient outcomes was evaluated. Results: Immunoreactive ANGPTL2 protein was expressed mainly in GC cell cytoplasm. Kaplan–Meier analysis showed that high expression of ANGPTL2 indicated significantly poorer overall disease-free survival (DFS). Among patients with curatively resected GC, multivariate analysis for DFS revealed that high ANGPTL2 expression (p<0.01), advanced T-stage (p=0.015), lymph node metastasis (p<0.01), and advanced Unio Internationalis Contra Cancrum (UICC) stage (p<0.01) were independent risk factors for poor DFS. Conclusion: High cytoplasmic ANGPTL2 expression in GC tissue was associated with tumor progression, invasion, metastasis, and poor prognosis. ANGPTL2 may be a useful marker for detecting early postoperative recurrence in patients with GC.
Gastric cancer (GC) is the leading cause of cancer-related death worldwide, despite a worldwide decline in its incidence (1, 2). Survival rates in patients with GC without distant metastasis have improved over the past decade as a result of developments in surgery and chemotherapy (3, 4). However, approximately one-third of patients with GC have disease at an advanced stage with distant metastasis at the time of diagnosis, and their prognosis is poor (5). Hence, there is continuing interest in prognostic factors that can permit more accurate patient stratification and improve clinical decision-making in patients with GC (6).
Chronic inflammation has important roles at various stages of cancer development, including inflammatory carcinogenesis, tumor progression, invasion, and metastasis (7). In the case of GC, chronic inflammation induced by Helicobacter pylori infection promotes gastric mucosal atrophy, metaplasia, and dysplasia, precursors to GC (8).
Angiopoietin-like protein-2 (ANGPTL2) is a secreted glycoprotein belonging to the ANGPTL family identified as a group of proteins containing a coiled-coil domain and a fibrinogen-like domain that are structurally similar to angiopoietin (9). Adipose tissue-derived ANGPTL2 activates an inflammatory cascade in endothelial cells, resulting in initiation and propagation of inflammation in adipose tissues and obesity-related metabolic diseases (10). Moreover, ANGPTL2 acts as an important rheumatoid synovium-derived inflammatory mediator in the pathogenesis of rheumatoid arthritis (11). Overall, ANGPTL2 acts as a mediator of chronic inflammation.
ANGPTL2 expression is also highly associated with the frequency of carcinogenesis and of lymph node metastasis and distant organ metastasis in a mouse model of chemically induced skin squamous cell carcinoma (SCC) (12). Overexpression of ANGPTL2 in tumor cells has been associated with tumor progression in several cancer types including breast cancer (13, 14), lung cancer (13, 15), prostate cancer (16), hepatocellular carcinoma (17), pancreatic ductal carcinoma (18), osteosarcoma (19), oesophageal cancer (20), and colorectal cancer (21). Serum ANGPTL2 expression is a potential diagnostic biomarker for patients with GC (22) but the association between ANGPTL2 expression in tumors and the clinicopathological features in patients with GC has not been fully investigated.
In this study, we evaluated the association of ANGPTL2 expression and clinicopathological characteristics in patients with GC using immunohistochemistry (IHC).
Materials and Methods
Patients. A total of 354 patients with GC who underwent surgical resection at the Department of Gastrointestinal and Paediatric Surgery of the Mie University Graduate School of Medicine from 2001 to 2011 were included in this study. Patients were excluded from this analysis because of incomplete clinical data, inadequate follow-up, or inadequate IHC analysis. Of the 354 patients, 251 were men and 103 were women, with a median age of 68 years (range=18-90 years). Staging was based on clinical assessment and histopathological analysis according to the International Union Against Cancer (UICC) TNM staging system (23). There were 155 patients with stage I disease, 68 with stage II, 70 with stage III, and 61 with stage IV. They were followed up every 2 weeks to 6 months after discharge from the hospital. The median follow-up time was 28.1 months (range=1.0-124.3 months). Adjuvant chemotherapy was administered to all patients with stage III and high-risk stage I and II disease. A total of 59 patients experienced a recurrence after curative surgery. Overall, 73 patients died of GC-related causes during this period. Written informed consent from all patients was obtained according to the local ethics guidelines. The Medical Ethics Committee of Mie University Graduate School of Medicine approved this retrospective study (NO.2215).
