Abstract
Background: Neolymphangiogenesis, a process of lymphatic vessel development in neoplastic tissue, may be a key event in the transmission of cancer cells into lymph nodes. The current study examined the relationship between lymphatic vessel density (LVD) measured by podoplanin (D2-40) expression, clinicopathological parameters and patient survival in gastric cancer. Materials and Methods: D2-40 expression was examined by immunohistochemistry in formalin-fixed paraffin-embedded tissue specimens obtained from 60 patients with gastric cancer. D2-40 immunoreactivity was analyzed in intratumoral and peritumoral compartments of tumors and correlated with tumor grade, type in Lauren's classification, lymph node status, distant metastasis, presence of ulceration, inflammatory infiltration, angio-invasion, lymphangio-invasion and patient survival using a Receiver Operating Characteristic (ROC) curve analysis to find cut-off points that enabled fair decision making in survival analysis. Results: The mean values of intratumoral and peritumoral LVD were 6.63 and 11.25, respectively. Enhanced intratumoral LVD measured by D2-40 immunoexpression was correlated with the presence of lymph node metastases (p=0.04). Our study revealed a statistically significant correlation between intratumoral LVD measured by D2-40 expression and survival of patients with gastric cancer: an intratumoral LVD higher than 4.68 is significantly correlated with unfavorable prognosis, with a probability of death of approximately 80%. No significant relationship was identified between peritumoral LVD, lymph node status and survival in patients with gastric cancer. Conclusion: A high intratumoral LVD measured by D2-40 expression in specimens from primary tumors is strongly associated with lymph node metastasis and predicts worse clinical outcome. Increased intratumoral D2-40 immunoreactivity is a putative predictor of aggressive gastric cancer behavior.
Neolymphangiogenesis as a process of lymphatic vessel development in neoplastic tissue may be a key event in the transmission of cancer cells into lymph nodes. The presence of lymph node metastases is one of the most important prognostic factors in gastric cancer (1, 2). Lymphangiogenesis also plays an important physiological role in metabolism, homeostasis and immunity. Additionally, lymphatic vessel formation has been implicated in a number of non-malignant conditions such as edema, rheumatoid arthritis and psoriasis (3).
The presence of lymphatic vessel invasion (LVI) in gastric cancer cells correlates with lymph node metastasis (2, 3). LVI is commonly described as the presence of tumor emboli within vessel-like structures that are lined with a single layer of endothelial cells. Its similarity to blood vessels and lymphatic vessels makes LVI almost impossible to distinguish by Hematoxylin and Eosin (HE) staining, which means that LVI evaluated by HE staining might include a blood vessel invaded by neoplastic cells (4). For this reason, a much more sophisticated method of LVI detection should be applied for the study of lymph node metastases that is specific only for lymphatic vessels. Podoplanin (D2-40) is considered to be an appropriate marker.
Clinicopathological characteristics of patients with gastric cancer.
D2-40, a monoclonal antibody against the carcino-embryonic antigen M2A, may be used successfully for immunohistochemical evaluation of lymphatic invasion by gastric and colonic cancer cells (5). D2-40 plays a crucial role in preventing cellular adhesion and is strongly involved in the maintenance of glomerular permeability. Many studies have revealed that the quantification of lymphatic vessel density (LVD) indicated by D2-40 is useful for predicting tumor growth, recurrence-free survival, and overall survival (1-6). Podoplanin as a marker of unfavorable prognosis is up-regulated in a variety of human tumors including squamous cell carcinoma of the oral cavity, of the cervix, the lung, the esophagus, mesothelioma, skin adnexal carcinomas and tumors of the central nervous system (7-10).
Correlations between clinicopathological features of gastric cancer specimens and lymphatic vessel density measured by D2-40 expression in intratumoral and peritumoral compartments.
The aim of the present study was to examine the relationship between LVD measured by D2-40 expression, clinicopathological parameters and patient survival using a Receiver Operating Characteristic (ROC) curve analysis to find cut-off points that enabled fair decision making in survival analysis.
Materials and Methods
Patients. Sixty patients with histologically-confirmed adenocarcinoma of the stomach operated on with curative intent (R0) entered the study. The group of patients consisted of 44 males and 16 females with a mean age of 63 years. The study was conducted on histopathological material obtained from patients operated for gastric cancer at the Second Department and Clinic of General and Oncological Surgery, Wroclaw Medical University, Poland. The stage of tumors was assessed according to the fifth edition of the TNM Classification of Malignant Tumors (11) (Table I).
All patients underwent elective total gastrectomy and D2 lymphadenectomy with curative intent (the mean number of dissected lymph nodes was 19). Follow-up visits were scheduled every three months during the first two years and then every six months. Chest X-ray, abdominal sonography, computed tomographic (CT) scan, as well as clinical and endoscopic examinations, were performed. Information about overall survival was obtained from the Lower-Silesian Regional Cancer Registry. The data were collected in a retrospective manner.
