ANTICANCER RESEARCH 25: 489-496 (2005)

Serum -C as a Potential Predictive Marker of in Non-small Cell Lung Cancer

TOSHIJI ISHIWATA, KAZUHISA TAKAHASHI, YURI SHIMANUKI , RINA OHASHI, RI CUI, FUMIYUKI TAKAHASHI, KAZUE SHIMIZU, KAYO MIURA and YOSHINOSUKE FUKUCHI

Department of Respiratory Medicine, Juntendo University, School of Medicine, 2-1-1 Hongo, Bunkyo-Ku, Tokyo 113-8421, Japan

Abstract. Background: Tenascin (Tn)-C is an extracellular Primary lung cancer is the leading cause of cancer-related matrix that is involved in tissue interactions during fetal deaths in Japan (1, 2). Non-small cell lung cancer (NSCLC) development and oncogenesis. However, the role of serum Tn- is the most frequent type of lung cancer, and is treated by C in non-small cell lung cancer (NSCLC) has not been surgical resection for patients diagnosed as early stage clarified. Patients and Methods: In this study, we determined carcinoma. Unfortunately, more than 50% of NSCLC the serum levels of Tn-C among NSCLC patients who patients who undergo surgery will eventually experience underwent surgery, as well as other factors implicated for local relapse or distant metastasis. This high metastatic angiogenesis, to address the clinical implications in NSCLC. potential of NSCLC may be due to the active angiogenesis Results and Conclusion: The median concentration of serum of the tumors, in which many angiogenic factors such as Tn-C in NSCLC patients was slightly higher than that of vascular endothelial growth factor (VEGF) and normal controls, but this difference was not statistically (OPN) play an important role (3, 4). Adjuvant combination significant. There was a positive correlation between serum Tn- using cisplatin and new drugs has recently C levels and microvessel density (MVD), serum osteopontin been approved and demonstrated to be effective in the (OPN) and vascular endothelial growth factor (VEGF). In improvement of patient survival (5, 6). However, not all contrast, there was no correlation between serum Tn-C levels patients can receive survival benefit by adjuvant and serum carcinoembryonic antigen (CEA) and sialyl lewis-X chemotherapy. These results indicate the need for a more (SLX) levels. The overall survival of patients with low Tn-C valuable surrogate serum marker for angiogenesis in order levels (<96 ng/ml) was significantly greater than that of to select an appropriate neoadjuvant therapy including anti- patients with high Tn-C levels (≥96 ng/ml). Intratumoral Tn- angiogenic treatment. C expression was co-localized with expression of microvessels Tenascin (Tn)-C is a large, secreted oligomeric extracellular in the stroma of the cancer cells by immunohistochemical matrix protein (ECM) that is expressed in the developing analysis. Moreover, enhanced in vitro migration of human brain, cartilage, and mesenchyme and is re-expressed in umbilical vascular endothelial cells (HUVEC) was induced by tumors, wound healing and inflammation (7). Tn-C consists recombinant Tn-C. Collectively, Tn-C may play an important of several isoforms derived from alternative splicing of its role in angiogenesis of patients with NSCLC, and the mRNA (8). It has recently been demonstrated that the Tn-C determination of serum Tn-C may be useful in predicting high molecular weight isoform regulates angiogenesis, intratumoral vasculature and patients’ prognosis. resulting in up-regulation of tumorigenesis of various cancers (8-14). Studies concerned with the relationship between Tn-C serum concentration and the clinical stage of the tumor for colon carcinoma, malignant melanoma and squamous cell Correspondence to: Kazuhisa Takahashi, M.D., Ph.D., Department carcinoma of the head and neck have been reported, but are of Respiratory Medicine, Juntendo University, School of Medicine, few in number (15-17). However, a limited number of reports 2-1-1 Hongo, Bunkyo-Ku, Tokyo 113-8421, Japan. Tel: +81-3- has focused on the involvement of Tn-C in lung cancer 5802-1063, Fax: +81-3-5802-1617, +81-3-3817-0706, e-mail: progression, particularly from the aspect of angiognesis (18- [email protected] 20). For instance, Kusugawa et al. (18) demonstrated the Key Words: Tenascin (Tn-C)-C, vascular endothelial growth factor expression of Tn-C and its degradation in human lung cancers. (VEGF), microvessel density (MVD), angiogenesis, non-small cell They investigated the correlation between the amount of lung cancer (NSCLC). protein and mRNA expression of Tn-C in surgically resected

0250-7005/2005 $2.00+.40 489 ANTICANCER RESEARCH 25: 489-496 (2005)

Table I. Relationship between clinicopathological findings and serum Tn- C level.

