Immunohistochemical Analysis of Foxp3+, CD4+, CD8+ Cell Infiltrates and PD-L1 in Oral Squamous Cell Carcinoma

Immunohistochemical Analysis of Foxp3+, CD4+, CD8+ Cell Infiltrates and PD-L1 in Oral Squamous... Pathol. Oncol. Res. (2018) 24:497–505 DOI 10.1007/s12253-017-0270-y ORIGINAL ARTICLE + + + Immunohistochemical Analysis of Foxp3 ,CD4 ,CD8 Cell Infiltrates and PD-L1 in Oral Squamous Cell Carcinoma 1 1 Olga Stasikowska-Kanicka & Małgorzata Wągrowska-Danilewicz & Marian Danilewicz Received: 16 September 2016 /Accepted: 22 June 2017 /Published online: 1 July 2017 The Author(s) 2017. This article is an open access publication Abstract The immunoexpression of the PD-L1 and the num- correlation between immunoexpression of PD-L1 and the ber of immune infiltrating cells have been shown to be a sig- number of CD4+ cells in OSCCPP group. In conclusion, our nificant prognostic factors in various human cancers. findings support the hypothesis of involvement of Tregs and Immunohistochemical method was used to examine the PD-L1 in OSCC development and progression. immunoexpression of PD-L1 and number of Foxp3+, CD4+ . . . , CD8+ cells in 78 cases of oral squamous cell carcinomas Keywords Treg FoxP3 PD-L1 Oral cancer (OSCCs): with better prognosis - OSCCBP (n = 37), and with poorer prognosis - OSCCPP (n = 41), and 18 cases of normal mucosa as a control. The immunoexpression of PD-L1 and the Introduction mean number of Foxp3+ cells was significantly increased in OSCCPP group in comparison to OSCCBP and control Cancer of the oral cavity is the sixth most common malignan- groups. The mean number of CD4+ cells was significantly cy reported worldwide [1]. It is the most common cancer in increased in OSCCPP group in comparison to OSCCBP and males and the third most common cancer in females [2]. Every control groups. CD8+ cells were significantly more numerous year, an estimated 3 million new cases occur worldwide, and in OSCCBP group in comparison to OSCCPP and control the overall 5-year survival rate for oral squamous cell carci- group. In both OSCCPP and OSCCBP groups there were noma (OSCC) is only 50%. Chronic inflammation, coupled positive significant correlations between number of Foxp3+ with alcohol, betel quid and cigarette consumption is associ- and CD4+ cells. We found positive correlations between the ated with oral squamous cell carcinoma [3]. immunoexpression of PD-L1 and numbers of Foxp3+ cells, Regulatory T cells (Treg) consist of functionally diverse and negative correlation between the immunoexpression of subsets of immunosuppressive T cells that play a crucial role PD-L1 and numbers of CD8+ cells in both OSCCPP and in the modulation of immune responses and the reduction of OSCCBP groups. We found also significant positive deleterious immune activation [4]. Several subsets of Treg cells have been identified and characterized, such as CD8 Treg cells, CD4 Treg cells, and γδ-TCR [5]. Tregs can be divided * Olga Stasikowska-Kanicka into two subpopulations: natural occurring (nTreg) which de- olga.stasikowska@umed.lodz.pl velop in the thymus, and adaptive (iTreg) generated in the periphery from CD4 naive T cells [6]. Tregs are capable of Małgorzata Wągrowska-Danilewicz migrating to inflammation sites and suppressing a broad range malgorzata.wagrowska@umed.lodz.pl of effector lymphocytes, particularly helper T (Th) cell subsets, Marian Danilewicz such as Th1, Th2, Th17, and follicular Th (Tfh) cells [7, 8]. marian.danilewicz@umed.lodz.pl Until now, Foxp3 (forkhead box P3) has been the most specific marker distinguishing Treg cells from T cells. Foxp3 is a mem- Department of Nephropathology, Medical University of Lodz, ul. Czechoslowacka 8/10, 92-216 Lodz, Poland ber of the forkhead/winged-helix family of transcription factors that are critically involved in the development and function of Department of Pathomorphology, Medical University of Lodz, Lodz, Poland Tregs [9]. The lack of Tregs due to the loss of Foxp3 function 498 Stasikowska-Kanicka O. et al. leads to autoimmune diseases whereas high prevalence of basis of their PD-L1 status and the presence or absence of Tregs in the peripheral blood due to the over-expression of tumor-infiltrating lymphocytes (TILs). These include type I + pos Foxp3 causes immunodeficiency [10]. Foxp3 Tregs can sup- (PD-L1 with TILs driving adaptive immune resistance), type press the activation, proliferation, and effector functions of nu- II (PD-L1 negative with no TIL indicating immune ignorance), + + pos merous cell types, including CD4 ,CD8 Tcells, dendritic type III (PD-L1 with no TIL indicating intrinsic induction) cells (DCs), B cells, and natural killer (NK) cells [11]. and type IV (PD-L1 negative with TIL indicating the role of Emerging evidence supports the notion that Treg popula- other suppressor(s) in promoting immune tolerance) [25]. tion plays a critical role in the suppression of anti-tumor im- Therefore, the objectives of this study were to evaluate the + + mune response and thus contributes to cancer progression. immunoexpression of PD-L1 and the number of CD4 ,CD8 , Tregs can mediate peripheral tolerance by suppressing self- Foxp3 cells in oral cancers. Another purpose was to find + + antigen reactive T cells [12]. Since most tumor antigens are possible association between number of Foxp3 ,CD4 , self-antigens, the suppression of tumor antigen reactive T lym- CD8 cells and immunoexpression of PD-L1. phocytes by Tregs is an important obstacle in antitumor im- munity [13]. Tumor cells can secrete soluble factors that pro- mote the induction, expansion, and recruitment of Treg cells Material and Methods to the tumor microenvironment. Therefore, tumors contribute to the generation and expansion of Treg cells in the tumor Patients microenvironment [6]. There are three ways of Tregs accumu- lation within the tumor microenvironment: increased migra- Seventy eight formalin-fixed, paraffin-embedded tissue spec- tion, preferential Treg cells expansion, and de novo conver- imens of oral squamous cell carcinomas (OSCC), and eigh- sion of Foxp3 T cells into Treg cells [14]. teen control cases (normal mucosa, non-cancer affected pa- The programmed death-ligand 1 (PD-L1) belongs to the B7 tients) were retrieved from archival material (Chair of superfamily, which also includes B7–1 (CD80), B7–2 Pathomorphology, Medical University of Lodz, Poland). (CD86), B7-DC (PDL2), B7-H2, B7-H3, B7-H4 and B7-H6 Paraffin-embedded tissue sections taken from postoperative [15]. PD-L1 has two known receptors, programmed death-1 material were diagnosed using a standard haematoxylin and (PD-1) and B7–1(CD80)[16]. PD-1 as a dominant receptor, eosin staining and the histological diagnoses were established belongs to the CD28 family and is expressed on T cells, den- according to the current standards [26]. The main criteria for dritic cells, natural killer cells, macrophages and B cells [15]. patients selection was the same anatomical localization of le- PD-L1 is constitutively expressed on murine T, B cells, DCs, sions (the floor of the mouth). To find the possible relationship macrophages, mesenchymal stem cells and cultured bone between the studied markers and clinical prognosis, patients marrow-derived mast cells [17]. PD-L1 is also expressed on with OSCC were additionally divided into two groups: with non-hematopoietic cells, including epithelial, vascular endo- better prognosis – OSCCBP (without metastases, n =37), and thelial, muscle cells, hepatocyte, pancreatic and astrocyte with poorer prognosis – OSCCPP, (with metastases to region- cells, in addition to its expression in the eye [17], lung, kidney, al lymph nodes or/and with distant metastases, n =41). The spleen, thymus, placenta, and the heart [18]. histopathological grade was classified into groups according PD-L1 is both inducible and constitutively expressed on cells to the WHO classification (for OSCCBP: G1 n =3, G2 n =33, of many solid and hematologic malignancies [19–23]. The ab- G3 n = 1, and for OSCCPP: G1 n =0, G2 n =36, G3 n =5). normal expression of PD-L1 has been linked with prognosis and The age range for OSCCBP group was from 28 to 75 years treatment response in multiple malignancies. An overexpression (mean ± SD = 59,24 ± 10, 89), for OSCCPP group was from of PD-L1 has been observed in different solid tumors including 40 to 84 (mean ± SD = 59,39 ± 11,16) and for control cases 15 melanoma [19], colorectal cancer [20], lung cancer [21], pancre- to 74 (mean ± SD = 47,05 ± 18,71). atic carcinoma [22] and hepatocellular carcinoma [23]. Mounting evidence suggests that the PD1:PD-L1 pathway Immunohistochemistry may play a central role in antigen-specific T cell response mediating PD-1-dependent immune suppression. When PD- Paraffin-embedded, 3-μm tissue sections were mounted onto 1 interacts with cells bearing one of its ligands, which can be SuperFrost slides (SuperFrost Plus, Gerhord Menzel GmbH, highly expressed on cancer cells, the ability of T cells to target Braunschweig, Germany), deparaffinized in xylene and ethanol the tumor cells can be effectively subverted [24]. Tumors can of graded concentrations. For antigen retrieval, the slides were thereby employ the PD-1:PD-L1 inhibitory pathway to silence treated in a microwave oven in a solution of TRS (Target the immune system. Thus, interrupting this interaction can Retrieval Solution, High pH, Dako, Denmark) for 30 min improve the ability of T cells to attack tumor cells. (2 × 6 minutes 360 W, 2 × 5 180 W, 2 × 4 