Salvage esophagectomy after definitive chemoradiotherapy for synchronous double cancers of the esophagus and head-and-neck.

Head-and-neck cancer is frequently associated with esophageal cancer. Because the operative procedures for these synchronous double cancers are too invasive, definitive chemoradiotherapy tends to be applied as an initial treatment. A salvage esophagectomy for either recurrent or residual disease after definitive chemoradiotherapy in patients with such double cancer has never been reported. We reviewed 21 patients with esophageal cancer who underwent a salvage esophagectomy after definitive chemoradiotherapy. Among them, the treatment course of five patients who underwent a salvage esophagectomy for patients with synchronous double cancers of the esophagus and head-and-neck region was analyzed. Because head-and-neck cancer was well controlled after chemoradiotherapy in all five patients, a salvage esophagectomy was indicated for either recurrent or residual esophageal cancer after definitive chemoradiotherapy. Anastomotic leakage developed in four patients; however, no other complications including pulmonary complications were recognized. All of them were discharged to home and three of them are still alive without any recurrence for 20-43 months. A salvage esophagectomy should be considered as a treatment option for either recurrent or residual esophageal cancer with well-controlled head-and-neck cancer after definitive chemoradiotherapy when complete resection of the esophagus is expected.

Acute lung injury following an esophagectomy for esophageal cancer, with special reference to the clinical factors and cytokine levels of peripheral blood and pleural drainage fluid.

Acute lung injury (ALI) is one of most serious complications to occur after an esophagectomy for esophageal cancer. However, the pathogenesis of ALI is still unclear. The cytokine levels of pleural drainage fluid as well as peripheral blood were measured in 27 patients who had undergone an extended radical esophagectomy. Both the clinical factors and cytokine levels were compared between 11 patients with (group I) and 16 without ALI (group II). ALI occurred more frequently in patients who underwent colon interposition than in those who received a gastric tube reconstruction (86%vs 25%, P = 0.009). The operation time of group I was significantly longer than that of group II. A logistic regression analysis revealed colon interposition to be an independent factor associated with the ALI (P < 0.05). Postoperative anastomotic leakage and systemic inflammatory response syndrome (SIRS) occurred more frequently in group I than in group II (P < 0.01). Both the serum interleukin-6 (IL-6) and IL-8 levels of group I were significantly higher than those of group II. IL-1beta and tumor necrosis factor-alpha were undetectable in the peripheral blood, whereas they were detectable in the pleural effusion. The IL-1beta of pleural effusion was higher in group I than group II. In conclusion, greater surgical stress, such as a longer operative time, is thus considered to be associated with the first attack of ALI. The adverse events developing in the extra-thoracic site, such as necrosis and local infection around anastomosis may therefore be the second attack. Furthermore, ALI may cause not only SIRS but also other complications such as anastomotic leakage.

Salvage esophagectomy after definitive chemoradiotherapy for esophageal cancer.

Salvage esophagectomy is performed for esophageal cancer after definitive chemoradiotherapy. The clinical significance and safety of salvage surgery has not been well established. We reviewed 14 cases of salvage esophagectomy following definitive chemoradiotherapy from 1994 through 2005 and investigated complication rates and outcomes. Seven of 14 cases were completely resected with salvage surgery. Operation time and bleeding were greater in patients who experienced incomplete resection (R1/R2). Anastomosis leakage, pulmonary dysfunction and heart failure were recognized in four, two and one patients, respectively. The postoperative complications were more frequent (71.4%) in patients with incomplete resection (R1/R2) than in patients with complete resection (R0) (28.4%). Two patients with complete resection (R0) showed long-term survival. Salvage esophagectomy may be indicated when the tumor can be resected completely after definitive chemotherapy. However, all cases of T4 cancer cannot be resected completely, resulting in a high risk for complications and poor survival.

Reduced tau expression in gastric cancer can identify candidates for successful Paclitaxel treatment.

