doi: 10.1111/jop.12381
J Oral Pathol Med © 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd wileyonlinelibrary.com/journal/jop
Global DNA hypomethylation is associated with the development and poor prognosis of tongue squamous cell carcinoma Hung-Chih Chen1,2, Cheng-Mei Yang1,2, Jiin-Tsuey Cheng3, Kuo-Wang Tsai4, Ting-Ying Fu5, Huei-Han Liou4, Hui-Hwa Tseng5, Jang-Hwa Lee5, Guan-Cheng Li4,6, Jyh-Seng Wang5,7, Yu-Yi Hou8,9, Ta-Jung Weng4,6, Luo-Ping Ger4,6 1
Department of Stomatology, Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan; 2Department of Dental Technology, Shu-Zen Junior College of Medicine and Management, Kaohsiung, Taiwan; 3Department of Biological Sciences, National Sun Yat-Sen University, Kaohsiung, Taiwan; 4Department of Medical Education and Research, Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan; 5 Department of Pathology and Laboratory Medicine, Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan; 6Institute of Biomedical Sciences, National Sun Yat-Sen University, Kaohsiung, Taiwan; 7School of Medicine, National Yang-Ming University, Taipei, Taiwan; 8 Department of Otorhinolaryngology-Head & Neck Surgery, Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan; 9Department of Nursing, Yuh-Ing Junior College of Health Care and Management, Kaohsiung, Taiwan
BACKGROUNDS: Oral cancer is the 4th leading cause of cancer death for males and the top cancer in young adult males in Taiwan. Tongue squamous cell carcinoma (TSCC) is a common oral cancer and generally associated with poor prognosis. Global DNA hypomethylation at the 5 position of cytosine (5mC) is a well-known epigenetic feature of cancer. Therefore, the purpose of this study was to investigate the relationship of the global 5mC content with the tumorigenesis and prognosis of patients with TSCC. METHODS: The levels of global 5mC were evaluated by immunohistochemistry using tissue microarray slides of 248 surgically resected TSCC and 202 corresponding tumor adjacent normal (TAN) tissues. RESULTS: We found that the level of 5mC in TSCC (P < 0.001) was significantly decreased as compared to TAN. Among TSCC tissues, decreased levels of 5mC were associated with female gender (P = 0.036). In addition, the global hypomethylation was associated with the poor disease-specific survival in TSCC patients (adjusted hazard ratio: 1.55, P = 0.043), especially for patients in older age group (> 50 years, P = 0.013), with moderate or poor cell differentiation (P = 0.044), early stage of disease (I-II, P = 0.046), small tumor size (T1–T2, P = 0.005), without lymph node involvement (P = 0.041), and ever received postoperative radiotherapy (P = 0.009).
Correspondence: Luo-Ping Ger, Department of Medical Education and Research, Kaohsiung Veterans General Hospital, Kaohsiung 813, Taiwan. Tel: +011 886 7 346 8356, Fax: +011 886 7 346 8056, E-mail: [email protected] H.-C. C. and C.-M. Y. contributed equally to this study. Accepted for publication September 28, 2015
CONCLUSIONS: Global hypomethylation was an independent biomarker for the development and poor prognosis of TSCC. J Oral Pathol Med (2015) Keywords: hypomethylation; immunohistochemistry; survival; tissue microarray; tongue squamous cell carcinoma
Introduction Oral cancer is the 6th most common malignancy worldwide (1). It is the fourth cause of cancer death for males and the top cancer in young male adults (30–50 years old) in Taiwan (2). Tongue squamous cell carcinoma (TSCC) is a common form of oral cancer and accounted for 27.3% of oral and pharyngeal cancer cases in 2010 (2). Tongue has a rich supply of lymphatic drainage and neurovascular bundles, which increases the probability of neck nodal metastasis (3). Grandi et al. (4) demonstrated that the presence of nodal metastases in patients with TSCC reduced the 5-year survival rate from 65 to 29%. Therefore, TSCC has a significantly worse prognosis than other tumors in oral cavity (5). Despite considerable advances in diagnostic and therapeutic techniques, TSCC continues to have a poor prognosis (6). Therefore, identification of biomarkers of tumorigenesis, clinical pathologic characteristics, and survival of TSCC might be helpful in early diagnosis of disease and guiding clinical decision-making in treatment. Epigenetic modification of DNA by cytosine methylation to produce 5-methylcytosine (5mC) is well-recognized as an important epigenetic process in many cancers (7–9).
