Diseases of the Esophagus (2015) 28, 188–195 DOI: 10.1111/dote.12156
Original article
Overexpression of melanoma-associated antigen D4 is an independent prognostic factor in squamous cell carcinoma of the esophagus H. Oya, M. Kanda, H. Takami, S. Hibino, D. Shimizu, Y. Niwa, M. Koike, S. Nomoto, S. Yamada, Y. Nishikawa, M. Asai, T. Fujii, G. Nakayama, H. Sugimoto, M. Fujiwara, Y. Kodera Department of Gastroenterological Surgery (Surgery II), Nagoya University Graduate School of Medicine, Nagoya, Japan
SUMMARY. To pursue an urgently needed treatment target for esophageal cancer (EC), we investigated the function of the recently discovered melanoma-associated antigen (MAGE)-D4 in squamous cell EC. MAGE-D4 messenger RNA (mRNA) expression was analyzed in nine EC cell lines using quantitative reverse transcription polymerase chain reaction. In 65 surgical specimens of squamous cell EC with no prior neoadjuvant therapy, MAGE-D4 mRNA expression in EC tissues and corresponding normal tissues was analyzed and compared, and evaluated in terms of clinicopathological factors. In representative cases, MAGE-D4 protein distribution was analyzed immunohistochemically. The heterogeneity of MAGE-D4 mRNA expression was confirmed in EC cell lines by quantitative reverse transcription polymerase chain reaction. In surgical specimens, MAGE-D4 mRNA expression was significantly higher in EC tissues than in corresponding normal tissues (P < 0.001). Patients with the highest MAGE-D4 mRNA expression in EC tissues (top quartile, n = 17) had significantly shorter overall survival than patients with low expression (2-year survival: 44% and 73%, respectively, P = 0.006). Univariate analysis identified age (≥65 years), lymphatic involvement, and high MAGE-D4 mRNA expression as significant prognostic factors; high MAGE-D4 mRNA expression was also an independent prognostic factor in multivariable analysis (hazard ratio: 2.194; P = 0.039) and was significantly associated with Brinkman index (P = 0.008) and preoperative carcinoembryonic antigen level (P = 0.002). Immunohistochemical MAGE-D4b expression was consistent with MAGE-D4 mRNA profiling. Our results suggest that MAGE-D4 overexpression influences tumor progression, and MADE-D4 can be a prognostic marker and a potential molecular target in squamous cell EC. KEY WORDS: esophageal cancer, expression, MAGE-D4, progression.
INTRODUCTION Worldwide, esophageal cancer (EC) is the eighth most common cancer and sixth in cancer mortality.1 In 2008, worldwide EC incidence was 482 000 patients, and its mortality was 406 000 cases.2 It is generally diagnosed at a late stage and has a poor prognosis, with a 5-year survival rate of less than 10%.3 In spite of improved multimodal treatment, prognosis remains unsatisfactory because effective systemic therapy has not been established for patients with advanced EC.4 Understanding mechanisms of
Address correspondence to: Dr Mitsuro Kanda, MD, PhD, Department of Gastroenterological Surgery (Surgery II), Nagoya University Graduate School of Medicine, University of Nagoya, 65 Tsurumai-cho, Showa-ku, Nagoya 466-8550, Japan. Email: [email protected] 188
carcinogenesis and tumor progression is necessary to develop novel therapeutic strategies, including molecular target agents. Currently, great attention has been paid to the expression of tumor-specific antigen genes and their association with tumorigenesis. The melanomaassociated antigen (MAGE) family, a type of tumorassociated antigens, is reportedly important in the progression of various malignant tumors.5–9 MAGE-A1 was first characterized by the expression of tumor antigens on a malignant melanoma recognized by autologous cytolytic T lymphocytes.10 Based on these characteristics of the MAGE-A1 subfamily, antitumor vaccination11–13 and other clinical applications (tumor detection and disease monitoring) have been attempted.14,15 Immunotherapy with MAGEA1 peptides showed tumor regression without significant side-effects in melanoma patients who were
© 2013 Wiley Periodicals, Inc. and the International Society for Diseases of the Esophagus
MAGE-D4 expression in esophageal cancer
vaccinated with the MAGE peptide.12 More than 60 MAGE genes have been identified. Many genes homologous to MAGE-A1 have been found on Xq 28, Xp21.3, Xq26, and Xp11.23,16–20 and are classified as MAGE-A, -B, -C, -D, and -H, respectively. MAGE genes include type I and type II genes based on differences in gene structure and tissue-specific gene expression.21 Type I MAGE genes are located on the X-chromosome and consist of MAGE-A, -B, and -C22; their gene products are expressed during germ cell development, and expression ceases in normal mature somatic cells.23,24 Notably, research on type I MAGE proteins is comparatively more developed; increased survival of proliferating cells, inhibited apoptosis, and modulated E3 ubiquitin ligase activity have been reported.25–28 By contrast, the localization and expression of type II MAGE proteins (MAGE-D, E, F, G, and H) are less elucidated. MAGE-D, a type II MAGE protein, is divided into MAGE-D1–4. MAGE-D4 reportedly contributes to proliferation, migration, and invasion of tumor cells in breast cancer and oral squamous cell carcinoma.29,30 However, the function and expression of MAGE-D4 in digestive tract cancers, including EC, has not been reported. MAGE-D4 reportedly has a unique distribution pattern in tissues, confirmed to differ from those of other MAGE family members.31,32 Therefore, evaluating MAGE-D4 expression independently of investigations of expression of other MAGE genes in EC would be worthwhile. As one of numerous attempts to identify novel cancer-related genes in gastroenterological malignancies and to clarify the underlying mechanisms of carcinogenesis and tumor progression,33,34 we focused on the MAGE-D4 gene as a candidate cancer-related gene, with the expectation that MAGE-D4 might be recognized as novel diagnostic marker and therapeutic target in squamous cell carcinoma of the esophagus.
