Prognostic Significance of Aberrant Methylation of Solute Carrier Gene Family 5A8 in Lung Adenocarcinoma Koei Ikeda, MD, PhD, Kenji Shiraishi, MD, PhD, Takafumi Koga, MD, Yamato Motooka, MD, Kosuke Fujino, MD, Hidekatsu Shibata, MD, PhD, Takeshi Mori, MD, PhD, and Makoto Suzuki, MD, PhD
Background. Solute carrier family 5 member A8 (SLC5A8) is a sodium-coupled transporter for several chemicals. The SLC5A8 gene has been reported to function as a tumor suppressor gene that contributes to carcinogenesis and tumor progression. The expression of SLC5A8 is silenced in colon neoplasia by hypermethylation of CpGrich islands located in exon 1. In this study, we assessed the significance of aberrant methylation of the SLC5A8 gene as a prognostic factor for lung adenocarcinoma (AD). Methods. We analyzed the methylation levels of a consecutive series of 143 node-negative
stage I and II lung AD samples using pyrosequencing. Results. The methylation level of exon 1 in the SLC5A8 gene was significantly associated with poor prognosis in cases of node-negative stage I and II lung AD. Conclusions. Gene silencing of SLC5A8 by hypermethylation was associated with poor prognosis in cases of node-negative stage I and II lung AD.
A
protein expression. Finally, we examined the prognostic significance of aberrant methylation of SLC5A8 in lung AD.
denocarcinoma (AD) is the most frequent subtype of non-small cell lung cancer [1]. Although major improvements in treatment strategies have increased the short-term survival for patients with non-small cell lung cancer, the impact on long-term survival has remained modest [2]. A better understanding of the molecular pathogenesis of lung AD is needed to identify biomarkers that would enable early detection and development of novel therapeutic targets. Solute carrier family 5 (SLC5) is a family of solutelinked carriers that contains 12 sodium-coupled transporters for nicotinate and nicotinic analogs, lactate, and short-chain fatty acids [3]. The SLC5A8 gene has been reported to function as a tumor-suppressor gene that contributes to carcinogenesis and tumor progression [4]. In several human malignancies, low expression of SLC5A8 and hypermethylation in the promoter region of this gene are associated with tumor aggressiveness and poor prognosis [5–11]. In this study, we quantified the promoter methylation of SLC5A8 in 143 samples from curatively resected, nodenegative, stage I and stage II lung AD using a bisulfite— polymerase chain reaction (PCR)—pyrosequencing. We also evaluated the correlations of promoter hypermethylation with clinicopathological factors or SLC5A8
Accepted for publication Feb 10, 2015. Address correspondence to Dr Ikeda, Department of Thoracic Surgery, Graduate School of Medical Sciences, Kumamoto University, 1-1-1 Honjo, Kumamoto 860-8556, Japan; e-mail: [email protected].
Ó 2015 by The Society of Thoracic Surgeons Published by Elsevier
(Ann Thorac Surg 2015;99:1755–60) Ó 2015 by The Society of Thoracic Surgeons
Material and Methods Patients and Tissue Samples Frozen samples from tumors and corresponding noncancerous lung tissue were obtained from a consecutive series of 143 patients with node-negative stage I or stage II lung AD who underwent curative resections without preoperative chemotherapy or radiotherapy at Kumamoto University Hospital between April 2010 and December 2012. The patients consisted of 69 men and 74 women, whose ages ranged from 50 to 87 years (mean, 68.8 9.0). Information on smoking history (with a packyear estimate) was available for all patients. Disease stage was determined in accordance with the seventh edition of the TNM classification for lung and pleural tumors [12]. Pathologic diagnoses were made in accordance with the International Association for the Study of Lung Cancer/ American Thoracic Society/European Respiratory Society’s international multidisciplinary classification system, as described by Travis and colleagues [13]. Venous invasion was evaluated by using Victoria blue stain. The follow-up schedule consisted of a clinic visit every 3 months for the first year after resection, and the once every 6 months thereafter. Follow-up procedures included a physical examination, chest radiography, and blood test (including assays for serum tumor markers). A computed tomography scan of the chest and abdomen was performed every 6 months for the first 3 years. When 0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2015.02.013
GENERAL THORACIC
Department of Thoracic Surgery, Graduate School of Medical Science, Kumamoto University, Kumamoto, Japan
1756
IKEDA ET AL SLC5A8 METHYLATION IN LUNG ADENOCARCINOMA
any symptoms or signs of recurrence were detected, magnetic resonance imaging of the brain and bone scintigraphy were performed. Twenty-two patients received adjuvant chemotherapy. Two patients received platinumbased chemotherapy (cisplatin plus gemcitabine), and 20 received administration of oral fluoropyrimidine. The study design was approved by the Ethics Review Board of Kumamoto University. Written informed consent for the research was obtained from each patient.
