The relationship of mast cell density in pulmonary, gastric and ovarian malignant epithelial tumors with tumor necrosis and vascularization

Introduction

Malignant epithelial tumors are an important group of tumors that can develop in many organs including the stomach, ovary, and primarily the lungs, and can cause death. Tumor tissue interacts with surrounding connective tissue elements and inflammatory cells that the immune system produces against the tumor. Mast cells (MC) are found to be one of these inflammatory cellular elements [1]. Mast cells are immune cells found in the bone marrow derived from hematopoietic stem cells (CD34+), all vascularized tissues and in the serous cavities where they reach maturity [2]. Mast cells are of bone marrow origin and have small nuclei and granular cytoplasm [3]. Mast cells are potentially a source of a range of cytokines/chemokines, and act as growth factors involved in both natural and adaptive immune responses mediating allergic disease [4]. Mast cells are effector cells involved in IgE mediated responses, but they can also mediate inflammation after activation by various molecules through Toll-like receptors, interleukin 1 (IL-1) receptors and neurotransmitter receptors without the intervention of IgE. Thus, binding of the antigen to these receptors can also activate mast cells without degranulation. There are two types of MCs in humans: mast cell tryptase and mast cell tryptase-chymase [5]. Interleukin 1, IL-18, IL-33 and IL-36 playing an important role in the immune response and inflammation are members of the IL-1 family cytokines. Cytokines such as TNF and IL-1 are produced by MCs and are also capable of activating mast cells [6]. Their proportions vary in different organs and in different malignant variations [7, 8, 9, 10]. The interaction of MCs with cancer tissue has been investigated in many studies. The interaction between MC presence and tumor tissue has been demonstrated in oral malignant and premalignant tumors [11, 12, 13, 14, 15], gastric carcinoma [16], breast carcinoma [17], endometrium carcinoma [18], cervical carcinoma [19] and basal cell carcinoma [20] tumors. Mast cell-associated VEGF [21] allows neovascularization [22] and cancer development [23]. In a study by Hiromatsu and Toda, which supports this condition, it has been reported that MCs are closely associated with angiogenesis or neovascularization, and histopathological examinations suggested an increase in the number of mast cells in areas of angiogenesis [24].
It has been suggested that degranulation of MCs induces tumor growth and spread via certain secreted molecules (angiogenic VEGF, collagenolytic enzymes, etc.), and it has also been shown that MCs inhibit tumor growth and cause tumor cell apoptosis by other secreted molecules such as TNF-α and IL-4 [25]. It is also mentioned that MCs exert an anti-tumor effect by causing cytotoxicity [27]. Although these features of MCs are investigated, it is not yet clear whether the increase in MC concentration causes tumor growth or inhibition, or whether this is caused by a simple defensive reaction [25]. Therefore, the role of MCs in terms of tumor tissue interaction and the changes they cause should be well known.
For these reasons, we aimed to investigate the relationship of MCs with angiogenesis between necrotizing and non-necrotizing groups in stomach, lung, and ovarian malignant epithelial tumors, which are commonly seen in the population. For this, we immunohistochemically compared anti-mast cell tryptase antibody and anti-CD34 antibody expression between necrotizing and non-necrotizing groups. A larger number of MCs and vascular structures in the non-necrotizing group compared to the necrotizing group will support the hypothesis that MCs affect neovascularization positively. This will in turn support the possibility that MCs may induce ischemic necrosis in tumor tissue by inhibition of the mechanism of action on neoangiogenesis. Thus, a new therapeutic approach can be obtained that restricts or prevents tumor tissue development. In the literature, there is no study demonstrating the relationship of MC and neovascularization between necrotizing and non-necrotizing groups.

Material and methods

For the present study, approval of Dicle University Medical Faculty Ethics Committee was obtained with protocol number 81-23.01.2015. Hospital records of the stomach, lung and ovarian resection materials and relevant preparations of the patients who were admitted to Diyarbakır Dicle University Faculty of Medicine, Pathology Department between 2006 and 2015 were examined by looking at the records. Cases diagnosed with malignant epithelial tumor were included in the study. All hematoxylin and eosin (HE) stained slides of the studied cases were re-examined by an expert pathologist for confirmation of diagnosis. Histological classification was based on the latest World Health Organization (WHO) classification. As a result of these investigations, a total of 165 malignant epithelial tumors (50 stomach, 56 lung, 59 ovarian) to be included in this study were detected. The cases were divided into two groups: necrotizing (necrosis+) and non-necrotizing (necrosis–).

