Small oral squamous cell carcinomas with nodal lymphogenic metastasis show increased infiltration of M2 polarized macrophages – An immunohistochemical analysis

Abstract

Background

In solid malignancies the influence of immunological parameters – especially of macrophages – on invasiveness, metastatic potential and prognosis has been shown. There are no studies quantitatively analysing the macrophage polarization in oral squamous cell carcinoma (oscc). The aim of this study was to correlate macrophage polarization in the epithelial and stromal compartment of oscc with histopathologic parameters.

Methods

T1 and T2 oscc samples (n = 34) were used. Automated immunohistochemical staining detected CD68, CD11c, CD163 and MRC1 positive cells. All samples were completely digitalized using whole slide imaging and the number of stained cells per area was assessed quantitatively.

Results

Primary tumours with lymphogenic metastasis (N+) showed a significantly (p < 0.05) increased count of CD68, CD11c, CD163 and MRC1 positive cells in the epithelial fraction compared to N0 tumours. The ratio of CD163 positive cells (M2 macrophages) to CD68 positive cells (M1 and M2 macrophages) was significantly (p < 0.05) increased in N+ tumours.

Conclusion

An increased macrophage infiltration and an increased M2 polarization in primary oral squamous cell carcinomas with lymphogenic metastasis was shown. Macrophages that migrated into the epithelial tumour fraction seem to be of special biological importance.

The results indicate a central role of macrophages in the progression of oscc.

Introduction

The oral squamous cell carcinoma (oscc) is the 8th most frequent tumour worldwide (Scully and Bagan, 2009) and leads to a high level of treatment-based morbidity with aesthetic and functional impairment (Lopez-Jornet et al., 2012). In spite of distinct advances in the surgical treatment options, the prognosis of this kind of cancer has not improved during the last 30 years (Kumagai et al., 2010).

In the 19th century, a connection between malignancy and inflammation was hypothesized. The search for molecular alterations in cancer cells, though, has been at the centre of scientific interest for the recent past (Balkwill and Mantovani, 2001, Mantovani et al., 2008). A possible role of impaired immune reactions on tumour spread and metastasis formation was not addressed for a long period. Today, a key influence of the immune system on the tumour biology is largely accepted (Mantovani et al., 2008).

For investigating tumour immunology an in vitro tumour-model will not be able to represent all the interactions between the tumour and immune system, and might therefore lead to unsatisfactory conclusions. For immunological reactions, the complex interplay of immune cells, endothelial cells, stromal cells, local cytokines and chemokines and many other factors are responsible. As these framework requirements are only present in vivo, in this study histological sections of tumour tissue have been investigated to display the in vivo situation.

The relevance of immunological markers in tumour tissue has been shown in several solid malignancies (Ino et al., 2013, Kurahara et al., 2011, Shah et al., 2011, Watanabe et al., 2010). Based on their role in linking innate with adaptive immunity and their immunoregulatory capabilities, macrophages are of special interest when investigating tumour immunology.

The literature distinguishes between two different activation states of macrophages. M1 macrophages are responsible for a type I inflammatory reaction. This is accompanied by elimination of pathogens, tissue destruction and tumour resistance (Mantovani et al., 2007, Mantovani et al., 2013, Sica and Mantovani, 2012). In contrast, M2 cells cause a type II inflammation and have immunoregulatory properties. They are responsible for wound healing, tissue remodelling and angiogenesis and also for tumour progression (Cao et al., 2011, Hirata et al., 2011, Kurahara et al., 2011, Mantovani et al., 2007, Mantovani et al., 2013, Murray and Wynn, 2011, Sica and Mantovani, 2012). CD68 is an established marker to stain monocytes and macrophages regardless of their polarization (Cao et al., 2011, Cho et al., 2012, Kawamura et al., 2009). M1 polarized tissue macrophages can be identified by staining the CD11c surface antigen (Cho et al., 2012, Hirata et al., 2011, Pejnovic et al., 2013). To detect M2 polarized macrophages the antigens CD163 (Aron-Wisnewsky et al., 2009, Cao et al., 2011, Kawamura et al., 2009) and MRC1 (Aron-Wisnewsky et al., 2009, Hirata et al., 2011) are commonly used.

When investigating macrophages using immunohistochemistry (CD68, CD163, CD204) in pancreatic cancer specimens, patients with a higher number of tumour infiltrating M2 macrophages showed more frequent occurrence of lymphatic metastasis leading to a worse prognosis (Kurahara et al., 2011).

Examining the oscc, macrophages also seem to be of importance. Previous studies show a positive correlation between macrophage infiltration in the tumour stroma and the invasive front of an oscc with a higher histological grading (El-Rouby, 2010). It has been shown that the number of tumour infiltrating macrophages is an independent prognostic factor for disease free survival and overall survival in oscc (Lu et al., 2010).

When predominantly T3 and T4 tumours were investigated, CD68 positive macrophage infiltration in metastatic oscc specimens was significantly higher than in non-metastatic and healthy control groups (Costa et al., 2012). Flow cytometry and RT-PCR showed an M2 like cytokine profile (IL-10 high and TGF-β high) in the oscc samples of 20 patients (Costa et al., 2012). For methodical reasons, a differential consideration of the various tissue types like tumour, stroma, lymphatic infiltration or necrosis was not possible.

Another study analysed the cancer associated fibroblasts and macrophages in T1 – T4 oscc stroma tissue (Fujii et al., 2012). These data showed that high CD163 levels correlate with poor outcome and are an independent prognostic factor (Fujii et al., 2012). One limitation of this study was that the authors used a semiquantitative method rather than assessing cell counts.

There is thus evidence that macrophage polarization influences progression of solid malignancies, but there are no studies quantitatively analysing the macrophage polarization in oscc specimens, taking into account the topographic distribution in the epithelial and stromal tumour compartment. The present study analyses T1 and T2 tumours because they are characterized by easier resectability and better prognosis compared to larger T3 and T4 tumours (O'Brien et al., 2003). Furthermore, T4 tumours could show a special immunological microenvironment due to their direct contact to bone marrow stem cells.

The aim of this study is to correlate the macrophage polarization in small (T1, T2) primary oral squamous cell carcinomas with histopathological parameters like TNM-status, grading, vein-, lymph-vessel-, and perineural-infiltration. Therefore, we performed immunohistochemical analysis of oscc specimens using a computer assisted quantitative cell count distinguishing between the epithelial and the stromal tumour fraction.