GENERAL THORACIC
DNA Extraction and Pyrosequencing Genomic DNA was obtained from primary tumors and nonmalignant tissues by using PureLink Genomic DNA kits (Invitrogen, Carlsbad, CA) to digest the DNA with proteinase K and phenol/chloroform (1:1) extraction. The DNA was treated with sodium bisulfite using an EpiTect Bisulfite kit (Qiagen, Valencia, CA) in accordance with the manufacturer’s instructions. Subsequent PCR and pyrosequencing for each gene were performed by using a PyroMark kit (Qiagen), as described previously [14]. The PCR conditions were 45 cycles of 95 C for 20 s, 50 C for 20 s, and 72 C for 20 s, followed by 72 C for 5 minutes. The biotinylated PCR products were purified and denatured before pyrosequencing with the Pyrosequencing Vacuum Prep Tool in the PyroMark Q96 MD System (Qiagen). The pyrosequencing analysis was designed for CpG-rich islands near the transcription start site in exon 1 of the gene. The sequences of forward biotinylated primer, reverse primer, and sequence primer were 50 -GGATTG GAGGAGTTTTTTAGGT-30 , 50 -CCAATTCTCATCTACT CAAATATCC-30 , and 50 -TACCAACCCTCACCC-30 respectively. The nucleotide dispensation order was ACGACTACAAGATATCGATACAAGAATCAGACAC. The amount of C relative to the sum of the amounts of C and T at each CpG site was calculated as a percentage. The average of the relative amounts of C in the CpG sites was used as the overall methylation level of each gene in a given tumor.
Immunohistochemistry Sections, 4 mm thick, obtained from the tissue microarray blocks of each patient were deparaffinized in xylene and rehydrated in a graded alcohol series. The SLC5A8 protein was determined by using a purchased antibody (goat, polyclonal, sc-34189; 1:100; Santa Cruz Biotechnology, Santa Cruz, CA). The primary antibody was incubated with the sections overnight at room temperature. Tumor cells with cytoplasmic or membrane-associated expression of SLC5A8, or both, were considered positive. The degree of staining was assessed by using three semiquantitative categories based on the percentage of stained (positive) tumor cells: absence of staining or less than 50% positive cells (low), 50% to 90% positive cells (moderate), and greater than 90% positive cells (high). The slides were examined independently by two observers (K.I and Y.H) blinded to both the clinical and pathologic data.
Statistical Methods Data are expressed as the mean SD for continuous data or numbers and percentages for categoric data. All
Ann Thorac Surg 2015;99:1755–60
statistical analyses were performed by using SPSS for Windows, version 15 (Texas Instruments, Schaumburg, IL). Differences between continuous variables were evaluated by using two-tailed Student’s t tests, and categoric data were compared by using c2 tests. KaplanMeier analysis was performed for survival curves, and statistical significance was assessed using the log rank test. To evaluate whether a biomarker was an independent prognostic factor of overall survival, a multivariate, stage-stratified Cox proportional hazard model was constructed to compute a hazard ratio that was based on the clinicopathologic parameters. All p values were based on the two-sided tests, and differences with p values of less than 0.05 were considered significant.
Results Methylation Levels of SLC5A8 in Resected Specimens of Patients With Lung Cancer The methylation levels of SLC5A8 in the resected lung cancer tissues and matched normal lung tissues were quantified using pyrosequencing. The mean methylation level of SLC5A8 in lung AD tissues was 9.6% 6.3%, which was significantly greater than that of matched normal lung tissues (5.8% 1.7%, p < 0.007 by the paired t test). The associations between clinical characteristics and the methylation levels of SLC5A8 in lung AD tissues are shown in Table 1. Only smoking status and histologic grade were significantly associated with a higher level of methylation (p ¼ 0.005 and p ¼ 0.05, respectively).