MATERIALS AND METHODS Ethics This study conforms to the ethical guidelines of the World Medical Association Declaration of Helsinki: Ethical Principles for Medical Research Involving Human Subjects. Written informed consent for usage of clinical samples and data, as required by the institutional review board at Nagoya University, Japan, was obtained from all patients. Cell lines and surgical specimens Nine EC cell lines (TE1, TE2, TE3, NUEC1, NUEC2, NUEC3, TT, TTn, and WSSC) were obtained from the American Type Culture Collection (Manassas, VA, USA), stored at −80°C with cell
189
preservative solution (Cell Banker®, Mitsubishi Chemical Medience Corporation, Tokyo, Japan) and cultured in RPMI-1640 (Sigma-Aldrich, St. Louis, MO, USA) supplemented with 10% fetal bovine serum and incubated in 5% CO2 at 37°C. Sixty-five primary EC tissues and corresponding non-cancerous tissues were obtained at Nagoya University Hospital, all during radical esophageal resection between December 2001 and April 2012. None of the patients underwent preoperative treatment including chemotherapy and radiation. Specimens were classified histologically using the Union for International Cancer Control (UICC) staging system for EC.35 Demographics, tobacco and alcohol consumption, preoperative serum tumor markers, tumor size, pathological findings including tumor differentiation, tumor depth, vascular invasion, and lymph node metastasis were investigated from the retrospective database. Tobacco consumption was estimated by the Brinkman index, which is defined as numbers of cigarette smoked per day times smoking years. Data on alcohol consumption was based on questioning patients; excessive alcohol consumption was defined as alcohol intake >210 g/week for ≥3 years.36 Median duration of patient follow up was 33.5 months (range: 1.5–125 months). Postoperative follow-up examinations included physical examination and measurement of serum tumor markers every 3 months, and enhanced computed tomography scan (chest and abdominal cavity) every 6 months. Adjuvant chemotherapy was applied to selected patients based on patient’s condition and physician’s discretion. Collected samples were stored immediately in liquid nitrogen at −80°C until analysis. Total RNA was obtained from these samples using the RNeasy kit (Qiagen, Hilden, Germany). Quantitative real-time reverse transcription-polymerase chain reaction Levels of MAGE-D4 messenger RNA (mRNA) expression were analyzed by quantitative real-time reverse transcription polymerase chain reaction (RTPCR). Total RNA (10 μg) was isolated from EC cell lines (TE1, TE2, TE3, NUEC1, NUEC2, NUEC3, TT, TTn, and WSSC), primary EC tissues, and corresponding non-cancerous tissues, and used to generate complementary DNAs, which were then amplified using PCR primers MAGE-D4; sense (S) (5′-GGC GATCTGAGGAAGCTCAT-3′ in exon 10) and antisense (AS) (5′-CATACTCAGGTGGGTTGCT GT-3′ in exon 11) to amplify a 91-base pair product. As this primer pair amplifies MAGE-D4a, -b, and -c, this reaction reflects the total amount of MAGE-D4 mRNA. Quantitative RT-PCR was performed with the SYBR-Green PCR core reagents kit (PerkinElmer, Applied Biosystems, Foster City, CA, USA) as follows: one cycle at 50°C for 2 minutes; one cycle
© 2013 Wiley Periodicals, Inc. and the International Society for Diseases of the Esophagus
190
Diseases of the Esophagus
at 95°C for 10 minutes; then 45 cycles at 95°C for 15 seconds; and 60°C for 30 seconds. Real-time detection of the SYBR-Green emission intensity was conducted with an ABI prism 7000 Sequence Detector (Perkin-Elmer, Applied Biosystems). mRNA expression was standardized to that of glyceraldehyde3-phosphate dehydrogenase (GAPDH) mRNA (TaqManR, GAPDH control reagents, Applied Biosystems). Quantitative RT-PCR was performed in triplicate and included no-template samples as negative controls. Expression levels for samples are shown as MAGE-D4 values standardized to MAGE-D4/GAPDH ratio. MAGE-D4 mRNA expression in tumor tissues and corresponding noncancerous tissues were compared and correlated with clinicopathological characteristics and prognoses.
Prognostic value of MAGE-D4 mRNA expression level was assessed by in a regression analysis. Overall survival (OS) rates were calculated using the Kaplan–Meier method; differences in survival curves were analyzed using the log-rank test. We performed multivariable regression analysis to detect prognostic factors using Cox proportional hazards models. Variables with P < 0.05 were entered into the final model. Associations between MAGE-D4 mRNA expression and clinicopathological parameters were evaluated using the χ2 test. Statistical analysis was performed using JMP® 10 software (SAS Institute Inc., Cary, NC, USA). P < 0.05 was considered significant.