Immunohistochemical staining

Tissues suitable for immunohistochemical examination were identified in HE stained slides, and the paraffin blocks in the archives of our department were accessed. Two 5 µm-thick sections were prepared on positively charged slides from each of the identified paraffin blocks. These preparations were dried in an incubator at 56°C for one hour and the sections were allowed to stick to the slide. Immunohistochemical staining was automatically performed on the Ventana Benchmark XT device (Ventana, Tucson, AZ), as appropriate for the device kit manual. Control blocks were first identified for proper immunohistochemical evaluation. For the anti-CD34 antibody, paraffin blocks containing tonsil tissue evaluated in our unit, and for the anti-mast cell tryptase antibody, paraffin blocks containing mastocytosis-diagnosed skin tissue were selected as the control blocks. Using these control blocks, it was determined whether the ready-to-use CD34 antibody worked, and in which dilution the anti-mast cell typing antibody (dilution type antibody) worked best. The preparations of all cases treated with immunohistochemical staining were examined under light microscopy. MCs were counted in intratumoral (IT) and peritumoral (PT) localization in 10 fields (using 40× magnification) showing the best positive staining by anti-mast cell tryptase antibody (clone: 1A1, DAKO, at a dilution rate of 1/300, with an incubation time of 0 h 44 min; Fig. 1-3). Vascular structures with monocellular and lumen appearance were counted in IT localization in 10 fields (using 40× magnification) showing the best positive staining by anti-CD34 antibody (clone; QBEnd/10, VENTANA/ROCHE, ready-to-use form, with an incubation time of 0 h 32 min; Fig. 1-3).
Statistical evaluation of the obtained data was performed through the statistical package program SPSS for Windows 15.0 (SPSS Inc., Chicago). Chi-square analysis was used to compare categorical data. The Kolmogorov-Smirnov test was used to check whether numerical data were normally distributed. Those with normal distribution were analyzed with one-way ANOVA (Tukey test) and Student’s t test, while those with non-normal distribution were analyzed with the Kruskal-Wallis test and Mann-Whitney U test. It was found that the difference was statistically significant with a p value < 0.05.

Results

Male/female (M/F) ratio in stomach and lung cases was 2.1 and 6.0 respectively. The mean age of stomach, lung and ovarian cases was 61.54 ±12.68, 58.02 ±13.26, and 54.37 ±15.85, respectively. Mean tumor size in stomach, lung, and ovarian cases was 5.23 ±2.23, 4.65 ±2.46, and 7.28 ±4.38, respectively.
The number and percentage of cases with and without necrosis in organs are shown in Table I. The number of cases without necrosis in gastric tumors is approximately 3 times the number of cases with necrosis. The numbers of cases with and without necrosis are similar for lung and ovary. While the number of cases without necrosis is higher than the number of cases with necrosis in gastric and ovarian cases, the opposite is the case in lung cases.
Anti-mast cell tryptase antibody and anti-CD34 antibody staining and statistical data of three organs are shown in Table II. In the gastric, lung and ovarian cases, mast cells are present in the intratumoral and peritumoral areas. In the gastric and lung cases, the group without necrosis contained more mast cells and vascular structures in total. In the ovarian cases, more mast cells but fewer vascular structures were counted in the group with necrosis. In addition, structures with a lumen in the intratumoral area were more common in all organs than vascular structures with a single-cell appearance. In the group without necrosis, the vascular structures with a single-cell appearance in the intratumoral area were more common and significantly higher.

