Acta Oto-Laryngologica. 2014; 134: 859–864
ORIGINAL ARTICLE
Acta Otolaryngol Downloaded from informahealthcare.com by Central Michigan University on 01/15/15 For personal use only.
Regulatory T-cell infiltration in tongue squamous cell carcinoma
HIROYUKI HANAKAWA1, YORIHISA ORITA1, YASUHARU SATO2, MAI TAKEUCHI2, KYOTARO OHNO2, YUKA GION2, KIYOAKI TSUKAHARA3, RYO TAMAMURA4, TOSHIHIRO ITO5, HITOSHI NAGATSUKA6, KAZUNORI NISHIZAKI1 & TADASHI YOSHINO2 1
Departments of Otolaryngology Head and Neck Surgery, 2Pathology, 3Department of Otolaryngology Head and Neck Surgery, Tokyo Medical University Hachioji Medical Center, Tokyo, Japan, 4Oral Pathology and Medicine and 5 Pathology and Experimental Medicine, Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama
Abstract Conclusion: In tongue squamous cell carcinoma (SCC), high levels of regulatory T-cell (Treg) infiltration in tumor nests are observed in the cases with poor prognosis. Objectives: The role of Tregs in head and neck cancers remains unclear. The aim of this study was to observe the distribution of Tregs in different stages of tongue SCC and estimate the effects on prognosis. Methods: Thirty-four cases with tongue SCC were examined immunohistochemically for CD4, CD8, and Forkhead box P3 (Foxp3). Immunoreactive cells were counted in cancer stroma and nest regions, and relationships between cell numbers and disease-free survival rates were analyzed. Results: In the 34 cases, univariate analysis for disease-free survival indicated high-level infiltration of Tregs (CD4+Foxp3+) into both cancer nests and stroma and presence of helper T (CD4+Foxp3–) cells in cancer stroma as potential predictors of significantly worse prognosis. In early-stage cases (stage I/II), high-level infiltration of Tregs in cancer nests correlated significantly with poor disease-free survival rate. Multivariate analysis for disease-free survival found no independent variables.
Keywords: Immunohistochemical study, Foxp3, disease-free survival, tumor nest
Introduction Numerous studies have documented the recruitment of regulatory T cells (Tregs) to human carcinomas, and Treg accumulation is predictive of poor outcome in many solid tumors [1]. The ability of Tregs to suppress functions of the effector T cells responsible for antitumor responses and thus contribute to tumor progression has been evaluated in patients with various kinds of cancers [2,3]. In head and neck cancers, high levels of Tregs in the peripheral blood correlate with a higher probability of early disease recurrence [4]. However, in terms of the infiltration of Tregs into the cancer microenvironment, their role in anticancer
immune responses and their influence on patient prognosis remain unclear. For example, in ovarian cancer and hepatocarcinoma, tumor-infiltrating Tregs negatively influence patient prognosis, whereas in Hodgkin’s disease, follicular lymphoma, and colorectal cancer, as well as head and neck cancer, infiltration appears to be associated with favorable outcomes [5]. In a previous study of 84 cases of head and neck squamous cell carcinoma (SCC), including 14 carcinomas of the oral cavity, levels of tumor-infiltrating Tregs correlated positively with better control of the primary lesion [6]. Bron et al. [5] studied 35 head and neck SCCs, including 16 cases with oral cavity carcinomas, and found
Correspondence: Yorihisa Orita, MD PhD, Department of Otolaryngology, Head and Neck Surgery, Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, 2-5-1, Shikata-Cho, Kita-Ku, 700-8558, Okayama, Japan. Tel: +81 86 235 7307. Fax: +81 86 235 7308. E-mail: [email protected]
(Received 6 March 2014; accepted 6 April 2014) ISSN 0001-6489 print/ISSN 1651-2251 online 2014 Informa Healthcare DOI: 10.3109/00016489.2014.918279
Acta Otolaryngol Downloaded from informahealthcare.com by Central Michigan University on 01/15/15 For personal use only.
860
H. Hanakawa et al.
that high levels of tumor-infiltrating Tregs correlated with absence of locoregional metastases. Since the roles of Tregs may differ according to the origin of the cancer, individual studies of Treg function in each type of cancer are needed. The present study focused on tongue SCC, and to the best of our knowledge, the number of cases (n = 34) in the present study makes this the largest case series of tongue SCCs yet studied regarding Treg distribution.