minutes 90 W). Tumors escape immune surveillance by a number of mech- After cooling down at room temperature, they were transferred anisms of which four groups have now been proposed on the to 0,3% hydrogen peroxide in methanol, for 30 min, to block + FoxP3 Cells and PD-L1 in Oral Squamous Carcinoma 499 endogenous peroxidase activities. Sections were rinsed with Results Tris-buffered saline (TBS, Dako, Denmark) and incubated from 30 to 60 min with monoclonal mouse primary antibodies The cytoplasmic, perinuclear and nuclear pattern of Foxp3 against: CD4 (Dako; clone: 4B12, dilution 1:40), CD8 (Dako; immunoexpression on tumor infiltrating cells was seen in clone: C8/144B, dilution 1:50), Foxp3 (Abcam; clone: all OSCC cases and 9 control cases. Tumor cells were de- 236A/E7, dilution 1:50), and rabbit polyclonal antibody against void of Foxp3 staining. CD4 and CD8 were clearly stained PD-L1 (Abcam; dilution 1:400). Immunoreactive proteins were in the cell membrane of infiltrating cells of control and visualized using adequate EnVision-HRP kit (Dako, OSCC cases. The immunoexpression of PD-L1 in cancer Carpinteria, CA, USA) according to the instructions of the cells was predominantly cytoplasmic although membra- manufacturer. Visualisation was performed by incubation the nous expression was also noted. Membranous PD-L1 sections in a solution of 3,3′-diaminobenzidine (Dako, immunoexpression was predominantly detected on infiltrat- Denmark). After washing, the sections were counterstained ingcells.Inour studyPD-L1wasexpressedoncancercells with Mayer’s hematoxylin and mounted. For each antibody and tumor-infiltrating lymphocytes as well as epithelial, and for each sample a negative control was processed. vascular endothelial and infiltrating cells of control cases. Negative controls were carried out by incubation in the absence The cytoplasmic immunoexpression of PD-L1 on cancer of the primary antibody and always yielded negative results. cells was noted in 37 of 41 cases of SCCPP group, in 25 In each specimen distribution and cytoplasmic staining in- of 37 cases of SCCBP group, and in 16 of 18 control cases. tensity of PD-L1 in cancer cells were recorded semiquantita- The semiquantitative and quantitative data on the + + + tively by two independent observers in 7–10 (depending on immunoexpression of PD-L1, Foxp3 ,CD4 and CD8 cells the specimen size) adjacent high power fields and graded from appear in Table 1. The immunoexpression of PD-L1 and the 0 (staining not detectable), 1 (weak immunostaining), 2 (mod- mean number of Foxp3 cells was significantly increased in erate immunostaining intensity) and 3 (strong staining). The OSCCPP group (Figs. 1 and 2), in comparison to OSCCBP mean grade was calculated by averaging grades assigned by (Figs. 3 and 4) and control groups (Figs. 5 and 6). We also the two authors and approximating the arithmetical mean to found significantly higher immunoexpression of PD-L1 and the nearest unity. number of Foxp3 cells in OSCCBP compared to control group. The mean number of CD4 cells was significantly in- creased in OSCCPP group (Fig. 7) in comparison to both Morphometry OSCCBP (Fig. 8) and control groups (Fig. 9). CD8 cells were significantly more frequent in OSCCBP group (Fig. 10)in + + + Foxp3 ,CD4 and CD8 cells were evaluated using computer comparison to OSCCPP (Fig. 11) and control group (Fig. 12). image analysis system consisting of a PC computer equipped In both OSCCPP and OSCCBP groups there were positive with a Pentagram graphic tablet, Indeo Fast card (frame grab- significant correlations between the number of Foxp3 and ber, true-color, real-time), produced by Indeo (Taiwan), and CD4 cells, whereas the correlations between the number of + + color TV camera Panasonic (Japan) coupled with Carl Zeiss Foxp3 and CD8 cells were not statistically significant microscope (Germany). This system was programmed (Table 2). The correlative study revealed in both OSCCPP (MultiScan 18.03 software, produced by Computer Scanning and OSCCBP groups, positive correlations between the Systems, Poland) to calculate the number of objects (semiau- immunoexpression of PD-L1 and numbers of Foxp3 cells, tomatic function). and negative correlation between the immunoexpression of + + + + The number of Foxp3 ,CD4 as well as CD8 cells was PD-L1 and numbers of CD8 cells. We found also significant estimated by counting all positive cells in 7–10 high power positive correlation between immunoexpression of PD-L1 and 2 + monitor fields (HPF) (0.029 mm each), marking the number of CD4 cells in OSCCPP group (Table 3). immunopositive cells (semiautomatic function). The results In control group all these correlations were weak and not were presented as a number of positive cells per HPF. significant (data not shown). Statistical Methods Discussion Differences between groups were tested using unpaired There is accumulating evidence that head and neck squamous Student’s t-test preceded by evaluation of normality and cell carcinoma (HNSCC) patients display increased levels of Levene’s test. The Mann-Whitney U test was used where ap- nTreg cells with greater suppressive activity, compared to propriate. Correlation coefficients were calculated using healthy controls [27, 28]. However, while some studies have Spearman’s method. Results were considered statistically sig- linked higher Treg cells levels to worse clinical outcome in nificant if p <0.05. HNSCC [27], others have provided conflicting results [28]. 500 Stasikowska-Kanicka O. et al. Table 1 The immunoexpression Groups PD-L1 (mean score) Foxp3 Cells/HPF CD4+ Cells/HPF CD8+ Cells/HPF of PD-L1, Foxp3, lymphocytes CD4+ and CD8+ in oral OSCCPP (n = 41) 2.33 ± 2.02 13.4 ± 10.2 18.6 ± 12.3 9.7 ± 2.9 squamous cell carcinomas with poorer prognosis (OSCCPP), in OSCCBP (n = 37) 1.32 ± 1.26 7.7 ± 6.2 12.2 ± 7.1 17.8 ± 11.1 oral squamous cell carcinomas Controls (n = 18) 0.63 ± 0.62 2.4 ± 1.8 9.5 ± 5.5 8.2 ± 4.3 with better prognosis (OSCCBP) OSCCPP vs OSCCBP p <0.011 p <0.005 p <0.007 p <0.001 and controls OSCCPP vs control p <0.002 p <0.001 p <0.004 p = 0.12 (NS) OSCCBP vs control p <0.033 p <0.05 p =0.16 (NS) p <0.001 NS not significant Foxp3 is known as the most specific marker distinguishing increased number of Foxp3 cells in the OSCCPP group com- Treg cells from T cells, and in our study Foxp3 was expressed pared to OSCCBP. An association between high intratumoral on tumor-infiltrating lymphocytes - tumor cells were entirely density of Foxp3 cells and poorer clinical prognosis can sug- negative. In contrary to above-mentioned results, Liang et al. gest that the presence of Foxp3 cells might play a role in [29] observed in tongue cancer, that Foxp3 can be expressed OSCC progression. Increased Foxp3 Treg infiltration has been by both tumor cells and tumor-infiltrating lymphocytes and known to be associated with worse clinical outcomes and that tumor cells were the major cell types expressing Foxp3 various poor prognostic factors in many cancers [33–35]. (59,3% of tongue squamous cell carcinomas). Similar positive Suzuki et al. [36] found in colorectal cancer that the number score for Foxp3 immunoexpression was observed in pancre- of intratumoral Foxp3 cells was positively associated with atic cancer cells (61%) [30], and breast cancer tissues (57% lymph node metastases. Furthermore, it has been reported that and 73%) [31]. Subcellular staining of Foxp3 was heteroge- high numbers of circulating Tregs are associated with rapid neous in the present study. Different expression level and tumor progression in experimental animal models of melano- complex post-translational modification of Foxp3 may be pos- ma and in patients with melanoma. In these patients, the pres- sible reasons. Chen et al. [32] demonstrated that in Tregs, ence of Foxp3 cells in primary tumor has also been associat- TCR-mediated post-translational modifications could mediate ed with a higher frequency of metastases in the sentinel lymph the regulation function, and influence the subcellular distribu- node [37]. Even so, the association of Foxp3 expression and tion of Foxp3. Authors revealed a change in the subcellular its impact on overall survival remains controversial. Kim et al. localization of Foxp3 from a more cytoplasmic/perinuclear to [38] observed that Foxp3 expression in tumor cells of colo- a nuclear expression pattern in Tregs activated with anti-CD3/ rectal cancer, but not in infiltrating Treg cells, were correlated anti-CD28 antibodies. with disease progression and poor prognosis. Moreover, liter- In accordance with previous studies in various malignant ature data demonstrate that tumor infiltration by Foxp3 Tregs diseases [33, 34], we found significantly higher numbers of is not always associated with a poor prognosis, but, on the infiltrating Foxp3 cells in both tested groups of OSCC com- contrary, can be associated with an improved prognosis in pared with controls. Additionally, we found significantly some cancer types. In colorectal cancer, high levels of Fig. 1 Cytoplasmic immunoexpression of PD-L1 in oral squamous cell Fig. 2 Nuclear and perinuclear immunoexpression of Foxp3 in oral carcinomas with poorer prognosis (OSCCPP). Immunohistochemistry. squamous cell carcinomas with poorer prognosis (OSCCPP). Total magnification × 100 Immunohistochemistry. Total magnification × 100 + FoxP3 Cells and PD-L1 in Oral Squamous Carcinoma 501 Fig. 3 Cytoplasmic immunoexpression of PD-L1 in oral squamous cell Fig. 5 Cytoplasmic immunoexpression