Inflammation and macrophage polarization in cutaneous melanoma: Histopathological and immunohistochemical study

Abstract

Tumor-associated macrophages (TAMs) are considered to affect tumor growth and progression. Macrophages can be classified into two states of polarized activation, namely classically activated M1 macrophages and alternatively activated M2 macrophages. The dynamic balance between TAMs and tumor cells has an important impact on tumor homeostasis and progression. The aim of this study was to characterize the phenotype of TAMs present in different subtypes of superficial spreading cutaneous melanoma and their relationship with the lymphocytic infiltrate in order to identify new histopathological tools for melanoma prognosis and suitable targets for melanoma therapy. We selected four groups of patients with malignant melanoma in order to analyze the profile of polarized macrophage activation using immunohistochemical methods. Histopathological analysis showed that the macrophage polarization state appears to be more related to the lymphocytic infiltrate than to the thickness of the lesions. Further studies are necessary to increase understanding of the immunopathological dynamic of melanoma that may be modulated by future targeted immunotherapies.

Introduction

Melanomas are malignant melanocytic neoplasms that can occur in any part of the body, but more than 90% of all melanomas arise in the skin. ¹ Malignant melanomas (MMs) are considered important immunogenic tumor types whose biological behavior is influenced by host immune cells and by the presence of inflammatory cells within the tumor microenvironment. ² Tumor-associated macrophages (TAMs) and other cells of the innate immune system play a key role in chronic inflammatory processes promoting tumor development and progression. ³ In MM, TAMs and their secreted mediators may be involved in all steps of invasion and metastasis, correlating in most cases with poor prognosis. ⁴ Macrophages are usually categorized into two states of polarized activation. Classically activated M1 macrophages show immunostimulatory, anti-tumorigenic, and anti-angiogenic functions, whereas alternatively activated M2 macrophages are important for tumor growth and angiogenesis.^{5 –8} On the other hand, the presence of tumor-infiltrating lymphocytes (TILs) in MM is considered a manifestation of the host immune response to cancer, having in some cases a positive impact on disease progression and patient’s survival even if there are no conclusive data about their actual role. ⁹ The aim of this study was to characterize the phenotype of TAMs present in different subtypes of superficial spreading MM and their relationship with the lymphocytic infiltrate in order to identify new histopathological tools for melanoma prognosis and potentially provide suitable targets for melanoma therapy.

Material and methods

Patients and specimens

We selected four groups of patients with different subtypes of superficial-spreading MMs, diagnosed at the Institute of Dermatology of University of Magna Graecia, between 2004 and 2014, in order to analyze the profile of polarized macrophages activation. The four patient groups were characterized as follows: (1) MM with Breslow thickness <1 mm and heavy lymphohistiocytic infiltrate (MM-thin-brisk); (2) MM Breslow thickness <1 mm with scarce lymphohistiocytic infiltrate (MM-thin-non-brisk; (3) MM Breslow thickness >1 mm and heavy lymphohistiocytic infiltrate (MM-thick-brisk); (4) MM Breslow thickness >1 mm with scarce lymphohistiocytic infiltrate (MM-thick-non-brisk).

In our study, archived formalin-fixed and paraffin-embedded tissues were used. Histopathological (surgical resection) specimens, diagnosed and excised from a total of 72 untreated patients (39 men, 33 women) at the Department of Pathology of the School of Medicine of the University of Catanzaro, were evaluated. The average age was 57 years (age range, 31–72 years). The study was approved by the local ethics committee and informed consent was obtained. Seriated, deparaffinized sections (4-μm thick) were used for staining procedures including standard hematoxylin-eosin (H&E) and specific immunohistochemical (IHC) evaluation. All procedures were carried out at room temperature.

Immunohistochemistry

All cases were tested using a 1:100 dilution of mouse anti-CD68 antibody (Dako Italia, Milan, Italy), 1:100 dilution of mouse anti-iNOS (Sigma Aldrich, Milan, Italy), and 1:200 dilution of mouse anti-CD163 (Leica Microsystems, Milan, Italy) by means of an automated immunostainer (Bond TM Max). CD68⁺ cells are considered to be macrophages M0, M1, and M2, iNOS⁺ cells are considered typical M1 macrophages when a consistent morphology was expressed, while CD163⁺ cells are recognized as M2 macrophages. For all cases, histopathological diagnosis and tumor grading were jointly confirmed by two pathologists by re-evaluation of the original H&E stained tumor sections.