Clinical findings Serum Tn-C (ng/ml) n mean±SE p-value total 63 89±7 gender male 45 92±8.6 0.444 female 18 81±9.2 histology adeno 42 90±8.0 0.580 squamous 16 81±15 c-stage I A,B 43 89±7.9 0.880 Figure 1. Comparison of serum Tn-C levels between healthy volunteers II A,B 8 92±19 (n=49) and the patients with NSCLC (n=63). Serum Tn-C levels in III A 11 89±18 NSCLC patients who underwent surgery were slightly higher than those of healthy volunteers, but this difference did not reach statistical significance adeno: adenocarcinoma, squamous:squamous cell carcinoma (p=0.1189).

tumor specimens and the histopathological stage, but did not Blood samples. Blood samples were drawn preoperatively, evaluate its role in angiogenesis in lung cancer. Moreover, this centrifuged at 2500 rpm for 15 min and aliquots were frozen at method is not capable of predicting intratumoral angiogenic –80ÆC until analysis. The protocol was approved by the Committee for Medical Ethics of Juntendo University, School of activity prior to surgery. To the best of our knowledge, there Medicine, and informed consent was obtained from all subjects are no studies reporting on the association between serum Tn- enrolled in this study. C concentration and angiogenesis in the primary tumor of NSCLC. Measurement of serum Tn-C and VEGF concentrations. Serum In this study, we revealed a positive correlation for concentrations of Tn-C and OPN were measured with preoperative concentration of serum Tn-C in reference to commercially available sandwich ELISA kits (IBL, Gumma, Japan), according to the manufacturer’s instructions. Serum intratumoral angiogenesis and serum levels of other samples were analyzed for VEGF with commercially available angiogenic factors (VEGF and OPN). Interestingly, no sandwich ELISA kits (R&D Systems, Inc., Minneapolis, MN, correlation was demonstrated between Tn-C serum level USA). The limits of sensitivity were 0.38 ng/ml for Tn-C, 0.6 ng/ml and the clinical stage. We also clarified that preoperative for OPN and 9 pg/ml for VEGF. serum Tn-C is a valuable marker for the prediction not only of intratumoral angiogenesis, but also patient survival. Measurement of serum CEA and SLX. Serum concentrations of carcinoembryonic antigen (CEA) and sialyl lewis-X (SLX) were determined using the chemiluminescent immunoassay (CLIA) for Patients and Methods the former and radioimmunoassay (RIA) for the latter, according to the manufacturers’ instructions, respectively (CEA: Abbot Patients. A total of 63 patients with primary NSCLC (registered at Laboratories, North Chicago, IL, USA; SLX, Otsuka Juntendo University Hospital, Japan, from 1992 to 2002) were Pharmacoceutical Co., Tokyo, Japan). evaluated in this study, including 43 males and 20 females with a mean age of 65±10 years (Table I). For all patients, the diagnosis Immunohistochemical staining and microvessel counting. of lung cancer was confirmed by histological examinations of either Microvessels within the tumor tissue and Tn-C expression were biopsy or cytological specimens. Staging was determined based on evaluated by immunohistochemical analysis using the the new international staging system. The procedure included chest streptavidin-biotin method (8, 21). Briefly, paraffin-embedded radiography, computer tomography (CT) scans of the chest and tumor tissue samples were sectioned at the thickness of 3 Ìm and abdomen, magnetic resonance imaging of the brain and bone deparaffinized in xylene. Endogenous peroxidase activity was scanning. Adenocarcinoma was diagnosed in 43 patients, squamous blocked with 0.3% hydrogen peroxidase in methanol. The cell carcinoma in 15 patients, adenosquamous cell carcinoma in 2 sections were treated by autoclaving for 15 min at 120ÆC in 10 patients, large cell carcinoma in 1 patient, carcinosarcoma in 1 mM citrate buffer, pH 6.0, to retrieve the antigen. These sections patient and pleomorphic carcinoma in 1 patient. Eighteen patients were incubated with normal horse serum for 15 min at room were classified as Stage IA, 26 patients as Stage IB, 8 patients as temperature, and then reacted with anti-CD34 monoclonal Stage IIB and 11 patients as Stage IIIA. Forty-nine healthy antibody (QBEnd 10, Medical Biological Laboratories, Nagoya, volunteers, with mean age of 49 ± 2 served as controls. Japan) for 1 h at room temperature. After washing with PBS, the