Immunohistochemical Analysis of SLC5A8 Successful immunohistochemical staining was achieved in 140 cases. Twenty-six tumor samples (18.6%) displayed low expression, 44 (31.4%) displayed moderate expression, and 70 (50.0%) displayed high expression. The cytoplasm and nucleus were both positively stained with anti-SLC5A8 antibodies in samples with high expression. Representative images of tissue samples with high and low expression are shown in Figure 1. The mean methylation levels of SLC5A8 in tumors in the low, moderate, and high expression groups were 15.3% 8.4%, 9.5% 5.7%, and 8.0% 4.6%, respectively. The average methylation level of SLC5A8 in tumors in the high expression group was significantly lower than those of the moderate and low expression groups (p ¼ 0.002 and p < 0.001, respectively, by unpaired t test).
Prognostic Significance of SLC5A8 Promoter Methylation Recurrence of cancer was observed in 11 of 143 patients with lung AD. The median follow-up period of patients was 33.0 months. Patients with stage II cancer (n ¼ 6) showed significantly shorter disease-free survival (DFS) than stage I patients (n ¼ 137, p < 0.001; Fig 2A). The 3year survival rates of patients with stage I and stage II cancer were 91.8% and 50.0%, respectively. For survival analysis, we divided the hypermethylation group (n ¼ 72) and nonhypermethylation group (n ¼ 71) by the median
Ann Thorac Surg 2015;99:1755–60
IKEDA ET AL SLC5A8 METHYLATION IN LUNG ADENOCARCINOMA
1757
Table 1. Correlations Between Clinicopathologic Factors and Methylation Rate of SLC5A8 Variable
Methylation Level (%)
p Value
143 74 69
8.9 5.7 10.3 6.8
0.2
69 74
10.5 6.8 8.7 5.6
0.08
75 68
8.2 5.4 11.2 6.9
0.004
108 35
9.0 5.6 11.4 7.9
0.05
128 15
9.4 6.3 11.5 6.3
0.2
125 18
9.5 6.2 10.6 7.1
0.5
108 35
9.6 6.5 9.6 5.7
1.0
137 6
6.2 0.5 9.7 3.7
0.5
115 28
9.6 6.6 9.6 5.0
1.0
76 67
10.1 6.0 9.1 6.6
0.4
EGFR ¼ epidermal growth factor receptor gene; carrier family 5 member A8.
SLC5A8 ¼ solute
value of methylation rate (7.75%) to predict postoperative recurrence of lung AD. The Kaplan-Meier curves (disease-free survival) of patients with and patients without a hypermethylated SLC5A8 survival in the two groups, meaning hypermethylation (7.75%) and nonhypermethylation (3 cm),
GENERAL THORACIC
Total Age, years 30 mm Pathologic stage I II Carcinoembryonic antigen 3.5 ng/mL), mutation status of the epidermal growth factor receptor gene (EGFR [positive]), adjuvant chemotherapy (positive), and methylation status of SLC5A8 (hypermethylation group). Vascular invasion, pleural invasion, pathologic stage, and methylation status emerged as prognostic factors predicting poor outcomes (Table 2). We also performed a Cox regression analysis of disease-free survival that included histologic grade, carcinoembryonic antigen, and SLC5A8 methylation status. Only methylation status of SLC5A8 (odds ratio 9.4, 95% confidence interval: 1.1 to 79.6; p ¼ 0.04) was an independent prognostic factor (Table 3).