A recent study disclosed that breast cancer cases with low 'tau' expression can predict susceptibility to Paclitaxel administration. In the current study, the clinical significance of tau expression in gastric cancer cases was established by identifying candidates with Paclitaxel administration. Tissue specimens from 20 cases of in-operable or noncuratively resected gastric cancer were examined. Subsequent to the administration of 80 mg m(-2) of Paclitaxel in six 3-h intravenous infusions, the clinical effectiveness of Paclitaxel was evaluated by the size of metastatic lesions with computed tomography. The status of the tau expression was determined by immunohistochemistry. Based on a previously reported classification scheme, six were classified as tau-negative expression (0, 1+) cases and 14 were classified as tau-positive expression (2+, 3+) cases. All six (100%) cases of tau-negative expression showed a favourable response (partial response or minor response) to Paclitaxel administration. However, 12 (86%) of the 14 cases of tau-positive expression showed progressive disease (n = 11) or no change (n = 1) after Paclitaxel administration. The serum carcinoembryonic antigen values of the six cases of tau-negative expression were markedly decreased in comparison to the 14 tau-positive cases. These data indicate that tau-negative expression can be used to select gastric cancer patients, which will favourably respond to Paclitaxel treatment.

Overexpression of HRad17 mRNA in human breast cancer: correlation with lymph node metastasis.

PURPOSE: A novel human gene, designated HRad17, was identified as the human homologue of the Rad17 of Schizosaccharomyces pombe and Rad24 of Saccharomyces cerevisiae. In yeast, these genes play a critical role in maintaining genomic stability. The aim of this study was to evaluate the expression of HRad17 in human breast cancer. EXPERIMENTAL DESIGN: We investigated HRad17 mRNA expression in 64 cases of human breast cancer by means of reverse-transcription-PCR, in situ hybridization, and immunohistochemistry. RESULTS: The HRad17 mRNA was overexpressed in 35 cases (54.7%). Twenty-four (68.6%) of 35 cases with HRad17 overexpression in cancer tissues were node-positive, whereas only 8 (27.6%) of 29 cases without HRad17 overexpressions were node-positive. The expression of HRad17 mRNA correlated with both lymph node metastasis (P = 0.001) and high Ki67 labeling index (P = 0.006). Although not significantly different, expression of HRad17 mRNA tended to correlate with tumor size (P = 0.06) and expression of mutant p53 protein (P = 0.10). Furthermore, expression of HRad17 mRNA was an independent predictor of axillary lymph node metastasis as well as of lymphatic permeation by multivariate analysis (P < 0.0001). CONCLUSIONS: Our study demonstrates that HRad17 might be related to the development of lymph node metastasis in human breast cancers. Although its function still remains unclear, the expression of HRad17 mRNA could open up a new window for the diagnostic staging and treatment of human breast cancers.

Expression of multiple cancer-testis antigen genes in gastrointestinal and breast carcinomas.

Cancer-testis antigens (CTAs) such as MAGE are selectively expressed in various types of human neoplasms but not in normal tissues other than testis. This characteristic feature of CTAs makes them promising antigens for cancer-specific immunotherapy. A critical requirement for this therapy is identification of promising antigens. In this study, we investigated the expression of 6 genes recently identified by serological analysis of antigens by recombinant expression (SEREX) libraries: NY-ESO-1, LAGE-1, SCP-1, SSX-1, SSX-2, and SSX-4, in many surgical samples of gastrointestinal and breast carcinomas using reverse transcription-polymerase chain reaction. We found relatively high expression of SCP-1 (23.5%) and SSX-4 (20.6%) in gastric carcinoma, LAGE-1 (39.1%) and NY-ESO-1 (23.9%) in oesophageal carcinoma, and SCP-1 (34.1%) in breast carcinoma. We also found frequent synchronous expression with MAGE, including LAGE-1 (46.2%) in oesophageal carcinoma, SSX-4 (46.7%) in gastric carcinoma, and SCP-1 (38.3%) in breast carcinoma. Immunohistochemical analysis of the tumour samples expressing both MAGE-4 and NY-ESO-1 genes demonstrated differences in distribution between MAGE-4 and NY-ESO-1 in serial sections. We concluded that NY-ESO-1, LAGE-1, SCP-1 and SSX-4 genes may be promising candidates for cancer-specific immunotherapy in addition to MAGE, and that polyvalent cancer vaccines may be useful in cases of heterogeneous expressions of CTA genes in gastrointestinal and breast carcinomas.

Dendritic cell vaccination with MAGE peptide is a novel therapeutic approach for gastrointestinal carcinomas.

