Quantitative assessment of tumor-associated tissue eosinophilia and mast cells in tumor proper and lymph nodes of oral squamous cell carcinoma.

BACKGROUND: Oral squamous cell carcinoma (OSCC) is the most common cancer of oral cavity. Tumor stage, thickness, lymph node metastasis (LNM), extranodal spread, perineural invasion, tumor differentiation, mutations, human papillomavirus infection, and tumor microenvironment are independent prognostic indicators of OSCC. However, clinically, among all factors, LNM is considered an important prognostic factor in OSCC as it not only determines the stage of disease but also the strongest independent factor which predicts recurrence of disease. Further research proves that there are several biologically important factors in tumor tissue and LNs which promote or defend LNM. While it is proposed that tumor-associated tissue eosinophils (TATE) and mast cells (MCs) have "immuno-protective" effect, this remains unproven and various researchers have conflicting opinion. AIM: The aim is to determine the presence of TATE and MCs in OSCC and to evaluate if any association exists between them and LNM. STUDY
DESIGN: It is a comparative-retrospective study between 2 groups including 35 OSCC cases positive and 35 negative for LNM. MATERIALS AND
METHODOLOGY: Quantification of cells was done by counting total number of cells in 10 high-power fields under ×40 objective lens using "zigzag" method and dividing it by total number of fields. Eosinophils stained bright red with carbol chromotrope and MCs purple-violet with toluidine blue. STATISTICS: Independent t-test and Pearson's correlation were done using STATA IC 0.2 software. Level of significance was at 5%. Comparison of eosinophil and MC infiltration was done based on gender, metastatic, nonmetastatic LN, and in tumor proper. RESULTS AND
CONCLUSION: This study showed weak positive correlation between mean eosinophils count in tumor and LNs. Recognition of TATE and MCs as integral to tumor biology opens an avenue for novel approaches to cancer therapies. We conclude that an increased number of immunological cells are a favorable prognostic indicator in OSCC.

INTRODUCTION

Oral squamous cell carcinoma (OSCC), although seen worldwide, is more common in India. Lymph node metastasis (LNM) is shown to be a strong prognostic factor in OSCC. Many histopathological and immunohistochemical markers have been studied to predict LNM.[1]

OSCC is graded by degree of differentiation and amount of keratinization. Tumor microenvironment comprises a range of inflammatory cells, except neoplastic cells. These are chiefly lymphocytes, macrophages, neutrophils, plasma cell, mast cells (MCs), and eosinophils. Stromal response to tumor is characterized by intensity of lymphoplasmacytic infiltration around tumor and dense lymphoplasmacytic invasion is presumably indicative of good host response to tumor. Role of MCs and eosinophils has been implicated in biology of tumors. Eosinophils are thought to become active following action of MCs which secrete histamine and eosinophils chemoattractant factor and attract eosinophils in tissues.[2]

Tissue eosinophilia is a regular finding in allergic and parasitic disorders, but their role still needs to be evaluated in SCCs.[3] Eosinophils are present in large numbers in some SCC of the oral cavity, cervix, lower colon, and anus.[4] Eosinophils release chemical substances under diverse stimuli, such as interleukins, chemokines (RANTES, endotoxin-1), eosinophil chemoattractant protein, major basic protein, eosinophil peroxidase, and eosinophil-derived neurotoxin.[5] These substances may induce inflammation and cell death and contribute to tumor microenvironment. MCs play a diverse role that may contribute to defense against tumors or tumor progression. Recent studies have shown an increase in MC density (MCD) in OSCC, being associated with tumor-favoring effects.[6] Moreover, in an experimental model of carcinogenesis, MCD was associated with carcinoma development by upregulation of angiogenesis.[7]

While it is proposed that they have an “immunoprotective” effect, this remains unproven. Contradictory reports may relate to inconsistencies in counting. Hence, this study is taken up to evaluate infiltration of these immunological cells (eosinophils and MCs) in OSCC.

Cervical LNM is a major factor of outcome in OSCC. Although histologic evaluation of invasiveness provides useful information, histopathologic diagnosis provides only partial information on neoplastic changes. Consequently, biomarkers which characterize tumor behavior and predict outcome have sought to enhance treatment planning in patients with OSCC.[8]

Although only few studies have been done to understand the role of tumor-associated tissue eosinophils (TATE) and MCs in OSCC, it still remains unclear. It is further obscured by studies showing TATE and MCs associated with an improved prognosis and also with poor prognosis. The need for this study is to determine the presence of TATE and MCs in OSCC and to evaluate if any relation exists between TATE and MCs with LNM.