of PD-L1 in control. carcinomas with better prognosis (OSCCBP). Immunohistochemistry. Immunohistochemistry. Total magnification × 100 Total magnification × 100 intratumoral CD4 cells in the group of OSCCPP compared infiltrating Treg cells were associated with early stage disease to OSCCBP and controls. An association between high and improved prognosis [39, 40]. intratumoral density of CD4 cells and poorer clinical prog- Tumor-derived CD4 Treg cells have been extensively nosis seems to be consistent with other findings [42]. In breast + + + studied in many different types of cancer. It has been strongly cancer, the frequency of CD4 CD25 FoxP3 regulatory T suggested that antigen-specific CD4 Tregs at tumor sites may cells was inversely correlated with clinical outcomes [42]. In significantly suppress immune responses, leading to immune a mouse model of human breast cancer, the depletion of + + + + tolerance of tumor cells. Among various CD4 T cell frac- CD4 CD25 T cells was shown to reduce CD4 CD25 T + + + tions, a particular subset with CD4 CD25 Foxp3 expression cell-mediated suppression, improve immunity, and enhance was previously described as regulatory T cells and was shown tumor regression [43]. to mediate suppression [41]. Increased number of CD4 cells We found significantly increased number of CD8 cells in and the positive correlation between number of Foxp3 and OSCCBP group in comparison to OSCCPP patients and con- CD4 cells was observed in both tested groups of OSCC, trols. These results seem to be consistent with other findings. suggesting that increased immune infiltration is associated Zhu et al. [44] observed that breast cancer patient survival was with an increased frequency of Treg cells within the infiltrate. associated with higher frequencies of CD8 cytotoxic T cells in Based on these data, we hypothesize also that the growth of infiltrating lymphocytes. Emerging evidence from clinical stud- OSCCs may induce the generation of CD4 Treg cells. ies emphasizes the role of CD8 T cells in the control of tumor Moreover, we described significantly increased number of growth and the prolongation of patient survival [45, 46]. Lack Fig. 4 Nuclear and perinuclear immunoexpression of Foxp3 in oral squamous cell carcinomas with better prognosis (OSCCBP). Fig. 6 Nuclear and perinuclear immunoexpression of Foxp3 in control. Immunohistochemistry. Total magnification × 100 Immunohistochemistry. Total magnification × 100 502 Stasikowska-Kanicka O. et al. Fig. 7 Membranous immunoexpression of CD4 in oral squamous cell Fig. 9 Membranous immunoexpressionof CD4incontrol. carcinomas with poorer prognosis (OSCCPP). Immunohistochemistry. Immunohistochemistry. Total magnification × 100 Total magnification × 100 of significant correlations between number of Foxp3 and cells and epithelial and vascular endothelial cells in control CD8 cells in both studied groups of cancers suggest that cases. Lyford-Pike et al. [49] suggest that in normal tissue, CD8 cells observed in our study may have other than regula- PD-L1 is induced in response to inflammatory cytokines such tory functions or represent non-traditional subpopulation of as IFN-γ. This system represents a major mechanism for tis- Tregs. Recent studies demonstrated that Foxp3 T cells are sue protection in the setting of T cell-mediated inflammation. heterogeneous with respect to phenotype, gene expression, It is well established that PD-L1 expression is up-regulated in and function, including suppressive and non-suppressive sub- solid tumors where it can provide direct tumor protection, and populations [47]. Miyara et al. [47] divided human Foxp3 + reduce activity of PD-1 expressing tumor-infiltrating effector cells into three functional subpopulations: effector, resting and CD4 and CD8 T cells [24, 50]. We observed a significantly non-suppressive cytokine-secreting Tregs. increased immonoexpression of PD-L1 in both tested groups In recent years, many studies have confirmed that cancer of OSCC compared to controls. Overexpression of PD-L1 has cells can evade host immune systems by expressing certain been identified in several cancers, including the head and neck ligands that down-regulate cytotoxic T lymphocytes through cancers [24, 48]. We found significantly higher immunoexpression of PD-L1 in OSCCPP compared to inhibitory pathways that are usually initiated by ligand- receptor interactions [48]. Currently, PD-1:PD-L1 pathway OSCCBP. Our results are in concordance with literature data seems to be a one of major mechanism of controlling tumor [48, 51]. Thompson et al. [51] reported statistically significant immunity. In our study PD-L1 was expressed on infiltrating association of PD-L1 expression with poor clinical outcome in Fig. 8 Membranous immunoexpression of CD4 in oral squamous cell Fig. 10 Membranous immunoexpression of CD8 in oral squamous cell carcinomas with better prognosis (OSCCBP). Immunohistochemistry. carcinomas with better prognosis (OSCCBP). Immunohistochemistry. Total magnification × 100 Total magnification × 100 + FoxP3 Cells and PD-L1 in Oral Squamous Carcinoma 503 Table 2 The correlations between mean number of Foxp3+ and CD4+, CD8+ cells in oral squamous cell carcinomas with poorer prognosis (OSCCPP), and oral squamous cell carcinomas with better prognosis (OSCCBP) Correlations OSCCPP (n = 41) OSCCBP (n =37) Foxp3+ vs r =0. 44, p <0.005 r =0.35, p < 0.04 CD4+ cells Foxp3+ vs r = − 0.21, p =0.18 (NS) r = − 0.27, p = 0.11(NS) CD8+ cells NS not significant infiltrating CD8 cells in both tested groups of OSCC. In contrary to our results, Thompson demonstrated that gastric cancer patients with higher CD8 T cell densities also have Fig. 11 Membranous immunoexpression of CD8 in oral squamous cell higher PD-L1 expression, indicating an adaptive immune re- carcinomas with poorer prognosis (OSCCPP). Immunohistochemistry. sistance mechanism may be occurring. Lyford-Pike et al. [49] Total magnification × 100 using quantitative RT-PCR found a significant increase in the expression of CD8 mRNA in PD-L1(+) as compared to PD- gastric cancer. Lin et al. [52] indicated that a higher PD-L1 L1(−) in oropharyngael cancer. On the other hand, Tokito expression level was correlated with several clinicopatholog- et al. [53] observed that lack of PD-L1 immunoexpression ical factors, such as distant metastasis. These authors sug- accompanied by increased CD8 cells density was significant- gested also that PD-L1 immunoexpression might be associat- ly associated with favourable survival in non-small cell lung ed with oral cancer development and progression. We found cancer. Increased number of CD8 cells in OSCCBP group, significant positive correlation between immunoexpression of lack of correlation of PD-L1 with number of Foxp3 cells and PD-L1 in tumor cells and the number of infiltrating Foxp3 negative correlation between number of CD8 cells and cells in both tested groups of OSCC and between PD-L1 and immunoexpression of PD-L1 seem to be consistent and sug- number of CD4 cells in OSCCPP group. Similar to melano- gest that tumor infiltrating CD8 cells may have other than ma [19], and in keeping with the proposed adaptive resistance suppressive function. hypothesis, in our study PD-L1 was not expressed uniformly Literature data and our results revealed the complicated within OSCCs, but rather at sites of lymphocyte infiltration. interactions within the tumor microenvironment and empha- Our results suggest that PD-L1 immunoexpression on cancer size that impact of individual types of immune cells may be cells is associated with Treg infiltration, and PD-L1 may be highly dependent on many factors. We hypothesize that the induced by an inflammatory microenvironment involving microenvironment of a tumor is critically important in deter- TILs. mining leukocyte phenotype and function. We speculate also We found significant negative correlation between that not only the number, type and localization of tumor infil- immunoexpression of PD-L1 on cancer cells and number of trating lymphocytes but activity (e.g. cytokine releasing pat- terns) of particular infiltrating cells can determine mutual re- lationship and has prognostic value. Although our findings support the hypothesis of involve- ment of Tregs and PD-L1 in OSCC development and progres- sion, further studies of the relationship between number and Table 3 The correlations between the immunoexpression of PD-L1 and mean number of Foxp3+, CD4+, CD8+ cells in oral squamous cell carcinomas with poorer prognosis (OSCCPP), and oral squamous cell carcinomas with better prognosis (OSCCBP) Correlation between OSCCPP (n =41) OSCCBP (n =37) PD-L1 vs Foxp3+ cells r =0.42, p <0.007 r =035, p <0.04 PD-L1 vs CD4+ cells r =0.31, p <0.002 r =0.25, p = 0.14 (NS) PD-L1 vs CD8+ cells r = − 0.47, p <0.05 r = − 0.33, p <0.05 Fig. 12 Membranous immunoexpression of CD8 in control. Immunohistochemistry. Total magnification × 100 NS not significant 504 Stasikowska-Kanicka O. et al. 17. Sharpe AH, Wherry EJ, Ahmed R et al (2007) The function of activity of immune infiltration cells and immunoexpression of programmed cell death 1 and its ligands in regulating autoimmunity PD-L1 on cancer cells are needed to better understand their and infection. Nat Immunol 8:239–245 role in oral carcinogenesis. 18. Freeman GJ, Long AJ, Iwai Yet al (2000) Engagement of the PD-1 immunoinhibitory receptor by a novel B7 family member leads to negative regulation of lymphocyte activation. J Exp Med 192: Acknowledgments This work was supported by grant of Medical 1027–1034 University of Lodz 503/6-038-01/503-61-002. 