Confocal microscopy

For confocal microscopy, tumor sections were deparaffinized before endogenous peroxidase quenching and heat-induced epitope retrieval. In order to characterize the distribution pattern of M1/M2-polarized macrophages, two distinct double staining procedures were performed to detect CD163/iNOS and CD68. After permeabilization and blocking with 0.5% Saponin and 10% BSA in PBS for 30 min, sections were incubated (1–2 h) with primary reagents in the same permeabilization buffer. After incubation, the slides were extensively washed with PBS, then incubated for 1 h, with 1:400 dilution of FITC-conjugated anti-mouse antibody (Santa Cruz Biotechnology, Milan, Italy) and successively washed again. Antibodies directed against CD163 and iNOS were directly added to slices, whereas for immunostaining with 1:100 rabbit anti-CD68 (Abcam, Milan, Italy), we used an amplification biotin/streptavidin based method. In this case, after permeabilization, the slides were sequentially incubated with streptavidin and biotin using reagents of the avidin/biotin blocking kit (Vector Laboratories) and then processed for immunofluorescence with 1:1000 anti-rabbit secondary biotinylated antibody (Sigma-Aldrich, Milan, Italy) and 1:400 Alexa-Fluor 555 conjugated streptavidin (Santa Cruz Biotechnology, Milan, Italy). The sections were counterstained with 2 µg/mL DAPI (Santa Cruz Biotechnology, Milan, Italy), mounted using an antifade mounting medium (Life Technologies, Monza, Italy) and observed with a Laser Confocal Scanning Microscopy (SP2 LSCM, Leica Microsystems, Milan, Italy). Single staining for CD68, CD163, and iNOS, as well as for negative controls, were also carried out. The double immunofluorescence staining procedure was repeated three times.

Integrated analysis of immunohistochemical and confocal microscopy results

The ratio between M1 and CD68⁺ macrophage populations and between M2 and CD68⁺ macrophage populations was expressed as a score according to the following scheme: score 0, <1%; score 1, 1–20%; score 2, 21–40%; score 3, 41–60%; score 4, 61–80%; score 5, 81–100%. The score was assessed by two pathologists at ×100 magnification.

Statistical analysis

In order to verify if the assigned scores, representing the percentage of M1 or M2 macrophages, respectively, were significantly different between the four different groups, an ANOVA test was performed. All tests were considered statistically significant when the P value was <0.05%. The main results for each group are summarized in Table 1 and the relative graph.

Table 1.

Statistical analysis (ANOVA).

Melanoma type	Number	Score iNOS/CD68Mean (SD)	Score CD163/CD68Mean (SD)
Thin brisk	20	2.55 (1.15)	2.70 (0.923)
Thin non-brisk	13	1.92 (0.760)	3.62 (1.04)
Thick brisk	19	2.68 (1.00)	2.79 (1.03)
Thick non-brisk	20	1.80 (0.894)	3.40 (0.995)
P value		0.015	0.021

SD, standard deviation.

Results

Our data (Table 1) allow us to reject the null hypothesis. Moreover, examining the data between the groups, we may affirm that in the superficial spreading type of MM, M1/CD68, and M2/CD68, ratios are more closely linked to the inflammation degree than to the thickness of lesions (Figures 1 and 2).

Figure 1.

Distribution of M1 and M2 activated macrophages in thin brisk (TB), thin non-brisk (TNB), thick brisk (TKB), and thick non-brisk (TKNB) melanoma type. Columns represent the averages of iNOS/CD68 ratios (black columns) and CD163/CD68 ratios (gray columns) of immunohistochemical scores. Bars above columns indicate Standard Error.

Figure 2.