490 Ishiwata et al: Serum Tenascin in Lung Cancer

Figure 2. The correlation between intratumoral microvessel density (MVD) (A), serum VEGF level (B) and serum OPN level (C) with serum Tn-C level in patients with NSCLC. A significant correlation was revealed between serum Tn-C level and intratumoral MVD (A: n=62, r=0.262, p=0.0469), serum VEGF level (B: n=62, r=0.457, P=0.0002) and serum OPN level (C: n=62, r=0.439, p=0.0004).

sections were treated with biotin-labelled horse anti-mouse previously described (22, 23). Briefly, 1 x 105 human umbilical immunoglobulin for 20 min. The sections were washed with PBS endothelial cells (HUVEC) were plated onto the inserts with 8 Ìm and streptavidin-peroxidase complex was added. Color pore membrane (Becton Dickinson, Franklin, NJ, USA). The development was performed with 3-3’ diaminobenzidine (Sigma, reverse sides of the membranes were coated with either St. Louis, MI, USA) in 0.03% hydrogen peroxide, and counter- recombinant Tn-C (100 Ìg/ml), VEGF (10 ng/ml) or bovine serum stained with hematoxylin. Tumor sections were microscopically albumin (BSA; 100 Ìg/ml). The lower chambers of the 24-well plate scanned and the areas with the highest microvessel density were filled with Opti-MEM containing 0.1% BSA. HUVEC were (MVD) were identified. Microvessel counts were performed in allowed to incubate at 37ÆC. After 6-h incubation, the cells were five areas with the highest vascular density at 100 magnification fixed with 10% neutralized buffered formalin for 15 min and were and expressed as a mean value. For double staining of CD34 and stained with 0.3% crystal violet for 10 min. After three washings Tn-C, sections stained with anti-CD34 antibody were washed with Milli-Q water, the remaining cells on the surface of the again with TBS (0.1M Tris-buffered saline, pH 8.2 containing membrane were wiped clean with cotton swabs. Stained cells were 0.05% Tween), followed by blocking with normal horse serum for calculated by low power magnification. 10 min at room temperature. These sections were incubated with anti-human Tn-C monoclonal antibody (4C8MC, IBL) for 1h at Statistical analysis. All data are presented as mean±SD. The room temperature. After washing with TBS, the sections were relationship between serum Tn-C and various clinicopathological treated with biotin-labelled horse anti-mouse immunoglobulin parameters were analyzed using the Mann-Whitney U-test with the for 20 min. The sections were washed with TBS and streptavidin- exception of c-stage. The Kruskal-Wallis method was used to alkaline phosphatase complex was added. Color development compare serum Tn-C levels with c-stage. Pearson’s correlation by was performed using Vector Blue alkaline phosphatase substrate rank was used to evaluate the correlations between serum Tn-C kit III (Vector Laboratories, Burlingame, CA, USA) in 0.03% levels and microvessel density (MVD), VEGF, osteopontin (OPN) hydrogen peroxide. or various other tumor markers. Overall patient survival was analyzed by Kaplan-Meier method using the log-rank test. The In vitro cell migration assay. In vitro cell proliferation assay was criterion of statistical significance was p<0.05. Statview version 5.0 performed with the Boyden chamber method modified, as (Abacus Conceps) was used for all analyses.

491 ANTICANCER RESEARCH 25: 489-496 (2005)

Figure 3. The correlation between serum Tn-C levels and serum tumor markers, CEA (A) and SLX (B) in patients with NSCLC. No correlation was demonstrated between serum CEA and SLX levels and serum Tn-C level (A: n=62, r=0.095, p=0.463; B: n=62, r=0.054, p=0.676).