RESULTS Immunohistochemical staining
Patient characteristics
Expression and distribution of MAGE-D4b protein, the putative dominant MAGE-D4 isoform, were analyzed immunohistochemically in representative cases and compared with mRNA expression patterns. Sections of formalin-fixed and paraffin-embedded tissues were dewaxed in xylene twice for 5 minutes, rehydrated in grading alcohols 100%, 90%, and 70% to H2O for 2 minutes each and subsequently treated with 3% H2O2 to inhibit endogenous peroxidase, and subjected to antigen retrieval using 10 mmol/L citrate buffer at 95°C for 10 minutes, five times. Sections were incubated with Histofine SAB-PO(R) (Nichirei, Tokyo, Japan) for 10 minutes, to limit non-specific reactivity, and then incubated for 1 hour with rabbit anti-MAGE-D4b polyclonal antibody (HPA003554, Sigma Aldrich) diluted 1:1000 in ChemMate antibody diluent (Dako, Carpinteria, CA, USA). Samples were then washed with phosphate-buffered saline, followed by a 10-minute incubation with biotinylated secondary antibody (Histofine SAB-PO(R), Nichirei). All antibody staining was developed for 2 minutes using liquid 3,3′-diaminobenzidine as substrate (Nichirei). Specimens were randomized and coded before analysis; analyses were conducted by two independent observers, who evaluated all specimens at least twice within a given time interval to minimize intra-observer variation. Expression level of MAGED4b protein was evaluated both in EC tissues and corresponding non-cancerous tissue. Components with ≥10% of stained cells were qualitatively defined as positive.
The mean age of the 65 patients was 64.1 ± 7.3 years (mean ± standard deviation; range: 46–80 years). The male-to-female ratio was 53:12. Fifty-three patients had preoperative symptoms, such as chest pain. Fifty patients had histories of excessive alcohol consumption; 28 patients had Brinkman index ≥1000. When classified by the seventh edition of the UICC classification, 5, 20, 30, and 10 patients were in stages I, II, III, and IV, respectively.
Statistical analysis Relative mRNA expression levels (MAGE-D4/ GAPDH) were calculated from quantified data. Differences in MAGE-D4 mRNA expression between cancerous tissues and corresponding non-cancerous tissues were analyzed by the Mann–Whitney U-test.
MAGE-D4 mRNA expression analyzed by quantitative RT-PCR Expression analysis of MAGE-D4 mRNA in EC cell lines was performed to figure out whether EC involves aberrant expression of MAGE-D4 or not. Expression levels of MAGE-D4 mRNA in nine EC cell lines were shown in Figure 1A with the median value and range of expression level in 65 normal esophageal tissues, demonstrating the heterogeneity of MAGE-D4 expression in EC cell lines. Mean expression of MAGE-D4 mRNA was significantly higher in EC tissues than in corresponding normal tissues (P < 0.001; Fig. 1B). Prognostic impact of overexpression of MAGE-D4 mRNA We first analyzed prognostic impact of MAGE-D4 mRNA expression level in EC tissues as a continuous variable in a regression analysis, and the expression level was significantly associated with OS (P = 0.011). Next, we compared two patents group categorized by the median value of MAGE-D4 mRNA expression level in EC tissues. Although the high expression group tended to have a shorter OS, there was no significant difference between the two groups (P = 0.192). Then we adopted the quartile analysis. Patients with the highest MAGE-D4 mRNA
© 2013 Wiley Periodicals, Inc. and the International Society for Diseases of the Esophagus
MAGE-D4 expression in esophageal cancer
Fig. 1 Analysis of melanoma-associated antigen (MAGE)-D4 messenger RNA (mRNA) in esophageal cancer cell lines. (A) Expression levels of MAGE-D4 mRNA in nine esophageal cancer cell lines were shown along with the median value in 65 normal epithelial tissues. The error bars represent the range. (B) In 65 surgical specimens, mean expression of MAGE-D4 mRNA was significantly higher in cancerous tissues than in corresponding normal tissues (P < 0.001). The error bars represent the standard deviation.
expression in EC tissues (fourth quartile, n = 17) had significantly shorter OS than patients with low (first– third quartiles, n = 48) MAGE-D4 mRNA expression (2-year OS: 44% and 73%, respectively, P = 0.006; Fig. 2A). Patients with the highest MAGE-D4 mRNA expression in EC tissues tended to have shorter recurrence-free survival (RFS) than patients with low MAGE-D4 mRNA expression (2-year RFS: 66% and 79%, respectively, P = 0.058; Fig. 2B). Patients were grouped by age, gender, preoperative symptom, Brinkman index (≥1000), excessive alcohol consumption, carcinoembryonic antigen (CEA; >5 ng/mL), squamous cell carcinoma-related antigen (>1.5 ng/mL), tumor size ( ≥ 5.0 cm), T factor (T3–4), tumor differentiation (poor), lymphatic involvement, vessel invasion, intra-epithelial spread, lymph node metastasis, and MAGE-D4 mRNA expression. Of these variables, univariate analysis identified age ≥65 years, lymphatic involvement, and high MAGE-D4 mRNA expression as significant prognostic factors. In multivariable analysis, high MAGE-D4 mRNA expression was identified as an independent prognostic factor in multivariable analy-
191
Fig. 2 Survival curves for 65 patients with esophageal cancer. Patients with high melanoma-associated antigen (MAGE)-D4 messenger RNA (mRNA) expression in cancerous tissues (top quartile, n = 17) had significantly shorter (A) overall survival (P = 0.006) and (B) recurrence-free survival (P = 0.058) than patients with low MAGE-D4 mRNA expression.
