Research ArticleExperimental Studies

Microarray Analysis Reveals Distinct Gene Set Profiles for Gastric and Intestinal Gastrointestinal Stromal Tumors

RYUHEI HARA, HIROTOSHI KIKUCHI, TOMOHIKO SETOGUCHI, SHINICHIRO MIYAZAKI, MASAYOSHI YAMAMOTO, YOSHIHIRO HIRAMATSU, KINJI KAMIYA, MANABU OHTA, SATOSHI BABA and HIROYUKI KONNO
Anticancer Research June 2015, 35 (6) 3289-3298;

Abstract

Aim: We sought to address the mechanisms by which intestinal gastrointestinal stromal tumors (GIST) have a markedly higher risk of recurrence than gastric GISTs. Materials and Methods: Gene expression levels were compared among six primary gastric, three intestinal and six metastatic liver GISTs using cDNA microarray. Protein levels of Slit homolog 2 (SLIT2) were analyzed by immunohistochemistry in 25 primary gastric and 10 intestinal GIST. Results: Intestinal GIST had gene expression profiles similar to clinically malignant and metastatic GIST. In gene set-enrichment analysis, the gene sets MITOTIC_CELL CYCLE and NEURON_DIFFERENTIATION were up-regulated and down-regulated, respectively, in intestinal GIST compared to gastric GIST. High-risk gastric GISTs and intestinal GIST, expressed similar levels of SLIT2 protein, which were lower than those of low-risk gastric GISTs. Conclusion: The gene-expression profile of intestinal GISTs was similar to that of metastatic liver GISTs. Besides higher proliferative activity, down-regulation of SLIT2 might be involved in clinically malignant phenotypes of intestinal GIST.

Gastrointestinal stromal tumors (GIST), originating from the interstitial cells of Cajal (ICC) or their progenitor cells, are the most common mesenchymal neoplasm in the human digestive tract (1, 2). The current consensus is that gain-of-function mutations in the mast/stem cell growth factor receptor Kit (c-KIT) or platelet-derived growth factor receptor alpha (PDGFRA) genes in ICC are the leading cause of GIST, which results in ligand-independent activation of the receptors and consequential tumor progression (3-5). GISTs arise in a variety of organs throughout the gastrointestinal tract, most commonly in the stomach (60%), jejunum and ileum (30%), and the duodenum and colorectum (5%) (6). Their clinical aggressiveness can be evaluated based on previously reported risk classification criteria (7-9). In those criteria, tumor size, mitotic count and tumor site are regarded as key factors that have a strong impact on the prognosis of patients with GIST. However, they do not clarify the biological mechanism underlying the clinical aggressiveness or malignant potential of the tumors. Although adjuvant imatinib therapy has proven to prolong recurrence-free survival and overall survival in patients who are at high risk for GIST recurrence following resection (10), the selection criteria for patients who receive adjuvant therapy may be improved by developing a novel classification of risk of recurrence based on biological markers.

Compared to gastric GIST, patients with intestinal GIST have markedly higher risk for recurrence, even after surgical resection (8, 9, 11, 12). It is assumed that approximately 40-50% of intestinal GISTs are clinically malignant, while 20-25% of gastric GISTs are thought to behave in an aggressive manner (12). Generally, intestinal GISTs are difficult to diagnose until they present symptoms, such as gastrointestinal bleeding or acute abdomen, and they tend to be larger in size by the time diagnoses are made (13, 14). Although tumor size and mitotic activity are regarded as the best predictive prognostic features of GIST, even when classified and compared on the basis of similar size and mitotic index, patients with intestinal GIST still have a worse prognosis than those with gastric GIST (8, 9, 13).

The malignant nature of intestinal GIST has not yet been characterized. In this study, we performed a microarray analysis to clarify the differences in gene-expression profiles between gastric and intestinal GIST, and to identify the genes that are closely related regarding the relatively malignant biological potential of intestinal GIST.

View this table:
Table I.

Clinicopathological and genetic findings of gastrointestinal stromal tumors used for microarray analysis.

View this table:
Table II.

Gene sets significantly enriched in intestinal gastrointestinal stromal tumors.

Materials and Methods

Patients and tumors. Surgically-resected frozen tissue specimens of six primary gastric GISTs (including one gastric GIST with metastasis to the liver), three primary intestinal GISTs and six liver-metastatic GIST (five from gastric GIST and one from intestinal GIST) were used for microarray analysis. None of the patients had received imatinib therapy before surgery. Formalin-fixed paraffin-embedded (FFPE) tissue specimens of 25 gastric GISTs and 10 intestinal GISTs from 34 patients who underwent surgery at the Hamamatsu University School of Medicine from September 2001 to April 2013 were included in the immunohistochemical analysis. Ethical approval for the study was obtained from the Institutional Review Board of Hamamatsu University School of Medicine (No.21-10).

