Article Text

Download PDFPDF

Clinicopathological significance of angiopoietin-like protein 4 expression in oesophageal squamous cell carcinoma
  1. Kenichiro Shibata1,
  2. Toshiyuki Nakayama2,
  3. Hiroshi Hirakawa3,
  4. Shigekazu Hidaka1,
  5. Takeshi Nagayasu1
  1. 1Department of Surgical Oncology, Nagasaki University Graduate School of Biomedical Sciences, Nagasaki, Japan
  2. 2Department of Pathology, Nagasaki University Graduate School of Biomedical Sciences, Nagasaki, Japan
  3. 3Department of Tumor and Diagnostic Pathology, Nagasaki University Graduate School of Biomedical Sciences, Nagasaki, Japan
  1. Correspondence to Toshiyuki Nakayama, Department of Pathology, Nagasaki University Graduate School of Biomedical Sciences, 1-12-4 Sakamoto, Nagasaki 852-8523, Japan; toshi-n{at}nagasaki-u.ac.jp

Abstract

Background Angiopoietin-like protein 4 (ANGPTL4) is involved in regulating glucose homeostasis, insulin sensitivity, angiogenesis and lipid metabolism, and also acts as an apoptosis survival factor for vascular endothelial cells. The protein is also known to be induced in hypoxic environments characteristic of cancer tissue. Recently, ANGPTL4 was shown in cancer cells to facilitate the transendothelial passage of the cells, resulting in distant metastasis. Clinically, venous invasion resulting in distant metastasis is crucial for oesophageal cancer progression.

Aims To determine ANGPTL4 expression and its association with clinicopathological factors and prognosis in human oesophageal squamous cell carcinoma (OSCC).

Methods 104 cases of surgically-resected OSCC specimens were examined by immunohistochemistry. The association of ANGPTL4 expression with clinicopathological characteristics and postoperative survival rate was statistically analysed.

Results Expression of ANGPTL4 was statistically correlated with the degree of differentiation, lymphatic invasion and venous invasion. Results of multivariate analysis, performed using multiple logistic regression, showed that lymph node metastasis, lymphatic invasion and ANGPTL4 expression were independent factors predicting venous invasion. Survival rates of patients with ANGPTL4-positive tumours tended to be statistically lower than those with ANGPTL4-negative tumours.

Conclusions ANGPTL4 may play an important role in metastasis through lymphovascular invasion, and may potentially affect prognosis.

  • Angiopoietin-like protein 4
  • human oesophageal squamous cell carcinoma
  • venous invasion
  • immunohistochemistry
  • cancer
  • histopathology
  • oesophagus

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

Oesophageal cancer is the sixth most common malignancy, and one of the leading causes of cancer-related deaths worldwide.1 The disease is characterised by late clinical presentation, rapid progression and poor survival. In spite of the application of modern surgical techniques combined with various adjuvant treatment modalities, such as radiotherapy and chemotherapy, the prognosis of oesophageal cancer patients remains poor. In fact, no ideal prognosis and long-term survival can be obtained for patients with progressing oesophageal cancer, and no significant progress has occurred in treatment. Poor prognosis of oesophageal cancer is due to not only resistance to chemotherapy and radiotherapy, but also the unpredictable dissemination of the cancer cells via lymphatic and venous vessels, resulting in lymph node and distant metastases.2–4 Thus, identification of the molecular factors related to cancer cell invasion of vessels is necessary and may contribute to improved therapy for oesophageal cancer patients.