Immunohistochemistry. All resected specimens were fixed in 10% formalin, embedded in paraffin, and stained with haematoxylin and eosin (H&E). All histopathological data were reviewed from the corresponding H&E-stained slides. Paraffin-embedded tissue sections (3 μm) were subjected to IHC. Whole-tissue sections were made and used to assess histology and tumor cell ANGPTL2 expression. The sections were deparaffinized by xylene, rehydrated in graded concentrations of ethanol, and pretreated in an autoclave at 121°C for 10 min in 10 mM citrate buffer (pH 6.0). Endogenous peroxidase activity was blocked by incubation for 10 min with 0.3% hydrogen peroxide in ethanol. Nonspecific binding sites were blocked with 1 mol/l phosphate-buffered saline (PBS) with 10% normal goat serum (Vector Laboratories, Burlingame, CA, USA). The sections were incubated with a primary mouse monoclonal antibody to ANGPTL2 (1:100; R&D Systems, Foster City, CA, USA) in PBS containing 1% bovine serum albumin for 16 h at 4°C. After washing with PBS, sections were loaded with secondary antibody coupled with peroxidase-conjugated polymers (ENVISION+ System, DakoCytomation A/S; Dako, Glostrup, Denmark) for 30 min. The primary antibodies were then detected using Dako ChemMate ENVISION Kit/HRP diaminobenzidine (DAB) (DakoCytomation A/S; Dako, Glostrup, Denmark). The sections were counterstained with haematoxylin and dehydrated in graded concentrations of ethanol and xylene.
Evaluation of ANGPTL2 protein expression. Each slide was observed by scanning the whole section at low (×40) and high (×200) magnification. Two independent researchers with no prior knowledge of the clinicopathological parameters evaluated ANGPTL2 immunoreactivity at the centre of the tumor based on the staining intensity and extent. Staining intensity was scored as 0 (negative) (Figure 2a), 1 (weak) (Figure 2b), 2 (medium) (Figure 2c), or 3 (strong) (Figure 2d). Staining extent was scored as 0 (0%), 1 (1-25%) (Figure 3a), 2 (26-50%) (Figure 3b), 3 (51-75%) (Figure 3c), or 4 (76-100%) (Figure 3d) according to the percentage of tumor cells with cytoplasmic ANGPTL2 expression. Final IHC scores were calculated by multiplying the intensity of staining by the extent. Consequently, IHC scores were in the range of 0-12.
Statistical analysis. All statistical analyses were performed using JMP version 10 (SAS Institute, Cary, NC, USA). Associations between ANGPTL2 cytoplasmic expression at the centre of the tumor and clinicopathological characteristics were evaluated using the Mann–Whitney U-test. The overall survival (OS) time was calculated from the date of the operation to the date of GC-specific death. Alternatively, it was censored at the time of death from other causes or at the end of follow-up. The disease-free survival (DFS) time was calculated in the cohort of GC patients with curative intent from the date of operation to the date of recurrence. The Kaplan–Meier method was used to determine the cumulative probability of OS and DFS. For these survival analyses, we defined the optimum cutoff value of ANGPTL2 IHC score as determined by receiver operating characteristic analysis. The log-rank test was used for comparisons of the OS and DFS rates in the patient groups. Univariate and multivariate analyses were performed using the Cox proportional hazards model to determine the factors affecting OS and DFS. Parameters with p<0.05 in the univariate analysis were used for the multivariate analysis. All values with p<0.05 were considered statistically significant.
Results
ANGPTL2 is highly expressed in GC compared with adjacent normal gastric mucosa. Immunoreactive ANGPTL2 protein was expressed mainly in the cytoplasm of GC cells (Figure 1a) but was weak in or absent from adjacent normal mucosa (Figure 1b). According to our IHC scoring system, the mean±SD ANGPTL2 IHC score for the 354 patients was 3.1±2.4. Overall, 21 patients had no ANGPTL2 expression (IHC score 0).