Relationship between serum Tn-C level and various clinical factors in patients with NSCLC. The relationship between clinicopathological findings and serum Tn-C concentrations in all patients with NSCLC is shown in Table I. There was no difference in serum Tn-C levels among gender and histological type. Interestingly, there was no significant correlation between serum Tn-C level and the clinical-stage classification.

Correlation between serum Tn-C level and microvessel density (MVD), VEGF, OPN levels and various tumor markers in the tumor tissue of NSCLC. We assessed the relationship between serum Tn-C level and intratumoral MVD by immunohistochemical analysis. There was a weak correlation Figure 4. Kaplan-Meier overall survival curves of NSCLC patients between serum Tn-C level and intratumoral MVD (Figure according to serum Tn-C levels. NSCLC patients were categorized into 2A, r=0.262, p=0.0469). The correlation between serum Tn- two groups according to serum Tn-C levels. The overall survival of the C level and serum VEGF, OPN levels and various tumor group with low Tn-C level (n=47, <96 ng/ml) was significantly greater markers (CEA and SLX) was also analyzed. As expected, than that of the group with high Tn-C levels (n=16, ≥96 ng/ml, p=0.0141). there was a significant correlation between serum Tn-C and both VEGF (Figure 2B) and OPN level (Figure 2C) (r=0.457, p=0.0002, r=0.439, p=0.0004, respectively). In contrast, the serum Tn-C level did not correlate with either serum CEA (Figure 3A) or SLX level (Figure 3B) (r=0.095, Results p=0.463, r=0.054, p=0.676, respectively).

Comparison between serum Tn-C levels in patients with Correlation of serum Tn-C levels with overall survival of the NSCLC and those of healthy volunteers. We first compared patients with NSCLC. We analyzed the influence of serum serum Tn-C levels of patients with NSCLC with those of the Tn-C levels on overall survival according to the Kaplan- healthy volunteers. As shown in Figure 1, the median Meier method. A statistically significant difference in the concentration of serum Tn-C in NSCLC patients who average overall survival between patients with high Tn-C underwent surgery was slightly higher than that of the value (upper quartile group; ≥96 ng/ml, n=16) and patients healthy volunteers, although this difference did not reach with low Tn-C value (<96 ng/ml, n=47) was revealed statistical significance (p=0.1189). (Figure 4, p=0.0141).