The MAGE gene is selectively expressed in cancer tissues such as melanoma or gastrointestinal carcinomas, whereas no expression is observed in normal tissues except testis. There are several reports of successful induction of HLA class I-restricted antitumor CTLs using MAGE peptides, and some clinical trials with these immunogenic peptides were reported as effective for some patients with malignant melanoma. However, there are no similar studies in gastrointestinal carcinomas, which are important neoplasms. Autologous dendritic cells (DCs) were generated ex vivo and were pulsed with MAGE-3 peptide, depending on the patient's HLA haplotype (HLA-A2 or A24). Patients were immunized with DC pulsed with MAGE-3 peptide every 3 weeks at four times. Twelve patients with advanced gastrointestinal carcinoma (six stomach, three esophagus, and three colon) were treated, and no toxic side effects were observed. Peptide-specific CTL responses after vaccination were observed in four of eight patients. Improvement in performance status was recognized in four patients. Tumor markers decreased in seven patients. In addition, minor tumor regressions evidenced by imaging studies were seen in three patients. These results suggested that DC vaccination with MAGE-3 peptide is a safe and promising approach in the treatment of gastrointestinal carcinomas.

Expression of Fas ligand in gastric carcinoma relates to lymph node metastasis.

Tumors may escape a host's immune response by means of various mechanisms. The Fas (CD95/APO-1)/Fas ligand (FasL) system is one of the major apoptotic pathways. Recently, it has been reported that tumor cells can express FasL, induce apoptosis in tumor infiltrating lymphocytes, and thus can escape host immune surveillance. In gastric carcinoma, tumor progression by way of the lymphatics is often seen, and lymph node metastasis is a critical factor influencing the recurrence of cancer and its prognosis. We, therefore, investigated the relationship between the expression of FasL and the lymphatic spread of gastric carcinoma. FasL-expression was examined by an immunohistochemical method using 100 surgically resected gastric carcinomas and 55 metastatic lymph nodes. Apoptotic cells among tumor infiltrating T lymphocytes were detected by T lymphocyte staining and the terminal deoxynucleotidyl transferase (TdT) mediated dUTP nick end labeling (TUNEL) method in a series of sections of metastatic lymph nodes. FasL-expression was detected in 86% of primary lesions and 71% of metastatic lymph nodes. In cases with high levels of FasL-expression, the observed expression of lymph node metastases was significant (p=0.047). Moreover, FasL-positive cases with both primary lesions and metastatic lymph nodes showed also distant lymph node metastasis beyond the regional lymph nodes (p=0.030). Apoptosis among tumor infiltrating T lymphocytes was more frequently seen in FasL-positive lesions (p=0.057). Furthermore, patients with FasL-negative primary lesions tended to exhibit longer survival times than patients with FasL-positive primary lesions. The results suggest tumor escape through the lymphatic pathway via FasL-expression in gastric carcinomas.

Expression and function of galectin-3, a beta-galactoside-binding protein in activated T lymphocytes.

A soluble beta-galactoside-binding lectin, galectin-3 has been shown to be involved in cell adhesion and activation of immune cells. Although galectin-3 is known to be expressed in various types of cells, it has not been shown whether galectin-3 is expressed in T lymphocytes. We present evidence here that galectin-3 is expressed in activated murine T lymphocytes including CD4+ and CD8+ T cells but not in resting T cells. Galectin-3 expression was induced by anti-CD3 mAb or mitogen and enhanced by common gamma-chain signaling cytokines, IL-2, IL-4, and IL-7, in activated T lymphocytes, whereas the inflammatory cytokines including TNF-alpha and IFN-gamma did not. Galectin-3 expression and proliferation were down-regulated by withdrawal of IL-2 and gamma irradiation. Antisense but not sense phosphorothioated oligonucleotides for galectin-3 inhibited galectin-3 expression and blocked proliferation of T cells significantly. This study suggests that up-regulation of galectin-3 plays an important role in proliferation of activated T lymphocytes.

Heparanase expression in clinical digestive malignancies.