Angiopoietin-like protein 4 (ANGPTL4) is also known as hepatic fibrinogen/angiopoietin-related protein (HFARP), peroxisome proliferator-activated receptor-γ (PPARγ) angiopoietin-related gene (PGAR) and fasting-induced adipose factor (FIAF).5–7 ANGPTL4 is involved in regulating glucose homeostasis, lipid metabolism and insulin sensitivity, and also acts as an apoptosis survival factor for vascular endothelial cells.6 8–11 ANGPTL4 expression is induced in hypoxic environments characteristic of cancer tissue, and this protein has been detected in the perinecrotic areas of cancers including lung and renal cancers.12 In addition, ANGPTL4 expression has been identified in various tumours (breast, liver, prostate, stomach and melanoma of skin). However, the role of ANGPTL4 in cancer progression has been controversial.12–17 Recently, Padua et al showed that ANGPTL4 is mechanistically involved in distant metastasis in a breast cancer model.18 They showed that tumour cell-derived ANGPTL4 could disrupt vascular endothelial cell–cell junctions, and facilitate the transendothelial passage of tumour cells, resulting in their subsequent retention at secondary sites. Vessel invasion and distant metastases are manifested at high rates in oesophageal cancer. These findings prompted us to investigate the expression and role ANGPTL4 in oesophageal cancer.

The goal of this study was to determine the clinicopathological significance of the expression of ANGPTL4 in oesophageal cancer. To this end, we evaluated the relationships between the expression of ANGPTL4 and clinicopathological characteristics of oesophageal cancer. We also performed survival analysis. To the best of our knowledge, the present study is the first to evaluate ANGPTL4 in oesophageal cancer.

Methods

Patients and pathological samples

Patients who underwent radical oesophagectomy at the Department of Surgical Oncology, Nagasaki University Hospital from 1995 to 2003 were selected for this study. We chose the patients whose clinicopathological information could be extracted and whose prognoses have been followed at our institution. Since the predominant histological type of oesophageal cancer is squamous cell carcinoma in Japan, we limited the cases for this study to this type. Further, we excluded cases with distant metastasis at preoperative examination and those who received anticancer treatment before surgery. Finally, 104 patients (95 men and 9 women, mean age 63.6±10.3 years) were enrolled for this study. All investigations were approved by our institutional review board, and informed consent was obtained from all participants prior to the study.

Table 1 summarises clinicopathological data according to the 6th edition International Union Against Cancer TNM classification.19 Histological classification of oesophageal squamous cell carcinoma (OSCC) was classified into three types—well, moderately and poorly differentiated—based on the predominant features according to the World Health Organization classification.20 The patients remained for a median follow-up period of 832 days, ranging from 19 to 3770 days. For all patients, periodic inspection with chest x-ray, CT scan and tumour marker assays was performed every several months, even if patients experienced no complaints or no symptoms. MRI, ultrasound and endoscopy were used to diagnose the site of recurrence. Twenty cases developed distant metastases: to the liver (n=10), lung (n=9), bone (n=3) and brain (n=1) (three cases had metastases to multiple organs).

Table 1

Patient and tumour characteristics for immunohistochemistry

Immunohistochemistry

We prepared 4 μm thick sections from formalin-fixed, paraffin-embedded tissue blocks, which were subjected to immunohistochemical analyses using rabbit polyclonal anti-ANGPTL4 (1:150; R&D Systems, Minneapolis, Minnesota, USA) as primary antibody and peroxidase-conjugated secondary antibodies (Simple Stain MAX-PO kit, Nichirei, Tokyo, Japan). To facilitate antigen retrieval, the sections were incubated in 10 mM of citric acid (pH 6.0) at 121°C for 15 min. A diaminobenzidine (DAB) kit (Histofine, Nichirei) (brown) was used for the detection of immunoreactions, and counterstained with haematoxylin. The control sections were incubated with isotype-matched IgG. Analysis of the immunohistochemical staining was performed independently. When immunoreactivity was noted in the cytoplasm of tumour cells, they were designated as positive. Since the invasive component of the primary tumour was more intensely stained than the superficial part in almost all cases of invasive carcinoma, the tumour specimens were deemed immunopositive when positively-stained cells were observed in more than 30% of the invasive edge of the primary tumours.

To identify venous and lymphatic invasions, we performed elastic van Gieson (EVG) staining and immunohistochemical staining for CD34 (1:50, Novo Castra Laboratories, Newcastle upon Tyne, UK) and podoplanin (clone: D2-40, 1:100, Nichirei) in addition to H&E staining. Lymphovascular invasion was considered to be present when at least one lymphatic vessel or vein was invaded by cancer cells.