sis (hazard ratio: 2.194; P = 0.039; Table 1), together with lymphatic involvement. Association between MAGE-D4 mRNA expression and clinicopathological factors Associations between MAGE-D4 mRNA expression and clinicopathological parameters in the 65 patients are shown in Table 2. High MAGE-D4 mRNA expression was significantly associated with Brinkman index ≥1000 (P = 0.008) and preoperative CEA level >5 ng/mL (P = 0.002). Immunohistochemical staining MAGE-D4b protein expression was also evaluated immunohistochemically in representative cases with MAGE-D4 mRNA overexpression, reduced expression or equivalent expression in EC tissues compared with the corresponding non-cancerous tissues. In accord with the mRNA results, MAGE-D4b protein expression was confirmed exclusively in the membrane and cytoplasm of EC tissues that overexpressed
© 2013 Wiley Periodicals, Inc. and the International Society for Diseases of the Esophagus
192
Diseases of the Esophagus
Table 1 Prognostic factors for 65 patients with squamous cell carcinoma of the esophagus Univariate
Multivariable
Variable
n
Hazard ratio
95% CI
P-value
Hazard ratio
95% CI
P-value
Age (≥65) Gender (male) Preoperative symptom Brinkman index (≥1000) Excessive alcohol consumption CEA (>5 ng/mL) SCC (>1.5 ng/mL) Tumor size (≥5.0 cm) T factor (T3-4) Tumor differentiation (poor) Lymphatic involvement Vessel invasion Intra-epithelial spread Lymph node metastasis High MAGE-D4 mRNA expression (fourth quartile)
29 53 53 28 50 8 21 39 47 11 54 26 32 44 17
2.26 2.27 0.95 1.96 0.77 1.57 0.69 0.75 1.24 2.33 5.08 1.11 1.02 1.89 2.62
1.11–4.67 0.89–7.68 0.43–2.39 0.98–4.02 0.36–1.84 0.58–3.57 0.29–1.47 0.37–1.54 0.59–2.83 0.98–5.01 1.53–31.4 0.54–2.23 0.51–2.06 0.88–4.49 1.26–5.32
0.024* 0.093 0.905 0.059 0.535 0.346 0.345 0.423 0.582 0.056 0.005* 0.773 0.956 0.104 0.011*
1.89 — — — — — — — — — 4.53 — — — 2.19
0.92–3.94 — — — — — — — — — 1.36–28.1 — — — 1.04–4.49
0.081 — — — — — — — — — 0.010* — — — 0.039*
Univariate analysis was performed using the log-rank test. Multivariate analysis was performed using the Cox proportional hazards model. *Statistically significant (P < 0.05). CEA, carcinoembryonic antigen; CI, confidence interval; MAGE, melanoma-associated antigen; mRNA, messenger RNA; SCC, squamous cell carcinoma-related antigen.
MAGE-D4 mRNA (Fig. 3A,B). Conversely, samples with low MAGE-D4 mRNA expression showed little MAGE-D4b protein expression in either EC or noncancerous components (Fig. 3C).