19. Hino R, Kabashima K, Kato Yet al (2010) Tumor cell expression of programmed cell death-1 ligand 1 is a prognostic factor for malig- nant melanoma. Cancer 116(7):1757–1766 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http:// 20. Shi SJ, Wang LJ, Wang GD et al (2013) B7-H1 expression is asso- creativecommons.org/licenses/by/4.0/), which permits unrestricted use, ciated with poor prognosis in colorectal carcinoma and regulates the distribution, and reproduction in any medium, provided you give appro- proliferation and invasion of HCT116 colorectal cancer cells. PLoS priate credit to the original author(s) and the source, provide a link to the One 8(10):e76012 Creative Commons license, and indicate if changes were made. 21. Velcheti V, Schalper KA, Carvajal DE et al (2014) Programmed death ligand-1 expression in non-small cell lung cancer. Lab Investig 94(1):107–116 22. Chen XL, Yuan SX, Chen C et al (2009) Expression of B7-H1 References protein in human pancreatic carcinoma tissues and its clinical sig- nificance. Chin J Cancer 28(12):1328–1332 1. Siegel R, Ma J, Zou Z et al (2014) Cancer statistics 2014. CA 23. Gao Q, Wang XY, Qiu SJ et al (2009) Overexpression of PD-L1 Cancer J Clin 64:9–29 significantly associates with tumor aggressiveness and postopera- tive recurrence in human hepatocellular carcinoma. Clin Cancer 2. Byakodi R, Byakodi S, Hiremath S et al (2012) Oral cancer in Res 15(3):971–979 India: an epidemiologic and clinical review. J Community Health 37:316–319 24. Azuma T, Yao S, Zhu G et al (2008) H1 is a ubiquitous antiapoptotic receptor on cancer cells. Blood 111(7):3635–3643 3. Scully C, Bagan J (2009) Oral squamous cell carcinoma overview. Oral Oncol 45:301–308 25. Teng MW, Ngiow SF, Ribas A et al (2015) Classifying cancers based on T-cell infiltration and PD-L1. Cancer Res 75:2139–2145 4. Facciabene A, Motz GT, Coukos G (2012) T-regulatory cells: key players in tumor immune escape and angiogenesis. Cancer Res 72: 26. Barnes L, Everson JW, Reichart P, et al (2005) World Health 2162–2171 Organization classification of Tumours. Pathology and Genetics Head and Neck Tumours IARC Press Lyon 168–176 5. Wang RF (2008) CD8 regulatory T cells, their suppressive mech- anisms, and regulation in cancer. Hum Immunol 69:811–814 27. Drennan S, Stafford ND, Greenman J et al (2013) Increased fre- quency and suppressive activity of CD127 Treg cells in the periph- 6. Zhang S, Ke X, Zeng S, et al (2015) Analysis of CD8+ Treg cells in eral circulation of patients with head and neck squamous cell car- patients with ovarian cancer: a possible mechanism for immune cinoma are associated with advanced stage and nodal involvement. impairment. Cell Mol Immunol Sep;12(5):580–591 Immunology. doi:10.1111/imm.12144 7. Chaudhry A, Rudra D, Treuting P et al (2009) CD4 regulatory T 28. Schott AK, Pries R, Wollenberg B (2010) Permanent up-regulation cells control TH17 responses in a Stat3-dependent manner. Science of regulatory T-lymphocytes in patients with head and neck cancer. 326(5955):98691 Int J Mol Med 26:67–75 8. Linterman MA, Pierson W, Lee SK et al (2011) Foxp3 follicular 29. Liang YJ, Liu HC, Su YX et al (2011) Foxp3 expressed by tongue regulatory T cells control the germinal center response. Nat Med squamous cell carcinoma cells correlates with clinicopathologic 17(8):975–982 features and overall survival in tongue squamous cell carcinoma 9. Sakaguchi S, Miyara M, Costantino CM et al (2010) FOXP3+ patients. Oral Oncol 47(7):566–570 regulatory T cells in the human immune system. Nat Rev 30. Hinz S, Pagerols-Raluy L, Oberg HH et al (2007) Foxp3 expression Immunol 10:490–500 in pancreatic carcinoma cells as a novel mechanism of immune 10. Hori S, Nomura T, Sakaguchi S (2003) Control of regulatory T evasion in cancer. Cancer Res 67(17):8344–8350 cell development by the transcription factor Foxp3. Science 31. Merlo A, Casalini P, Carcangiu ML et al (2009) FOXP3 expression 299:1057–1061 and overall survival in breast cancer. J Clin Oncol 27(11):1746–1752 11. Von Boehmer H (2005) Mechanisms of suppression by suppressor 32. Chen CX, Rowell WA, Thomas RM et al (2006) Transcriptional reg- T cells. Nat Immunol 6(4):338–344 ulation by Foxp3 is associated with direct promoter occupancy and 12. DiPaolo RJ, Glass DD, Bijwaard KE et al (2005) CD4+CD25+ T modulation of histone acetylation. J Biol Chem 281:36828–36834 cells prevent the development of organ-specific autoimmune dis- 33. Hiraoka N, Onozato K, Kosuge T et al (2006) Prevalence of ease by inhibiting the differentiation of autoreactive effector T cells. FOXP3+ regulatory T cells increases during the progression of J Immunol 175:7135–7142 pancreatic ductal adenocarcinoma and its premalignant lesions. 13. Sakaguchi S (2005) Naturally arising Foxp3-expressing CD25+ Clin Cancer Res 12:5423–5434 CD4+ regulatory T cells in immunological tolerance to self and 34. Kobayashi N, Hiraoka N, Yamagami W et al (2007) FOXP3+ reg- non-self. Nat Immunol 6:345–352 ulatory T cells affect the development and progression of 14. Oleinika K, Nibbs RJ, Graham GJ et al (2013) Suppression, sub- hepatocarcinogenesis. Clin Cancer Res 13:902–911 version and escape: the role of regulatory T cells in cancer progres- 35. Bohling SD, Allison KH (2008) Immunosuppressive regulatory T sion. Clin Exp Immunol 171:36–45 cells are associated with aggressive breast cancer phenotypes: a 15. Chen L (2004) Co-inhibitory molecules of the B7-CD28 family in potential therapeutic target. Mod Pathol 21:1527–1532 the control of T-cell immunity. Nat Rev Immunol 4(5):336–347 36. Suzuki H, Chikazawa N, Tasaka Tet al (2010) Intratumoral CD8(+) T/ 16. Park JJ, Omiya R, Matsumura Y et al (2010) B7–H1/CD80 inter- FOXP3 (+) cell ratio is a predictive marker for survival in patients with action is required for the induction and maintenance of peripheral T- colorectal cancer. Cancer Immunol Immunother 59(5):653–661 cell tolerance. Blood 116(8):1291–1298 + FoxP3 Cells and PD-L1 in Oral Squamous Carcinoma 505 37. Ma MW, Medicherla RC, Qian M et al (2012) Immune response in 46. Dhodapkar MV, Sznol M, Zhao B et al (2014) Induction of antigen-specific immunity with a vaccine targeting NY-ESO- melanoma: an in-depth analysis of the primary tumor and corre- sponding sentinel lymph node. Mod Pathol 25:1000–1010 1 to the dendritic cell receptor DEC-205. Sci Transl Med 6: 38. Kim M, Grimmig T, Grimm M et al (2013) Expression of Foxp3 in 232–251 colorectal cancer but not in Treg cells correlates with disease pro- 47. Miyara M, Sakaguchi S (2011) Human FoxP3(+)CD4(+) regulatory T gression in patients with colorectal cancer. PLoS One 8(1):e53630 cells: their knowns and unknowns. Immunol Cell Biol 89(3):346–351 39. Salama P, Phillips M, Grieu F et al (2009) Tumor-infiltrating 48. Ohigashi Y, Sho M, Yamada Y et al (2005) Clinical significance of FOXP3+ T regulatory cells show strong prognostic significance programmed death-1 ligand-1 and programmed death-1 ligand-2 ex- in colorectal cancer. J Clin Oncol 27(2):186–192 pression in human esophageal cancer. Clin Cancer Res 11:2947–2953 40. Correale P, Rotundo MS, Del Vecchio MT et al (2010) Regulatory 49. Lyford-Pike S, Peng S, Young GD, et al (2013) Evidence for a role (FoxP3+) T-cell tumor infiltration is a favorable prognostic factor in of the PD-1:PD-L1 pathway in immune resistance of HPV- advanced colon cancer patients undergoing chemo or associated head and neck squamous cell carcinoma. Cancer Res chemoimmunotherapy. J Immunother 33:435–441 15;73(6):1733-1741 41. Viglietta V, Baecher-Allan C, Weiner HL et al (2004) Loss of func- 50. Hirano F, Kaneko K, Tamura H et al (2005) Blockade of B7–H1 tional suppression by CD4+CD25+ regulatory T cells in patients and PD-1 by monoclonal antibodies potentiates cancer therapeutic with multiple sclerosis. J Exp Med 199:971–979 immunity. Cancer Res 65(3):1089–1096 42. Watanabe MAE, Oda JMM, Amarante MK et al (2010) Regulatory 51. Thompson ED, Zahurak M, Murphy A et al (2016) Patterns of PD-L1 T cells and breast cancer: implications for immunopathogenesis. expression and CD8 T cell infiltration in gastric adenocarcinomas and Cancer Metastasis Rev 29:569–579 associated immune stroma. Gut. doi:10.1136/gutjnl-2015-310839 43. Knutson KL, Dang Y, Lu H et al (2006) IL-2 immunotoxin therapy 52. Lin YM, Sung WW, Hsieh MJ, et al (2015) High PD-L1 expression modulates tumor-associated regulatory T cells and leads to lasting correlates with metastasis and poor prognosis in oral squamous cell immune-mediated rejection of breast cancers in neu-transgenic carcinoma. PLoS One 12;10(11): e0142656. doi:10.1371/journal. mice. J Immunol 177:84–91 pone.0142656 44. Zhu S, Lin J, Qiao G et al (2015) Differential regulation and func- 53. Tokito T, Azuma K, Kawahara A et al (2016) Predictive relevance tion of tumor-infiltrating T cells in different stages of breast cancer of PD-L1 expression combined with CD8+ TIL density in stage III patients. Tumour Biol 36(10):7907–7913 non-small cell lung cancer patients receiving concurrent chemora- 45. Mellman I, Coukos G, Dranoff G (2011) Cancer immunotherapy diotherapy. Eur J Cancer 55:7–14 comes of age. Nature 480:480–489 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Pathology & Oncology Research Springer Journals

Immunohistochemical Analysis of Foxp3+, CD4+, CD8+ Cell Infiltrates and PD-L1 in Oral Squamous Cell Carcinoma

Free
9 pages

Loading next page...