Statistical analysis

The Stat View II program (Abacus Concepts, Berkeley, California, USA) was used for statistical analyses. Associations between ANGPTL4 and categorical variables were analysed by the χ2 test for independence, Fisher's exact probability test or the Mann–Whitney U test. To determine the factors leading to lymphovascular invasion, we used multiple logistic regression to estimate the adjusted ORs and 95% CIs. We evaluated ANGPTL4 immunoreactivity for prognostic value of affecting survival using univariate Kaplan–Meier analysis (p value for log rank test) for overall survival (OS) and disease-free survival (DFS). Subjects who neither died nor had recurrence were censored at time of their last follow-up. A p-value of <0.05 was considered statistically significant.

Results

Expression of ANGPTL4 in oesophageal squamous cell carcinoma and normal mucosa

Figure 1 shows representative results of the immunohistochemistry for ANGPTL4 in normal and OSCC tissues. Positive staining was detected in the cytoplasm of cancer cells. In normal oesophageal mucosa, the immunoreactivity of ANGPTL4 was not observed in the stratified squamous epithelium. Sixty-five of the 104 samples (62.5%) were categorised as immunopositive cases (table 2).

Figure 1

Immunohistochemical staining of normal squamous epithelium and squamous cell carcinoma of oesophagus. Representative images of normal mucosa (A), positive staining for angiopoietin-like protein 4 (ANGPTL4) in cytoplasm of human oesophageal cancer cells (B), and negative staining for ANGPTL4 (C). Bar, 100 μm.

Table 2

Relationship between angiopoietin-like protein 4 (ANGPTL4) and clinicopathological factors

Correlation between clinicopathological characteristics and ANGPTL4 expression

Table 2 shows the correlations between ANGPTL4 and clinicopathological characteristics. Statistically, the expression of ANGPTL4 was correlated with lymphatic invasion (p=0.026) and venous invasion (p<0.005). Lymphovascular invasion was detected more frequently in ANGPTL4-positive tumours than in ANGPTL4-negative tumours as shown by univariate analysis. Results also showed that ANGPTL4 was correlated with the degree of differentiation, such that the expression of ANGPTL4 in a poorly differentiated tumour was statistically higher than that in a well differentiated tumour (p=0.031). However, no statistical correlation was found between ANGPTL4 expression and patient age, gender, depth of invasion, lymph node metastasis and distant metastasis (metastasis to non-regional lymph nodes) in pathological assessment (staging) on primary surgery.

Correlation between ANGPTL4 expression and venous invasion

By univariate analysis, in addition to the expression of ANGPTL4 in cancer cells, the following parameters were shown to be associated with venous invasion: depth of invasion, lymphatic invasion, lymph node metastasis, and distant metastasis in pathological assessment (staging) on primary surgery (table 3).

Table 3

Correlation between clinicopathological characteristics and venous invasion

Table 4 shows the results of multiple logistic regression. Lymph node metastasis, lymphatic invasion, and ANGPTL4 expression were shown to be individually independent factors predicting venous invasion as shown by multivariate analysis. Adjusted OR was 2.538 (95% CI 1.005 to 6.410) in ANGPTL4-positive cases.

Table 4

Multivariate ORs and 95% CIs for the association between angiopoietin-like protein 4 (ANGPTL4) expression and independent variables

Expression of ANGPTL4 in OSCC cells caused lymphovascular obstruction (tumour emboli)

ANGPTL4-positivity was often identified in cancer cells, which invaded into blood vessels and lymph ducts. ANGPTL4-positive cancer cells often pervaded the vessels causing tumour emboli. Figure 2 shows a representative image of ANGPTL4-positive cancer cells causing tumour embolus.

Figure 2

Venous invasion in oesophageal squamous cell carcinoma. Cancer cells which caused tumour embolus were immunopositive for angiopoietin-like protein 4 (ANGPTL4) by immunostaining using elastic van Gieson (EVG) (A) and ANGPTL4 (B).

Survival analyses

Figure 3 shows the graphical results for OS and DFS. The expression of ANGPTL4 was correlated with both OS (p=0.0003) and DFS (p=0.0022). Kaplan–Meier analysis showed that both OS and DFS were statistically shorter in patients with ANGPTL4-positive immunoreactivity.