ANGPTL2 expression increases according to GC tumor progression. High cytoplasmic ANGPTL2 expression was significantly associated with male sex (p<0.01), large (>35 mm) tumor size (p<0.01), advanced T-stage (p<0.01), lymphovascular invasion (p<0.01), lymph node metastasis (p<0.01), synchronous liver and distant organ metastasis (p<0.01), and advanced UICC stage (p<0.01) in patients with GC (Table I).
High ANGPTL2 in GC patients is significantly associated with a poor prognosis. Patients were divided into two groups: patients with an ANGPTL2 IHC score ≥5 (n=73) (high expression group) and those with an ANGPTL2 IHC score ≤4 (n=281) (low expression group). Kaplan–Meier analysis showed that the patients in the high ANGPTL2 expression group had significantly poorer OS than the patients in the low-expression group (p<0.01) (Figure 4). Univariate analysis for OS showed that older age (p<0.01), large tumor size (p<0.01), advanced T-stage (p<0.01), lymphovascular invasion (p<0.01), lymph node metastasis (p<0.01), synchronous liver, peritoneal, and distant organ metastasis (p<0.01), advanced UICC stage, and high ANGPTL2 expression (p<0.01) were risk factors for poor OS. However, the multivariate analysis did not retain high expression of ANGPTL2 as an independent risk factor [hazard ratio (HR)=1.28, 95% confidence interval (CI)=0.79-2.06; p=0.31] (Table II).
ANGPTL2 expression is an independent predictor for early GC recurrence. Patients were divided into two groups: patients with ANGPTL2 IHC score ≥5 (n=54) (high-expression group) and those with ANGPTL2 IHC score ≤4 (n=239) (low-expression group). Kaplan–Meier analysis showed that the patients in the high-expression group had significantly poorer DFS than the patients in the low-expression group (p<0.01) (Figure 5). Univariate analysis for DFS showed that male sex (p=0.02), older age (p<0.01), large tumor size (p<0.01), advanced T stage (p<0.01), lymphovascular invasion (p<0.01), lymph node metastasis (p<0.01), advanced UICC stage (p<0.01), and high ANGPTL2 expression (p<0.01) were risk factors for poor DFS (p<0.01) (Table III). Multivariate analysis revealed that ANGPTL2 high expression (HR=2.21, 95% CI=1.26-3.90; p<0.01), as well as advanced T-stage (HR=3.91, 95% CI=1.30-13.59; p=0.015), lymph node metastasis (HR=4.58, 95% CI=1.67-14.62; p<0.01), and advanced UICC stage (HR=2.44, 95% CI=1.23-5.24; p<0.01) were independent risk factors for poor DFS (Table III).
Discussion
This study, for the first time, demonstrates that high cytoplasmic ANGPTL2 expression was associated with progression of the primary tumor site, lymph node metastasis, and liver and distant metastasis in patients with GC. In addition, we reveal that ANGPTL2 was an independent risk factor for poor DFS using a multivariate Cox hazard model, suggesting that ANGPTL2 might be a useful predictive marker for detecting early recurrence after curative surgery in patients with GC.
ANGPTL2 plays a significant role in increasing tumor cell motility and invasion capacity. It promotes tumor angiogenesis, lymphangiogenesis, and metastasis in vitro. Aoi et al. reported that skin tissues expressing a high level of ANGPTL2 exhibited a high frequency of carcinogenesis in a mouse model of chemically induced skin SCC. They also reported that ANGPTL2 derived from SCC increased tumor cell lymph node metastasis and distant metastasis through tumor angiogenesis and lymphangiogenesis, acquiring epithelial–mesenchymal transition-related invasive properties (12). In addition, Endo et al. noted that human lung cancer cells expressing ANGPTL2 exhibited high metastatic potential for lymph nodes and distant organs by acquiring highly invasive and motility phenotypes in an autocrine/paracrine manner (13). High expression of ANGPTL2 in osteosarcoma cell lines was also associated with increased tumor cell intravasation, mediated by integrin a5b1, p38, and matrix metalloproteinases, resulting in poor survival of xenografts (19).