Global hypomethylation and tongue cancer Chen et al.
2
Epigenetic methylation of gene promoter regions is known to silence the expression of associated genes and aberrant promoter methylation is associated with altered behavior in oncogenes and tumor suppressor genes (10). In head and neck squamous cell carcinoma (HNSCC), epigenetic silencing by promoter hypermethylation (also known as regional hypermethylation) was reported to be involved in the inactivation of genes involved in cell cycle regulation [p16INK4A (11), p14ARF (11, 12)], DNA repair {O6-methylguanine-DNA methyltransferase [MGMT] (11), FHIT (13)}, apoptosis (DAPK) (11), cell adhesion (Ecadherin [CDH1] (14)), and other cellular processes (15, 16). Despite the frequently observed cancer-associated increases of regional hypermethylation (17, 18), the prevalence of non-specific (or global) DNA hypomethylation was also found in many types of human cancer (19–22). To date, there have been limited number of experiments investigating the genomewide methylation for head and neck squamous cell carcinoma (HNSCC) (23–27) and their results are still inconclusive. In this study, we evaluated the expression of 5mC in tumor adjacent normal tissue and TSCC tissue by immunohistochemistry assay. Then, we evaluated the impact of 5mC content on the tumorigenesis and prognosis of patients with TSCC.
Materials and methods Tissue specimens A total of 248 surgically resected TSCC tissues and 202 paired tumor adjacent normal (TAN) tissues were obtained from Department of Pathology, Kaohsiung Veterans General Hospital between 1993 and 2006. Patients with base of tongue cancer or surgical margin < 2 mm were excluded in this study. The survival time was estimated from the time of operation to October 2010. Pathologic TNM classification was determined at the time of the initial resection of the tumor in accordance with the guidelines of the 2002 American Joint Committee on Cancer (AJCC) system. The study protocol was independently reviewed and approved by the Institutional Review Board of the Kaohsiung Veterans General Hospital (IRB number: VGHKS13-CT10-04). Tissue microarray construction A tissue microarray (TMA) block consisted of 129 cores, 1.5 mm in diameter, including 43 trios, with each trio containing 2 cores from the tumor tissue and 1 core from the non-cancer epithelia of the same patient. Because some of the tumor tissue samples did not include non-cancer epithelia, non-cancer cores could not be obtained for some patients; therefore, only 6 TMA blocks were constructed in all. After construction, TMA blocks were cut into 4-lm paraffin sections using standard techniques. All the representative areas of tumor and non-cancer epithelia tissues were selected by a senior oral cancer pathologist from hematoxylin–eosinstained sections for coring cylindrical tissues from paraffinembedded tissues. The histological contents of 496 tumor cores (for 248 patients) contained 484 cores (97.6%) with only tumor epithelia and 12 cores with incorrect content. Within the 12 cores, 6 (1.2%) were without any tumor epithelia or with too few tumor epithelia (< 20 cells per core), 3 (0.6%) were with only normal epithelia, and 3 (0.6%) were missing or J Oral Pathol Med
without good quality to be scored. Although these 12 cores cannot be scored, the staining score of the alternative core can represent their staining. Therefore, all 248 patients were included in this study. In addition, 235 corresponding tumor adjacent normal (TAN) cores (for 235 patients) were stained in this study. A total of 202 cores were with normal epithelia only and were scored, but 33 cores with incorrect content were not scored and excluded from this study. They were 17 cores (7.2%) without any normal epithelia or with too few normal epithelia (< 20 cells), 12 cores (n = 12, 5.1%) without any normal epithelia but tumor epithelia, and 4 cores (n = 4, 1.7%) with missing or without good quality to be scored. Immunohistochemistry Novolink max polymer detection system (Leica, Newcastle Upon Tyne, UK) was used for IHC in this study. The slides were deparaffinized in xylene and rehydrated in grade alcohols. Antigen retrieval performed by immersion in Tris-EDTA (10 mM, pH 9.0) for 10 min at 125°C in the pressure boiler. Endogenous peroxidase activity was blocked by incubation of the slides for 30 min with 3% hydrogen peroxide in methanol. After blocking at room temperature, primary antibodies were immediately applied and slides were incubated overnight at 4°C in a wet chamber. The primary antibody, mouse monoclonal anti5mC (1:500; Abcam, Cambridge, MA, USA), was used in this study. After phosphate-buffered saline washes, the slides were incubated with horseradish peroxidase-labeled secondary antibody for 10 min at room temperature and the sections were counterstained with hematoxylin. Immunohistochemical analysis and scoring At beginning, an oral cancer pathologist (Dr. Ting-Ying Fu) accompanied a senior technician (Mr. Guan-Cheng Li) to evaluate slides until all the discrepancies were resolved. Then, this technician independently reviewed all slides, except those cores with incorrect or uncertain contents, which must be scored by the pathologist. Subsequently, 15% core samples were randomly selected for re-evaluation by this pathologist. If the pathologist and technician agreement of IHC scores were < 95%, this pathologist accompanied the technician to evaluate slides again until all the discrepancies were resolved. Then, the senior technician should rereview and rescore all slides until another random core samples (5–20%) resulted in scoring agreement ≧ 95%. Around 42% core samples were scored (for those incorrect, special, or uncertain cores) or re-evaluated (for the random core samples) by pathologist. In the final reevaluation, the pathologist and technician agreement of 5mC IHC score was 100% in this study. During the evaluation process, both of them were blind of the clinical outcomes of patients. We graded the immunoreactivity by a semiquantitative approach. Marker scores for nuclear staining were calculated based on staining intensity (0, no signal; 1, mild; 2, moderate; and 3, strong; Fig. 1A–D) and percentage of tumor cells staining at each intensity level (0, 75%). The intensity score and percentage of the positive cells were then multiplied to produce the final scores. For survival analysis, the expression levels were dichotomized as low expression and high expression with the cutoff set at the
Global hypomethylation and tongue cancer Chen et al.
3
Figure 1 Immunoreactivity of 5mC in oral tongue SCC. (A) Negative (B) Weak intensity (C) Moderate intensity (D) Strong intensity.
25th percentile based on the distribution of marker score. The cutoff values were 4 for 5mC in TSCC tissues. Statistical analysis The expression differences of 5mC between the paired tissues (TSCC vs. corresponding TAN) were analyzed by the Wilcoxon matched-pairs signed-rank test, but differences between the independent tissues (TSCC vs. TAN) were evaluated by Mann–Whitney U-test. Kruskal–Wallis one-way ANOVA and Mann–Whitney U-test were used to evaluate the correlation of 5mC expression levels of TSCC tissues with the clinicopathologic parameters. In addition, overall survival was calculated from the date of initial resection of the primary tumor to the date of death from any cause or last follow-up. Disease-specific survival was measured from the time of initial resection of the primary tumor to the date of cancer-specific death or last follow-up. Disease-free survival was calculated from the date of initial resection of the primary tumor to date of recurrence or last follow-up. The cumulative survival curves were estimated using the Kaplan–Meier method. The comparison of the survival curves was performed by the log-rank test. A Cox proportional hazards model was used to determine independent predictors of survival using factors significant on univariate analysis as covariates. A value of P < 0.05 (2sided) was considered significant.