MATERIALS AND METHODOLOGY

This study was approved by the institutional ethical committee. It was a comparative-retrospective study of 12-month duration. Criteria for sample selection were to include histopathologically diagnosed OSCC cases which were surgically excised with concomitant neck dissections and to exclude patients with known primary other than oral cavity. Using 95% confidence interval and 80% power (Mann–Whitney test), formalin-fixed paraffin-embedded tissue blocks of seventy OSCC resection cases with concomitant neck dissections, comprising 35 cases positive and 35 cases negative for LNM were retrieved from archives of the Department of Oral Pathology, MPDC, Vadodara.

Three sections of 4-micron thickness were made for each case using soft-tissue microtome and stained with hematoxylin and eosin, carbol chromotrope, and toluidine blue stains, respectively.

Eosinophilic granules stained bright red with carbol chromotrope [Figure 1], and MC granules stained purple violet with toluidine blue [Figure 2]. These cells were then observed under compound microscope. Quantification was done by randomly selecting 10 high-power fields in each slide which showed high density of these cells. Each field was screened under ×40 objective lens using “zigzag” method for the evaluation of TATE and MCs. Total numbers of cells were counted and divided by total number of fields to obtain an average number of cells. Cells were counted, data tabulated, and statistical analysis was done.

Figure 1

Eosinophils stained by carbol chromotrope in nonmetastatic lymph node (×400)

Figure 2

Mast cells stained by toluidine blue in tumor stroma (×400)

Statistical analysis

It was done with Stata IC version 13.0, (StataCorp LLC, Texas, USA) using independent t-test and Pearson's correlation. Level of significance was set at 5%. Comparison of eosinophil and MC infiltration was done based on gender, metastatic, and nonmetastatic LNs and in tumor proper.

RESULTS

Among total metastatic cases, 77.1% (27) were male whereas 22.9% (8) were female. Among total nonmetastatic cases, 65.7% (23) were male whereas 34.3% (12) were female. There was no significant difference in gender distribution between two groups (P = 0.290).

Among metastatic cases, 40% (14) were from tongue, followed by 22.9% (8) from buccal mucosa, 11.4% (4) from lower alveolus, 8.6% (3) from palate, 5.7% (2) from lower sulcus, and 2.9% (1) each from corner of mouth, floor of mouth, maxilla, and retromolar triangle. Among nonmetastatic cases, 51.4% (18) were from tongue, followed by 25.7% (9) from buccal mucosa, 8.6% (3) from lower alveolus, 5.7% (2) each from floor of mouth and retromolar triangle, and 2.9% (1) from lower lip.

The mean age among metastatic group was 50.06 ± 13.86, whereas among nonmetastatic group was 49.06 ± 11.30. There was no significant difference in mean age between two groups (P = 0.742).

The mean eosinophil count in tumor proper of metastatic group was 4.03 ± 3.05 and of nonmetastatic group was 8.71 ± 4.40. There was a significant difference in mean eosinophil count in tumor proper between two groups (P < 0.001) [Table 1a and b].

Table 1a

Distribution of tumor-associated tissue eosinophils between metastatic and nonmetastatic groups (independent t-test)

Table 1b

Evaluation of P value

The mean eosinophil count in LNs of metastatic group was 5.06 ± 3.97 and of nonmetastatic group was 7.76 ± 3.74. There was a significant difference in mean eosinophil count in LNs between two groups (P = 0.005) [Table 1a and b].

The relationship between mean eosinophil count in tumor and mean eosinophil count in LNs suggest a weak positive correlation. The correlation value was 0.234 for metastatic group and 0.067 for nonmetastatic group [Table 2a and b].

Table 2a

Correlation of distribution of mean eosinophil count in primary tumor and in lymph nodes (group - metastatic)

Table 2b

Correlation of distribution of mean eosinophil count in primary tumor and in lymph nodes (group - nonmetastatic)

The mean MC count in tumor proper of metastatic group was 6.91 ± 3.84 and of nonmetastatic group was 8.48 ± 2.83. There was no significant difference in mean MC count in tumor proper between two groups (P = 0.056) [Table 3a and b].