Nest
Material and methods
Stroma
Patients This study examined paraffin-embedded samples of 34 primary tongue SCCs from patients initially treated between January 2003 and December 2009 at Okayama University Hospital, Okayama, Japan. Immunohistochemical staining Serial sections (4 mm) were cut from each paraffinembedded tissue block, and several sections were stained with hematoxylin and eosin. Appropriate sections underwent immunohistochemical doublestaining with either CD4 (1:40 1F6; Nichirei, Tokyo, Japan) and Forkhead box P3 (Foxp3) (1:100 236A/ E7; Abcam, Cambridge, UK), or CD8 (1:40 1A5; Nichirei) and Foxp3, using an automated Bond Max stainer (Leica Biosystems, Melbourne, Australia). Foxp3 is a protein responsible for the regulation of Treg function and development, and has been used as the best marker for detecting Tregs. Evaluation of staining Immunohistochemically stained cells were classified into two groups: (a) those showing infiltration into cancer cell nests, and (b) those showing infiltration intocancerstroma(Figure1).Cancernestsweredefined as ‘cancer nests without fibroblasts and vasculature’ and cancer stroma was defined as ‘connective tissues surrounding cancer nests without any cancer cells’ [7]. The number of immunoreactive cells in a 200 microscopic field (0.933 mm2) was then counted. Counting of immunoreactive cells was performed by investigators (H.H. and Y.O.) blinded to the clinical data of patients. Five areas showing the most abundant distribution were selected, and the mean number of immunoreactive cells was calculated for each case. CD4 and CD8 staining was observed in the cell membraneand Foxp3inthenucleioflymphocytes(Figure2). CD8+Foxp3+ cells were seldom identified, so the number of CD4+Foxp3+ cells was used as the number of Tregs in this study. The median count was used for
Figure 1. Tongue squamous cell carcinoma after immunohistochemical double-staining for CD4 and Foxp3. Cancer nests were defined as ‘cancer tissue without fibroblasts and vasculature’ and cancer stroma as ‘connective tissue surrounding cancer nests without any cancer cells.’ Magnification, 200.
the number of each kind of cell, and in cases of large discrepancies, a consensus was reached by both investigators using a multi-headed microscope. Statistical analysis Since most patients with early tongue SCC in this study remain alive as of the time of analysis, determination and comparison of disease-specific survival was difficult. The relationship between disease-free survival and variables including the number of immunoreactive cells was assessed using Cox proportional hazards regression modeling for both univariate and multivariate analyses. The chi-squared test was used to test for relationships between categorical variables. All analyses were performed using SPSS version 21.0J software (SPSS, Armonk, NY, USA) and STATA/SE 12.0J software (Stata Corp., College Station, TX, USA).
Figure 2. Tongue squamous cell carcinoma after immunohistochemical double-staining for CD4 and Foxp3. CD4 staining was observed in the cell membrane and Foxp3 in the nuclei of lymphocytes. CD4+Foxp3+ cells were regarded as regulatory T cells (arrows). Magnification, 400.
Tregs in tongue carcinoma Results
Table I. Univariate analysis of outcomes for 34 patients with tongue carcinoma: characteristics and outcomes.
Acta Otolaryngol Downloaded from informahealthcare.com by Central Michigan University on 01/15/15 For personal use only.
Overall outcomes The 34 patients with tongue SCC comprised 27 men and 7 women, with a median age at the first presentation of 62.5 years (range 26–79 years). The median duration of follow-up after detection of tongue carcinoma was 45.5 months (range 1–111 months). Union for International Cancer Control (UICC) staging classification was adopted to stage cancer, as follows: 9 cases with stage I, 11 with stage II, 3 with stage III, and 11 with stage IV. All 34 patients underwent surgery, and 10 received neoadjuvant chemotherapy (NAC) using cisplatin. Although the choice to administer NAC was made at the discretion of the attending physician, this treatment tended to be used in late T2 or stage III/IV cases. Of the 34 patients, 24 survived and 10 died of the disease during the observation period. Among the 20 patients with early (stage I/II) tongue carcinoma, 3 cases died during the observation period.
861
Characteristics
Number of patients (n = 34)
5-year disease-free survival (%)
p value (log-rank test)
27 (79%)
36.7
0.256
7 (21%)
66.7
Sex Male Female
Age (mean, 62.5 years) ‡60
21 (62%)
27.8
ORIGINAL ARTICLE
Acta Otolaryngol Downloaded from informahealthcare.com by Central Michigan University on 01/15/15 For personal use only.