DISCUSSION MAGE-D4 is a novel and unique member of the MAGE family that was discovered in 2001 and is specifically expressed in brain and ovary.31,32
MAGE-D4 is reportedly overexpressed in some human malignancies, including gliomas, non-smallcell lung cancers, breast cancer, and oral squamous cell carcinomas.29–31 Among MAGE family members, MAGE-D1 (also known as neutrophin receptorinteracting MAGE homologue) showed the greatest homology to MAGE-D4.37 MAGE-D1 has a 78% similarity to MAGE-D4, but lacks the N-terminal region, which implies that MAGE-D4 has a function in cell division and proliferation different from that of MAGE-D1. Previous experimental studies showed
Fig. 3 Immunohistochemical staining of melanoma-associated antigen (MAGE)-D4b in representative cases. (A, B) Specimen in which MAGE-D4 messenger RNA (mRNA) overexpression was detected by quantitative reverse transcription polymerase chain reaction (RT-PCR). Strong MAGE-D4b protein expression was confirmed in membranes and cytoplasm of cancerous components (A; upper 40×; lower 200×; B; intramural metastasis, 100×). (C) Specimen with equivalent expression of MAGE-D4 mRNA. Neither cancerous nor normal components showed MAGE-D4b protein expression (40×). © 2013 Wiley Periodicals, Inc. and the International Society for Diseases of the Esophagus
MAGE-D4 expression in esophageal cancer
193
Table 2 Association between MAGE-D4 mRNA expression and clinicopathological parameters in 65 patients with squamous cell carcinoma of the esophagus
Clinicopathological parameters Age 1.5 Tumor size
Original article
Overexpression of melanoma-associated antigen D4 is an independent prognostic factor in squamous cell carcinoma of the esophagus H. Oya, M. Kanda, H. Takami, S. Hibino, D. Shimizu, Y. Niwa, M. Koike, S. Nomoto, S. Yamada, Y. Nishikawa, M. Asai, T. Fujii, G. Nakayama, H. Sugimoto, M. Fujiwara, Y. Kodera Department of Gastroenterological Surgery (Surgery II), Nagoya University Graduate School of Medicine, Nagoya, Japan
SUMMARY. To pursue an urgently needed treatment target for esophageal cancer (EC), we investigated the function of the recently discovered melanoma-associated antigen (MAGE)-D4 in squamous cell EC. MAGE-D4 messenger RNA (mRNA) expression was analyzed in nine EC cell lines using quantitative reverse transcription polymerase chain reaction. In 65 surgical specimens of squamous cell EC with no prior neoadjuvant therapy, MAGE-D4 mRNA expression in EC tissues and corresponding normal tissues was analyzed and compared, and evaluated in terms of clinicopathological factors. In representative cases, MAGE-D4 protein distribution was analyzed immunohistochemically. The heterogeneity of MAGE-D4 mRNA expression was confirmed in EC cell lines by quantitative reverse transcription polymerase chain reaction. In surgical specimens, MAGE-D4 mRNA expression was significantly higher in EC tissues than in corresponding normal tissues (P < 0.001). Patients with the highest MAGE-D4 mRNA expression in EC tissues (top quartile, n = 17) had significantly shorter overall survival than patients with low expression (2-year survival: 44% and 73%, respectively, P = 0.006). Univariate analysis identified age (≥65 years), lymphatic involvement, and high MAGE-D4 mRNA expression as significant prognostic factors; high MAGE-D4 mRNA expression was also an independent prognostic factor in multivariable analysis (hazard ratio: 2.194; P = 0.039) and was significantly associated with Brinkman index (P = 0.008) and preoperative carcinoembryonic antigen level (P = 0.002). Immunohistochemical MAGE-D4b expression was consistent with MAGE-D4 mRNA profiling. Our results suggest that MAGE-D4 overexpression influences tumor progression, and MADE-D4 can be a prognostic marker and a potential molecular target in squamous cell EC. KEY WORDS: esophageal cancer, expression, MAGE-D4, progression.
INTRODUCTION Worldwide, esophageal cancer (EC) is the eighth most common cancer and sixth in cancer mortality.1 In 2008, worldwide EC incidence was 482 000 patients, and its mortality was 406 000 cases.2 It is generally diagnosed at a late stage and has a poor prognosis, with a 5-year survival rate of less than 10%.3 In spite of improved multimodal treatment, prognosis remains unsatisfactory because effective systemic therapy has not been established for patients with advanced EC.4 Understanding mechanisms of
Address correspondence to: Dr Mitsuro Kanda, MD, PhD, Department of Gastroenterological Surgery (Surgery II), Nagoya University Graduate School of Medicine, University of Nagoya, 65 Tsurumai-cho, Showa-ku, Nagoya 466-8550, Japan. Email: [email protected] 188
carcinogenesis and tumor progression is necessary to develop novel therapeutic strategies, including molecular target agents. Currently, great attention has been paid to the expression of tumor-specific antigen genes and their association with tumorigenesis. The melanomaassociated antigen (MAGE) family, a type of tumorassociated antigens, is reportedly important in the progression of various malignant tumors.5–9 MAGE-A1 was first characterized by the expression of tumor antigens on a malignant melanoma recognized by autologous cytolytic T lymphocytes.10 Based on these characteristics of the MAGE-A1 subfamily, antitumor vaccination11–13 and other clinical applications (tumor detection and disease monitoring) have been attempted.14,15 Immunotherapy with MAGEA1 peptides showed tumor regression without significant side-effects in melanoma patients who were
© 2013 Wiley Periodicals, Inc. and the International Society for Diseases of the Esophagus
MAGE-D4 expression in esophageal cancer
vaccinated with the MAGE peptide.12 More than 60 MAGE genes have been identified. Many genes homologous to MAGE-A1 have been found on Xq 28, Xp21.3, Xq26, and Xp11.23,16–20 and are classified as MAGE-A, -B, -C, -D, and -H, respectively. MAGE genes include type I and type II genes based on differences in gene structure and tissue-specific gene expression.21 Type I MAGE genes are located on the X-chromosome and consist of MAGE-A, -B, and -C22; their gene products are expressed during germ cell development, and expression ceases in normal mature somatic cells.23,24 Notably, research on type I MAGE proteins is comparatively more developed; increased survival of proliferating cells, inhibited apoptosis, and modulated E3 ubiquitin ligase activity have been reported.25–28 By contrast, the localization and expression of type II MAGE proteins (MAGE-D, E, F, G, and H) are less elucidated. MAGE-D, a type II MAGE protein, is divided into MAGE-D1–4. MAGE-D4 reportedly contributes to proliferation, migration, and invasion of tumor cells in breast cancer and oral squamous cell carcinoma.29,30 However, the function and expression of MAGE-D4 in digestive tract cancers, including EC, has not been reported. MAGE-D4 reportedly has a unique distribution pattern in tissues, confirmed to differ from those of other MAGE family members.31,32 Therefore, evaluating MAGE-D4 expression independently of investigations of expression of other MAGE genes in EC would be worthwhile. As one of numerous attempts to identify novel cancer-related genes in gastroenterological malignancies and to clarify the underlying mechanisms of carcinogenesis and tumor progression,33,34 we focused on the MAGE-D4 gene as a candidate cancer-related gene, with the expectation that MAGE-D4 might be recognized as novel diagnostic marker and therapeutic target in squamous cell carcinoma of the esophagus.