Table 3a

Distribution of mast cells between metastatic and nonmetastatic groups (independent t-test)

Table 3b

Evaluation of P value

The mean MC count in LNs of metastatic group was 4.17 ± 2.89 and of nonmetastatic group was 5.81 ± 3.41. There was a significant association in mean MC count in LNs between two groups (P = 0.034) [Table 3a and b].

The relationship between mean number of MCs in tumor and mean number of MCs in LNs suggest a weak positive correlation. The correlation value was 0.237 for metastatic group and 0.091 for nonmetastatic group [Table 4a and b].

Table 4a

Correlation of distribution of mean mast cell count in primary tumor and in lymph nodes (group - metastatic)

Table 4b

Correlation of distribution of mean mast cell count in primary tumor and in lymph nodes (group - nonmetastatic)

DISCUSSION

Cancer terminates thousands of lives daily. Despite numerous attempts to find a permanent cure, overall survival has not increased, reason being restricted knowledge of tumor microenvironment. Therefore, this study was done in search of new prognostic markers for OSCC.

Tissue eosinophilia in SCC has long been known, but its role in tumor development remains debatable. Studies have reported both favorable and unfavorable prognoses for patients with tumors exhibiting TATE. Eosinophil infiltration in malignant tumor is seen in variety of tissues. It does not relate to site or etiology of tumor, nor to an idiosyncrasy of patients in whom it occurs.[9]

Various studies exist determining the role of TATE in carcinomas of nasopharynx, esophagus, breast, stomach, and cervix. However, very few have been done in OSCC. Moreover, various parameters have been researched in relation to TATE including tumor size, distant metastasis, etc. However, not many studies relate TATE with LNM.

The results of the present study show no statistical significance in relation to site. Among metastatic cases, 40% were from tongue, followed by 22.9% from buccal mucosa, 11.4% from lower alveolus, 8.6% from palate, 5.7% from lower sulcus, and 2.9% each from corner of mouth, floor of mouth, maxilla, and retromolar triangle. Among nonmetastatic cases, 51.4% were from tongue, followed by 25.7% from buccal mucosa, 8.6% from lower alveolus, 5.7% each from floor of mouth and retromolar triangle, and 2.9% from lower lip. This goes in concordance with studies done by Dorta et al. and Oliveira et al. who found no statistically significant differences in the distribution of eosinophils among these sites.[1011]

Our study suggested a weak positive correlation between mean eosinophil count in tumor and in LNs. The correlation value is 0.234 for metastatic group and 0.067 for nonmetastatic group. Contrary to this, Looi (2006) observed that TATE in primary tumor was not always associated with eosinophilia in the metastases.[12]

Our study also demonstrates a higher count of TATE in nonmetastatic cases of OSCC when compared to metastatic cases, which is in concordance with studies done by Ohashi et al. and Ishibashi et al. Findings indicate the cruciality of TATE in biological behavior of OSCC. The number of tumor-associated eosinophils was significantly higher in cases without LNM. This suggests possible correlation between TATE and a less aggressive biological behavior of tumor.[313]

Regarding the role of TATE various studies were conducted by Goldsmith et al. (1987), Goldsmith et al. (1992), Iwasaki et al. (2006), and Debta et al. (2011 and 2012). All these studies suggest that increased number of TATE is associated with antitumoral role and shows good prognosis. These are in concordance with our study which also shows an increase in TATE in nonmetastatic cases than metastatic cases, hence favoring good prognosis. Mean eosinophil count in tumor proper of metastatic group is 4.03 ± 3.05 and of nonmetastatic group is 8.71 ± 4.40. Thus, there is statistically significant difference in mean eosinophil count in tumor proper between two groups (P < 0.001). Mean eosinophil count in LNs of metastatic group is 5.06 ± 3.97, and nonmetastatic group is 7.76 ± 3.74. There is also a significant difference in mean eosinophil count in LNs between two groups (P = 0.005).

However, many other studies suggest a tumor-promoting role of eosinophils like those done by David Lowe (1984), Van Driel et al. (1996), Wong et al. (1999), and Alrawi et al. (2005). These studies are in contrast with the present study which suggests a favorable prognostic implication of TATE. These studies suggest that an elevated TATE in SCC is associated with aggressive tumor biology and presence of higher number of eosinophils in an excisional specimen should indicate the need for additional therapeutic measures and close surveillance to detect earlier locoregional recurrence and possible distant metastasis.