Results The content of 5mC in TSCC tissues and the paired tumor adjacent normal (TAN) tissues To assess the impact of 5mC on TSCC development, the content of 5mC in the tongue TMA containing TSCC and
TAN tissues was examined (Table 1). A total of 202 TSCC patients’ tumor and paired tumor adjacent normal (TAN) tissues as well as another 46 patients’ tumor-only tissues (without paired TAN) were assayed in TMA. As 202 TAN and 248 (202 + 46) tumor tissues were treated as two independent samples, Mann–Whitney U-test was used to evaluate the impact of 5mC. We found that the staining of 5mC was found only in cell nucleus and the levels of 5mC in tumor tissues were significantly lower than those in TAN tissues (P < 0.001). As 202 tumor and paired TAN tissues were treated as dependent samples, Wilcoxon signed-rank test was used to evaluate the pairwise comparison of 5mC. We found that the content of 5mC was also significantly reduced in the tumor tissues as compared to the paired TAN tissues (P < 0.001, Table 1). The demographic and clinicopathologic characteristics and their impact on survival of patients with TSCC A total of 248 TSCC patients were eligible and studied (Table 2). Most TSCC patients were males (88%), middleaged adults (41–60 years: 60%; mean age = 51.25 years), with moderate grade of cell differentiation (Grade 2: 83%), locally advanced stage of disease (II + III: 57%), small size of tumor (T1 + T2: 79%), and without involvement of lymph node (N0: 79%). In addition, only 66 (26.6%) and 5 (2.0%) patients of TSCC had ever received postoperative RT and CT, respectively. With a median follow-up of 48.53 (range, 1.6–236.30) months, the overall survival rate was 56.65 3.24% at 5 years, 43.80 3.75% at 10 years, and 29.38 5.53% at 15 years. The disease-specific survival rate was 60.71 3.20% at 5 years, 51.60 3.72% at 10 years, and
Table 1 The comparisons of 5mC content between two different histological groups for tongue squamous cell carcinoma patients (tongue base cases were excluded) Tumor adjacent normal Variables
Mean SD
Independent samples
(n = 8.89 (n = 8.89
Dependent samples
202) 3.07 202) 3.07
Median (IQR) 8.00 (8.00–12.00) 8.00 (8.00–12.00)
Tumor Mean SD (n = 7.46 (n = 7.57
248) 2.86 202) 2.86
Median (IQR)
Statistics
P-value
8.00 (4.00–8.00)
Z=
5.037
J Oral Pathol Med © 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd wileyonlinelibrary.com/journal/jop
Global DNA hypomethylation is associated with the development and poor prognosis of tongue squamous cell carcinoma Hung-Chih Chen1,2, Cheng-Mei Yang1,2, Jiin-Tsuey Cheng3, Kuo-Wang Tsai4, Ting-Ying Fu5, Huei-Han Liou4, Hui-Hwa Tseng5, Jang-Hwa Lee5, Guan-Cheng Li4,6, Jyh-Seng Wang5,7, Yu-Yi Hou8,9, Ta-Jung Weng4,6, Luo-Ping Ger4,6 1
Department of Stomatology, Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan; 2Department of Dental Technology, Shu-Zen Junior College of Medicine and Management, Kaohsiung, Taiwan; 3Department of Biological Sciences, National Sun Yat-Sen University, Kaohsiung, Taiwan; 4Department of Medical Education and Research, Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan; 5 Department of Pathology and Laboratory Medicine, Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan; 6Institute of Biomedical Sciences, National Sun Yat-Sen University, Kaohsiung, Taiwan; 7School of Medicine, National Yang-Ming University, Taipei, Taiwan; 8 Department of Otorhinolaryngology-Head & Neck Surgery, Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan; 9Department of Nursing, Yuh-Ing Junior College of Health Care and Management, Kaohsiung, Taiwan
BACKGROUNDS: Oral cancer is the 4th leading cause of cancer death for males and the top cancer in young adult males in Taiwan. Tongue squamous cell carcinoma (TSCC) is a common oral cancer and generally associated with poor prognosis. Global DNA hypomethylation at the 5 position of cytosine (5mC) is a well-known epigenetic feature of cancer. Therefore, the purpose of this study was to investigate the relationship of the global 5mC content with the tumorigenesis and prognosis of patients with TSCC. METHODS: The levels of global 5mC were evaluated by immunohistochemistry using tissue microarray slides of 248 surgically resected TSCC and 202 corresponding tumor adjacent normal (TAN) tissues. RESULTS: We found that the level of 5mC in TSCC (P < 0.001) was significantly decreased as compared to TAN. Among TSCC tissues, decreased levels of 5mC were associated with female gender (P = 0.036). In addition, the global hypomethylation was associated with the poor disease-specific survival in TSCC patients (adjusted hazard ratio: 1.55, P = 0.043), especially for patients in older age group (> 50 years, P = 0.013), with moderate or poor cell differentiation (P = 0.044), early stage of disease (I-II, P = 0.046), small tumor size (T1–T2, P = 0.005), without lymph node involvement (P = 0.041), and ever received postoperative radiotherapy (P = 0.009).