Figure 1.
Figure 1.

cDNA microarray analysis of primary gastric and intestinal gastrointestinal stromal tumors (GIST) and liver-metastatic GIST. A: Hierarchical clustering. Fifteen GISTs were classified into two groups according to their gene expressions. One group consisted of clinically malignant GIST, intestinal GIST and liver-metastatic GIST (cluster A), and the other consisted of all five primary gastric GISTs without postoperative recurrence and one liver-metastatic GIST from the stomach (cluster B). Of 28,869 genes analyzed, 194 genes were up-regulated and 274 genes were down-regulated in cluster A. B: Principal component analysis. Clinically malignant GIST, intestinal GIST and liver-metastatic GIST had similar gene-expression profiles with common components that are distinct from primary gastric GIST without postoperative recurrence. S: Stomach; I: intestine.

Microarray analysis. Total RNA was extracted from the frozen tumor samples using an RNeasy Mini Kit (Qiagen, Valencia, CA, USA). The intensity and quantity of total RNA was assessed with a Nanodrop ND-1000 spectrometer (Nanodrop Technologies, Wilmington, DE, USA). The gene-expression profiles of primary gastric GIST, primary intestinal GIST and metastatic liver GIST were determined using the oligonucleotide microarray Human Gene 1.0 ST (Affymetrix, Santa Clara, CA, USA) according to the manufacturer's protocol. Data were analyzed using the GeneSpringGX11 software (Agilent, Santa Clara, CA, USA). Hierarchical clustering analysis was performed with at least a two-fold change and p-value of less than 0.05.

We also used the Gene Set Enrichment Analysis (GSEA) application (Broad Institute of Massachusetts Institute of Technology and Harvard University, http://www.broad.mit.edu/gsea) to define gene sets that were differentially expressed between primary gastric and intestinal GIST. The gene set database c5.bp.v4.0 gene_set.symbols.gmt was used for the analysis. Gene sets with false-discovery rates of <25% and nominal p-values of <0.01 were defined as being significant.

Immunohistochemical analysis. For evaluation of Slit homolog 2 (SLIT2) protein expression in GIST, immunohistochemical analysis was performed using 3-μm sections of FFPE surgically-resected tissue specimens. Tumor sections were de-paraffinized with successive xylene and ethanol treatment, as well as rehydration. Antigen retrieval was conducted by heating the samples at 95°C for 40 min in Target Retrieval Solution, pH 9 (DAKO, Glostrup, Denmark). Slides were rinsed with phosphate-buffered saline (PBS) and endogenous peroxidase was blocked by incubation in 3% hydrogen peroxide in absolute methanol for 10 min. Sections were washed in PBS, incubated with rabbit polyclonal antibody for human SLIT2 (1:25 dilution; Sigma-Aldrich, St. Louis, MO, USA) at 4°C overnight. Sections were then incubated with EnVision (DAKO) for 30 min. Signals were developed using 3,3’-diamonobenzidine (Nichirei, Tokyo, Japan) and counterstained with hematoxylin for 1 min then mounted with a permanent mounting medium. Human normal colonic mucosa was used for positive control, and immunostaining specificity was examined by omission of primary antibodies.

Figure 2.
Figure 2.
Figure 2.

cDNA microarray analysis comparing primary gastric and intestinal gastrointestinal stromal tumors (GIST). A: Gene Set Enrichment Analysis. Gene sets enriched and down-regulated in intestinal GIST were mainly those related to cell proliferation or mitosis (upper panels) and those related to the development or differentiation of neuronal cells (lower panels) respectively. B; Gene-expression clustering. The majority of genes up-regulated and down-regulated in intestinal GISTs were those involved in cell proliferation or mitosis (upper panel) and those involved in neuronal differentiation (lower panel), respectively.

SLIT2 staining intensity was evaluated using a light microscope under ×200 magnification in 10 high power fields from every tumor specimen by two independent researchers with no knowledge of the clinical data.

Statistical analysis. Data are presented as mean±SE. Statistical significance of differences were assessed using the Mann–Whitney U-test. For comparisons among three or more groups, data were statistically evaluated using one-way analysis of variance, followed by the Holm–Sidak multiple comparisons test. Statistical analysis was performed using GraphPad Prism version 6.0 for Windows (GraphPad Software, San Diego, CA). A p-value of less than 0.05 was considered statistically significant.