MATERIALS AND METHODS Ethics This study conforms to the ethical guidelines of the World Medical Association Declaration of Helsinki: Ethical Principles for Medical Research Involving Human Subjects. Written informed consent for usage of clinical samples and data, as required by the institutional review board at Nagoya University, Japan, was obtained from all patients. Cell lines and surgical specimens Nine EC cell lines (TE1, TE2, TE3, NUEC1, NUEC2, NUEC3, TT, TTn, and WSSC) were obtained from the American Type Culture Collection (Manassas, VA, USA), stored at −80°C with cell
189
preservative solution (Cell Banker®, Mitsubishi Chemical Medience Corporation, Tokyo, Japan) and cultured in RPMI-1640 (Sigma-Aldrich, St. Louis, MO, USA) supplemented with 10% fetal bovine serum and incubated in 5% CO2 at 37°C. Sixty-five primary EC tissues and corresponding non-cancerous tissues were obtained at Nagoya University Hospital, all during radical esophageal resection between December 2001 and April 2012. None of the patients underwent preoperative treatment including chemotherapy and radiation. Specimens were classified histologically using the Union for International Cancer Control (UICC) staging system for EC.35 Demographics, tobacco and alcohol consumption, preoperative serum tumor markers, tumor size, pathological findings including tumor differentiation, tumor depth, vascular invasion, and lymph node metastasis were investigated from the retrospective database. Tobacco consumption was estimated by the Brinkman index, which is defined as numbers of cigarette smoked per day times smoking years. Data on alcohol consumption was based on questioning patients; excessive alcohol consumption was defined as alcohol intake >210 g/week for ≥3 years.36 Median duration of patient follow up was 33.5 months (range: 1.5–125 months). Postoperative follow-up examinations included physical examination and measurement of serum tumor markers every 3 months, and enhanced computed tomography scan (chest and abdominal cavity) every 6 months. Adjuvant chemotherapy was applied to selected patients based on patient’s condition and physician’s discretion. Collected samples were stored immediately in liquid nitrogen at −80°C until analysis. Total RNA was obtained from these samples using the RNeasy kit (Qiagen, Hilden, Germany). Quantitative real-time reverse transcription-polymerase chain reaction Levels of MAGE-D4 messenger RNA (mRNA) expression were analyzed by quantitative real-time reverse transcription polymerase chain reaction (RTPCR). Total RNA (10 μg) was isolated from EC cell lines (TE1, TE2, TE3, NUEC1, NUEC2, NUEC3, TT, TTn, and WSSC), primary EC tissues, and corresponding non-cancerous tissues, and used to generate complementary DNAs, which were then amplified using PCR primers MAGE-D4; sense (S) (5′-GGC GATCTGAGGAAGCTCAT-3′ in exon 10) and antisense (AS) (5′-CATACTCAGGTGGGTTGCT GT-3′ in exon 11) to amplify a 91-base pair product. As this primer pair amplifies MAGE-D4a, -b, and -c, this reaction reflects the total amount of MAGE-D4 mRNA. Quantitative RT-PCR was performed with the SYBR-Green PCR core reagents kit (PerkinElmer, Applied Biosystems, Foster City, CA, USA) as follows: one cycle at 50°C for 2 minutes; one cycle
© 2013 Wiley Periodicals, Inc. and the International Society for Diseases of the Esophagus
190
Diseases of the Esophagus
at 95°C for 10 minutes; then 45 cycles at 95°C for 15 seconds; and 60°C for 30 seconds. Real-time detection of the SYBR-Green emission intensity was conducted with an ABI prism 7000 Sequence Detector (Perkin-Elmer, Applied Biosystems). mRNA expression was standardized to that of glyceraldehyde3-phosphate dehydrogenase (GAPDH) mRNA (TaqManR, GAPDH control reagents, Applied Biosystems). Quantitative RT-PCR was performed in triplicate and included no-template samples as negative controls. Expression levels for samples are shown as MAGE-D4 values standardized to MAGE-D4/GAPDH ratio. MAGE-D4 mRNA expression in tumor tissues and corresponding noncancerous tissues were compared and correlated with clinicopathological characteristics and prognoses.
Prognostic value of MAGE-D4 mRNA expression level was assessed by in a regression analysis. Overall survival (OS) rates were calculated using the Kaplan–Meier method; differences in survival curves were analyzed using the log-rank test. We performed multivariable regression analysis to detect prognostic factors using Cox proportional hazards models. Variables with P < 0.05 were entered into the final model. Associations between MAGE-D4 mRNA expression and clinicopathological parameters were evaluated using the χ2 test. Statistical analysis was performed using JMP® 10 software (SAS Institute Inc., Cary, NC, USA). P < 0.05 was considered significant.