Discussion

Our findings support the theory that MCs induce neoangiogenesis since we found that there were more MCs and vascular structures in non-necrotizing cases in lung and stomach tumors. However, this relationship is absent in ovarian tumors.
A clear and precise understanding of the tumor growth mechanism is highly important. It is vital for prevention of tumor development and treatment. So far, many scientific studies have contributed to this process. With the definition of microenvironment in the 1980s, the interaction of cancer tissue and normal surrounding tissue has become a hot topic. MCs are one of the cellular elements that can be found in the tumor microenvironment. However, it is difficult to recognize MCs in the histomorphological evaluation of tumors. To make MCs visible, toluidine blue can be used histochemically, and anti-mast cell tryptase and anti-CD117 antibodies can be used immunohistochemically. In our study, we used anti- mast cell tryptase antibody to stain MCs.
In some studies in the relevant field, researchers used MC/tumor ratio to determine tissue mast cell density (MCD), while others accepted ≥ 20 under 10× magnification of tissue sample [27], or > 3 detected in every 30 immersion levels under high magnification (40× objective) as a high MC count [28]. In our study, we considered the number of mast cells instead of mast cell density. We evaluated total mast cell count as IT and PT by counting 10 fields with 40× magnification. There is no study in the literature analyzing the superiority of existing methods over each other. We believed that specifying the number of cells would give a better result.
Studies conducted by Ying-An Jiang et al. [29] and Açıkalın et al. [30] showed no significant relationship between tumor size and mast cell density. In our study, the relationship between tumor size and MC number was not evaluated.
Neoangiogenesis is essential for tumor growth and spread. When the tumor size exceeds 0.5 mm3, tumor nourishment is dependent on angiogenesis. Tumors smaller than 0.5 mm3 can obtain oxygen and nutrients via diffusion [31]. For this reason, in tumors larger than 0.5 mm3, if neoangiogenesis does not occur, the existing tumoral tissue nutrition will be restricted and necrosis and cell death will occur in tumor tissue as a result of hypoxia. Cell injury and necrotic substances will perform the collection of inflammatory cells into this area after many complex processes. Mast cells, one of the inflammatory cells, are effector cells involved in IgE-mediated responses (with FcεRI on their surface), also after activation by various molecules via Toll-like receptors, interleukin 1 (IL-1) receptors and neurotransmitter receptors without the interference of IgE can also mediate inflammation [5]. Thus, they can be activated by a series of stimuli, both via the FcεRI receptor on the mast cell surface and via receptors other than the FcεRI receptor [2, 3, 4, 5, 6]. Interleukin 1 produced by MCs also has the ability to activate mast cells [6]. Neoangiogenesis is stimulated by some substances released by activated mast cells [26]. With neoangiogenesis, tumor tissue can be fed adequately and avoids necrosis due to hypoxia. With sufficient angiogenesis, the tumor can grow rapidly and have the ability to metastasize to close or long distances. Clinical data suggest that metastatic potential and prognosis are related to the severity of angiogenesis. Therefore, the current level of angiogenesis needs to be determined. Some methods used in this regard include detection of microvessel density, measurement of angiogenic factors in blood and urine, and determination of tissue levels of angiogenic proteins. In our study, we determined the increase in angiogenesis by counting the vascular structures with luminal and monocellular appearance in the IT field. To visualize vascular structures, we performed immunohistochemical anti-CD34 antibody staining.
It has been proposed that tryptase secreted by MCs breaks down the connective tissue to open the necessary area for neoangiogenesis and is a potent angiogenic stimulant [32]. Many studies in the literature support this hypothesis. In one, mast cell concentration in the intratumoral and peritumoral area and microvessel density in oral squamous cell carcinoma were compared using immunohistochemical markers (anti-CD31 antibody, anti-tryptase anti- body). The authors found a statistically significant increase in mast cell concentration and microvessel density relative to normal mucosa in the IT and PT fields [33]. In another study on odontogenic tumors, a significant correlation was found between MCs and microvessels in odontogenic tumors [34]. In a study investigating angiogenesis and mast cell density in relation to the clinical behavior and morphological characteristics of pheochromocytomas, the difference between benign and malignant cases was not significant but vascular structures were correlated with mast cell counts [35]. A direct correlation was found between MCD and angiogenesis during tumor growth in esophagus, lung cancer, and melanoma, and this was thought to be important in prognosis [36]. In our study, non-necrotizing cases contained more MCs and vascular structures in gastric and lung malignant epithelial tumors compared to necrotizing cases. This finding strongly supports the positive effect of MCs on neovascularization.
Some mediators secreted by MCs (chymosin, granzyme, TNF-α, IL-4, etc.) suppress tumor growth. For example, MC expression has been shown to increase in several types of cancer, such as breast, ovary, lung, and colorectal carcinoma, and to accompany good prognosis. Furthermore, it has been argued that MCD can inhibit proliferation and spread of tumors in gastric cancer, breast cancer, and soft tissue sarcomas. Higher MCD has been reported to be a good prognostic marker in invasive breast cancer and prostate cancer [37]. In our study, there were more mast cells in the necrotizing group of ovarian cases. However, more vascular structures were counted in the non-necrotizing group. This finding supports the hypothesis that MCs suppress tumor growth.
There are certain limitations of this study. More MCs and vascular structures were counted in the lung and stomach cases compared to ovarian cases. The limitation regarding the evaluation of tumor stroma due to more abundant necrotic material in tumors and dense glandular structures in tumor tissue observed in microscopic examination of ovarian cases may explain this difference. In addition, although 10 large magnification fields were counted in each sample, fewer PT fields in some cases compared to other tissues may have affected the number of existing mast cells and vascular structures. Mean tumor sizes were not compared between necrotizing and non-necrotizing groups. For this reason, it is not possible to comment on the relationship between vascular structure, MC count and tumor size.
While new methods and treatment methods are being investigated every day in tumor therapy, it can be seen that inhibition of angiogenesis via mechanisms related to tumor angiogenesis is a topic of high interest. A much better understanding of this topic through further and extensive research can open new horizons in tumor treatment.
The relationship between MCs and neoangiogenesis has been investigated in many types of tumors. However, there is no study investigating this relationship in tumors between necrotizing and non-necrotizing groups. In this regard, our study is the first in the literature. In conclusion, our findings have demonstrated that the increase in number of MCs in the IT and PT fields in the non-necrotizing group in stomach and lung cases accompanied an increase in the number of vascular structures in the tumor tissue. Based on this finding, we believe that ischemic necrosis can be induced, limiting or inhibiting neoangiogenesis in tumor tissue on the basis of MH inhibition, especially in lung and gastric malignant epithelial tumors. If this can be achieved, we believe that a novel approach for tumor treatment can be developed.

The authors declare no conflict of interest.