Comment In an attempt to identify iodide transporters that may play functional roles in the physiology of the thyroid gland, SLC5A8 was originally cloned by Rodriguez and colleagues [15]. The SLC5A8 gene is located on human chromosome 12q13. The protein product of SLC5A8, which is expressed specifically in the apical membrane of
1758
IKEDA ET AL SLC5A8 METHYLATION IN LUNG ADENOCARCINOMA
Ann Thorac Surg 2015;99:1755–60
GENERAL THORACIC
Table 2. Univariate Analyses of Disease-Free Survival Variable
p Value
Sex (male) Older age (70 years) Smoking (plus) Histologic grade (II) Vascular invasion Pleural invasion Tumor size (>3 cm) Stage (II) Carcinoembryonic antigen (>3.5) EGFR mutation Adjuvant chemotherapy Hypermethylation of SLC5A8 (7.75%)
0.61 0.57 0.54 0.07 0.003 0.007 0.31
Background. Solute carrier family 5 member A8 (SLC5A8) is a sodium-coupled transporter for several chemicals. The SLC5A8 gene has been reported to function as a tumor suppressor gene that contributes to carcinogenesis and tumor progression. The expression of SLC5A8 is silenced in colon neoplasia by hypermethylation of CpGrich islands located in exon 1. In this study, we assessed the significance of aberrant methylation of the SLC5A8 gene as a prognostic factor for lung adenocarcinoma (AD). Methods. We analyzed the methylation levels of a consecutive series of 143 node-negative
stage I and II lung AD samples using pyrosequencing. Results. The methylation level of exon 1 in the SLC5A8 gene was significantly associated with poor prognosis in cases of node-negative stage I and II lung AD. Conclusions. Gene silencing of SLC5A8 by hypermethylation was associated with poor prognosis in cases of node-negative stage I and II lung AD.
A
protein expression. Finally, we examined the prognostic significance of aberrant methylation of SLC5A8 in lung AD.
denocarcinoma (AD) is the most frequent subtype of non-small cell lung cancer [1]. Although major improvements in treatment strategies have increased the short-term survival for patients with non-small cell lung cancer, the impact on long-term survival has remained modest [2]. A better understanding of the molecular pathogenesis of lung AD is needed to identify biomarkers that would enable early detection and development of novel therapeutic targets. Solute carrier family 5 (SLC5) is a family of solutelinked carriers that contains 12 sodium-coupled transporters for nicotinate and nicotinic analogs, lactate, and short-chain fatty acids [3]. The SLC5A8 gene has been reported to function as a tumor-suppressor gene that contributes to carcinogenesis and tumor progression [4]. In several human malignancies, low expression of SLC5A8 and hypermethylation in the promoter region of this gene are associated with tumor aggressiveness and poor prognosis [5–11]. In this study, we quantified the promoter methylation of SLC5A8 in 143 samples from curatively resected, nodenegative, stage I and stage II lung AD using a bisulfite— polymerase chain reaction (PCR)—pyrosequencing. We also evaluated the correlations of promoter hypermethylation with clinicopathological factors or SLC5A8
Accepted for publication Feb 10, 2015. Address correspondence to Dr Ikeda, Department of Thoracic Surgery, Graduate School of Medical Sciences, Kumamoto University, 1-1-1 Honjo, Kumamoto 860-8556, Japan; e-mail: [email protected].
Ó 2015 by The Society of Thoracic Surgeons Published by Elsevier
(Ann Thorac Surg 2015;99:1755–60) Ó 2015 by The Society of Thoracic Surgeons
Material and Methods Patients and Tissue Samples Frozen samples from tumors and corresponding noncancerous lung tissue were obtained from a consecutive series of 143 patients with node-negative stage I or stage II lung AD who underwent curative resections without preoperative chemotherapy or radiotherapy at Kumamoto University Hospital between April 2010 and December 2012. The patients consisted of 69 men and 74 women, whose ages ranged from 50 to 87 years (mean, 68.8 9.0). Information on smoking history (with a packyear estimate) was available for all patients. Disease stage was determined in accordance with the seventh edition of the TNM classification for lung and pleural tumors [12]. Pathologic diagnoses were made in accordance with the International Association for the Study of Lung Cancer/ American Thoracic Society/European Respiratory Society’s international multidisciplinary classification system, as described by Travis and colleagues [13]. Venous invasion was evaluated by using Victoria blue stain. The follow-up schedule consisted of a clinic visit every 3 months for the first year after resection, and the once every 6 months thereafter. Follow-up procedures included a physical examination, chest radiography, and blood test (including assays for serum tumor markers). A computed tomography scan of the chest and abdomen was performed every 6 months for the first 3 years. When 0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2015.02.013
GENERAL THORACIC
Department of Thoracic Surgery, Graduate School of Medical Science, Kumamoto University, Kumamoto, Japan
1756
IKEDA ET AL SLC5A8 METHYLATION IN LUNG ADENOCARCINOMA
any symptoms or signs of recurrence were detected, magnetic resonance imaging of the brain and bone scintigraphy were performed. Twenty-two patients received adjuvant chemotherapy. Two patients received platinumbased chemotherapy (cisplatin plus gemcitabine), and 20 received administration of oral fluoropyrimidine. The study design was approved by the Ethics Review Board of Kumamoto University. Written informed consent for the research was obtained from each patient.