492 Ishiwata et al: Serum Tenascin in Lung Cancer

Expression of Tn-C in NSCLC with immunohistochemical chemotherapy could be selected. However, only a limited staining. To investigate Tn-C expression in NSCLC tissues, number of studies have focused on the clinical immunohistochemical analysis was performed. implication of circulating Tn-C in serum (15-17, 26, 27). Immunoreactivity of Tn-C was observed mainly in the For instance, Burchardt et al. (16) have recently revealed stroma, but not in cancer cells. As shown in Figure 5, the that a significant increase in Tn-C serum level was majority of Tn-C was co-expressed with CD34-positive observed in patients with stage IV melanoma in endothelial cells in the stroma, suggesting the involvement comparison to healthy controls. Pauli et al. (17) reported of Tn-C in angiogenesis. on elevated levels of serum Tn-C in patients with higher tumor stages or recurrent squamous cell carcinoma of the In vitro HUVEC migration. As shown in Figure 6, HUVEC head and neck. In colorectal carcinoma, the preoperative migrated toward the reverse side of the membrane coated levels of serum Tn-C were demonstrated to reflect on the with not only recombinant VEGF but also Tn-C to a greater total tumor burden and correlate with metastatic disease degree than those coated with bovine serum albumin. activity (15). These results suggest that serum Tn-C may Addition of Tn-C (100 Ìg/ml) into the upper chamber be a potential tumor marker in certain cancers. However, abrogated the Tn-C-mediated HUVEC migration (data not there have been no studies defining the association of shown). These results suggest that Tn-C is directly involved circulating Tn-C with either intratumoral angiogenesis or in endothelial cell migration. prognosis of NSCLC. In this study, we first revealed a positive correlation of Tn-C serum levels with not only Discussion intratumoral MVD, but also with other angiogenic factors, VEGF and OPN. In contrast, we could not Tn-C has been reported to regulate angiogenesis in tumor demonstrate any relationship between Tn-C serum levels through its role in the regulation of VEGF expression and either CEA, SLX or clinical-stage. Collectively, the (24). Moreover, both Tn-C and VEGF are expressed Tn-C serum level may not reflect on the total tumor during the peak of angiogensis, and disappear when burden in NSCLC, and thus it may not be categorized as angiogensis ceases (13). In fact, several researchers have a tumor marker. This idea is supported by our demonstrated that Tn-C is involved in neoplastic immunohistochemical findings that intense and diffuse angiogenesis in various cancers including breast cancers, Tn-C staining was observed in the stroma but not in astrocytoma, and oral and pharyngeal squamous cell cancer cells of NSCLC. carcinoma (10-14, 25). However, no reports have ever Many researchers have investigated whether Tn-C investigated its clinical role in the development of lung expression in tissue specimens could be a potential cancer, particularly from the point of view of prognostic marker in various cancers (15, 25, 28-36). angiogenesis. In this study, we determined the serum However, the result still remains controversial. In some concentration of Tn-C, several angiogenic factors and cancers, high Tn-C expression in tumor has been reported tumor markers in patients with operable NSCLC. The to correlate with poor prognosis (28, 29, 31, 32). In contrast, relationship between serum Tn-C levels and the levels of high Tn-C expression has been associated with better angiogenic factors (e.g. VEGF and OPN) were revealed to prognosis in some other cancers (30, 34, 37). Recently, Cai be significantly correlated. Serum concentrations of Tn-C et al. (19) reported that degradation of Tn-C is associated correlated with intratumoral MVD, which represents with tumor recurrence in early stage NSCLC. This appears neoplastic angiogenic activity. Moreover, the to be the first paper describing the relationship between Tn- immunohistochemical analysis revealed that Tn-C is co- C expression and clinical outcome in lung cancer. However, localized with CD34-positive endothelial cells. These they did not mention the merit of serum Tn-C level on results, together with the in vitro finding that HUVEC patient survival. In this study, we first revealed that the strongly migrated toward Tn-C, suggest that Tn-C may overall survival of patients with high Tn-C serum levels play an important role in angiogenesis in NSCLC. To the (upper quartile group) was significantly shorter than that of best of our knowledge, this may be the first report in which patients with low serum Tn-C levels. Since this difference is the serum level of Tn-C was demonstrated to be a useful subtle, further large-scale studies are necessary to confirm predictive marker of intratumoral angiogenesis in NSCLC. the result. Compared to immunohistochemical analysis, the In conclusion, we revealed that the preoperative measurement of serum Tn-C using ELISA can be easily concentration of serum Tn-C is a valuable marker for the performed. Moreover, the blood sample can be obtained prediction of intratumoral angiogenesis and prognosis in prior to surgery. If neoplastic angiogenic activity of the NSCLC. Determination of serum Tn-C may be of value tumor and patient prognosis could be predicted prior to when neoadjuvant therapy, including anti-angiogenic surgery, patients who should undergo neoadjuvant therapy, is considered for patients with NSCLC.

493 ANTICANCER RESEARCH 25: 489-496 (2005)

Figure 5. Histological finding of Tn-C and CD34 expressions in the tumor tissue specimen of adenocarcinoma with immunohistochemical analysis. Note that the majority of Tn-C was expressed in the stroma of the tumor and co-localized with CD34-positive endothelial cells (x400). Tn-C is stained blue, while CD34 is stained brown in color. Interestingly, there was no Tn-C-positive staining in the cancer cells.