Association between Angiopoietin-like Protein 2 expression and clinicopathological variables.
Based on mechanistic analyses of ANGPTL2 acting as an oncogene, accumulating evidence demonstrates that ANGPTL2 expression is associated with tumor progression and poor outcomes in patients with several types of malignancies. DFS status of patients with non-small-cell lung carcinoma (NSCLC) with high ANGPTL2 expression was poorer than in patients with lower expression (13). Similarly, high ANGPTL2 mRNA expression was correlated with lymph node metastasis in 110 patients with surgically resected NSCLC (15).
We also demonstrated the clinical importance of ANGPTL2 expression in the field of gastrointestinal malignancies including oesophageal cancer (OEC) and colorectal cancer (CRC). Ide et al. revealed that high expression of ANGPTL2 in OEC tissues was significantly associated with tumorigenesis, poor prognosis, and early recurrence (20). Toiyama et al. demonstrated that serum ANGPTL2 was a novel biomarker for early diagnosis and recurrence of CRC (21). Although serum ANGPTL2 has already been reported to be a potential diagnostic biomarker distinguishing patients with GC from healthy controls (22), our data provide novel evidence that ANGPTL2 expression in GC tissues is an independent marker for predicting early postoperative recurrence in patients undergoing curative surgery for GC. Hence, these patients require intensive monitoring and possibly adjuvant chemotherapy.
Angiopoietin-like Protein-2 expression in gastric cancer tissue and adjacent normal mucosa. a: Immunoreactive ANGPTL2 protein is expressed mainly in the cytoplasm of gastric cancer cells. b: ANGPTL2 protein expression is weak in or absent from adjacent normal gastric mucosa.
Angiopoietin-like Protein-2 immunohistochemistry staining intensity. ANGPTL2 staining intensity was scored as: 0 (negative) (a);1 (weak) (b); 2 (medium) (c); 3 (strong) (d).
Angiopoietin-like Protein 2 staining extent. ANGPTL2 staining extent was scored as 0 (0%); 1 (1-25%) (a); 2 (26-50%) (b); 3 (51-75% (c); 4 (76-100%) (d).
Conversely, ANGPTL2 was frequently silenced in ovarian cancer cell lines through DNA methylation in sequences around the CpG-island exhibiting promoter activity. Lower ANGPTL2 protein levels were frequently observed in primary ovarian carcinomas (i.e. acting as a putative tumor suppressor). Thus, the clinical significance of ANGPTL2 expression might differ among disease stages (24). Although that report included results contradictory to those in previous reports and of our study, it has been suggested that ANGPTL2 expression may be regulated not only by methylation status but also by upstream signaling pathways, including the transforming growth factor β-signalling pathway, exerting both tumor-suppressive activity and invasive/metastatic activity through epithelial–mesenchymal transition of epithelial cells during cancer progression and in a tumor cell type-dependent manner (25).
In conclusion, we provide novel evidence that ANGPTL2 expression in GCs promotes tumor progression and metastasis and portends a poor prognosis. ANGPTL2 may be a significant predictive marker for detecting recurrence after surgery in patients with GC.
Kaplan–Meier analysis of overall survival by Angiopoietin-like Protein-2 expression in gastric cancer tissue (p<0.01, log-rank test).
Kaplan–Meier analysis of disease-free survival Angiopoietin-like Protein-2 expression in gastric cancer tissue (p<0.01, log-rank test).
Univariate and multivariate analyses for overall survival.
Univariate and multivariate analyses for disease-free survival.
Footnotes
Conflicts of Interest
All Authors declare no conflicts of interest with regard to this study.
- Received May 15, 2015.
- Revision received June 19, 2015.
- Accepted June 22, 2015.
- Copyright© 2015 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.