A Role for Mast Cell-Mediated Antibodies in the Formation of Cholesteatoma and Cholesteatoma-Induced Bone Erosion

The study aimed to evaluate the effects and relationships between mast cells in the matrix, mast cell enzymes tryptase and chymase, epithelial proliferation, microvascular density, and bone destruction in cholesteatoma. Thirty-five biopsies diagnosed with cholesteatoma and seven healthy skin tissues taken from the retro-auricular region for control were evaluated. Immunohistochemical studies were performed with CD117, CD34, Ki-67, chymase, and tryptase antibodies, in a single session for all cases and the control group. The relationship between erosion size and antibody load was determined. The mean cholesteatoma epithelium Ki-67 was higher than the control group (p < 0.001). CD117-positive mast cells, chymase-positive mast cells, tryptase-positive mast cells, and microvessel density were significantly higher in the cholesteatoma matrix compared to the control group (p < 0.002, p < 0.001, p < 0.005). In the group with bone erosion scores of two and above, immunohistochemical markers tended to be higher. A positive correlation was found between CD117 and chymase, tryptase, and microvessel density; between tryptase, chymase, and microvessel density; and between chymase and microvessel density. CD117-positive mast cells and chymase-positive mast cells stimulate angiogenesis, increase the epithelium’s proliferative capacity in the cholesteatoma matrix, and form cholesteatoma. The increased proliferation of cholesteatoma epithelium and increased vascular density in the matrix exacerbate bone erosion.


Introduction
Cholesteatoma is the accumulation of keratinised squamous epithelium in the middle ear. It is important to note that cholesteatoma, despite its benign nature, can cause several complications inside the skull, including hearing loss, facial nerve paralysis, and various complications associated with blood vessels [1]. Even though its incidence is not known with certainty, it occurs in approximately 6 out of every 100,000 individuals. Three types have been defined: congenital, acquired, and unclassifiable [2]. Histomorphology consists of three layers: a cystic part, a matrix, and a perimatrix. A cystic part consists of dead keratinocytes. The matrix layer consists of the stratified keratinised epithelium; this epithelium has a high proliferation capacity. The stroma is referred to as the perimatrix layer; this stroma consists of connective tissue, vessels, polymorphous leukocytes, macrophages, and mast cells [3,4]. The etiopathogenesis of cholesteatoma is still controversial, and its development's cellular and molecular mechanisms are still unclear [5]. However, it has been demonstrated that the matrix and perimatrix are involved in developing destructive cholesteatoma. The increased proliferation of keratinised epithelium facilitates bone erosion by promoting the development and progression of cholesteatoma [6,7]. immunohistochemical staining. For each antibody in the study, four micron-thick sections from the tissues in the paraffin blocks were included in the research, and we took the control group on poly-L-lysine-coated slides. In the IHC studies, the antigen retrieval technique was used, and the avidin-biotin-peroxidase complex method was applied. The antibodies were examined in a Leica band max automatic immunohistochemical device. The Bond Polymer Refine Detection kit (DS9800, Leica Biosystems, Buffalo Grove, IL) was used for each antibody. The necessary staining procedure on the data sheet was applied, and appropriate positive and negative controls were used for each antibody. The characteristics of the primary anticipation used in the immunohistochemical study are listed in Table 1. The prepared samples were examined under an Olympus BX51 model microscope after the coverslip was covered with the ultra-mount.

Immunohistochemical Assessment
To determine the amount of microvessel density, CD117-positive mast cells, tryptasepositive mast cells, and chymase-positive mast cells in cholesteatoma and control skin tissue, we first determined the densest area in three areas with low power magnification (×100 and ×200) in each biopsy. Then 4-5 fields at a magnification of ×400 under an ocular grid (0.075 mm 2 ) were counted. Then, the average value in each mm 2 area was determined with the following formula (Number of cells or microvessel per mm 2 = Number of cells or microvessel × magnification of 400/Area of Ocular Grid (0.075 mm 2 ). Formula validated in Sharma and colleagues' study [27].
Any endothelial lined-up vessel lumen or endothelial cell cluster appearing brown and separate from an adjacent cluster was a single countable microvessel.