Recently, mammalian heparanase was cloned, and its probable function in tumor progression was reported. However, its expression in human clinical cancers has not been fully studied. Thus we determined the heparanase mRNA expression in 30 esophageal cancer cell lines and 144 clinical samples including 38 esophageal squamous cell carcinomas, 71 gastric adenocarcinomas, and 35 colorectal adenocarcinomas. The fresh surgical specimens of cancer tissue (T) and its paired normal tissue (N) were used. The heparanase mRNA was evaluated by reverse transcriptase-polymerase chain reaction, and the T/N expression ratio was determined in clinical cases. All 30 esophageal cancer cell lines expressed heparanase mRNA. The T/N ratio was determined as high (> or =1.3), equal (0.8 approximately 1.2) or low (< or = 0.7) in each clinical case. In cases of esophageal cancer, 7 showed high, 10 equal and 21 low expression. In cases of colorectal cancer, 3 showed high, 16 equal and 16 low expression. On the other hand, 42 showed high, 22 equal and 7 low expression in cases of gastric cancer. The frequency of high expression cases was greater in gastric cancer than in esophageal or colorectal cancers (p < 0.05). There were no differences in clinicopathologic factors including prognosis between high and low expression cases of each cancer. Our mRNA study of heparanase indicated that its expression status was different among gastrointestinal cancers. The clinical and pathological impact was low compared to other proteinases, although further studies are recommended for final conclusion.

Expression of MAGE-B genes in esophageal squamous cell carcinoma.

The MAGE-B (MAGE-B1, -B2, -B3, and -B4) genes share strong homology with the MAGE-A gene family. MAGE-B1 and -B2 encode common tumor-specific peptide antigens. There is, however, still very little information about the expression of these genes in human gastro-intestinal carcinomas. We investigated the expression of MAGE-B1 and -B2 genes in 29 cell lines and 53 clinical tumor samples of esophageal squamous cell carcinoma by reverse transcription polymerase chain reaction (RT-PCR). MAGE-B1 and -B2 gene transcripts were detected by RT-PCR in 1 (3%) and 6 (21%) cell lines, and in 9 (17%) and 17 (32%) clinical samples, respectively. Among them, 7 / 29 (24%) cell lines and 19 / 53 (36%) clinical samples expressed at least either MAGE-B1 or -B2. A significant correlation was found between negative MAGE-B gene expression and vascular invasion (P = 0.008). In 45 out of 53 esophageal carcinoma RNA samples, the MAGE-A1, -A2, and -A3 genes were detected in 27 (60%), 23 (51%), and 30 (67%) samples, respectively, while the MAGE-B genes were detected in 18 (40%) samples. The frequency of MAGE-B gene expression in esophageal carcinoma was relatively higher than that observed for gastric or colorectal carcinomas (12% and 2%, respectively). Therefore, the MAGE-B genes could be used as targets in specific immunotherapy of esophageal squamous cell carcinomas.

Prognostic impact of tissue inhibitor of matrix metalloproteinase-1 in esophageal carcinoma.

Tissue inhibitor of metalloproteinase-1 (TIMP-1) inhibits the activity of matrix metalloproteinase, which may play an important role in carcinoma invasion and metastasis. TIMP-1 is thus considered to inhibit carcinoma invasion and metastasis. However, TIMP-1 possesses another important function, cell growth promotion. The clinical significance of TIMP-1 expression has not been fully determined in esophageal carcinoma. We thus examined the expression of TIMP-1 mRNA in tumor (T) and corresponding normal (N) tissues of 85 esophageal carcinoma cases by RT-PCR. The T:N ratio of TIMP-1 mRNA expression in each case was evaluated semi-quantitatively with adjustment by an internal control gene. The mean T:N ratio was 2.0 (range 0.2-6.5). When comparing high-expression cases (T:N > 2.0, n = 37) with low-expression cases (T:N < or = 2.0, n = 48), the former showed a significantly higher frequency of lymph vessel invasion, vascular vessel invasion, lymph node metastasis and advanced-stage disease. The former cases showed a poorer prognosis than the latter. Multivariate analysis disclosed that TIMP-1 expression status was an independent determining factor for prognosis. Our findings suggest that TIMP-1 expression correlates with tumor extension of esophageal carcinoma and might, if validated, prove useful as a novel prognostic marker for esophageal carcinoma.

MAGE-specific cytotoxic T lymphocytes require non-specific CD54 adherence receptors on carcinoma cells.

We investigated several factors that could regulate the susceptibility of carcinoma cells to MAGE-2-specific CTL mediated lysis. Cytofluorometric analysis showed that the cytolysis correlated with the expression of CD54. IFN-gamma treatment induced the TAP-1 and LMP-2 genes, continuously up-regulated the HLA class I expression and increased cytolysis. Although HLA class I were highly induced in MRKnu1 cells, CD54 was not induced and the cytolysis was minimal. Cytotolysis of IFN-gamma-treated MKN-1 cells was completely inhibited by a monoclonal anti-CD54 antibody. These results suggest that HLA-restricted CTL lysis requires non-specific CD54 adherence receptors in addition to specific TCR signals.