 
/lp/springer_journal/immunohistochemical-analysis-of-foxp3-cd4-cd8-cell-infiltrates-and-pd-3aS0KvZtlK
Publisher
Springer Journals
Copyright
Copyright © 2017 by The Author(s)
Subject
Biomedicine; Cancer Research; Oncology; Pathology; Immunology; Biomedicine, general
ISSN
1219-4956
eISSN
1532-2807
D.O.I.
10.1007/s12253-017-0270-y
Publisher site
See Article on Publisher Site

Abstract

Pathol. Oncol. Res. (2018) 24:497–505 DOI 10.1007/s12253-017-0270-y ORIGINAL ARTICLE + + + Immunohistochemical Analysis of Foxp3 ,CD4 ,CD8 Cell Infiltrates and PD-L1 in Oral Squamous Cell Carcinoma 1 1 Olga Stasikowska-Kanicka & Małgorzata Wągrowska-Danilewicz & Marian Danilewicz Received: 16 September 2016 /Accepted: 22 June 2017 /Published online: 1 July 2017 The Author(s) 2017. This article is an open access publication Abstract The immunoexpression of the PD-L1 and the num- correlation between immunoexpression of PD-L1 and the ber of immune infiltrating cells have been shown to be a sig- number of CD4+ cells in OSCCPP group. In conclusion, our nificant prognostic factors in various human cancers. findings support the hypothesis of involvement of Tregs and Immunohistochemical method was used to examine the PD-L1 in OSCC development and progression. immunoexpression of PD-L1 and number of Foxp3+, CD4+ . . . , CD8+ cells in 78 cases of oral squamous cell carcinomas Keywords Treg FoxP3 PD-L1 Oral cancer (OSCCs): with better prognosis - OSCCBP (n = 37), and with poorer prognosis - OSCCPP (n = 41), and 18 cases of normal mucosa as a control. The immunoexpression of PD-L1 and the Introduction mean number of Foxp3+ cells was significantly increased in OSCCPP group in comparison to OSCCBP and control Cancer of the oral cavity is the sixth most common malignan- groups. The mean number of CD4+ cells was significantly cy reported worldwide [1]. It is the most common cancer in increased in OSCCPP group in comparison to OSCCBP and males and the third most common cancer in females [2]. Every control groups. CD8+ cells were significantly more numerous year, an estimated 3 million new cases occur worldwide, and in OSCCBP group in comparison to OSCCPP and control the overall 5-year survival rate for oral squamous cell carci- group. In both OSCCPP and OSCCBP groups there were noma (OSCC) is only 50%. Chronic inflammation, coupled positive significant correlations between number of Foxp3+ with alcohol, betel quid and cigarette consumption is associ- and CD4+ cells. We found positive correlations between the ated with oral squamous cell carcinoma [3]. immunoexpression of PD-L1 and numbers of Foxp3+ cells, Regulatory T cells (Treg) consist of functionally diverse and negative correlation between the immunoexpression of subsets of immunosuppressive T cells that play a crucial role PD-L1 and numbers of CD8+ cells in both OSCCPP and in the modulation of immune responses and the reduction of OSCCBP groups. We found also significant positive deleterious immune activation [4]. Several subsets of Treg cells have been identified and characterized, such as CD8 Treg cells, CD4 Treg cells, and γδ-TCR [5]. Tregs can be divided * Olga Stasikowska-Kanicka into two subpopulations: natural occurring (nTreg) which de- olga.stasikowska@umed.lodz.pl velop in the thymus, and adaptive (iTreg) generated in the periphery from CD4 naive T cells [6]. Tregs are capable of Małgorzata Wągrowska-Danilewicz migrating to inflammation sites and suppressing a broad range malgorzata.wagrowska@umed.lodz.pl of effector lymphocytes, particularly helper T (Th) cell subsets, Marian Danilewicz such as Th1, Th2, Th17, and follicular Th (Tfh) cells [7, 8]. marian.danilewicz@umed.lodz.pl Until now, Foxp3 (forkhead box P3) has been the most specific marker distinguishing Treg cells from T cells. Foxp3 is a mem- Department of Nephropathology, Medical University of Lodz, ul. Czechoslowacka 8/10, 92-216 Lodz, Poland ber of the forkhead/winged-helix family of transcription factors that are critically involved in the development and function of Department of Pathomorphology, Medical University of Lodz, Lodz, Poland Tregs [9]. The lack of Tregs due to the loss of Foxp3 function 498 Stasikowska-Kanicka O. et al. leads to autoimmune diseases whereas high prevalence of basis of their PD-L1 status and the presence or absence of Tregs in the peripheral blood due to the over-expression of tumor-infiltrating lymphocytes (TILs). These include type I + pos Foxp3 causes immunodeficiency [10]. Foxp3 Tregs can sup- (PD-L1 with TILs driving adaptive immune resistance), type press the activation, proliferation, and effector functions of nu- II (PD-L1 negative with no TIL indicating immune ignorance), + + pos merous cell types, including CD4 ,CD8 Tcells, dendritic type III (PD-L1 with no TIL indicating intrinsic induction) cells (DCs), B cells, and natural killer (NK) cells [11]. and type IV (PD-L1 negative with TIL indicating the role of Emerging evidence supports the notion that Treg popula- other suppressor(s) in promoting immune tolerance) [25]. tion plays a critical role in the suppression of anti-tumor im- Therefore, the objectives of this study were to evaluate the + + mune response and thus contributes to cancer progression. immunoexpression of PD-L1 and the number of CD4 ,CD8 , Tregs can mediate peripheral tolerance by suppressing self- Foxp3 cells in oral cancers. Another purpose was to find + + antigen reactive T cells [12]. Since most tumor antigens are possible association between number of Foxp3 ,CD4 , self-antigens, the suppression of tumor antigen reactive T lym- CD8 cells and immunoexpression of PD-L1. phocytes by Tregs is an important obstacle in antitumor im- munity [13]. Tumor cells can secrete soluble factors that pro- mote the induction, expansion, and recruitment of Treg cells Material and Methods to the tumor microenvironment. Therefore, tumors contribute to the generation and expansion of Treg cells in the tumor Patients microenvironment [6]. There are three ways of Tregs accumu- lation within the tumor microenvironment: increased migra- Seventy eight formalin-fixed, paraffin-embedded tissue spec- tion, preferential Treg cells expansion, and de novo conver- imens of oral squamous cell carcinomas (OSCC), and eigh- sion of Foxp3 T cells into Treg cells [14]. teen control cases (normal mucosa, non-cancer affected pa- The programmed death-ligand 1 (PD-L1) belongs to the B7 tients) were retrieved from archival material (Chair of superfamily, which also includes B7–1 (CD80), B7–2 Pathomorphology, Medical University of Lodz, Poland). (CD86), B7-DC (PDL2), B7-H2, B7-H3, B7-H4 and B7-H6 Paraffin-embedded tissue sections taken from postoperative [15]. PD-L1 has two known receptors, programmed death-1 material were diagnosed using a standard haematoxylin and (PD-1) and B7–1(CD80)[16]. PD-1 as a dominant receptor, eosin staining and the histological diagnoses were established belongs to the CD28 family and is expressed on T cells, den- according to the current standards [26]. The main criteria for dritic cells, natural killer cells, macrophages and B cells [15]. patients selection was the same anatomical localization of le- PD-L1 is constitutively expressed on murine T, B cells, DCs, sions (the floor of the mouth). To find the possible relationship macrophages, mesenchymal stem cells and cultured bone between the studied markers and clinical prognosis, patients marrow-derived mast cells [17]. PD-L1 is also expressed on with OSCC were additionally divided into two groups: with non-hematopoietic cells, including epithelial, vascular endo- better prognosis – OSCCBP (without metastases, n =37), and thelial, muscle cells, hepatocyte, pancreatic and astrocyte with poorer prognosis – OSCCPP, (with metastases to region- cells, in addition to its expression in the eye [17], lung, kidney, al lymph nodes or/and with distant metastases, n =41). The spleen, thymus, placenta, and the heart [18]. histopathological grade was classified into groups according PD-L1 is both inducible and constitutively expressed on cells to the WHO classification (for OSCCBP: G1 n =3, G2 n =33, of many solid and hematologic malignancies [19–23]. The ab- G3 n = 1, and for OSCCPP: G1 n =0, G2 n =36, G3 n =5). normal expression of PD-L1 has been linked with prognosis and The age range for OSCCBP group was from 28 to 75 years treatment response in multiple malignancies. An overexpression (mean ± SD = 59,24 ± 10, 89), for OSCCPP group was from of PD-L1 has been observed in different solid tumors including 40 to 84 (mean ± SD = 59,39 ± 11,16) and for control cases 15 melanoma [19], colorectal cancer [20], lung cancer [21], pancre- to 74 (mean ± SD = 47,05 ± 18,71). atic carcinoma [22] and hepatocellular carcinoma [23]. Mounting evidence suggests that the PD1:PD-L1 pathway Immunohistochemistry may play a central role in antigen-specific T cell response mediating PD-1-dependent immune suppression. When PD- Paraffin-embedded, 3-μm tissue sections were mounted onto 1 interacts with cells bearing one of its ligands, which can be SuperFrost slides (SuperFrost Plus, Gerhord Menzel GmbH, highly expressed on cancer cells, the ability of T cells to target Braunschweig, Germany), deparaffinized in xylene and ethanol the tumor cells can be effectively subverted [24]. Tumors can of graded concentrations. For antigen retrieval, the slides were thereby employ the PD-1:PD-L1 inhibitory pathway to silence treated in a microwave oven in a solution of TRS (Target the immune system. Thus, interrupting this