For Ki-67 at ×400 magnification, each biopsy's cholesteatoma epithelium and the control skin tissue epithelium were the most intense. We calculated the number of positive cells among these cells by counting 200 cells [28].

The Assessment of Temporal Bone and Ossicular Erosion
The assessment of temporal bone and ossicular erosion was based on the operative notes and the preoperative CT scans. For the quantification of ossicular erosion, the operated ears were classified according to a modification of the STAMCO classification [22]: 0 (no ossicles eroded), 1 (one ossicle eroded), 2 (two ossicles eroded) or 3 (three ossicles eroded).

Statistical Assessment
Obtained data were evaluated with descriptive statistics (arithmetic mean, standard deviation, percentage distributions, and minimum and maximum values), and they were assessed with the Shapiro-Wilk test if continuous data did not fit normal distribution (p < 0.05). The Mann-Whitney U test was used to compare the paired groups. The Spearman Rank Differences Correlation Coefficient test was used to evaluate the correlation of two measurement data. Statistical significance level p was accepted as <0.05. Bonferronitype adjustment was applied to reduce the chance of making a Type I error when multiple measures were tested. Adjusted p level (p adjusted ) was calculated with formula "p/the number of comparisons".
In comparing the immunohistochemical evaluations between the cholesteatoma and control groups: the mean Ki-67 detected in the epithelium in the control group was 14.43 ± 2.37 and the mean in the perimatrix in the cholesteatoma group was 41.23 ± 23.19. The Ki-67 proliferative index was significantly higher in the cholesteatoma epithelium compared to the epithelium of the control group (p < 0.001) (Table 1, Figure 1). adjustment was applied to reduce the chance of making a Type I error when multiple measures were tested. Adjusted p level (padjusted) was calculated with formula "p/the number of comparisons".
In comparing the immunohistochemical evaluations between the cholesteatoma and control groups: the mean Ki-67 detected in the epithelium in the control group was 14.43 ± 2.37 and the mean in the perimatrix in the cholesteatoma group was 41.23 ± 23.19. The Ki-67 proliferative index was significantly higher in the cholesteatoma epithelium compared to the epithelium of the control group (p < 0.001) (Table 1, Figure 1). A comparison of CD117-positive mast cell count: the mean CD117 value was 71.43 ± 19.52/mm 2 in the subepithelial area (dermis) in the control group, and the mean CD117 value was 194.80 ± 92.54/mm 2 in the cholesteatoma matrix. The difference between the two groups regarding the mean of CD117-positive mast cells was statistically significant (p < 0.002).
Based on comparisons between chymase-positive mast cells and tryptase-positive mast cells in the subepithelial area and cholesteatoma matrix of the control group, our findings are as follows: The mean of chymase-positive mast cells was 48.57 ± 12.25/mm 2 in the control group and 133.14 ± 59.62/mm 2 in the cholesteatoma matrix; the difference was statistically significant (p < 0.001) (Table 2, Figure 1). A comparison of CD117-positive mast cell count: the mean CD117 value was 71.43 ± 19.52/mm 2 in the subepithelial area (dermis) in the control group, and the mean CD117 value was 194.80 ± 92.54/mm 2 in the cholesteatoma matrix. The difference between the two groups regarding the mean of CD117-positive mast cells was statistically significant (p < 0.002).
Based on comparisons between chymase-positive mast cells and tryptase-positive mast cells in the subepithelial area and cholesteatoma matrix of the control group, our findings are as follows: The mean of chymase-positive mast cells was 48.57 ± 12.25/mm 2 in the control group and 133.14 ± 59.62/mm 2 in the cholesteatoma matrix; the difference was statistically significant (p < 0.001) (Table 2, Figure 1). The mean of tryptase-positive mast cells was 35.57 ± 6.68/mm 2 in the control group and 91.89 ± 41.68/mm 2 in the cholesteatoma matrix; the difference was statistically significant (p < 0.003).
Chymase-positive and tryptase-positive mast cells are more prevalent in the cholesteatoma matrix than in the control group. The mean number of chymase-positive mast cells was also dominant.