Results

Microarray analysis of primary gastric, intestinal and liver-metastatic GIST. To address the mechanisms underlying the clinically malignant phenotype of intestinal GIST, gene expression profiles in six primary gastric GIST, three intestinal GIST and six liver-metastatic GIST were analyzed using a cDNA microarray. Patient profiles are shown in Table I. Unsupervised hierarchical clustering classified them into two groups according to their gene expressions (Figure 1A). One group consisted of clinically malignant GIST, intestinal GIST and liver-metastatic GIST (Figure 1A, cluster A), and the other consisted of all five primary gastric GISTs without postoperative recurrence and one liver-metastatic gastric GIST (Figure 1A, cluster B). Out of 28,869 genes analyzed, 194 genes were up-regulated and 274 genes were down-regulated in cluster A.

View this table:
Table III.

Gene sets significantly down-regulated in intestinal gastrointestinal stromal tumors.

Principal component analysis further revealed that clinically malignant GIST, intestinal GIST and metastatic liver GIST have similar gene-expression profiles, with common components that are distinct from primary gastric GIST without postoperative recurrence (Figure 1B). These findings suggest that intestinal GISTs are more likely to express gene sets related to the malignant phenotype of GIST, and potentially explain mechanisms by which intestinal GISTs differ from gastric GISTs in their clinically malignant phenotype.

GSEA comparing primary gastric and intestinal GIST. We next addressed details of distinct gene-expression profiles for gastric and intestinal GIST using the GSEA application. GSEA comparing primary gastric and intestinal GIST revealed that 23 gene sets were significantly enriched (Table II) and 16 gene sets significantly down-regulated (Table III) in intestinal GIST, with a false-discovery rates of <25% and nominal p-values of <0.01. Gene sets enriched in intestinal GIST were mainly those related in the gene ontology terms to cell proliferation or mitosis, such as “MITOTIC CELL CYCLE” and “M PHASE OF MITOTIC CELL CYCLE” (Figure 2A, upper panels). Conversely, gene sets related to the development or differentiation of neuronal cells such as “NEURON DIFFERENTIATION” and “NEUROGENESIS”, were noticeably down-regulated in intestinal GIST (Figure 2A, lower panels). Gene-expression clustering further indicated that the majority of genes up-regulated in intestinal GISTs are those involved in cell proliferation or mitosis (Figure 2B, upper panel) and those genes down-regulated are those involved in neural differentiation (Figure 2B, lower panel).

Immunohistochemical analysis of gastric and intestinal GIST. In the gene sets down-regulated in intestinal GIST, we further focused on SLIT2, which is involved in axon guidance and known as a tumor-suppressor gene. We performed immunohistochemical staining for SLIT2 using the specimens from 25 gastric GISTs and 10 intestinal GISTs. The staining intensity of SLIT2 was scored as 0 (none), 1+ (weak), 2+ (moderate), or 3+ (strong) (Figure 3). Compared with intestinal GIST, gastric GIST exhibited more intensive staining for the SLIT2 protein (Figure 4A). Furthermore, staining intensities of SLIT2 protein in high-risk gastric GIST were weaker than those at lower risk, but not significantly different from intestinal GIST (Figure 4B). These findings suggest that SLIT2 expression might be suppressed in GIST with higher malignant potential.

SLIT2 expression in primary gastric and intestinal GIST, and clinicopathological findings. To examine whether SLIT2 expression in primary GIST is correlated with clinicopathological findings, the dominant SLIT2 staining property of each tumor was compared to tumor size, mitotic counts and the Ki-67 (MIB-1) index. There was an inverse tendency between SLIT2 expression and tumor size (Figure 5A), whereas the mitotic count and the MIB-1 index were not significantly different among the three groups, according to SLIT2 intensity (Figure 5B and C).

Figure 3.
Figure 3.

Representative photographs of immunohistochemical staining for Slit homolog 2 (SLIT2). The staining intensity of SLIT2 was scored as 0 (none), 1+ (weak), 2+ (moderate), or 3+ (strong).

Discussion

GISTs are known to arise from the ICC or their progenitor cells, and are the most common mesenchymal neoplasms in the human gastrointestinal tract (1-3). Although surgical resection is widely accepted as the therapy-of-choice, suitable risk classification for tumor recurrence is very important for the application of adjuvant therapy and prediction of overall outcome. The known risk classifications of GIST are based mainly on tumor size, tumor site and number of mitoses (7-9). However, they do not clearly show or reflect the biological mechanisms underlying the nature of GISTs.