GENERAL THORACIC
DNA Extraction and Pyrosequencing Genomic DNA was obtained from primary tumors and nonmalignant tissues by using PureLink Genomic DNA kits (Invitrogen, Carlsbad, CA) to digest the DNA with proteinase K and phenol/chloroform (1:1) extraction. The DNA was treated with sodium bisulfite using an EpiTect Bisulfite kit (Qiagen, Valencia, CA) in accordance with the manufacturer’s instructions. Subsequent PCR and pyrosequencing for each gene were performed by using a PyroMark kit (Qiagen), as described previously [14]. The PCR conditions were 45 cycles of 95 C for 20 s, 50 C for 20 s, and 72 C for 20 s, followed by 72 C for 5 minutes. The biotinylated PCR products were purified and denatured before pyrosequencing with the Pyrosequencing Vacuum Prep Tool in the PyroMark Q96 MD System (Qiagen). The pyrosequencing analysis was designed for CpG-rich islands near the transcription start site in exon 1 of the gene. The sequences of forward biotinylated primer, reverse primer, and sequence primer were 50 -GGATTG GAGGAGTTTTTTAGGT-30 , 50 -CCAATTCTCATCTACT CAAATATCC-30 , and 50 -TACCAACCCTCACCC-30 respectively. The nucleotide dispensation order was ACGACTACAAGATATCGATACAAGAATCAGACAC. The amount of C relative to the sum of the amounts of C and T at each CpG site was calculated as a percentage. The average of the relative amounts of C in the CpG sites was used as the overall methylation level of each gene in a given tumor.
Immunohistochemistry Sections, 4 mm thick, obtained from the tissue microarray blocks of each patient were deparaffinized in xylene and rehydrated in a graded alcohol series. The SLC5A8 protein was determined by using a purchased antibody (goat, polyclonal, sc-34189; 1:100; Santa Cruz Biotechnology, Santa Cruz, CA). The primary antibody was incubated with the sections overnight at room temperature. Tumor cells with cytoplasmic or membrane-associated expression of SLC5A8, or both, were considered positive. The degree of staining was assessed by using three semiquantitative categories based on the percentage of stained (positive) tumor cells: absence of staining or less than 50% positive cells (low), 50% to 90% positive cells (moderate), and greater than 90% positive cells (high). The slides were examined independently by two observers (K.I and Y.H) blinded to both the clinical and pathologic data.
Statistical Methods Data are expressed as the mean SD for continuous data or numbers and percentages for categoric data. All
Ann Thorac Surg 2015;99:1755–60
statistical analyses were performed by using SPSS for Windows, version 15 (Texas Instruments, Schaumburg, IL). Differences between continuous variables were evaluated by using two-tailed Student’s t tests, and categoric data were compared by using c2 tests. KaplanMeier analysis was performed for survival curves, and statistical significance was assessed using the log rank test. To evaluate whether a biomarker was an independent prognostic factor of overall survival, a multivariate, stage-stratified Cox proportional hazard model was constructed to compute a hazard ratio that was based on the clinicopathologic parameters. All p values were based on the two-sided tests, and differences with p values of less than 0.05 were considered significant.
Results Methylation Levels of SLC5A8 in Resected Specimens of Patients With Lung Cancer The methylation levels of SLC5A8 in the resected lung cancer tissues and matched normal lung tissues were quantified using pyrosequencing. The mean methylation level of SLC5A8 in lung AD tissues was 9.6% 6.3%, which was significantly greater than that of matched normal lung tissues (5.8% 1.7%, p < 0.007 by the paired t test). The associations between clinical characteristics and the methylation levels of SLC5A8 in lung AD tissues are shown in Table 1. Only smoking status and histologic grade were significantly associated with a higher level of methylation (p ¼ 0.005 and p ¼ 0.05, respectively).