References

1 Kaneko S, Ishikawa K, Yoshimi I, Marugame T, Hamashima C, Kamo K, Mizuno S and Sobue T: Projection of lung cancer mortality in Japan. Cancer Sci 94: 919-923, 2003. 2 Marugome T and Mizuno S: Mortality trend of lung cancer in Japan: 1960-2000. Jpn J Clin Oncol 33: 148-149, 2003. 3 Folkman J, Watson K, Ingber D and Hanahan D: Induction of angiogenesis during the transition from hyperplasia to neoplasia. Nature 339: 58-61, 1989. 4 Weidner N, Folkman J and Pozza F: Tumor angiogenesis: a new significant and independent prognostic indicator in early-stage breast carcinoma. J Natl Cancer Inst 84: 1875- 1887, 1992. 5 Johnson B: Adjuvant chemotherapy for non-small cell lung cancer: the end of the beginning. J Natl Cancer Inst 95: 1422- 1424, 2003. Figure 6. In vitro HUVEC migration toward recombinant Tn-C. HUVEC 6 Evans T: Highlights from the Tenth World Conference on Lung (1 x 105) were placed in the upper chamber and were allowed to migrate Cancer. Oncologist 9: 232-238, 2004. toward the reverse side of the membrane coated with Tn-C, VEGF or 7 Jones F and Jones P: The tenascin family of ECM BSA. HUVEC migrated toward recombinant Tn-C or VEGF to a greater glycoproteins: structure, function, and regulation during degree than BSA. Data are expressed as mean + S.D. of migrated cell embryonic development and tissue remodeling. Dev Dyn 218: number of five different microscopic fields observed, magnification x100. 235-259, 2000. *p=0.019 vs. BSA **p=0.0002 vs. BSA