RESULTS Immunohistochemical staining
Patient characteristics
Expression and distribution of MAGE-D4b protein, the putative dominant MAGE-D4 isoform, were analyzed immunohistochemically in representative cases and compared with mRNA expression patterns. Sections of formalin-fixed and paraffin-embedded tissues were dewaxed in xylene twice for 5 minutes, rehydrated in grading alcohols 100%, 90%, and 70% to H2O for 2 minutes each and subsequently treated with 3% H2O2 to inhibit endogenous peroxidase, and subjected to antigen retrieval using 10 mmol/L citrate buffer at 95°C for 10 minutes, five times. Sections were incubated with Histofine SAB-PO(R) (Nichirei, Tokyo, Japan) for 10 minutes, to limit non-specific reactivity, and then incubated for 1 hour with rabbit anti-MAGE-D4b polyclonal antibody (HPA003554, Sigma Aldrich) diluted 1:1000 in ChemMate antibody diluent (Dako, Carpinteria, CA, USA). Samples were then washed with phosphate-buffered saline, followed by a 10-minute incubation with biotinylated secondary antibody (Histofine SAB-PO(R), Nichirei). All antibody staining was developed for 2 minutes using liquid 3,3′-diaminobenzidine as substrate (Nichirei). Specimens were randomized and coded before analysis; analyses were conducted by two independent observers, who evaluated all specimens at least twice within a given time interval to minimize intra-observer variation. Expression level of MAGED4b protein was evaluated both in EC tissues and corresponding non-cancerous tissue. Components with ≥10% of stained cells were qualitatively defined as positive.
The mean age of the 65 patients was 64.1 ± 7.3 years (mean ± standard deviation; range: 46–80 years). The male-to-female ratio was 53:12. Fifty-three patients had preoperative symptoms, such as chest pain. Fifty patients had histories of excessive alcohol consumption; 28 patients had Brinkman index ≥1000. When classified by the seventh edition of the UICC classification, 5, 20, 30, and 10 patients were in stages I, II, III, and IV, respectively.
Statistical analysis Relative mRNA expression levels (MAGE-D4/ GAPDH) were calculated from quantified data. Differences in MAGE-D4 mRNA expression between cancerous tissues and corresponding non-cancerous tissues were analyzed by the Mann–Whitney U-test.
MAGE-D4 mRNA expression analyzed by quantitative RT-PCR Expression analysis of MAGE-D4 mRNA in EC cell lines was performed to figure out whether EC involves aberrant expression of MAGE-D4 or not. Expression levels of MAGE-D4 mRNA in nine EC cell lines were shown in Figure 1A with the median value and range of expression level in 65 normal esophageal tissues, demonstrating the heterogeneity of MAGE-D4 expression in EC cell lines. Mean expression of MAGE-D4 mRNA was significantly higher in EC tissues than in corresponding normal tissues (P < 0.001; Fig. 1B). Prognostic impact of overexpression of MAGE-D4 mRNA We first analyzed prognostic impact of MAGE-D4 mRNA expression level in EC tissues as a continuous variable in a regression analysis, and the expression level was significantly associated with OS (P = 0.011). Next, we compared two patents group categorized by the median value of MAGE-D4 mRNA expression level in EC tissues. Although the high expression group tended to have a shorter OS, there was no significant difference between the two groups (P = 0.192). Then we adopted the quartile analysis. Patients with the highest MAGE-D4 mRNA
© 2013 Wiley Periodicals, Inc. and the International Society for Diseases of the Esophagus
MAGE-D4 expression in esophageal cancer
Fig. 1 Analysis of melanoma-associated antigen (MAGE)-D4 messenger RNA (mRNA) in esophageal cancer cell lines. (A) Expression levels of MAGE-D4 mRNA in nine esophageal cancer cell lines were shown along with the median value in 65 normal epithelial tissues. The error bars represent the range. (B) In 65 surgical specimens, mean expression of MAGE-D4 mRNA was significantly higher in cancerous tissues than in corresponding normal tissues (P < 0.001). The error bars represent the standard deviation.
expression in EC tissues (fourth quartile, n = 17) had significantly shorter OS than patients with low (first– third quartiles, n = 48) MAGE-D4 mRNA expression (2-year OS: 44% and 73%, respectively, P = 0.006; Fig. 2A). Patients with the highest MAGE-D4 mRNA expression in EC tissues tended to have shorter recurrence-free survival (RFS) than patients with low MAGE-D4 mRNA expression (2-year RFS: 66% and 79%, respectively, P = 0.058; Fig. 2B). Patients were grouped by age, gender, preoperative symptom, Brinkman index (≥1000), excessive alcohol consumption, carcinoembryonic antigen (CEA; >5 ng/mL), squamous cell carcinoma-related antigen (>1.5 ng/mL), tumor size ( ≥ 5.0 cm), T factor (T3–4), tumor differentiation (poor), lymphatic involvement, vessel invasion, intra-epithelial spread, lymph node metastasis, and MAGE-D4 mRNA expression. Of these variables, univariate analysis identified age ≥65 years, lymphatic involvement, and high MAGE-D4 mRNA expression as significant prognostic factors. In multivariable analysis, high MAGE-D4 mRNA expression was identified as an independent prognostic factor in multivariable analy-
191
Fig. 2 Survival curves for 65 patients with esophageal cancer. Patients with high melanoma-associated antigen (MAGE)-D4 messenger RNA (mRNA) expression in cancerous tissues (top quartile, n = 17) had significantly shorter (A) overall survival (P = 0.006) and (B) recurrence-free survival (P = 0.058) than patients with low MAGE-D4 mRNA expression.