Correspondence: Luo-Ping Ger, Department of Medical Education and Research, Kaohsiung Veterans General Hospital, Kaohsiung 813, Taiwan. Tel: +011 886 7 346 8356, Fax: +011 886 7 346 8056, E-mail: [email protected] H.-C. C. and C.-M. Y. contributed equally to this study. Accepted for publication September 28, 2015
CONCLUSIONS: Global hypomethylation was an independent biomarker for the development and poor prognosis of TSCC. J Oral Pathol Med (2015) Keywords: hypomethylation; immunohistochemistry; survival; tissue microarray; tongue squamous cell carcinoma
Introduction Oral cancer is the 6th most common malignancy worldwide (1). It is the fourth cause of cancer death for males and the top cancer in young male adults (30–50 years old) in Taiwan (2). Tongue squamous cell carcinoma (TSCC) is a common form of oral cancer and accounted for 27.3% of oral and pharyngeal cancer cases in 2010 (2). Tongue has a rich supply of lymphatic drainage and neurovascular bundles, which increases the probability of neck nodal metastasis (3). Grandi et al. (4) demonstrated that the presence of nodal metastases in patients with TSCC reduced the 5-year survival rate from 65 to 29%. Therefore, TSCC has a significantly worse prognosis than other tumors in oral cavity (5). Despite considerable advances in diagnostic and therapeutic techniques, TSCC continues to have a poor prognosis (6). Therefore, identification of biomarkers of tumorigenesis, clinical pathologic characteristics, and survival of TSCC might be helpful in early diagnosis of disease and guiding clinical decision-making in treatment. Epigenetic modification of DNA by cytosine methylation to produce 5-methylcytosine (5mC) is well-recognized as an important epigenetic process in many cancers (7–9).
Global hypomethylation and tongue cancer Chen et al.
2
Epigenetic methylation of gene promoter regions is known to silence the expression of associated genes and aberrant promoter methylation is associated with altered behavior in oncogenes and tumor suppressor genes (10). In head and neck squamous cell carcinoma (HNSCC), epigenetic silencing by promoter hypermethylation (also known as regional hypermethylation) was reported to be involved in the inactivation of genes involved in cell cycle regulation [p16INK4A (11), p14ARF (11, 12)], DNA repair {O6-methylguanine-DNA methyltransferase [MGMT] (11), FHIT (13)}, apoptosis (DAPK) (11), cell adhesion (Ecadherin [CDH1] (14)), and other cellular processes (15, 16). Despite the frequently observed cancer-associated increases of regional hypermethylation (17, 18), the prevalence of non-specific (or global) DNA hypomethylation was also found in many types of human cancer (19–22). To date, there have been limited number of experiments investigating the genomewide methylation for head and neck squamous cell carcinoma (HNSCC) (23–27) and their results are still inconclusive. In this study, we evaluated the expression of 5mC in tumor adjacent normal tissue and TSCC tissue by immunohistochemistry assay. Then, we evaluated the impact of 5mC content on the tumorigenesis and prognosis of patients with TSCC.