In the present study, we showed for the first time the differences in gene-expression profiles of primary gastric GIST and intestinal GIST using a microarray analysis. Firstly, we compared gene-expression profiles of gastric, intestinal and liver-metastatic GIST using microarray. The analysis showed that intestinal GISTs possess gene profiles that are more similar to those of malignant GIST and liver-metastatic GIST, but distinct from gastric GIST. These findings suggest that intestinal GISTs potentially express genes involved in the malignant transformation of GIST, perhaps from early in the development of intestinal GIST. In our previous studies, we compared primary gastric GIST and liver-metastatic GIST immunohistochemically and genetically, and reported that the loss of heterozygosity of the c-KIT gene or the loss of chromosome 4q could be responsible for GIST liver metastasis (15, 16). Furthermore, we performed a microarray analysis and an immunohistochemical analysis to reveal that versican and CD9 are novel prognostic markers in gastric GIST and may serve as potential therapeutic targets (17). Besides c-KIT gene and PDGFRA gene mutations, other potential genetic and epigenetic changes that regulate the biological behavior of GIST might contribute to the prognoses of patients with GIST.

In addition, we performed a GSEA of our microarray data to focus on the important biological processes regulated in intestinal GIST and found that 23 gene sets were significantly enriched and 16 gene sets were significantly down-regulated in intestinal GIST. In intestinal GIST, gene sets involved in mitosis or cell-cycle regulation were enriched, suggesting that these tumors have more aggressive proliferative activities than those arising in the stomach. As previously reported, the MIB-1 labeling index has been widely used as a relevant marker of cell proliferation (18-20) and mitotic index is the key factor for risk classifications of GIST. In the present study, we focused on biological processes other than proliferation or cell-cycle regulation in order to find out mitosis-independent mechanisms underlying the clinically malignant phenotype of intestinal GIST. Interestingly, gene sets related to the development or differentiation of neuronal cells were down-regulated in intestinal GISTs compared with gastric GISTs. Although further studies are needed, intestinal GIST might consist of ICC that are undifferentiated.

Figure 4.
Figure 4.

Immunohistochemical analysis of 25 gastric gastrointestinal stromal tumors (GIST) and 10 intestinal GISTs. A: Staining intensities of Slit homolog 2 (SLIT2) protein were significantly stronger in gastric GISTs than in intestinal GISTs. B: Staining intensities of SLIT2 protein in high-risk gastric GISTs were weaker than those at lower risk, but similar to those of intestinal GISTs. *p<0.01;**p<0.0001; n.s., p>0.05.

Figure 5.
Figure 5.

Features of primary gastric and intestinal gastrointestinal stromal tumors (GIST) according to staining score for Slit homolog 2 (SLIT2) expression: A: Tumor size; B: mitotic count; C: MIB-1 index. There was an inverse tendency between SLIT2 expression and the tumor size. Mitotic counts and MIB-1 index were not significantly different among the three groups according to SLIT2 intensity score. *p<0.05; n.s., p>0.05.

SLIT proteins are secreted glycoproteins known as molecular cues for axon guidance in developing neuronal tissue. They are the main ligands for the transmembrane protein roundabout (ROBO) receptors and have an evolutionarily-conserved role in repulsive axon guidance and neural tube development (21-23). In the development of the central nervous system, SLIT/ROBO signals control axon out-growth and neuronal cell migration through modulation of cytoskeleton dynamics (24). To date, three SLITs (SLIT1-3) and four ROBO receptors (ROBO1-4) have been characterized in mammals (25). Importantly, SLIT2 has been reported to be involved in human cancer and down-regulation of SLIT2 is considered to be associated with aggressive phenotypes in glioma (26), acute lymphatic leukemia (27), and breast, lung and colorectal cancer (28-30). In the present study, immunohistochemical evaluation of SLIT2 revealed that staining intensities were relatively higher in gastric GIST than intestinal GIST, and low SLIT2 expression was correlated with a trend for an increase in risk classification of gastric GIST. These data suggest that suppression of the SLIT/ROBO axis in GIST cells may be involved in the malignant transformation of GIST, possibly via acquisition of an undifferentiated phenotype.