Figure 3

Survival based on the expression of angiopoietin-like protein 4 (ANGPTL4) in human oesophageal cancer (104 cases). Kaplan–Meier survival analysis for (A) overall survival, (B) disease-free survival.

Discussion

Members of the angiopoietin-like protein family (ANGPTLs) have structure similar to the angiopoietin family, consisting of an N-terminal signal sequence, a unique region of variable length, a coiled-coil domain, and a large fibrinogen/angiopoietin-like domain.10 21 22 Angiopoietins function through the Tie2 receptor, whose signalling is critical to regulate vascular stabilisation and remodelling. It has been reported that the angiopoietin/Tie2 signal is involved in survival and migration of endothelial cells and regulates vascular remodelling and maintenance of vascular integrity. Interestingly, ANGPTLs also function in angiogenesis through regulating survival and migration of endothelial cells, although ANGPTLs do not bind the angiopoietin receptor Tie2.23 To date, the receptor of ANGPTL4 has not yet been identified and the role of ANGPTL4 is not fully understood, especially in cancer tissue. Padua et al showed that ANGPTL4 potentially affects disruption of vascular endothelial cell–cell junctions, which may eventually lead to distant metastasis.18 Cancer cell dissemination via infiltration into lymphovascular vessels, resulting in lymph node and distant recurrence is the predominant contributing factor for poor prognosis of oesophageal cancer patients. Thus, we hypothesised that ANGPTL4 has a role in oesophageal cancer progression.

Here, we first characterised the role of ANGPTL4 in OSCC by analysing and correlating its expression with clinicopathological data. The results of this study showed that the expression of ANGPTL4 was statistically correlated with the degree of differentiation (p=0.031), lymphatic invasion (p=0.026) and venous invasion (p<0.005). Furthermore, multivariate analysis revealed that lymph node metastasis, lymphatic invasion and ANGPTL4 expression were independent factors for prediction of venous invasion. Further, ANGPTL4-positivity was often identified in cancer cells which invaded into blood vessels. In addition, from all the investigated medical records, 18 of 20 cases who relapsed in distant organs showed ANGPTL4 immunopositivity (data not shown). These data showed that in OSCC, ANGPTL4 was associated with distant recurrence. Recurrence in distant organ and lymph node is due to cancer cell dissemination into vessels. It has been reported that ANGPTL4 affects the permeability of blood vessels. In OSCC patients, the permeability-altering function of ANGPTL4 may contribute to recurrence. In the present study, we showed the clinicopathological significance of ANGPTL4 in OSCC. However, some reports have indicated that ANGPTL4 inhibits cell motility and invasiveness.13 17 Thus the varying roles of ANGPTL4 may be due to differential cellular and pathological mechanisms of metastasis between tumours.24 25

Although univariate analysis showed that ANGPTL4 expression is correlated with lymphatic invasion, we could not show that ANGPTL4 expression is an independent factor for prediction of lymphatic invasion.

To our knowledge, the present study is the first to evaluate ANGPTL4 expression and role in OSCC. Our data revealed the clinicopathological significance of ANGPTL4 expression in OSCC specimens. We showed that ANGPTL4 may play an important role in the progression of OSCC, possibly by facilitating transendothelial passage of cancer cells, resulting in distant metastasis and recurrence. Further, the Kaplan–Meier survival analysis showed that ANGPTL4 levels may potentially affect prognosis of OSCC patients. However, further studies are needed to elucidate the biochemical mechanisms through which ANGPTL4 affects venous invasion in OSCC.

Take-home messages

  • This is the first report to identify the expression of angiopoietin-like protein 4 (ANGPTL4) in oesophageal squamous cell carcinoma (OSCC).

  • The expression of ANGPTL4 was shown to be an independent factor for predicting venous invasion of OSCC.

  • The expression of ANGPTL4 may have an important role in the progression of OSCC.

References

Footnotes

  • Competing interests None.

  • Ethics approval This study was conducted with the approval of Nagasaki University.

  • Provenance and peer review Not commissioned; externally peer reviewed.