Materials and methods Tissue specimens A total of 248 surgically resected TSCC tissues and 202 paired tumor adjacent normal (TAN) tissues were obtained from Department of Pathology, Kaohsiung Veterans General Hospital between 1993 and 2006. Patients with base of tongue cancer or surgical margin < 2 mm were excluded in this study. The survival time was estimated from the time of operation to October 2010. Pathologic TNM classification was determined at the time of the initial resection of the tumor in accordance with the guidelines of the 2002 American Joint Committee on Cancer (AJCC) system. The study protocol was independently reviewed and approved by the Institutional Review Board of the Kaohsiung Veterans General Hospital (IRB number: VGHKS13-CT10-04). Tissue microarray construction A tissue microarray (TMA) block consisted of 129 cores, 1.5 mm in diameter, including 43 trios, with each trio containing 2 cores from the tumor tissue and 1 core from the non-cancer epithelia of the same patient. Because some of the tumor tissue samples did not include non-cancer epithelia, non-cancer cores could not be obtained for some patients; therefore, only 6 TMA blocks were constructed in all. After construction, TMA blocks were cut into 4-lm paraffin sections using standard techniques. All the representative areas of tumor and non-cancer epithelia tissues were selected by a senior oral cancer pathologist from hematoxylin–eosinstained sections for coring cylindrical tissues from paraffinembedded tissues. The histological contents of 496 tumor cores (for 248 patients) contained 484 cores (97.6%) with only tumor epithelia and 12 cores with incorrect content. Within the 12 cores, 6 (1.2%) were without any tumor epithelia or with too few tumor epithelia (< 20 cells per core), 3 (0.6%) were with only normal epithelia, and 3 (0.6%) were missing or J Oral Pathol Med
without good quality to be scored. Although these 12 cores cannot be scored, the staining score of the alternative core can represent their staining. Therefore, all 248 patients were included in this study. In addition, 235 corresponding tumor adjacent normal (TAN) cores (for 235 patients) were stained in this study. A total of 202 cores were with normal epithelia only and were scored, but 33 cores with incorrect content were not scored and excluded from this study. They were 17 cores (7.2%) without any normal epithelia or with too few normal epithelia (< 20 cells), 12 cores (n = 12, 5.1%) without any normal epithelia but tumor epithelia, and 4 cores (n = 4, 1.7%) with missing or without good quality to be scored. Immunohistochemistry Novolink max polymer detection system (Leica, Newcastle Upon Tyne, UK) was used for IHC in this study. The slides were deparaffinized in xylene and rehydrated in grade alcohols. Antigen retrieval performed by immersion in Tris-EDTA (10 mM, pH 9.0) for 10 min at 125°C in the pressure boiler. Endogenous peroxidase activity was blocked by incubation of the slides for 30 min with 3% hydrogen peroxide in methanol. After blocking at room temperature, primary antibodies were immediately applied and slides were incubated overnight at 4°C in a wet chamber. The primary antibody, mouse monoclonal anti5mC (1:500; Abcam, Cambridge, MA, USA), was used in this study. After phosphate-buffered saline washes, the slides were incubated with horseradish peroxidase-labeled secondary antibody for 10 min at room temperature and the sections were counterstained with hematoxylin. Immunohistochemical analysis and scoring At beginning, an oral cancer pathologist (Dr. Ting-Ying Fu) accompanied a senior technician (Mr. Guan-Cheng Li) to evaluate slides until all the discrepancies were resolved. Then, this technician independently reviewed all slides, except those cores with incorrect or uncertain contents, which must be scored by the pathologist. Subsequently, 15% core samples were randomly selected for re-evaluation by this pathologist. If the pathologist and technician agreement of IHC scores were < 95%, this pathologist accompanied the technician to evaluate slides again until all the discrepancies were resolved. Then, the senior technician should rereview and rescore all slides until another random core samples (5–20%) resulted in scoring agreement ≧ 95%. Around 42% core samples were scored (for those incorrect, special, or uncertain cores) or re-evaluated (for the random core samples) by pathologist. In the final reevaluation, the pathologist and technician agreement of 5mC IHC score was 100% in this study. During the evaluation process, both of them were blind of the clinical outcomes of patients. We graded the immunoreactivity by a semiquantitative approach. Marker scores for nuclear staining were calculated based on staining intensity (0, no signal; 1, mild; 2, moderate; and 3, strong; Fig. 1A–D) and percentage of tumor cells staining at each intensity level (0, 75%). The intensity score and percentage of the positive cells were then multiplied to produce the final scores. For survival analysis, the expression levels were dichotomized as low expression and high expression with the cutoff set at the
Global hypomethylation and tongue cancer Chen et al.