Although it is not well-understood how SLIT2 affects tumor progression, SLIT2 is considered to function mainly through mechanisms that are related to tumor cell migration and adhesion (31). Werbowetski-Ogilvie et al. reported that SLIT2 inhibited medulloblastoma cell invasion rate through down-regulating activated cell division cycle (CDC) 42 without affecting cell direction or proliferation (32). Recent reports showed that low expression of SLIT2 correlates with poor prognosis and promotes metastasis in esophageal squamous cell carcinoma, which may be regulated by the CDC 42-mediated pathways (33) and that inactivation of SLIT2 and/or ROBO1 is one of the early events in development of dysplastic lesions of head and neck and has prognostic importance (34). Yiin et al. also reported that SLIT2 was expressed at lower levels in primary glioma specimens and invasive glioma cells, compared with normal neuronal cells or astrocytes, and ectopic expression of SLIT2 inhibited glioma cell migration and invasion (26). Prasad et al. reported that overexpression of SLIT2 led to decreased proliferative and migratory capabilities in breast cancer cells. It has also been reported that SLIT2 overexpression increases localization of E-cadherin in cell borders at cell-cell adhesion and translocation of β-catenin to the membrane. Furthermore, SLIT2 inhibited phosphorylation of AKT/ glycogen synthase kinase-3β, resulting in an increase in the phosphorylation of β-catenin (35). Thus, SLIT2 may induce its tumor suppressive effect by regulating E-cadherin-mediated cell-to-cell adhesion. Considering these findings, it is possible that the loss of SLIT2 expression enables a GIST cell to lose its adhesiveness and gain migratory capacity, resulting in a higher risk for metastasis or recurrence.

In the present study, we also found that there was an inverse tendency between SLIT2 expression and tumor size, whereas mitotic counts and the MIB-1 index were not significantly different among the three groups classified according to SLIT2 intensity score. These results show that SLIT2 has less impact on the tumor's proliferative potential, implying that it acts in a mitosis-independent manner rather than regulating proliferative activities of GIST. Although it is unclear how tumor size affects SLIT2 expression, it could reflect the loss of neural characteristics of GISTs as they differentiate. The origin and developmental process of GISTs have been sought with much interest and it is understood that they arise from the ICC, or their progenitor cells in the gastrointestinal wall. One possibility is that the different characteristics of ICC may be responsible for regulating the distinct clinical differences between gastric and intestinal GIST, although this hypothesis needs to be addressed in future study.

In conclusion, besides higher proliferative activity, down-regulation of SLIT2 might be involved in clinically malignant phenotypes of intestinal GIST. Although the functional mechanism of SLIT2 in GIST remains to be elucidated, it could be helpful as a biological marker to evaluate the recurrence risk and even to reveal the developmental process of GIST.

Acknowledgements

This work was supported in part by the Ministry of Education, Culture, Sports, Science, and Technology of Japan grants-in-aid 24591937 (H. Kikuchi) and 24390312 (H. Konno). No potential conflicts of interest were disclosed. The Authors thank Drs. Kiyotaka Kurachi, Takanori Sakaguchi, and other laboratory members, for their helpful discussions.

  • Received February 16, 2015.
  • Revision received March 1, 2015.
  • Accepted March 3, 2015.