Regulatory T-cell infiltration in tongue squamous cell carcinoma
HIROYUKI HANAKAWA1, YORIHISA ORITA1, YASUHARU SATO2, MAI TAKEUCHI2, KYOTARO OHNO2, YUKA GION2, KIYOAKI TSUKAHARA3, RYO TAMAMURA4, TOSHIHIRO ITO5, HITOSHI NAGATSUKA6, KAZUNORI NISHIZAKI1 & TADASHI YOSHINO2 1
Departments of Otolaryngology Head and Neck Surgery, 2Pathology, 3Department of Otolaryngology Head and Neck Surgery, Tokyo Medical University Hachioji Medical Center, Tokyo, Japan, 4Oral Pathology and Medicine and 5 Pathology and Experimental Medicine, Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama
Abstract Conclusion: In tongue squamous cell carcinoma (SCC), high levels of regulatory T-cell (Treg) infiltration in tumor nests are observed in the cases with poor prognosis. Objectives: The role of Tregs in head and neck cancers remains unclear. The aim of this study was to observe the distribution of Tregs in different stages of tongue SCC and estimate the effects on prognosis. Methods: Thirty-four cases with tongue SCC were examined immunohistochemically for CD4, CD8, and Forkhead box P3 (Foxp3). Immunoreactive cells were counted in cancer stroma and nest regions, and relationships between cell numbers and disease-free survival rates were analyzed. Results: In the 34 cases, univariate analysis for disease-free survival indicated high-level infiltration of Tregs (CD4+Foxp3+) into both cancer nests and stroma and presence of helper T (CD4+Foxp3–) cells in cancer stroma as potential predictors of significantly worse prognosis. In early-stage cases (stage I/II), high-level infiltration of Tregs in cancer nests correlated significantly with poor disease-free survival rate. Multivariate analysis for disease-free survival found no independent variables.
Keywords: Immunohistochemical study, Foxp3, disease-free survival, tumor nest
Introduction Numerous studies have documented the recruitment of regulatory T cells (Tregs) to human carcinomas, and Treg accumulation is predictive of poor outcome in many solid tumors [1]. The ability of Tregs to suppress functions of the effector T cells responsible for antitumor responses and thus contribute to tumor progression has been evaluated in patients with various kinds of cancers [2,3]. In head and neck cancers, high levels of Tregs in the peripheral blood correlate with a higher probability of early disease recurrence [4]. However, in terms of the infiltration of Tregs into the cancer microenvironment, their role in anticancer
immune responses and their influence on patient prognosis remain unclear. For example, in ovarian cancer and hepatocarcinoma, tumor-infiltrating Tregs negatively influence patient prognosis, whereas in Hodgkin’s disease, follicular lymphoma, and colorectal cancer, as well as head and neck cancer, infiltration appears to be associated with favorable outcomes [5]. In a previous study of 84 cases of head and neck squamous cell carcinoma (SCC), including 14 carcinomas of the oral cavity, levels of tumor-infiltrating Tregs correlated positively with better control of the primary lesion [6]. Bron et al. [5] studied 35 head and neck SCCs, including 16 cases with oral cavity carcinomas, and found
Correspondence: Yorihisa Orita, MD PhD, Department of Otolaryngology, Head and Neck Surgery, Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, 2-5-1, Shikata-Cho, Kita-Ku, 700-8558, Okayama, Japan. Tel: +81 86 235 7307. Fax: +81 86 235 7308. E-mail: [email protected]
(Received 6 March 2014; accepted 6 April 2014) ISSN 0001-6489 print/ISSN 1651-2251 online 2014 Informa Healthcare DOI: 10.3109/00016489.2014.918279
Acta Otolaryngol Downloaded from informahealthcare.com by Central Michigan University on 01/15/15 For personal use only.
860
H. Hanakawa et al.
that high levels of tumor-infiltrating Tregs correlated with absence of locoregional metastases. Since the roles of Tregs may differ according to the origin of the cancer, individual studies of Treg function in each type of cancer are needed. The present study focused on tongue SCC, and to the best of our knowledge, the number of cases (n = 34) in the present study makes this the largest case series of tongue SCCs yet studied regarding Treg distribution.