Panel 1: (a, b) a TB lesion immunostained for CD68. Please note the diffuse, red-stained macrophage component (arrowheads). (c–f) Immunohistochemical staining of CD163 in the same lesion; arrowheads indicate CD163 positive cells. In (f), a small group of CD163⁺ cells nearby melanoma cells is indicated (arrowheads); (c, ×40; a, d, ×100; b, e, ×200; f, ×400). Panel 2: (a, b) a TNB lesion immunostained for iNOS showed a weak positivity in some melanoma cells (arrowheads). (c–f) A TKB lesion. H&E staining showing the melanoma morphology (c). Immunochistochemical staining for iNOS showed many iNOS⁺, red stained cells (d); at the edge of the lesion M1 polarized iNOS⁺ TAMs and melanophages are present (e, f); (a, c, e, ×100; b, d, ×400; f, ×200). Panel 3: TB melanoma. Macrophages immunopositive for CD163 and iNOS were observed (arrowheads); note the presence of CD68⁺/iNOS^– and CD68⁺/CD163^– non-polarized macrophages (arrows). Panel 4: TNB melanoma. CD163⁺ macrophages (arrowheads; upper panel) outnumber iNOS⁺ macrophages (lower panel); note the presence of CD68⁺/iNOS^– and CD68⁺/CD163^– non-polarized macrophages (arrows). Panel 5: TKB melanoma. In this field, iNOS⁺ TAMs (arrowheads; lower panel) outnumber CD163⁺ macrophages (arrowhead; upper panel); non-polarized macrophages are also observed (arrows). Panel 6: TKNB melanoma. In this field, iNOS⁺ TAMs are nearly absent, whereas CD163⁺ macrophages (arrowheads; upper panel) are scattered in a very scarce inflammatory infiltrate. A small number of non-polarized macrophages is also observed (arrows).

The iNOS score may be partially overestimated in a few cases for the presence of iNOS⁺ neoplastic cells (Figure 1, Figure 2 [Panel 2a, b]). However, this error can be compensated for, in the ANOVA intergroup analysis, and confocal microscopy demonstrated that actually it is low. Qualitative histopathological analysis showed that distribution of TAMs was mostly peripheral in thin melanoma and central in thick lesions. Many iNOS⁺ cells, corresponding to macrophages, show a peripheral distribution (Figure 1, Figure 2 [Panel 2c–f]).

Confocal microscopy analysis confirms immunohistochemical data (Figure 1, Figure 2 [Panels 3–6]) but it also showed that not all CD68⁺ cells are phenotypically polarized (Figure 1, Figure 2 [Panel 5]).

Discussion

In 1969, Clark et al. first described the lymphocytic infiltration of primary cutaneous melanoma. ¹⁰ Elder et al. differentiated the lymphocytic infiltrate into brisk and non-brisk, according to its intensity, and demonstrated that TILs were of prognostic significance. ¹¹ Since then, such a datum is required in histopathological reports. ¹²

Macrophages are often greatly present in the tumor microenviroment as TAMs, mainly in the M2 state of polarized activation that balance the potential anti-tumoral and pro-inflammatory effects of M1 macrophages.^13,14

Using confocal microscopy in our study, we demonstrated that a population of non-polarized CD68⁺/CD163^–/iNOS^– macrophages (probably M0) is always present, suggesting that TAMs in melanoma are a complex, dynamic population. ¹⁵

We analyzed the profile of polarized macrophage activation in various subtypes of superficial spreading melanoma, using immunohistochemistry and correlated these preliminary data with the occurrence of the lymphocytic infiltrate. Our findings demonstrated that M1 macrophages are a key component of TAMs population both in thin melanoma as well as in thick melanoma. Immunohistochemical and confocal microscopy evaluation showed that in melanomas, levels of iNOS⁺ M1 macrophages were comparable to levels of CD163⁺ M2 macrophage when a brisk inflammatory infiltrate is present. On the other hand, in groups of thick and thin non-brisk superficial spreading MM, M2 macrophages density were much higher than M1 ones. Therefore, the macrophage polarization state appears to be related to the extent of lymphocytic infiltrate. Overall, our results suggest that M1 TAMs have a driver role in the regulation of TIL density, which constitute an important prognostic factor in histopathological reports.