References

  1. ↵
    1. Kindblom LG,
    2. Remotti HE,
    3. Aldenborg F,
    4. Meis-Kindblom JM
    : Gastrointestinal pacemaker cell tumor (GIPACT): gastrointestinal stromal tumors show phenotypic characteristics of the interstitial cells of Cajal. Am J Pathol 152(5): 1259-1269, 1998.
    OpenUrlPubMed
  2. ↵
    1. O'leary T,
    2. Berman JJ
    : Gastrointestinal stromal tumors: answers and questions. Hum Pathol 33(5): 456-458, 2002.
    OpenUrlPubMed
  3. ↵
    1. Hirota S,
    2. Isozaki K,
    3. Moriyama Y,
    4. Hashimoto K,
    5. Nishida T,
    6. Ishiguro S,
    7. Kawano K,
    8. Hanada M,
    9. Kurata A,
    10. Takeda M,
    11. Muhammad Tunio G,
    12. Matsuzawa Y,
    13. Kanakura Y,
    14. Shinomura Y,
    15. Kitamura Y
    : Gain-of-function mutations of c-KIT in human gastrointestinal stromal tumors. Science 279(5350): 577-580, 1998.
    OpenUrlAbstract/FREE Full Text
    1. Hirota S,
    2. Ohashi A,
    3. Nishida T,
    4. Isozaki K,
    5. Kinoshita K,
    6. Shinomura Y,
    7. Kitamura Y
    : Gain-of-function mutations of platelet-derived growth factor receptor alpha gene in gastrointestinal stromal tumors. Gastroenterology 125(3): 660-667, 2003.
    OpenUrlCrossRefPubMed
  5. ↵
    1. Heinrich MC,
    2. Corless CL,
    3. Duensing A,
    4. Mcgreevey L,
    5. Chen CJ,
    6. Joseph N,
    7. Singer S,
    8. Griffith DJ,
    9. Haley A,
    10. Town A,
    11. Demetri GD,
    12. Fletcher CD,
    13. Fletcher JA
    : PDGFRA activating mutations in gastrointestinal stromal tumors. Science 299(5607): 708-710, 2003.
    OpenUrlAbstract/FREE Full Text
  6. ↵
    1. Patil DT,
    2. Rubin BP
    : Gastrointestinal stromal tumor: advances in diagnosis and management. Arch Pathol Lab Med 135(10): 1298-1310, 2011.
    OpenUrlCrossRefPubMed
  7. ↵
    1. Fletcher CD,
    2. Berman JJ,
    3. Corless C,
    4. Gorstein F,
    5. Lasota J,
    6. Longley BJ,
    7. Miettinen M,
    8. O'leary TJ,
    9. Remotti H,
    10. Rubin BP,
    11. Shmookler B,
    12. Sobin LH,
    13. Weiss SW
    : Diagnosis of gastrointestinal stromal tumors: A consensus approach. Hum Pathol 33(5): 459-465, 2002.
    OpenUrlCrossRefPubMed
  8. ↵
    1. Miettinen M,
    2. Lasota J
    : Gastrointestinal stromal tumors: pathology and prognosis at different sites. Semin Diagn Pathol 23(2): 70-83, 2006.
    OpenUrlCrossRefPubMed
  9. ↵
    1. Joensuu H,
    2. Vehtari A,
    3. Riihimaki J,
    4. Nishida T,
    5. Steigen SE,
    6. Brabec P,
    7. Plank L,
    8. Nilsson B,
    9. Cirilli C,
    10. Braconi C,
    11. Bordoni A,
    12. Magnusson MK,
    13. Linke Z,
    14. Sufliarsky J,
    15. Federico M,
    16. Jonasson JG,
    17. Dei Tos AP,
    18. Rutkowski P
    : Risk of recurrence of gastrointestinal stromal tumour after surgery: an analysis of pooled population-based cohorts. Lancet Oncol 13(3): 265-274, 2012.
    OpenUrlCrossRefPubMed
  10. ↵
    1. Joensuu H,
    2. Eriksson M,
    3. Sundby Hall K,
    4. Hartmann JT,
    5. Pink D,
    6. Schutte J,
    7. Ramadori G,
    8. Hohenberger P,
    9. Duyster J,
    10. Al-Batran SE,
    11. Schlemmer M,
    12. Bauer S,
    13. Wardelmann E,
    14. Sarlomo-Rikala M,
    15. Nilsson B,
    16. Sihto H,
    17. Monge OR,
    18. Bono P,
    19. Kallio R,
    20. Vehtari A,
    21. Leinonen M,
    22. Alvegard T,
    23. Reichardt P
    : One vs three years of adjuvant imatinib for operable gastrointestinal stromal tumor: a randomized trial. JAMA 307(12): 1265-1272, 2012.
    OpenUrlCrossRefPubMed
  11. ↵
    1. Joensuu H
    : Risk stratification of patients diagnosed with gastrointestinal stromal tumor. Hum Pathol 39(10): 1411-1419, 2008.
    OpenUrlCrossRefPubMed
  12. ↵
    1. Miettinen M,
    2. Lasota J
    : Gastrointestinal stromal tumors: review on morphology, molecular pathology, prognosis, and differential diagnosis. Arch Pathol Lab Med 130(10): 1466-1478, 2006.