The mean MVD was significantly higher in cholesteatoma than in the control group (p < 0.005); the mean of the control group was 10.71 ± 4.15/mm 2 , while the mean of the cholesteatoma group was 123.26 ± 50.32/mm 2 .   The mean of tryptase-positive mast cells was 35.57 ± 6.68/mm 2 in the control group and 91.89 ± 41.68/mm 2 in the cholesteatoma matrix; the difference was statistically significant (p < 0.003).
Chymase-positive and tryptase-positive mast cells are more prevalent in the cholesteatoma matrix than in the control group. The mean number of chymase-positive mast cells was also dominant.
The mean MVD was significantly higher in cholesteatoma than in the control group (p < 0.005); the mean of the control group was 10.71 ± 4.15/mm 2 , while the mean of the cholesteatoma group was 123.26 ± 50.32/mm 2 . Figures 2-7 compare the haematoxylin-eosin (HE) findings and immunohistochemical findings in control and cholesteatoma tissues.            For the quantification of ossicular erosion, the operated ears were classified according to a modification of the STAMCO classification; the distribution of our 35 cases was as follows: The number of cases with a score of 0 (no bone erosion) was four (11.42%), the number of cases with a score of 1 (erosion in 1 bone) was five (14.28%), and the number of cases with a score of 2 (erosion in 2 bones) was twenty-six (74.28%). There was no case with a score of 3 (erosion in 3 bones). To perform group analysis, the 0 and 1 groups were combined into one "mild ossicular erosion" group, and the 2 and 3 groups were combined into one "advanced ossicular erosion" group. Accordingly, there was no statistically significant difference between the two groups in the Ki-67 index regarding the cholesteatoma epithelium, CD117-positive mast cells in the cholesteatoma matrix, tryptase-positive mast cells, chymase-positive mast cells, and microvessel density. Nevertheless, the mean of all studied immunohistochemical markers was higher in the group with a bone erosion score of 2 or above than in the other group. With increasing bone erosion scores, the means increased. (Table 3). Assuming the correlation of immunohistochemical markers with each other: A positive correlation was found between CD117 and chymase (r = 0.993, p < 0.001), tryptase (r For the quantification of ossicular erosion, the operated ears were classified according to a modification of the STAMCO classification; the distribution of our 35 cases was as follows: The number of cases with a score of 0 (no bone erosion) was four (11.42%), the number of cases with a score of 1 (erosion in 1 bone) was five (14.28%), and the number of cases with a score of 2 (erosion in 2 bones) was twenty-six (74.28%). There was no case with a score of 3 (erosion in 3 bones). To perform group analysis, the 0 and 1 groups were combined into one "mild ossicular erosion" group, and the 2 and 3 groups were combined into one "advanced ossicular erosion" group. Accordingly, there was no statistically significant difference between the two groups in the Ki-67 index regarding the cholesteatoma epithelium, CD117-positive mast cells in the cholesteatoma matrix, tryptase-positive mast cells, chymase-positive mast cells, and microvessel density. Nevertheless, the mean of all studied immunohistochemical markers was higher in the group with a bone erosion score of 2 or above than in the other group. With increasing bone erosion scores, the means increased. (Table 3). Assuming the correlation of immunohistochemical markers with each other: A positive correlation was found between CD117 and chymase (r = 0.993, p < 0.001), tryptase (r = 0.871, p < 0.001), and microvessel density (r = 0.775, p < 0.001). A positive correlation was found between tryptase and chymase (r = 0.837, p < 0.001) and microvessel density (r = 0.681, p < 0.001). A positive correlation was observed between chymase and microvessel density (r = 0.838, p < 0.001) ( Table 4).

Discussion
Cruvelhhier first described cholesteatoma in 1829 as a pearl tumour based on its macroscopic appearance. First described in the 1800s, the pathogenesis of the destructive behaviour of this benign lesion remains unclear. In its modern definition, cholesteatoma is a cluster of keratins developed from keratinised squamous cell epithelium within the air cells of the temporal bone. Despite its beginning histological nature, cholesteatoma is hazardous because of intense inflammation and bone erosion in the surrounding bone tissue [15].
According to the literature, which is compatible with our study, most adult cholesteatoma cases appear in males in their 40s [29].