MCs play a role in tumor microenvironment, and increased MCD has been demonstrated in OSCC. Serum tryptase levels are elevated with some malignant tumors and may thus be a useful parameter. However, there are no data available about OSCC.[14] MCs are found to accumulate around and within many types of solid cancers and recently, MC function in developing tumors have been extensively reviewed, with varying suggestions that they may shift the balance either in favor of or against tumor growth.[15] A study done by Tanooka et al. in mice, support the hypothesis that MCs are involved in tumor suppression. Another study on effects of long-term administration of cancer-promoting substances on oral subepithelial MCs in rat done by Sand et al. suggest that MCs play a role in immunological cell defense against chemical carcinogens.[1617]

In a study done by Tomita et al., MC count was significantly higher in nonmetastatic nodes than in metastatic nodes. Same observation was noted in our study with an increase in MC count in nonmetastatic LNs when compared to metastatic nodes. Mean MC count in LNs of metastatic group was 4.17 ± 2.89 and of nonmetastatic group was 5.81 ± 3.41. There was a significant association in mean MC count in LNs between two groups (P = 0.034).[18]

Dabiri et al. showed that the presence of MCs in peritumoral stroma correlates with a good prognosis in breast cancers with long-term follow-up, supporting an important role for host MCs in breast cancer. This goes in concordance with the present study where mean MC count in tumor proper of metastatic group is 6.91 ± 3.84 and of nonmetastatic group is 8.48 ± 2.83 (P = 0.056).[19]

The purpose of a study done by Samoszuk et al. was to test the hypothesis that MCs present in fibrotic regions of cancer can suppress the growth of tumor cells through an indirect mechanism involving peritumoral fibroblasts. Degranulating MCs are restricted to peritumoral fibrous tissue, and MC heparin is a powerful inhibitor of clonogenic growth of tumor cells co-cultured with fibroblasts. These results may help to explain the well-known ability of heparin to inhibit the growth of primary and metastatic tumors.[20]

Various studies done by Tanooka et al., Sand et al., Tomita et al., Dabiri et al., Samoszuk et al., Alkhabuli (2007), Sinnamon et al. (2008), Debta et al. (2011), and Divyarani et al. (2014) suggested an inverse relationship between number of MCs and the amount of tumor tissue. They are suggestive of antitumoral role of MCs and its correlation with good prognosis. Our study is in concordance with these studies and suggests MCs favor good prognosis. The relationship between mean number of MCs in tumor and mean number of MCs in LNs suggest a weak positive correlation in the present study. Correlation value is 0.237 for metastatic group and 0.091 for nonmetastatic group.

However, various other studies suggest that MCs play a tumor-promoting role. These studies done by Yano et al. (1999), Imada et al. (2000), by Elpek et al. (2001), Iamaroon et al., Rojas et al., Madhuri Ankle (2007), Fakhrjou (2013), and Anuradha et al. (2014) are in contrast to the present study which shows higher MC count in nonmetastatic cases, both in tumor proper and in LNs; suggesting a negative role of MCs in tumor growth and metastasis.

CONCLUSION

Recognition of TATE and MCs as integral to tumor growth opens an avenue for novel approaches to cancer therapies to decrease tumor growth and metastasis.[15] We conclude that both immunological cells (TATE and MCs) have an effect on OSCC. Thus, quantitative assessments of these cells are important aspects of microscopic OSCC evaluation. For proper evaluation of these cells, special stains are an important tool that is budget-friendly and give an acceptable rapid result. With the results of our study, we conclude that an increased number of TATE and MCs were found to be a favorable prognostic indicator in OSCC with an increased mean cell count observed in nonmetastatic OSCC cases when compared to metastatic cases. A decrease in these cells possibly reflects an important modification in microenvironment during tumor initiation and progression. Currently, exact functional relevance of these cells in tumors is perplexing. Their role needs to be further validated using larger samples that include recurrent cases and follow-up studies. Hence, in search of new prognostic and predictive factors for OSCC, we conclude that an increase in infiltration of TATE and MCs is associated with favorable prognosis in OSCC. Thus, quantitative assessment of eosinophils and MCs are the most essential aspects of microscopic evaluation of OSCC.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.