    OpenUrlCrossRefPubMed
  13. ↵
    1. Miettinen M,
    2. Makhlouf H,
    3. Sobin LH,
    4. Lasota J
    : Gastrointestinal stromal tumors of the jejunum and ileum: a clinicopathologic, immunohistochemical, and molecular genetic study of 906 cases before imatinib with long-term follow-up. Am J Surg Pathol 30(4): 477-489, 2006.
    OpenUrlCrossRefPubMed
  14. ↵
    1. Miettinen M,
    2. Sobin LH,
    3. Lasota J
    : Gastrointestinal stromal tumors of the stomach: a clinicopathologic, immunohistochemical, and molecular genetic study of 1765 cases with long-term follow-up. Am J Surg Pathol 29(1): 52-68, 2005.
    OpenUrlCrossRefPubMed
  15. ↵
    1. Kikuchi H,
    2. Yamashita K,
    3. Kawabata T,
    4. Yamamoto M,
    5. Hiramatsu Y,
    6. Kondo K,
    7. Baba M,
    8. Ohta M,
    9. Kamiya K,
    10. Tanaka T,
    11. Suzuki S,
    12. Kitagawa K,
    13. Kitagawa M,
    14. Sugimura H,
    15. Konno H
    : Immunohistochemical and genetic features of gastric and metastatic liver gastrointestinal stromal tumors: sequential analyses. Cancer Sci 97(2): 127-132, 2006.
    OpenUrlPubMed
  16. ↵
    1. Kikuchi H,
    2. Yamamoto M,
    3. Hiramatsu Y,
    4. Baba M,
    5. Ohta M,
    6. Kamiya K,
    7. Tanaka T,
    8. Suzuki S,
    9. Sugimura H,
    10. Kitagawa M,
    11. Kanai T,
    12. Kitayama Y,
    13. Kanda T,
    14. Nishikura K,
    15. Konno H
    : Effect of loss of heterozygosity of the c-KIT gene on prognosis after hepatectomy for metastatic liver gastrointestinal stromal tumors. Cancer Sci 98(11): 1734-1739, 2007.
    OpenUrlPubMed
  17. ↵
    1. Setoguchi T,
    2. Kikuchi H,
    3. Yamamoto M,
    4. Baba M,
    5. Ohta M,
    6. Kamiya K,
    7. Tanaka T,
    8. Baba S,
    9. Goto-Inoue N,
    10. Setou M,
    11. Sasaki T,
    12. Mori H,
    13. Sugimura H,
    14. Konno H
    : Microarray analysis identifies versican and CD9 as potent prognostic markers in gastric gastrointestinal stromal tumors. Cancer Sci 102(4): 883-889, 2011.
    OpenUrlCrossRefPubMed
  18. ↵
    1. Miettinen M,
    2. El-Rifai W,
    3. HLS L,
    4. Lasota J
    : Evaluation of malignancy and prognosis of gastrointestinal stromal tumors: a review. Hum Pathol 33(5): 478-483, 2002.
    OpenUrlCrossRefPubMed
    1. Hasegawa T,
    2. Matsuno Y,
    3. Shimoda T,
    4. Hirohashi S
    : Gastrointestinal stromal tumor: consistent CD117 immunostaining for diagnosis, and prognostic classification based on tumor size and MIB-1 grade. Hum Pathol 33(6): 669-676, 2002.
    OpenUrlCrossRefPubMed
  20. ↵
    1. Zhao WY,
    2. Xu J,
    3. Wang M,
    4. Zhang ZZ,
    5. Tu L,
    6. Wang CJ,
    7. Lin TL,
    8. Shen YY,
    9. Liu Q,
    10. Cao H
    : Prognostic value of Ki67 index in gastrointestinal stromal tumors. Int J Clin Exp Pathol 7(5): 2298-2304, 2014.
    OpenUrlPubMed
  21. ↵
    1. Li HS,
    2. Chen JH,
    3. Wu W,
    4. Fagaly T,
    5. Zhou L,
    6. Yuan W,
    7. Dupuis S,
    8. Jiang ZH,
    9. Nash W,
    10. Gick C,
    11. Ornitz DM,
    12. Wu JY,
    13. Rao Y
    : Vertebrate slit, a secreted ligand for the transmembrane protein roundabout, is a repellent for olfactory bulb axons. Cell 96(6): 807-818, 1999.
    OpenUrlCrossRefPubMed
    1. Brose K,
    2. Bland KS,
    3. Wang KH,
    4. Arnott D,
    5. Henzel W,
    6. Goodman CS,
    7. Tessier-Lavigne M,
    8. Kidd T
    : Slit proteins bind Robo receptors and have an evolutionarily conserved role in repulsive axon guidance. Cell 96(6): 795-806, 1999.
    OpenUrlCrossRefPubMed
  23. ↵
    1. Wang G,
    2. Li Y,
    3. Wang XY,
    4. Han Z,
    5. Chuai M,
    6. Wang LJ,
    7. Ho Lee KK,
    8. Geng JG,
    9. Yang X
    : Slit/Robo1 signaling regulates neural tube development by balancing neuroepithelial cell proliferation and differentiation. Exp Cell Res 319(8): 1083-1093, 2013.
    OpenUrlCrossRefPubMed
  24. ↵