Expression of tissue inhibitor of matrix metalloproteinase-1 in human breast carcinoma.

Tissue inhibitor of metalloproteinase-1 (TIMP-1) is an inhibitor of matrix metalloproteinase (MMP) in human carcinomas. TIMP-1 is thus considered to inhibit carcinoma invasion and metastasis. On the other hand, recent reports have disclosed that TIMP-1 also possesses a growth-promoting function. The clinical significance of TIMP-1 expression has not been fully determined in breast carcinoma. We thus examined the expression of TIMP-1 mRNA in tumor tissues of 100 breast carcinoma cases by a reverse transcriptase-polymerase chain reaction assay. The expression of TIMP-1 mRNA in each case was evaluated semi-quantitatively with adjustment for the TIMP-1 expression in a control breast carcinoma cell line, MCF7. There was a significant inverse correlation between the TIMP-1 expression and lymph node metastasis (p<0.05). A multivariate analysis disclosed that TIMP-1 expression status was an independent determinant factor for lymph node metastasis. In addition, the tumors with positive estrogen or progesterone receptors showed a higher TIMP-1 mRNA expression than those without the receptors, but this did not reach statistical significance. The findings suggest that the breast tumors with high TIMP-1 expression might show less malignant potential than those with low TIMP-1 expression.

RT-PCR detection of breast cancer cells in sentinel lymph modes.

Sentinel lymph node (SLN) biopsy is being evaluated in breast cancer patients to improve detection of metastases and to guide therapy with minimal morbidity. The aim of this study was to increase the sensitivity of tumor cell detection in SLNs using superior reverse transcription polymerase chain reaction (RT-PCR) for carcinoembryonic antigen (CEA) and mammaglobin (MMG) analysis rather than current methods which fail to identify clinically relevant disease in many patients. In seventy stage I and II breast cancer patients dye-guided lymph node mapping was performed and the SLNs were divided alternately for RT-PCR or hematoxylin and eosin staining (H&E). RT-PCR and H&E diagnosis of SLNs were compared. SLNs were identified in 66/70 (94.3%) patients. Seventeen patients (26. 2%) had histological metastasis in SLNs. CEA was expressed in 25.0% of 48 patients with H&E negative SLNs, and MMG was expressed in 20. 8%. SLNs could predict axillary lymph node status with 95.4% accuracy and 6.3% false negative rate by H&E. Moreover, RT-PCR improved these to 98.5% and 2.8%, respectively. SLN diagnosis using RT-PCR is a powerful and sensitive method, which increases the accuracy of clinical staging and may provide more informed choices for appropriate therapeutic management of breast cancer patients.

The heterogeneous expression of MAGE-3 protein: difference between primary lesions and metastatic lymph nodes in gastric carcinoma.

We evaluated the expression of MAGE-3 protein in gastric carcinoma by immunohistochemistry, and compared the frequency between primary lesions and metastatic lymph nodes. Primary lesions and metastatic lymph nodes of 38 cases of gastric carcinoma were analyzed using anti-MAGE-3 monoclonal antibody. Expression of MAGE-3 protein was detected in 17 (45%) primary lesions and 12 (32%) metastatic lymph nodes of 38 cases. In 70% of positive cases of primary lesions, the percentage of stained cells with MAGE-3 decreased in metastatic lymph nodes. The heterogeneous expression of MAGE-3 should be considered when we perform antigen specific immunotherapy for patients with gastric carcinoma.

Local secretion of IFN-gamma induces an antitumor response: comparison between T cells plus IL-2 and IFN-gamma transfected tumor cells.