494 Ishiwata et al: Serum Tenascin in Lung Cancer

8 Tsunoda T, Inada H, Kalembeyi I, Imanaka-Yoshida K, 23 Hirama M, Takahashi F, Takahashi K, Akutagawa S, Shimizu Sakakibara M, Okada R, Katsuta K, Sakakura T, Majima Y and K, Soma S, Shimanuki Y, Nishio K and Fukuchi Y: Osteopontin Yoshida K: Involvement of large Tenascin-C splice variants in overproduced by tumor cells acts as a potent angiogenic factor breast cancer progression. Am J Pathol 162: 1857-1867, 2003. contributing to tumor growth. Cancer Lett 198: 107-117, 2003. 9 Mighell A, Thompson J, Hume W, Markham A and Robinson 24 Tanaka K, Hiraiwa N, Hashimoto H, Yamazaki Y and P: Human tenascin-C: identification of a novel type III repeat Kusakabe M: Tenascin-C regulates angiogenesis in tumor in oral cancer and of novel splice variants in normal, malignant, through the regulation of vascular endothelial growth factor. Int and reative oral mucosae. Int J Cancer 72: 236-240, 1997. J Cancer 108: 31-40, 2004. 10 Atula T, Hedstrom J, Finne P, Leivo I, Markkanen-Leppanen 25 Tokes A, Hortovanyi E, Kulka J, Jackel M, Kerenyi T and M and Haglund C: Tenascin-C expression and its prognostic Kadar A: Tenascin expression and angiogenesis in breast significance in oral and pharyngeal squamous cell carcinoma. cancers. Pathol Res Pract 195: 821-828, 1999. Anticancer Res 23: 3051-3056, 2003. 26 Ylatupa S, Mertaniemi P, Haglund C and Partanen P: Enzyme 11 Mai J, Sameni M, Mikkelsen T and Sloane B: Degradation of immunoassay for quantification of tenascin in biologic samples. protein tenascin-C by cathepsin B: an Clin Biochem 28: 263-268, 1995. interaction involved in the progression of gliomas. Biol Chem 27 Yamauchi M, Mizuhara Y, Maezawa Y and Toda G: Serum 383: 1407-1413, 2002. tenascin levels in chronic liver disease. Liver 14: 148-153, 1994. 12 Zagzag D, Friedlander D, Dosik J, Chikramane S, Chan W, Greco 28 Juhasz A, Bardos H, Repassy G and Adany R: Characteristic M, Allen J, Dorovini-Zis K and Grumet M: Tenascin-C expression distribution patterns of tenascin in laryngeal and by angiogenic vessels in human astrocytomas and by human brain hypopharyngeal cancers. Laryngoscope 110: 84-92, 2000. endothelial cells in vitro. Cancer Res 56: 182-189, 1996. 29 Korshunov A, Golanov A, Ozerov S and Sycheva R: 13 Zagzag D and Capo V: Angiogenesis in the central nervous Prognostic value of tumor-associated antigens immuno- system: a role for vascular endothelial growth factor/vascular reactivity and apoptosis in medulloblastomas. Brain Tumor permeability factor and tenascin-C. Common molecular Pathol 16: 37-44, 1999. effectors in cerebral neoplastic and non-neoplastic "angiogenic 30 Pilch H, Schaffer U, Schlenger K, Lautz A, Tanner B, diseases". Histol Histopathol 17: 301-321, 2002. Hockel M and Knapstein P: Expression of tenascin in human 14 Kim C, Bak K, Kim Y, Kim J, Ko Y, Oh S, Kim K and Hong cervical cancer--association of tenascin expression with E: Expression of tenascin-C in astrocytic tumors: its clinicopathological parameters. Gynecol Oncol 73: 415-421, relevance to proliferation and angiogenesis. Surg Neurol 54: 1999. 235-240, 2000. 31 Jahkola T, Toivonen T, Virtanen I, von Smitten K, Nordling S, 15 Riedl S, Bodenmuller H, Hinz U, Holle R, Moller P, Schlag P, von Boguslawski K, Haglund C, Nevanlinna H and Blomqvist Herfarth C and Faissner A: Significance of tenascin serum level C: Tenascin-C expression in invasion border of early breast as tumor marker in primary colorectal carcinoma. Int J Cancer cancer: a predictor of local and distant recurrence. Br J Cancer 20: 65-69, 1995. 78: 1507-1513, 1998. 16 Burchardt E, Hein R and Bosserhoff A: , 32 Ishihara A, Yoshida T, Tamaki H and Sakakura T: Tenascin hyaluronan, tenascin-C and type VI levels in sera expression in cancer cells and stroma of human breast cancer from patients with malignant melanoma. Clin Exp Dermatol and its prognostic significance. Clin Cancer Res 1: 1035-1041, 28: 515-520, 2003. 1995. 17 Pauli C, Stieber P, Schmitt U, Andratschke M, Hoffmann K 33 Iskaros B, Hu X, Sparano J and Fineberg S: Tenascin pattern of and Wollenberg B: The significance of Tenascin-C serum level expression and established prognostic factors in invasive breast as tumor marker in squamous cell carcinoma of the head and carcinoma. J Surg Oncol 68: 107-112, 1998. neck. Anticancer Res 22: 3093-3097, 2002. 34 Iskaros B, Tanaka K, Hu X, Kadish A and Steinberg J: 18 Kusagawa H, Onoda K, Namikawa S, Yada I, Okada A, Morphologic pattern of tenascin as a diagnostic biomarker in Yoshida T and Sakakura T: Expression and degeneration of colon cancer. J Surg Oncol 64: 98-101, 1997. tenascin-C in human lung cancers. Br J Cancer 77: 98-102, 1998. 35 Kressner U, Lindmark G, Tomasini-Johansson B, Bergstrom R, 19 Cai M, Onoda K, Takao M, Kyoko I, Shimpo H, Yoshida T and Gerdin B, Pahlman L and Glimelius B: Stromal tenascin Yada I: Degradation of tenascin-C and activity of matrix distribution as a prognostic marker in colorectal cancer. Br J -2 are associated with tumor recurrence in Cancer 76(4): 526-30: 526-530, 1997. early stage non-small cell lung cancer. Clin Cancer Res 8: 1152- 36 Jahkola T, Toivonen T, von Smitten K, Blomqvist C and 1156, 2002. Virtanen I: Expression of tenascin in invasion border of early 20 Soini Y, Paakko P, Nuorva K, Kamel D, Linnala A, Virtanen I breast cancer correlates with higher risk of distant metastasis. and Lehto V: Tenascin immunoreactivity in lung tumors. Am J Int J Cancer 69: 445-447, 1996. Clin Pathol 100: 145-150, 1993. 37 Shoji T, Kamiya T, Tsubura A, Hamada Y, Hatano T, Hioki K 21 Dhar D, Ono T, Yamanoi A, Soda Y, Yamaguchi E, Rahman and Morii S: Tenascin staining positivity and the survival of M, Kohno H and Nagasue N: Serum endostatin predicts patients with invasive breast carcinoma. J Surg Res 55: 295- tumor vascularity in hepatocellular carcinoma. Cancer 95: 297, 1993. 2188-2195, 2002. 22 Zagzag D, Shiff B, Jallo G, Greco M, Blanco C, Cohen H, Hukin J, Allen J and Friedlander D: Tenascin-C promotes microvascular cell migration and phosphorylation of focal Received September 24, 2004 adhesion kinase. Cancer Res 62: 2660-2668, 2002. Accepted December 20, 2004

495