Tissue specimens. Tissue samples were fixed in 10% buffered formalin and embedded in paraffin. In each case, HE-stained slides were evaluated histopathologically by two pathologists. The analyzed clinicopathological data included TNM stage, histopathological type, Lauren's type, presence of angio- and lymphangio-invasion, ulceration, inflammation and long-term survival rate.
Intratumoral lymphatic vessel density (a, b) and peritumoral lymphatic vessel density (c, d) as reflected by D2-40 expression in gastric cancer specimens (×200, hematoxylin).
Finding an optimal cut-off point for intratumoral lymphatic vessel density (a) and peritumoral lymphatic vessel density (c) by D2-40 expression using Receiver Operating Characteristic (ROC) curve analysis. Kaplan–Meier estimate comparison for two groups of patients based on the results of ROC curve analysis for intratumoral (b) and peritumoral (d) lymphatic vessel density.
Immunohistochemistry. Formalin-fixed, paraffin-embedded tissue was freshly cut (4 μm). The sections were mounted on superfrost slides (Menzel Glaser, Germany), de-waxed with xylene, and gradually hydrated. The activity of endogenous peroxidase was blocked by 5 min exposure to 3% H2O2. All studied sections were boiled for 15 min at 250 W in Antigen Retrieval Solution (DakoCytomation, Glostrup, Denmark). Immunohistochemical reactions were performed using mouse monoclonal antibody detecting D2-40 (dilution 1:100; DakoCytomation) and rabbit anti-human antibody detecting Human Epidermal Growth Factor Receptor 2 (HER2) (optimally prediluted) (DakoCytomation). The tested sections were incubated with antibodies for 1 h at room temperature. The subsequent incubations involved biotinylated antibodies (15 min, room temperature) and streptavidin-biotinylated peroxidase complex (15 min, room temperature) (LSAB+, HRP; DakoCytomation). NovaRed (Vector Laboratories, Peterborough, UK) was used as a chromogen (10 min, at room temperature). All sections were counterstained with Meyer's hematoxylin. In every case, control reactions were included, in which specific antibody was substituted by the Primary Mouse Negative Control (DakoCytomation).
Evaluation of reaction intensity. Protein expression was assessed under a microscope (Olumpus BX41; Olympus, Warsaw, Poland) using computer microscopic image analysis (software: AnalySIS DOCU, ver. 3.2 for Windows 95/98/NT; Soft Imaging, license 100 7557; Olympus). All preparations were examined by two pathologists who independently performed immunopathological evaluation based on the analysis of large tissue slides (approximately 2 cm in diameter). In doubtful cases, re-evaluation was performed using a double-headed microscope and staining was discussed until consensus was achieved.
D2-40 protein expression was evaluated using a quantitative method, namely a modified Weidner's method (12, 13). Vessel density was evaluated within neoplastic tissues (intratumoral LVD) and within healthy tissues outside the tumor (peritumoral LVD). A microscopic image from one preparation was magnified 200 times and transferred to a computer program which automatically counted all positively stained vessels from three random fields of view with the highest density. Each individual vessel or cluster of endothelial cells was regarded as a hot spot and was counted as one microvessel. The average number of vessels from the three fields of view was then calculated. The procedure of evaluation of HER2 expression in gastric cancer cells was described in detail previously (14).
Statistical analysis. The main aim of the analysis was to find correlations between different parameters and to determine which factors have a significant impact on survival rate. In order to reveal correlations between variables, the χ2 test of independence, as well as the Mann–Whitney U-test, were performed. We also analyzed whether any particular parameter may be used to determine patients' prognoses. Therefore, for each parameter, a ROC curve was calculated and best cut-off points were analyzed. In the case of intratumoral LVD, ROC curve analysis revealed an interesting decision rule. Furthermore, patients were divided into two groups with respect to best cut-off point for intratumoral LVD and Kaplan–Meier estimator was obtained in order to detect significant differences in survival rate between the two groups.
Results
D2-40 immunostaining in gastric cancer specimens. The mean intratumoral LVD was 6.63 (minimum, 0; maximum, 21.67) (Figure 1a). D2-40-stained lymphatic vessels were also found in the peritumoral zone; their density was significantly higher than within the tumor, with an average of 11.25 vessels within one field of view. The smallest number of vessels was 1.5 and the largest 24.17 (Figure 1b).
Correlations between intratumoral and peritumoral LVD and clinicopathological features of patients with gastric cancer. Increased number of lymphatic vessels in the tumoral compartment measured by D2-40 expression was associated with the presence of lymph node metastases (p=0.04). Interestingly, statistical analysis revealed a significant correlation between enhanced intratumoral LVD and overexpression of HER2 in immunohistochemistry (p=0.028) (Table II). We did not find any significant correlations between intratumoral LVD and other analyzed clinicopathological parameters such as tumor size, distant metastasis, type in Lauren's classification, inflammatory infiltration, histopathological grade, presence of ulceration and lymphangioinvasion (Table II). No significant relationship was found between increased peritumoral LVD and the clinicopathological parameters analyzed (Table II).