interaction can Retrieval Solution, High pH, Dako, Denmark) for 30 min improve the ability of T cells to attack tumor cells. (2 × 6 minutes 360 W, 2 × 5 180 W, 2 × 4 minutes 90 W). Tumors escape immune surveillance by a number of mech- After cooling down at room temperature, they were transferred anisms of which four groups have now been proposed on the to 0,3% hydrogen peroxide in methanol, for 30 min, to block + FoxP3 Cells and PD-L1 in Oral Squamous Carcinoma 499 endogenous peroxidase activities. Sections were rinsed with Results Tris-buffered saline (TBS, Dako, Denmark) and incubated from 30 to 60 min with monoclonal mouse primary antibodies The cytoplasmic, perinuclear and nuclear pattern of Foxp3 against: CD4 (Dako; clone: 4B12, dilution 1:40), CD8 (Dako; immunoexpression on tumor infiltrating cells was seen in clone: C8/144B, dilution 1:50), Foxp3 (Abcam; clone: all OSCC cases and 9 control cases. Tumor cells were de- 236A/E7, dilution 1:50), and rabbit polyclonal antibody against void of Foxp3 staining. CD4 and CD8 were clearly stained PD-L1 (Abcam; dilution 1:400). Immunoreactive proteins were in the cell membrane of infiltrating cells of control and visualized using adequate EnVision-HRP kit (Dako, OSCC cases. The immunoexpression of PD-L1 in cancer Carpinteria, CA, USA) according to the instructions of the cells was predominantly cytoplasmic although membra- manufacturer. Visualisation was performed by incubation the nous expression was also noted. Membranous PD-L1 sections in a solution of 3,3′-diaminobenzidine (Dako, immunoexpression was predominantly detected on infiltrat- Denmark). After washing, the sections were counterstained ingcells.Inour studyPD-L1wasexpressedoncancercells with Mayer’s hematoxylin and mounted. For each antibody and tumor-infiltrating lymphocytes as well as epithelial, and for each sample a negative control was processed. vascular endothelial and infiltrating cells of control cases. Negative controls were carried out by incubation in the absence The cytoplasmic immunoexpression of PD-L1 on cancer of the primary antibody and always yielded negative results. cells was noted in 37 of 41 cases of SCCPP group, in 25 In each specimen distribution and cytoplasmic staining in- of 37 cases of SCCBP group, and in 16 of 18 control cases. tensity of PD-L1 in cancer cells were recorded semiquantita- The semiquantitative and quantitative data on the + + + tively by two independent observers in 7–10 (depending on immunoexpression of PD-L1, Foxp3 ,CD4 and CD8 cells the specimen size) adjacent high power fields and graded from appear in Table 1. The immunoexpression of PD-L1 and the 0 (staining not detectable), 1 (weak immunostaining), 2 (mod- mean number of Foxp3 cells was significantly increased in erate immunostaining intensity) and 3 (strong staining). The OSCCPP group (Figs. 1 and 2), in comparison to OSCCBP mean grade was calculated by averaging grades assigned by (Figs. 3 and 4) and control groups (Figs. 5 and 6). We also the two authors and approximating the arithmetical mean to found significantly higher immunoexpression of PD-L1 and the nearest unity. number of Foxp3 cells in OSCCBP compared to control group. The mean number of CD4 cells was significantly in- creased in OSCCPP group (Fig. 7) in comparison to both Morphometry OSCCBP (Fig. 8) and control groups (Fig. 9). CD8 cells were significantly more frequent in OSCCBP group (Fig. 10)in + + + Foxp3 ,CD4 and CD8 cells were evaluated using computer comparison to OSCCPP (Fig. 11) and control group (Fig. 12). image analysis system consisting of a PC computer equipped In both OSCCPP and OSCCBP groups there were positive with a Pentagram graphic tablet, Indeo Fast card (frame grab- significant correlations between the number of Foxp3 and ber, true-color, real-time), produced by Indeo (Taiwan), and CD4 cells, whereas the correlations between the number of + + color TV camera Panasonic (Japan) coupled with Carl Zeiss Foxp3 and CD8 cells were not statistically significant microscope (Germany). This system was programmed (Table 2). The correlative study revealed in both OSCCPP (MultiScan 18.03 software, produced by Computer Scanning and OSCCBP groups, positive correlations between the Systems, Poland) to calculate the number of objects (semiau- immunoexpression of PD-L1 and numbers of Foxp3 cells, tomatic function). and negative correlation between the immunoexpression of + + + + The number of Foxp3 ,CD4 as well as CD8 cells was PD-L1 and numbers of CD8 cells. We found also significant estimated by counting all positive cells in 7–10 high power positive correlation between immunoexpression of PD-L1 and 2 + monitor fields (HPF) (0.029 mm each), marking the number of CD4 cells in OSCCPP group (Table 3). immunopositive cells (semiautomatic function). The results In control group all these correlations were weak and not were presented as a number of positive cells per HPF. significant (data not shown). Statistical Methods Discussion Differences between groups were tested using unpaired There is accumulating evidence that head and neck squamous Student’s t-test preceded by evaluation of normality and cell carcinoma (HNSCC) patients display increased levels of Levene’s test. The Mann-Whitney U test was used where ap- nTreg cells with greater suppressive activity, compared to propriate. Correlation coefficients were calculated using healthy controls [27, 28]. However, while some studies have Spearman’s method. Results were considered statistically sig- linked higher Treg cells levels to worse clinical outcome in nificant if p <0.05. HNSCC [27], others have provided conflicting results [28]. 500 Stasikowska-Kanicka O. et al. Table 1 The immunoexpression Groups PD-L1 (mean score) Foxp3 Cells/HPF CD4+ Cells/HPF CD8+ Cells/HPF of PD-L1, Foxp3, lymphocytes CD4+ and CD8+ in oral OSCCPP (n = 41) 2.33 ± 2.02 13.4 ± 10.2 18.6 ± 12.3 9.7 ± 2.9 squamous cell carcinomas with poorer prognosis (OSCCPP), in OSCCBP (n = 37) 1.32 ± 1.26 7.7 ± 6.2 12.2 ± 7.1 17.8 ± 11.1 oral squamous cell carcinomas Controls (n = 18) 0.63 ± 0.62 2.4 ± 1.8 9.5 ± 5.5 8.2 ± 4.3 with better prognosis (OSCCBP) OSCCPP vs OSCCBP p <0.011 p <0.005 p <0.007 p <0.001 and controls OSCCPP vs control p <0.002 p <0.001 p <0.004 p = 0.12 (NS) OSCCBP vs control p <0.033 p <0.05 p =0.16 (NS) p <0.001 NS not significant Foxp3 is known as the most specific marker distinguishing increased number of Foxp3 cells in the OSCCPP group com- Treg cells from T cells, and in our study Foxp3 was expressed pared to OSCCBP. An association between high intratumoral on tumor-infiltrating lymphocytes - tumor cells were entirely density of Foxp3 cells and poorer clinical prognosis can sug- negative. In contrary to above-mentioned results, Liang et al. gest that the presence of Foxp3 cells might play a role in [29] observed in tongue cancer, that Foxp3 can be expressed OSCC progression. Increased Foxp3 Treg infiltration has been by both tumor cells and tumor-infiltrating lymphocytes and known to be associated with worse clinical outcomes and that tumor cells were the major cell types expressing Foxp3 various poor prognostic factors in many cancers [33–35]. (59,3% of tongue squamous cell carcinomas). Similar positive Suzuki et al. [36] found in colorectal cancer that the number score for Foxp3 immunoexpression was observed in pancre- of intratumoral Foxp3 cells was positively associated with atic cancer cells (61%) [30], and breast cancer tissues (57% lymph node metastases. Furthermore, it has been reported that and 73%) [31]. Subcellular staining of Foxp3 was heteroge- high numbers of circulating Tregs are associated with rapid neous in the present study. Different expression level and tumor progression in experimental animal models of melano- complex post-translational modification of Foxp3 may be pos- ma and in patients with melanoma. In these patients, the pres- sible reasons. Chen et al. [32] demonstrated that in Tregs, ence of Foxp3 cells in primary tumor has also been associat- TCR-mediated post-translational modifications could mediate ed with a higher frequency of metastases in the sentinel lymph the regulation function, and influence the subcellular distribu- node [37]. Even so, the association of Foxp3 expression and tion of Foxp3. Authors revealed a change in the subcellular its impact on overall survival remains controversial. Kim et al. localization of Foxp3 from a more cytoplasmic/perinuclear to [38] observed that Foxp3 expression in tumor cells of colo- a nuclear expression pattern in Tregs activated with anti-CD3/ rectal cancer, but not in infiltrating Treg cells, were correlated anti-CD28 antibodies. with disease progression and poor prognosis. Moreover, liter- In accordance with previous studies in various malignant ature data demonstrate that tumor infiltration by Foxp3 Tregs diseases [33, 34], we found significantly higher numbers of is not always associated with a poor prognosis, but, on the infiltrating Foxp3 cells in both tested groups of OSCC com- contrary, can be associated with an improved prognosis in pared with controls. Additionally, we found significantly some cancer types. In colorectal cancer, high levels of Fig. 1 Cytoplasmic immunoexpression of PD-L1 in oral squamous cell Fig. 2 Nuclear and perinuclear immunoexpression of Foxp3 in oral carcinomas with poorer prognosis (OSCCPP). Immunohistochemistry. squamous cell carcinomas with poorer prognosis (OSCCPP). Total magnification × 100 Immunohistochemistry. Total magnification × 100 + FoxP3 Cells