    1. Chedotal A
    : Slits and their receptors. Adv Exp Med Biol 621(65-80, 2007.
    OpenUrlCrossRefPubMed
  25. ↵
    1. Ypsilanti AR,
    2. Zagar Y,
    3. Chedotal A
    : Moving away from the midline: new developments for Slit and Robo. Development 137(12): 1939-1952, 2010.
    OpenUrlAbstract/FREE Full Text
  26. ↵
    1. Yiin JJ,
    2. Hu B,
    3. Jarzynka MJ,
    4. Feng H,
    5. Liu KW,
    6. Wu JY,
    7. Ma HI,
    8. Cheng SY
    : Slit2 inhibits glioma cell invasion in the brain by suppression of Cdc42 activity. Neuro Oncol 11(6): 779-789, 2009.
    OpenUrlAbstract/FREE Full Text
  27. ↵
    1. Dunwell TL,
    2. Dickinson RE,
    3. Stankovic T,
    4. Dallol A,
    5. Weston V,
    6. Austen B,
    7. Catchpoole D,
    8. Maher ER,
    9. Latif F
    : Frequent epigenetic inactivation of the SLIT2 gene in chronic and acute lymphocytic leukemia. Epigenetics 4(4): 265-269, 2009.
    OpenUrlCrossRefPubMed
  28. ↵
    1. Tseng RC,
    2. Lee SH,
    3. Hsu HS,
    4. Chen BH,
    5. Tsai WC,
    6. Tzao C,
    7. Wang YC
    : SLIT2 attenuation during lung cancer progression deregulates beta-catenin and E-cadherin and associates with poor prognosis. Cancer Res 70(2): 543-551, 2010.
    OpenUrlAbstract/FREE Full Text
    1. Dallol A,
    2. Morton D,
    3. Maher ER,
    4. Latif F
    : SLIT2 axon guidance molecule is frequently inactivated in colorectal cancer and suppresses growth of colorectal carcinoma cells. Cancer Res 63(5): 1054-1058, 2003.
    OpenUrlAbstract/FREE Full Text
  30. ↵
    1. Dallol A,
    2. Da Silva NF,
    3. Viacava P,
    4. Minna JD,
    5. Bieche I,
    6. Maher ER,
    7. Latif F
    : SLIT2, a human homologue of the Drosophila slit2 gene, has tumor-suppressor activity and is frequently inactivated in lung and breast cancers. Cancer Res 62(20): 5874-5880, 2002.
    OpenUrlAbstract/FREE Full Text
  31. ↵
    1. Mehlen P,
    2. Delloye-Bourgeois C,
    3. Chedotal A
    : Novel roles for slits and netrins: axon guidance cues as anticancer targets? Nat Rev Cancer 11(3): 188-197, 2011.
    OpenUrlCrossRefPubMed
  32. ↵
    1. Werbowetski-Ogilvie TE,
    2. Seyed Sadr M,
    3. Jabado N,
    4. Angers-Loustau A,
    5. Agar NY,
    6. Wu J,
    7. Bjerkvig R,
    8. Antel JP,
    9. Faury D,
    10. Rao Y,
    11. Del Maestro RF
    : Inhibition of medulloblastoma cell invasion by slit. Oncogene 25(37): 5103-5112, 2006.
    OpenUrlPubMed
  33. ↵
    1. Tseng RC,
    2. Chang JM,
    3. Chen JH,
    4. Huang WR,
    5. Tang YA,
    6. Kuo IY,
    7. Yan JJ,
    8. Lai WW,
    9. Wang YC
    : Deregulation of SLIT2-Mediated Cdc42 Activity Is Associated with Esophageal Cancer Metastasis and Poor Prognosis. J Thorac Oncol 10(1): 189-198, 2015.
    OpenUrlPubMed
  34. ↵
    1. Maiti GP,
    2. Ghosh A,
    3. Mondal P,
    4. Ghosh S,
    5. Chakraborty J,
    6. Roy A,
    7. Roychowdhury S,
    8. Panda CK
    : Frequent inactivation of SLIT2 and ROBO1 signaling in head and neck lesions: clinical and prognostic implications. Oral Surg Oral Med Oral Pathol Oral Radiol 119(2): 202-212, 2015.
    OpenUrlPubMed
  35. ↵
    1. Prasad A,
    2. Paruchuri V,
    3. Preet A,
    4. Latif F,
    5. Ganju RK
    : Slit-2 induces a tumor-suppressive effect by regulating beta-catenin in breast cancer cells. J Biol Chem 283(39): 26624-26633, 2008.
    OpenUrlAbstract/FREE Full Text
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Reprints and Permissions
Microarray Analysis Reveals Distinct Gene Set Profiles for Gastric and Intestinal Gastrointestinal Stromal Tumors
RYUHEI HARA, HIROTOSHI KIKUCHI, TOMOHIKO SETOGUCHI, SHINICHIRO MIYAZAKI, MASAYOSHI YAMAMOTO, YOSHIHIRO HIRAMATSU, KINJI KAMIYA, MANABU OHTA, SATOSHI BABA, HIROYUKI KONNO
Anticancer Research Jun 2015, 35 (6) 3289-3298;
del.icio.us logo Digg logo Reddit logo Twitter logo Facebook logo Google logo Mendeley logo

Jump to section

Keywords