sis (hazard ratio: 2.194; P = 0.039; Table 1), together with lymphatic involvement. Association between MAGE-D4 mRNA expression and clinicopathological factors Associations between MAGE-D4 mRNA expression and clinicopathological parameters in the 65 patients are shown in Table 2. High MAGE-D4 mRNA expression was significantly associated with Brinkman index ≥1000 (P = 0.008) and preoperative CEA level >5 ng/mL (P = 0.002). Immunohistochemical staining MAGE-D4b protein expression was also evaluated immunohistochemically in representative cases with MAGE-D4 mRNA overexpression, reduced expression or equivalent expression in EC tissues compared with the corresponding non-cancerous tissues. In accord with the mRNA results, MAGE-D4b protein expression was confirmed exclusively in the membrane and cytoplasm of EC tissues that overexpressed
© 2013 Wiley Periodicals, Inc. and the International Society for Diseases of the Esophagus
192
Diseases of the Esophagus
Table 1 Prognostic factors for 65 patients with squamous cell carcinoma of the esophagus Univariate
Multivariable
Variable
n
Hazard ratio
95% CI
P-value
Hazard ratio
95% CI
P-value
Age (≥65) Gender (male) Preoperative symptom Brinkman index (≥1000) Excessive alcohol consumption CEA (>5 ng/mL) SCC (>1.5 ng/mL) Tumor size (≥5.0 cm) T factor (T3-4) Tumor differentiation (poor) Lymphatic involvement Vessel invasion Intra-epithelial spread Lymph node metastasis High MAGE-D4 mRNA expression (fourth quartile)
29 53 53 28 50 8 21 39 47 11 54 26 32 44 17
2.26 2.27 0.95 1.96 0.77 1.57 0.69 0.75 1.24 2.33 5.08 1.11 1.02 1.89 2.62
1.11–4.67 0.89–7.68 0.43–2.39 0.98–4.02 0.36–1.84 0.58–3.57 0.29–1.47 0.37–1.54 0.59–2.83 0.98–5.01 1.53–31.4 0.54–2.23 0.51–2.06 0.88–4.49 1.26–5.32
0.024* 0.093 0.905 0.059 0.535 0.346 0.345 0.423 0.582 0.056 0.005* 0.773 0.956 0.104 0.011*
1.89 — — — — — — — — — 4.53 — — — 2.19
0.92–3.94 — — — — — — — — — 1.36–28.1 — — — 1.04–4.49
0.081 — — — — — — — — — 0.010* — — — 0.039*
Univariate analysis was performed using the log-rank test. Multivariate analysis was performed using the Cox proportional hazards model. *Statistically significant (P < 0.05). CEA, carcinoembryonic antigen; CI, confidence interval; MAGE, melanoma-associated antigen; mRNA, messenger RNA; SCC, squamous cell carcinoma-related antigen.
MAGE-D4 mRNA (Fig. 3A,B). Conversely, samples with low MAGE-D4 mRNA expression showed little MAGE-D4b protein expression in either EC or noncancerous components (Fig. 3C).
DISCUSSION MAGE-D4 is a novel and unique member of the MAGE family that was discovered in 2001 and is specifically expressed in brain and ovary.31,32
MAGE-D4 is reportedly overexpressed in some human malignancies, including gliomas, non-smallcell lung cancers, breast cancer, and oral squamous cell carcinomas.29–31 Among MAGE family members, MAGE-D1 (also known as neutrophin receptorinteracting MAGE homologue) showed the greatest homology to MAGE-D4.37 MAGE-D1 has a 78% similarity to MAGE-D4, but lacks the N-terminal region, which implies that MAGE-D4 has a function in cell division and proliferation different from that of MAGE-D1. Previous experimental studies showed
Fig. 3 Immunohistochemical staining of melanoma-associated antigen (MAGE)-D4b in representative cases. (A, B) Specimen in which MAGE-D4 messenger RNA (mRNA) overexpression was detected by quantitative reverse transcription polymerase chain reaction (RT-PCR). Strong MAGE-D4b protein expression was confirmed in membranes and cytoplasm of cancerous components (A; upper 40×; lower 200×; B; intramural metastasis, 100×). (C) Specimen with equivalent expression of MAGE-D4 mRNA. Neither cancerous nor normal components showed MAGE-D4b protein expression (40×). © 2013 Wiley Periodicals, Inc. and the International Society for Diseases of the Esophagus
MAGE-D4 expression in esophageal cancer
193
Table 2 Association between MAGE-D4 mRNA expression and clinicopathological parameters in 65 patients with squamous cell carcinoma of the esophagus
Clinicopathological parameters Age 1.5 Tumor size