Immunohistochemical Analysis of SLC5A8 Successful immunohistochemical staining was achieved in 140 cases. Twenty-six tumor samples (18.6%) displayed low expression, 44 (31.4%) displayed moderate expression, and 70 (50.0%) displayed high expression. The cytoplasm and nucleus were both positively stained with anti-SLC5A8 antibodies in samples with high expression. Representative images of tissue samples with high and low expression are shown in Figure 1. The mean methylation levels of SLC5A8 in tumors in the low, moderate, and high expression groups were 15.3% 8.4%, 9.5% 5.7%, and 8.0% 4.6%, respectively. The average methylation level of SLC5A8 in tumors in the high expression group was significantly lower than those of the moderate and low expression groups (p ¼ 0.002 and p < 0.001, respectively, by unpaired t test).
Prognostic Significance of SLC5A8 Promoter Methylation Recurrence of cancer was observed in 11 of 143 patients with lung AD. The median follow-up period of patients was 33.0 months. Patients with stage II cancer (n ¼ 6) showed significantly shorter disease-free survival (DFS) than stage I patients (n ¼ 137, p < 0.001; Fig 2A). The 3year survival rates of patients with stage I and stage II cancer were 91.8% and 50.0%, respectively. For survival analysis, we divided the hypermethylation group (n ¼ 72) and nonhypermethylation group (n ¼ 71) by the median
Ann Thorac Surg 2015;99:1755–60
IKEDA ET AL SLC5A8 METHYLATION IN LUNG ADENOCARCINOMA
1757
Table 1. Correlations Between Clinicopathologic Factors and Methylation Rate of SLC5A8 Variable
Methylation Level (%)
p Value
143 74 69
8.9 5.7 10.3 6.8
0.2
69 74
10.5 6.8 8.7 5.6
0.08
75 68
8.2 5.4 11.2 6.9
0.004
108 35
9.0 5.6 11.4 7.9
0.05
128 15
9.4 6.3 11.5 6.3
0.2
125 18
9.5 6.2 10.6 7.1
0.5
108 35
9.6 6.5 9.6 5.7
1.0
137 6
6.2 0.5 9.7 3.7
0.5
115 28
9.6 6.6 9.6 5.0
1.0
76 67
10.1 6.0 9.1 6.6
0.4
EGFR ¼ epidermal growth factor receptor gene; carrier family 5 member A8.
SLC5A8 ¼ solute
value of methylation rate (7.75%) to predict postoperative recurrence of lung AD. The Kaplan-Meier curves (disease-free survival) of patients with and patients without a hypermethylated SLC5A8 survival in the two groups, meaning hypermethylation (7.75%) and nonhypermethylation (3 cm),
GENERAL THORACIC
Total Age, years 30 mm Pathologic stage I II Carcinoembryonic antigen 3.5 ng/mL), mutation status of the epidermal growth factor receptor gene (EGFR [positive]), adjuvant chemotherapy (positive), and methylation status of SLC5A8 (hypermethylation group). Vascular invasion, pleural invasion, pathologic stage, and methylation status emerged as prognostic factors predicting poor outcomes (Table 2). We also performed a Cox regression analysis of disease-free survival that included histologic grade, carcinoembryonic antigen, and SLC5A8 methylation status. Only methylation status of SLC5A8 (odds ratio 9.4, 95% confidence interval: 1.1 to 79.6; p ¼ 0.04) was an independent prognostic factor (Table 3).
Comment In an attempt to identify iodide transporters that may play functional roles in the physiology of the thyroid gland, SLC5A8 was originally cloned by Rodriguez and colleagues [15]. The SLC5A8 gene is located on human chromosome 12q13. The protein product of SLC5A8, which is expressed specifically in the apical membrane of
1758
IKEDA ET AL SLC5A8 METHYLATION IN LUNG ADENOCARCINOMA
Ann Thorac Surg 2015;99:1755–60
GENERAL THORACIC
Table 2. Univariate Analyses of Disease-Free Survival Variable
p Value
Sex (male) Older age (70 years) Smoking (plus) Histologic grade (II) Vascular invasion Pleural invasion Tumor size (>3 cm) Stage (II) Carcinoembryonic antigen (>3.5) EGFR mutation Adjuvant chemotherapy Hypermethylation of SLC5A8 (7.75%)
0.61 0.57 0.54 0.07 0.003 0.007 0.31