and PD-L1 in Oral Squamous Carcinoma 501 Fig. 3 Cytoplasmic immunoexpression of PD-L1 in oral squamous cell Fig. 5 Cytoplasmic immunoexpression of PD-L1 in control. carcinomas with better prognosis (OSCCBP). Immunohistochemistry. Immunohistochemistry. Total magnification × 100 Total magnification × 100 intratumoral CD4 cells in the group of OSCCPP compared infiltrating Treg cells were associated with early stage disease to OSCCBP and controls. An association between high and improved prognosis [39, 40]. intratumoral density of CD4 cells and poorer clinical prog- Tumor-derived CD4 Treg cells have been extensively nosis seems to be consistent with other findings [42]. In breast + + + studied in many different types of cancer. It has been strongly cancer, the frequency of CD4 CD25 FoxP3 regulatory T suggested that antigen-specific CD4 Tregs at tumor sites may cells was inversely correlated with clinical outcomes [42]. In significantly suppress immune responses, leading to immune a mouse model of human breast cancer, the depletion of + + + + tolerance of tumor cells. Among various CD4 T cell frac- CD4 CD25 T cells was shown to reduce CD4 CD25 T + + + tions, a particular subset with CD4 CD25 Foxp3 expression cell-mediated suppression, improve immunity, and enhance was previously described as regulatory T cells and was shown tumor regression [43]. to mediate suppression [41]. Increased number of CD4 cells We found significantly increased number of CD8 cells in and the positive correlation between number of Foxp3 and OSCCBP group in comparison to OSCCPP patients and con- CD4 cells was observed in both tested groups of OSCC, trols. These results seem to be consistent with other findings. suggesting that increased immune infiltration is associated Zhu et al. [44] observed that breast cancer patient survival was with an increased frequency of Treg cells within the infiltrate. associated with higher frequencies of CD8 cytotoxic T cells in Based on these data, we hypothesize also that the growth of infiltrating lymphocytes. Emerging evidence from clinical stud- OSCCs may induce the generation of CD4 Treg cells. ies emphasizes the role of CD8 T cells in the control of tumor Moreover, we described significantly increased number of growth and the prolongation of patient survival [45, 46]. Lack Fig. 4 Nuclear and perinuclear immunoexpression of Foxp3 in oral squamous cell carcinomas with better prognosis (OSCCBP). Fig. 6 Nuclear and perinuclear immunoexpression of Foxp3 in control. Immunohistochemistry. Total magnification × 100 Immunohistochemistry. Total magnification × 100 502 Stasikowska-Kanicka O. et al. Fig. 7 Membranous immunoexpression of CD4 in oral squamous cell Fig. 9 Membranous immunoexpressionof CD4incontrol. carcinomas with poorer prognosis (OSCCPP). Immunohistochemistry. Immunohistochemistry. Total magnification × 100 Total magnification × 100 of significant correlations between number of Foxp3 and cells and epithelial and vascular endothelial cells in control CD8 cells in both studied groups of cancers suggest that cases. Lyford-Pike et al. [49] suggest that in normal tissue, CD8 cells observed in our study may have other than regula- PD-L1 is induced in response to inflammatory cytokines such tory functions or represent non-traditional subpopulation of as IFN-γ. This system represents a major mechanism for tis- Tregs. Recent studies demonstrated that Foxp3 T cells are sue protection in the setting of T cell-mediated inflammation. heterogeneous with respect to phenotype, gene expression, It is well established that PD-L1 expression is up-regulated in and function, including suppressive and non-suppressive sub- solid tumors where it can provide direct tumor protection, and populations [47]. Miyara et al. [47] divided human Foxp3 + reduce activity of PD-1 expressing tumor-infiltrating effector cells into three functional subpopulations: effector, resting and CD4 and CD8 T cells [24, 50]. We observed a significantly non-suppressive cytokine-secreting Tregs. increased immonoexpression of PD-L1 in both tested groups In recent years, many studies have confirmed that cancer of OSCC compared to controls. Overexpression of PD-L1 has cells can evade host immune systems by expressing certain been identified in several cancers, including the head and neck ligands that down-regulate cytotoxic T lymphocytes through cancers [24, 48]. We found significantly higher immunoexpression of PD-L1 in OSCCPP compared to inhibitory pathways that are usually initiated by ligand- receptor interactions [48]. Currently, PD-1:PD-L1 pathway OSCCBP. Our results are in concordance with literature data seems to be a one of major mechanism of controlling tumor [48, 51]. Thompson et al. [51] reported statistically significant immunity. In our study PD-L1 was expressed on infiltrating association of PD-L1 expression with poor clinical outcome in Fig. 8 Membranous immunoexpression of CD4 in oral squamous cell Fig. 10 Membranous immunoexpression of CD8 in oral squamous cell carcinomas with better prognosis (OSCCBP). Immunohistochemistry. carcinomas with better prognosis (OSCCBP). Immunohistochemistry. Total magnification × 100 Total magnification × 100 + FoxP3 Cells and PD-L1 in Oral Squamous Carcinoma 503 Table 2 The correlations between mean number of Foxp3+ and CD4+, CD8+ cells in oral squamous cell carcinomas with poorer prognosis (OSCCPP), and oral squamous cell carcinomas with better prognosis (OSCCBP) Correlations OSCCPP (n = 41) OSCCBP (n =37) Foxp3+ vs r =0. 44, p <0.005 r =0.35, p < 0.04 CD4+ cells Foxp3+ vs r = − 0.21, p =0.18 (NS) r = − 0.27, p = 0.11(NS) CD8+ cells NS not significant infiltrating CD8 cells in both tested groups of OSCC. In contrary to our results, Thompson demonstrated that gastric cancer patients with higher CD8 T cell densities also have Fig. 11 Membranous immunoexpression of CD8 in oral squamous cell higher PD-L1 expression, indicating an adaptive immune re- carcinomas with poorer prognosis (OSCCPP). Immunohistochemistry. sistance mechanism may be occurring. Lyford-Pike et al. [49] Total magnification × 100 using quantitative RT-PCR found a significant increase in the expression of CD8 mRNA in PD-L1(+) as compared to PD- gastric cancer. Lin et al. [52] indicated that a higher PD-L1 L1(−) in oropharyngael cancer. On the other hand, Tokito expression level was correlated with several clinicopatholog- et al. [53] observed that lack of PD-L1 immunoexpression ical factors, such as distant metastasis. These authors sug- accompanied by increased CD8 cells density was significant- gested also that PD-L1 immunoexpression might be associat- ly associated with favourable survival in non-small cell lung ed with oral cancer development and progression. We found cancer. Increased number of CD8 cells in OSCCBP group, significant positive correlation between immunoexpression of lack of correlation of PD-L1 with number of Foxp3 cells and PD-L1 in tumor cells and the number of infiltrating Foxp3 negative correlation between number of CD8 cells and cells in both tested groups of OSCC and between PD-L1 and immunoexpression of PD-L1 seem to be consistent and sug- number of CD4 cells in OSCCPP group. Similar to melano- gest that tumor infiltrating CD8 cells may have other than ma [19], and in keeping with the proposed adaptive resistance suppressive function. hypothesis, in our study PD-L1 was not expressed uniformly Literature data and our results revealed the complicated within OSCCs, but rather at sites of lymphocyte infiltration. interactions within the tumor microenvironment and empha- Our results suggest that PD-L1 immunoexpression on cancer size that impact of individual types of immune cells may be cells is associated with Treg infiltration, and PD-L1 may be highly dependent on many factors. We hypothesize that the induced by an inflammatory microenvironment involving microenvironment of a tumor is critically important in deter- TILs. mining leukocyte phenotype and function. We speculate also We found significant negative correlation between that not only the number, type and localization of tumor infil- immunoexpression of PD-L1 on cancer cells and number of trating lymphocytes but activity (e.g. cytokine releasing pat- terns) of particular infiltrating cells can determine mutual re- lationship and has prognostic value. Although our findings support the hypothesis of involve- ment of Tregs and PD-L1 in OSCC development and progres- sion, further studies of the relationship between number and Table 3 The correlations between the immunoexpression of PD-L1 and mean number of Foxp3+, CD4+, CD8+ cells in oral squamous cell carcinomas with poorer prognosis (OSCCPP), and oral squamous cell carcinomas with better prognosis (OSCCBP) Correlation between OSCCPP (n =41) OSCCBP (n =37) PD-L1 vs Foxp3+ cells r =0.42, p <0.007 r =035, p <0.04 PD-L1 vs CD4+ cells r =0.31, p <0.002 r =0.25, p = 0.14 (NS) PD-L1 vs CD8+ cells r = − 0.47, p <0.05 r = − 0.33, p <0.05 Fig. 12 Membranous immunoexpression of CD8 in control. Immunohistochemistry. Total magnification × 100 NS not significant 504 Stasikowska-Kanicka O. et al. 17. Sharpe AH, Wherry EJ, Ahmed R et al (2007) The function of activity of immune infiltration cells and immunoexpression of programmed cell death 1 and its ligands in regulating autoimmunity PD-L1 on cancer cells are needed to better understand their and infection. Nat Immunol 8:239–245 role in oral carcinogenesis. 