Cholesteatoma epithelium proliferative capacity, keratinocyte features, and intercellular interactions may contribute to the pathogenesis. A significant factor contributing to the destructive capacity of cholesteatoma is the proliferation capacity of the stratified squamous epithelium that forms the perimatrix [6]. Studies report Ki-67 (an essential marker for detecting proliferation in neoplastic or nonneoplastic epithelium) overexpression [7,26]. Similarly, in our study, the cholesteatoma epithelium's proliferative capacity was considerably higher than that of normal skin.
The cholesteatoma matrix is similar to granulation tissue both structurally and in terms of cell diversity: the matrix consists of vessels, fibroblasts, and a wide variety of inflammatory cells [19]. The mast cells are among these: mast cells produce fibroblast growth factor, keratinocyte growth factor, and angiogenic and matrix remodelling factors [30]. In their investigation on acquired cholesteatoma, Hamed et al. found that basic fibroblast growth factors positively stained basal and parabasal keratinocytes. In addition, specific staining was observed in the basal columnar middle-ear epithelium and mast cell membrane [4]. Very few studies show the presence of mast cells in the cholesteatoma matrix. While mast cells have been demonstrated in previous studies to be present in cholesteatomas, the functions of these cells in developing cholesteatomas are unclear [31]. Our study, too, revealed a high number of mast cells in the cholesteatoma matrix compared to the control tissue.
Tryptase and chymase have been investigated and were found to be responsible for tumour angiogenesis [24]. When tryptase is secreted from activated mast cell granules, it causes the degradation of the extracellular matrix through its proteolytic action. The degradation of the extracellular matrix is an important step in the early phase of angiogenesis [32]. Tryptase, which is mitogenic for endothelial cells [33], also promotes the chemotaxis of inflammatory cells that express cytokines (IL-1, IL-6, IL-8, stem cell factor, TNF-α, and other inflammatory mediators) that stimulate endothelial cells [34]. Tryptase can activate MMP-9, which releases angiogenic factors in the extracellular matrix [35]. It is known that human recombinant tryptase is induced by a well-known angiogenic cytokine, namely VEGF secretion [36]. Few studies with small sample sizes investigate mast cell enzymes in the cholesteatoma matrix. Chymase-positive mast cell accumulation was reported in the cholesteatoma matrix in a study including five cholesteatoma cases diagnosed with chronic otitis media [37]. Similarly, we detected chymase-positive mast cell numbers in 35 cases we included in our study, and chymase-positive mast cell accumulation was higher than the tryptase-positive mast cell number. In addition, numbers of both chymase-and tryptase-positive mast cells were significantly higher than in the control tissue. Unlike previous studies [37], our study's tryptase-positive mast cell count is not low. The small number of cases in the previous study is likely responsible for the difference. In contrast to solid organ tumours, mast cells with chymase-positive stains are higher than those with tryptase-positive stains in cholesteatoma.
The process of angiogenesis is defined as the development of new vessels during the development of new tissues, and embryogenesis is an essential process in the development of tumour growth, invasion, and metastasis, as well as normal physiological processes such as tissue repair and wound healing [16,38]. Studies examining MVD in cholesteatoma are few, and the number of cases in the studies is still very limited. In a study conducted with 14 cases, while the mean microvessel was 5.44/mm 2 in the control group, it was found to be 22.79 ± 11.2/mm 2 in cholesteatoma cases [11]. In our study, like in the study above, MVD was significantly higher in the cholesteatoma group compared to the control group; 10.71 ± 4.15/mm 2 in the control group and 123.26 ± 50.32/mm 2 in cholesteatoma groups. Like granulation tissue, cholesteatoma exhibits intense vascularisation. Increasing vascularisation seems to be effective in developing cholesteatoma. Our study found the mean of microvessels in the control group and cholesteatoma cases higher than in the study above. It is possible that this difference can be attributed to a variety of factors, including the demographic differences in the control group, the duration of the lesion in the cholesteatoma group, demographic differences between the patients, and the differences in immune system response.
Even though cholesteatomas are benign, they grow to destroy temporal bone, much like malignant tumours, with a mechanism that is still unclarified [39]. Bone resorption of cholesteatoma is affected by the Ki-67 weighted proliferation index [25], epidermal growth factor [40], matrix metalloproteinase-9 [41], bone morphogenic protein [34], and cytokines such as TNF-alpha [42]. A previous study observed a moderately positive correlation between the measured Ki-67 proliferative index in the cholesteatoma epithelium and bone destruction [26].