3
Figure 1 Immunoreactivity of 5mC in oral tongue SCC. (A) Negative (B) Weak intensity (C) Moderate intensity (D) Strong intensity.
25th percentile based on the distribution of marker score. The cutoff values were 4 for 5mC in TSCC tissues. Statistical analysis The expression differences of 5mC between the paired tissues (TSCC vs. corresponding TAN) were analyzed by the Wilcoxon matched-pairs signed-rank test, but differences between the independent tissues (TSCC vs. TAN) were evaluated by Mann–Whitney U-test. Kruskal–Wallis one-way ANOVA and Mann–Whitney U-test were used to evaluate the correlation of 5mC expression levels of TSCC tissues with the clinicopathologic parameters. In addition, overall survival was calculated from the date of initial resection of the primary tumor to the date of death from any cause or last follow-up. Disease-specific survival was measured from the time of initial resection of the primary tumor to the date of cancer-specific death or last follow-up. Disease-free survival was calculated from the date of initial resection of the primary tumor to date of recurrence or last follow-up. The cumulative survival curves were estimated using the Kaplan–Meier method. The comparison of the survival curves was performed by the log-rank test. A Cox proportional hazards model was used to determine independent predictors of survival using factors significant on univariate analysis as covariates. A value of P < 0.05 (2sided) was considered significant.
Results The content of 5mC in TSCC tissues and the paired tumor adjacent normal (TAN) tissues To assess the impact of 5mC on TSCC development, the content of 5mC in the tongue TMA containing TSCC and
TAN tissues was examined (Table 1). A total of 202 TSCC patients’ tumor and paired tumor adjacent normal (TAN) tissues as well as another 46 patients’ tumor-only tissues (without paired TAN) were assayed in TMA. As 202 TAN and 248 (202 + 46) tumor tissues were treated as two independent samples, Mann–Whitney U-test was used to evaluate the impact of 5mC. We found that the staining of 5mC was found only in cell nucleus and the levels of 5mC in tumor tissues were significantly lower than those in TAN tissues (P < 0.001). As 202 tumor and paired TAN tissues were treated as dependent samples, Wilcoxon signed-rank test was used to evaluate the pairwise comparison of 5mC. We found that the content of 5mC was also significantly reduced in the tumor tissues as compared to the paired TAN tissues (P < 0.001, Table 1). The demographic and clinicopathologic characteristics and their impact on survival of patients with TSCC A total of 248 TSCC patients were eligible and studied (Table 2). Most TSCC patients were males (88%), middleaged adults (41–60 years: 60%; mean age = 51.25 years), with moderate grade of cell differentiation (Grade 2: 83%), locally advanced stage of disease (II + III: 57%), small size of tumor (T1 + T2: 79%), and without involvement of lymph node (N0: 79%). In addition, only 66 (26.6%) and 5 (2.0%) patients of TSCC had ever received postoperative RT and CT, respectively. With a median follow-up of 48.53 (range, 1.6–236.30) months, the overall survival rate was 56.65 3.24% at 5 years, 43.80 3.75% at 10 years, and 29.38 5.53% at 15 years. The disease-specific survival rate was 60.71 3.20% at 5 years, 51.60 3.72% at 10 years, and
Table 1 The comparisons of 5mC content between two different histological groups for tongue squamous cell carcinoma patients (tongue base cases were excluded) Tumor adjacent normal Variables
Mean SD
Independent samples
(n = 8.89 (n = 8.89
Dependent samples
202) 3.07 202) 3.07
Median (IQR) 8.00 (8.00–12.00) 8.00 (8.00–12.00)
Tumor Mean SD (n = 7.46 (n = 7.57
248) 2.86 202) 2.86
Median (IQR)
Statistics
P-value
8.00 (4.00–8.00)
Z=
5.037