18. Freeman GJ, Long AJ, Iwai Yet al (2000) Engagement of the PD-1 immunoinhibitory receptor by a novel B7 family member leads to negative regulation of lymphocyte activation. J Exp Med 192: Acknowledgments This work was supported by grant of Medical 1027–1034 University of Lodz 503/6-038-01/503-61-002. 19. Hino R, Kabashima K, Kato Yet al (2010) Tumor cell expression of programmed cell death-1 ligand 1 is a prognostic factor for malig- nant melanoma. Cancer 116(7):1757–1766 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http:// 20. Shi SJ, Wang LJ, Wang GD et al (2013) B7-H1 expression is asso- creativecommons.org/licenses/by/4.0/), which permits unrestricted use, ciated with poor prognosis in colorectal carcinoma and regulates the distribution, and reproduction in any medium, provided you give appro- proliferation and invasion of HCT116 colorectal cancer cells. PLoS priate credit to the original author(s) and the source, provide a link to the One 8(10):e76012 Creative Commons license, and indicate if changes were made. 21. Velcheti V, Schalper KA, Carvajal DE et al (2014) Programmed death ligand-1 expression in non-small cell lung cancer. Lab Investig 94(1):107–116 22. Chen XL, Yuan SX, Chen C et al (2009) Expression of B7-H1 References protein in human pancreatic carcinoma tissues and its clinical sig- nificance. Chin J Cancer 28(12):1328–1332 1. Siegel R, Ma J, Zou Z et al (2014) Cancer statistics 2014. CA 23. Gao Q, Wang XY, Qiu SJ et al (2009) Overexpression of PD-L1 Cancer J Clin 64:9–29 significantly associates with tumor aggressiveness and postopera- tive recurrence in human hepatocellular carcinoma. Clin Cancer 2. Byakodi R, Byakodi S, Hiremath S et al (2012) Oral cancer in Res 15(3):971–979 India: an epidemiologic and clinical review. J Community Health 37:316–319 24. Azuma T, Yao S, Zhu G et al (2008) H1 is a ubiquitous antiapoptotic receptor on cancer cells. Blood 111(7):3635–3643 3. Scully C, Bagan J (2009) Oral squamous cell carcinoma overview. Oral Oncol 45:301–308 25. Teng MW, Ngiow SF, Ribas A et al (2015) Classifying cancers based on T-cell infiltration and PD-L1. Cancer Res 75:2139–2145 4. Facciabene A, Motz GT, Coukos G (2012) T-regulatory cells: key players in tumor immune escape and angiogenesis. Cancer Res 72: 26. Barnes L, Everson JW, Reichart P, et al (2005) World Health 2162–2171 Organization classification of Tumours. Pathology and Genetics Head and Neck Tumours IARC Press Lyon 168–176 5. Wang RF (2008) CD8 regulatory T cells, their suppressive mech- anisms, and regulation in cancer. Hum Immunol 69:811–814 27. Drennan S, Stafford ND, Greenman J et al (2013) Increased fre- quency and suppressive activity of CD127 Treg cells in the periph- 6. Zhang S, Ke X, Zeng S, et al (2015) Analysis of CD8+ Treg cells in eral circulation of patients with head and neck squamous cell car- patients with ovarian cancer: a possible mechanism for immune cinoma are associated with advanced stage and nodal involvement. impairment. Cell Mol Immunol Sep;12(5):580–591 Immunology. doi:10.1111/imm.12144 7. Chaudhry A, Rudra D, Treuting P et al (2009) CD4 regulatory T 28. Schott AK, Pries R, Wollenberg B (2010) Permanent up-regulation cells control TH17 responses in a Stat3-dependent manner. Science of regulatory T-lymphocytes in patients with head and neck cancer. 326(5955):98691 Int J Mol Med 26:67–75 8. Linterman MA, Pierson W, Lee SK et al (2011) Foxp3 follicular 29. Liang YJ, Liu HC, Su YX et al (2011) Foxp3 expressed by tongue regulatory T cells control the germinal center response. Nat Med squamous cell carcinoma cells correlates with clinicopathologic 17(8):975–982 features and overall survival in tongue squamous cell carcinoma 9. Sakaguchi S, Miyara M, Costantino CM et al (2010) FOXP3+ patients. Oral Oncol 47(7):566–570 regulatory T cells in the human immune system. Nat Rev 30. Hinz S, Pagerols-Raluy L, Oberg HH et al (2007) Foxp3 expression Immunol 10:490–500 in pancreatic carcinoma cells as a novel mechanism of immune 10. Hori S, Nomura T, Sakaguchi S (2003) Control of regulatory T evasion in cancer. Cancer Res 67(17):8344–8350 cell development by the transcription factor Foxp3. Science 31. Merlo A, Casalini P, Carcangiu ML et al (2009) FOXP3 expression 299:1057–1061 and overall survival in breast cancer. J Clin Oncol 27(11):1746–1752 11. Von Boehmer H (2005) Mechanisms of suppression by suppressor 32. Chen CX, Rowell WA, Thomas RM et al (2006) Transcriptional reg- T cells. Nat Immunol 6(4):338–344 ulation by Foxp3 is associated with direct promoter occupancy and 12. DiPaolo RJ, Glass DD, Bijwaard KE et al (2005) CD4+CD25+ T modulation of histone acetylation. J Biol Chem 281:36828–36834 cells prevent the development of organ-specific autoimmune dis- 33. Hiraoka N, Onozato K, Kosuge T et al (2006) Prevalence of ease by inhibiting the differentiation of autoreactive effector T cells. FOXP3+ regulatory T cells increases during the progression of J Immunol 175:7135–7142 pancreatic ductal adenocarcinoma and its premalignant lesions. 13. Sakaguchi S (2005) Naturally arising Foxp3-expressing CD25+ Clin Cancer Res 12:5423–5434 CD4+ regulatory T cells in immunological tolerance to self and 34. Kobayashi N, Hiraoka N, Yamagami W et al (2007) FOXP3+ reg- non-self. Nat Immunol 6:345–352 ulatory T cells affect the development and progression of 14. Oleinika K, Nibbs RJ, Graham GJ et al (2013) Suppression, sub- hepatocarcinogenesis. Clin Cancer Res 13:902–911 version and escape: the role of regulatory T cells in cancer progres- 35. Bohling SD, Allison KH (2008) Immunosuppressive regulatory T sion. Clin Exp Immunol 171:36–45 cells are associated with aggressive breast cancer phenotypes: a 15. Chen L (2004) Co-inhibitory molecules of the B7-CD28 family in potential therapeutic target. Mod Pathol 21:1527–1532 the control of T-cell immunity. Nat Rev Immunol 4(5):336–347 36. Suzuki H, Chikazawa N, Tasaka Tet al (2010) Intratumoral CD8(+) T/ 16. Park JJ, Omiya R, Matsumura Y et al (2010) B7–H1/CD80 inter- FOXP3 (+) cell ratio is a predictive marker for survival in patients with action is required for the induction and maintenance of peripheral T- colorectal cancer. Cancer Immunol Immunother 59(5):653–661 cell tolerance. Blood 116(8):1291–1298 + FoxP3 Cells and PD-L1 in Oral Squamous Carcinoma 505 37. Ma MW, Medicherla RC, Qian M et al (2012) Immune response in 46. Dhodapkar MV, Sznol M, Zhao B et al (2014) Induction of antigen-specific immunity with a vaccine targeting NY-ESO- melanoma: an in-depth analysis of the primary tumor and corre- sponding sentinel lymph node. Mod Pathol 25:1000–1010 1 to the dendritic cell receptor DEC-205. Sci Transl Med 6: 38. Kim M, Grimmig T, Grimm M et al (2013) Expression of Foxp3 in 232–251 colorectal cancer but not in Treg cells correlates with disease pro- 47. Miyara M, Sakaguchi S (2011) Human FoxP3(+)CD4(+) regulatory T gression in patients with colorectal cancer. PLoS One 8(1):e53630 cells: their knowns and unknowns. Immunol Cell Biol 89(3):346–351 39. Salama P, Phillips M, Grieu F et al (2009) Tumor-infiltrating 48. Ohigashi Y, Sho M, Yamada Y et al (2005) Clinical significance of FOXP3+ T regulatory cells show strong prognostic significance programmed death-1 ligand-1 and programmed death-1 ligand-2 ex- in colorectal cancer. J Clin Oncol 27(2):186–192 pression in human esophageal cancer. Clin Cancer Res 11:2947–2953 40. Correale P, Rotundo MS, Del Vecchio MT et al (2010) Regulatory 49. Lyford-Pike S, Peng S, Young GD, et al (2013) Evidence for a role (FoxP3+) T-cell tumor infiltration is a favorable prognostic factor in of the PD-1:PD-L1 pathway in immune resistance of HPV- advanced colon cancer patients undergoing chemo or associated head and neck squamous cell carcinoma. Cancer Res chemoimmunotherapy. J Immunother 33:435–441 15;73(6):1733-1741 41. Viglietta V, Baecher-Allan C, Weiner HL et al (2004) Loss of func- 50. Hirano F, Kaneko K, Tamura H et al (2005) Blockade of B7–H1 tional suppression by CD4+CD25+ regulatory T cells in patients and PD-1 by monoclonal antibodies potentiates cancer therapeutic with multiple sclerosis. J Exp Med 199:971–979 immunity. Cancer Res 65(3):1089–1096 42. Watanabe MAE, Oda JMM, Amarante MK et al (2010) Regulatory 51. Thompson ED, Zahurak M, Murphy A et al (2016) Patterns of PD-L1 T cells and breast cancer: implications for immunopathogenesis. expression and CD8 T cell infiltration in gastric adenocarcinomas and Cancer Metastasis Rev 29:569–579 associated immune stroma. Gut. doi:10.1136/gutjnl-2015-310839 43. Knutson KL, Dang Y, Lu H et al (2006) IL-2 immunotoxin therapy 52. Lin YM, Sung WW, Hsieh MJ, et al (2015) High PD-L1 expression modulates tumor-associated regulatory T cells and leads to lasting correlates with metastasis and poor prognosis in oral squamous cell immune-mediated rejection of breast cancers in neu-transgenic carcinoma. PLoS One 12;10(11): e0142656. doi:10.1371/journal. mice. J Immunol 177:84–91 pone.0142656 44. Zhu S, Lin J, Qiao G et al (2015) Differential regulation and func- 53. Tokito T, Azuma K, Kawahara A et al (2016) Predictive relevance tion of tumor-infiltrating T cells in different stages of breast cancer of PD-L1 expression combined with CD8+ TIL density in stage III patients. Tumour Biol 36(10):7907–7913 non-small cell lung cancer patients receiving concurrent chemora- 45. Mellman I, Coukos G, Dranoff G (2011) Cancer immunotherapy diotherapy. Eur J Cancer 55:7–14 comes of age. Nature 480:480–489

Journal

Pathology & Oncology ResearchSpringer Journals

Published: Jul 1, 2017

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off