Nest
Material and methods
Stroma
Patients This study examined paraffin-embedded samples of 34 primary tongue SCCs from patients initially treated between January 2003 and December 2009 at Okayama University Hospital, Okayama, Japan. Immunohistochemical staining Serial sections (4 mm) were cut from each paraffinembedded tissue block, and several sections were stained with hematoxylin and eosin. Appropriate sections underwent immunohistochemical doublestaining with either CD4 (1:40 1F6; Nichirei, Tokyo, Japan) and Forkhead box P3 (Foxp3) (1:100 236A/ E7; Abcam, Cambridge, UK), or CD8 (1:40 1A5; Nichirei) and Foxp3, using an automated Bond Max stainer (Leica Biosystems, Melbourne, Australia). Foxp3 is a protein responsible for the regulation of Treg function and development, and has been used as the best marker for detecting Tregs. Evaluation of staining Immunohistochemically stained cells were classified into two groups: (a) those showing infiltration into cancer cell nests, and (b) those showing infiltration intocancerstroma(Figure1).Cancernestsweredefined as ‘cancer nests without fibroblasts and vasculature’ and cancer stroma was defined as ‘connective tissues surrounding cancer nests without any cancer cells’ [7]. The number of immunoreactive cells in a 200 microscopic field (0.933 mm2) was then counted. Counting of immunoreactive cells was performed by investigators (H.H. and Y.O.) blinded to the clinical data of patients. Five areas showing the most abundant distribution were selected, and the mean number of immunoreactive cells was calculated for each case. CD4 and CD8 staining was observed in the cell membraneand Foxp3inthenucleioflymphocytes(Figure2). CD8+Foxp3+ cells were seldom identified, so the number of CD4+Foxp3+ cells was used as the number of Tregs in this study. The median count was used for
Figure 1. Tongue squamous cell carcinoma after immunohistochemical double-staining for CD4 and Foxp3. Cancer nests were defined as ‘cancer tissue without fibroblasts and vasculature’ and cancer stroma as ‘connective tissue surrounding cancer nests without any cancer cells.’ Magnification, 200.
the number of each kind of cell, and in cases of large discrepancies, a consensus was reached by both investigators using a multi-headed microscope. Statistical analysis Since most patients with early tongue SCC in this study remain alive as of the time of analysis, determination and comparison of disease-specific survival was difficult. The relationship between disease-free survival and variables including the number of immunoreactive cells was assessed using Cox proportional hazards regression modeling for both univariate and multivariate analyses. The chi-squared test was used to test for relationships between categorical variables. All analyses were performed using SPSS version 21.0J software (SPSS, Armonk, NY, USA) and STATA/SE 12.0J software (Stata Corp., College Station, TX, USA).
Figure 2. Tongue squamous cell carcinoma after immunohistochemical double-staining for CD4 and Foxp3. CD4 staining was observed in the cell membrane and Foxp3 in the nuclei of lymphocytes. CD4+Foxp3+ cells were regarded as regulatory T cells (arrows). Magnification, 400.
Tregs in tongue carcinoma Results
Table I. Univariate analysis of outcomes for 34 patients with tongue carcinoma: characteristics and outcomes.
Acta Otolaryngol Downloaded from informahealthcare.com by Central Michigan University on 01/15/15 For personal use only.
Overall outcomes The 34 patients with tongue SCC comprised 27 men and 7 women, with a median age at the first presentation of 62.5 years (range 26–79 years). The median duration of follow-up after detection of tongue carcinoma was 45.5 months (range 1–111 months). Union for International Cancer Control (UICC) staging classification was adopted to stage cancer, as follows: 9 cases with stage I, 11 with stage II, 3 with stage III, and 11 with stage IV. All 34 patients underwent surgery, and 10 received neoadjuvant chemotherapy (NAC) using cisplatin. Although the choice to administer NAC was made at the discretion of the attending physician, this treatment tended to be used in late T2 or stage III/IV cases. Of the 34 patients, 24 survived and 10 died of the disease during the observation period. Among the 20 patients with early (stage I/II) tongue carcinoma, 3 cases died during the observation period.
861
Characteristics
Number of patients (n = 34)
5-year disease-free survival (%)
p value (log-rank test)
27 (79%)
36.7
0.256
7 (21%)
66.7
Sex Male Female
Age (mean, 62.5 years) ‡60
21 (62%)
27.8