Intratumoral LVD and patient survival. While analyzing ROC curves for different parameters, we discovered that there was a good cut-off point in the case of intratumoral LVD. The cut-off point was set to 4.68 (Figure 2a). It was revealed that intratumoral LVD might be used as an indicator of poor prognosis as the probability that the patient would die within the first 60 months after surgery if their intratumoral LVD was greater than 4.68 is approximately 0.8
The patients were further divided into two groups: the first group included patients with an intratumoral LVD of 4.68 or lower, whereas the second group comprised all other patients. The Kaplan–Meier estimator was calculated to determine if there was a significant difference in survival rate between the two groups (Figure 2b) but unfortunately no significant difference was found (p=0.22). The main reason for such a finding is the fact that intratumoral LVD could only be used as an indicator of poor prognosis. It is worth pointing out that the probability that the patient would live more than 60 months after surgery if their intratumoral LVD was 4.68 or less is only 0.43. This means that if the patient's intratumoral LVD was 4.68 or less, it is very hard to predict their clinical outcome. It can only be stated that an intratumoral LVD greater them 4.68 is associated with an unfavorable prognosis.
Similar analysis was performed for peritumoral LVD. The ROC curve was not significantly different from a diagonal, which means no good cut-off point could be derived and no satisfactory decision rule could be obtained (Figure 2c). Therefore, we did not observe a significant difference in the Kaplan–Meier analysis (p=0.85) between the two groups of patients (cut-off point was set to 4.88) (Figure 2d).
Discussion
In the present study we examined the relationship between the density of lymphatic vessels measured by D2-40 expression, clinicopathological parameters and patient survival using a ROC curve analysis to find cut-off points that enabled fair decision making in survival analysis. Our study revealed a statistically significant correlation between intratumoral LVD measured by D2-40 expression and survival in patients with gastric cancer: intratumoral LVD higher than 4.68 is significantly correlated with unfavorable prognosis, with a probability of death of approximately 80%.
A role of D2-40 in tumor invasion and metastasis has been widely discussed (15-18). Interestingly, it was reported that D2-40-positive neoplastic cells (not only lymphatic vessels stained by D2-40) were found in several types of human malignant tumors, including oral squamous cell carcinoma (15, 16), squamous lung cancer (17), and mesothelioma (18). The molecular mechanism involved may be associated with promoting tumor invasion through cytoskeletal actin remodeling in tumor cells by D2-40, contributing to their elevated motility (19, 20). Moreover, D2-40 also binds ezrin/radixin/moesin proteins through its cytoplasmic domain and this interaction is crucial for D2-40-mediated activation of the small GTPase RhoA and its associated kinase ROCK, thereby promoting an epithelial–mesenchymal transition (19, 21). In our study, we did not observe D2-40 immunoreactivity in gastric cancer cells themselves. The most important aspect of D2-40 research includes its role as a specific marker of lymphatic vessels. The literature provides no clear answer as to whether there is a correlation between intratumoral LVD and patient survival. Nakamura et al. found a correlation between the density of lymphatic vessels stained with D2-40 and patient survival and prognosis, but provided no distinction between intratumoral and peritumoral compartments (22). Coskun et al. described only peritumoral LVD as a factor strongly associated with progression and poor prognosis in patients with gastric cancer (23). They observed intratumorally-stained vessels (in 38/65 cases, 58%), but did not correlate this type of D2-40 reactivity with clinicopathological parameters (23). Similar results were obtained by Yu et al., again with no distinction between intratumoral and peritumoral compartments (24).
We revealed that enhanced intratumoral reactivity of D2-40 was strongly correlated with HER2 overexpression, as measured by immunohistochemistry. Many studies have investigated the role of HER2 immunoexpression in patients with gastric cancer but the clinical significance of its expression in the context of lymphangiogenesis remains unclear. To the best of our knowledge, we are the first to report the relationship between increased intratumoral LVD and overexpression of HER2 in patients with gastric cancer.
Our conclusion that increased intratumoral LVD is a putative predictor of aggressive gastric cancer behavior is in agreement with the results obtained by Lee et al. (25). They performed separate analyses for intratumoral and peritumoral LVD in early gastric cancer cases and observed that intratumoral LVD might be a useful, independent predictor for lymph node metastasis in patients with early gastric cancer (25). On the other hand, Wang et al. revealed that intratumoral LVD had no prognostic significance in patients with gastric cancer – only enhanced peritumoral LVD was strongly associated with lymph node metastasis and shorter overall survival (26).
Due to discrepancies related to the clinical significance of intratumoral LVD, it is necessary to conduct further studies into lymphangiogenesis and its influence on the development and progression of cancer on a larger group of patients with gastric cancer.
Footnotes
-
Conflicts of Interest
All Authors declare that they have no conflicts of interest.
- Received February 2, 2014.
- Revision received April 4, 2014.
- Accepted April 7, 2014.
- Copyright© 2014 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.