The effect and mechanism of angiogenesis on bone destruction in cholesteatoma are unclear [12]. Furthermore, although there is evidence of an increase in mast cells in the cholesteatoma matrix [31], no link has been proven between this increase and bone destruction. The subepithelial connective tissue of cholesteatomas [perimatrix] exhibits angiogenesis. The process of angiogenesis facilitates and sustains the migration of keratinocytes into the middle ear cavity, their increased proliferation, and the expansion of cholesteatomas. As a result, angiogenesis is considered one of this disease's most destructive aspects [25]. In their investigation, Hamed et al. found that fibroblast growth factors positively stained not only the epithelium but also the mast cell membrane, which means that its expression in the mast cell membrane supports its role in bone resorption activity [4]. The results of our study indicated that the Ki-67 proliferative index of the cholesteatoma epithelium, the number of CD117-positive mast cells, the number of chymase-positive mast cells, the number of tryptase-positive mast cells, and the number of microvessels were not statistically significant in comparing the groups with mild bone erosion and advanced bone erosion. Regardless, we found that the mean of all parameters compared in the group with advanced bone erosion was higher than in the group with mild bone erosion. According to our study, the averages for all the markers examined were high in those with advanced bone erosion, suggesting that all these markers may have some effect on bone erosion. The validity of this claim requires further research with a larger sample size.
Mast cells regulate the tumour microenvironment in solid organ tumours, interact with invasive tumour cells, increase tumour angiogenesis, and contribute to invasion and survival [43]. Many of these settings shown to have potential functions as mast cells are capable of secreting, upon appropriate activation by a variety of immune or nonimmune stimuli, a wide variety of cytokines (including many chemokines) and growth factors, which may have autocrine, paracrine, local, and systemic effects [44]. Based on the correlation of immunohistochemical markers in our study, CD117 was positively correlated with chymase, tryptase, and MVD. Tryptase was positively correlated with MVD and chymase. A positive correlation was found between chymase levels and MVD. As a whole, except for the Ki-67 proliferative index, the positive correlation of other markers with each other and with MVD suggests that they promote angiogenesis in the cholesteatoma matrix, as well as in solid organ tumours and wound healing. Whether the mechanism for the upregulation of these enzymes is potentiation of each other or whether the specific one initiates the cascade, angiogenesis, and epithelial proliferation is not yet apparent. Our team will continue to work on this issue. It will be possible to determine the treatment goal and histological follow-up of therapy if we understand the mechanisms underlying cholesteatoma development and destruction.

The Strength of the Study
To the best of our knowledge, this is the first study to examine the effects of Ki67, CD117, tryptase, chymase, and CD34 on cholesteatoma progression and cholesteatomarelated bone loss and damage. In addition, we presented a relatively wide case population in our study.

Conclusions
CD117-positive mast cells, particularly chymase-positive mast cells, are found in the cholesteatoma matrix and can increase the proliferative capacity of the cholesteatoma epithelium and MVD (angiogenesis) in the cholesteatoma matrix. Thanks to these effects, they may contribute to the formation of cholesteatoma.
Interestingly, the proportion of chymase-positive mast cells in the cholesteatoma matrix is higher than that of tryptase-positive mast cells. In view of the fact that cholesteatoma is a benign condition, we may have detected it.
As a result of the increased proliferation of cholesteatoma epithelium and increased MVD via chymase-positive mast cells and tryptase-positive mast cells in the matrix, bone erosion was exacerbated. Bone erosion was associated with higher levels of Ki-67, CD117, tryptase, chymase, and CD34, suggesting that all these markers contribute to bone erosion.
Although cholesteatoma is a benign condition, its complications can have serious consequences. In treating cholesteatoma, mast cells can be selected as a target in addition to treatments that reduce the proliferative capacity of the epithelium, and treatments that will reduce MVD prevent bone erosion. This study is the first to examine the relationship between all these immunohistochemical markers and bone erosion.
There is, however, a need for larger studies to elucidate the mediators that contribute to the proliferation of epithelium and the enlargement of microvessels by mast cells.