Previously we described that the adoptive transfer of tumor-infiltrating lymphocytes (TIL) + interleukin-2 (IL-2) leads to eradication of established methylcholanthrene (MCA)-105 fibrosarcoma pulmonary metastases in a congenic murine model. The in vivo efficacy of TIL was associated with their ability to secrete interferon-gamma (IFN-gamma), and to a lesser extent granulocyte-macrophage colony-stimulating factor. The local secretion of these cytokines resulted in recruitment of naive host immune cells to the tumor and eventually in a successful host antitumor immune response. In the present study, to further evaluate the role of IFN-gamma in the induction of a host antitumor immune response, we compared the treatment efficacy of adoptively transferred T cells and IFN-gamma gene transfected tumor cells (MCA-105/IFN-gamma) as delivery systems of IFN-gamma. Treatment with TIL-IL-2 or irradiated MCA-105/IFN-gamma induced a similar reduction in pulmonary metastases of MCA-105 tumor. In contrast, irradiated wild-type MCA-105 or TIL from IFN-gamma gene knockout mice did not cause tumor eradication. MCA-105 tumor-bearing mice treated with MCA-205/IFN-gamma showed a partial reduction in the number of pulmonary metastases. Histologically, lungs of successfully treated mice showed that initially activated macrophages expressing inducible nitric oxide synthase (iNOS) and dendritic cells infiltrated the tumor bed. Subsequently, CD4+ and CD8+ T cells infiltrated tumors. The therapeutic efficacy of IFN-gamma transfected tumor cells was eliminated when either CD4+ T cells or CD8+ T cells were depleted. These results suggest that local secretion of IFN-gamma induces a tumor-specific host antitumor immune response mediated through activated macrophages, dendritic cells, and tumor-specific T cells. This may be a common component of successful immunotherapy.

Flow cytometric DNA analysis is useful in detecting multiple genetic alterations in squamous cell carcinoma of the esophagus.

BACKGROUND: Although flow cytometric DNA analysis has been recognized to be a useful prognostic indicator for patients with squamous cell carcinoma of the esophagus, the biologic significance of DNA aneuploidy remains to be elucidated. METHODS: Twenty-five patients with squamous cell carcinoma of the esophagus who underwent a curative subtotal esophagectomy were divided into 2 groups according to the DNA ploidy pattern. Multiple genetic changes, including the gene amplification of bcl-1, epidermal growth factor receptor, and c-myc, and the loss of heterozygosity of multiple tumor suppressor genes, including retinoblastoma, mutated in colorectal carcinoma, adenomatous polyposis coli, and deleted in colorectal carcinoma, in each case were investigated and the frequency of genetic alterations compared between both groups. In addition, the clinical outcome of these patients was also investigated. RESULTS: Eleven of 15 cases in the aneuploid group demonstrated at least 1 genetic change (73.3%) whereas only 2 of 10 cases in the diploid group did so (20.0%) (P<0.05). Both cases in the diploid group with genetic alterations had only 1 genetic change of 7 tested genes whereas 9 of 11 cases in the aneuploid group had multiple genetic alterations. Patients in the aneuploid group also showed a more unfavorable prognosis than patients in the diploid group. CONCLUSIONS: Based on the findings of the current study, flow cytometric DNA analysis is considered to be useful for both detecting multiple genetic alterations and predicting the prognosis of patients with carcinoma of the esophagus.

Expression of the MAGE gene family in human hepatocellular carcinoma.

BACKGROUND: The 12 members of the MAGE gene family encode tumor specific antigens that are recognized by autologous cytotoxic T lymphocytes (CTL). The MAGE genes are expressed not only in melanoma but in the other malignant tumors as well. There is, however, little information on their expression in hepatocellular carcinoma (HCC). The authors thus studied the expression of the MAGE gene family in human HCC and discuss the possibility of specific immunotherapy using MAGE peptides. METHODS: Tumor tissue samples of HCC and paired nontumor tissues of the liver were obtained from 22 HCC patients. Total RNA was extracted and cDNA was synthesized. Polymerase chain reaction amplification using each MAGE gene specific primer was then performed to detect the expression of each MAGE gene. Immunoblotting and immunohistochemical analysis were performed to confirm the expression of MAGE-3 gene product in HCC. RESULTS: The expression rate of each MAGE gene was as follows: MAGE-1 and -3 were expressed in approximately 68% of the tumors; MAGE-8 was expressed in 46%; and MAGE-2, -6, -10, -11, and -12 were expressed in approximately 30%. Nineteen (86%) of 22 tumors expressed at least 1 MAGE gene. On the other hand, no expression was detected in the noncarcinomatous liver tissue specimens. Actual expression of the gene product of MAGE-3 was detected in 50% of tumors. Clinicopathologic data on the MAGE positive and negative cases were compared. Significant differences were observed between MAGE expression status and a few clinicopathologic factors; however, further investigation is required to elucidate these correlations completely. CONCLUSIONS: These findings demonstrated that MAGE gene expression is frequent in HCC, thus suggesting that HCC patients may be good candidates for specific immunotherapy using MAGE peptides.