| | Cell kinetics and genetic instabilities in differentiated type early gastric cancers with different mucin phenotype☆Accepted 17 September 2002. Abstract To clarify the biological impact and molecular pathogenesis of cellular phenotype in differentiated-type gastric cancers (DGCs), we investigated cell kinetics and genetic instabilities in early stage of DGCs. A total of 43 early gastric cancers (EGCs) were studied. EGCs were divided into 3 phenotypic categories: gastric (G type, n = 11), ordinary (O type, n = 20), and complete intestinal (CI type, n = 12) based on the combination of HGM, ConA, MUC2, and CD10. Proliferative index (PI), apoptotic index (AI), and p53 overexpression were investigated by immunohistochemical staining with anti-Ki-67, the terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick-end labeling method, and p53 antibody, respectively. Using a high-resolution fluorescent microsatellite analysis system, microsatellite instability (MSI) and loss of heterozygosity (LOH) were examined. Frameshift mutation analysis of transforming growth factor-β type II receptor (TGF-βRII) and bcl-2–associated X (BAX) in cancers with MSI was also performed. The mean AI/PI ratio values were 0.04 for G-type, 0.10 for O-type, and 0.13 for CI-type cancers—significantly lower in G type than in O and CI types (P = 0.02 and P = 0.001, respectively). No difference in the incidence of MSI and LOH was seen among the 3 cellular phenotypes. However, the major pattern of MSI, which showed drastic and widely dispersed changes and is related to an increased risk for cancer, was significantly higher in G and O types than in CI type (P <0.005). No frame shift mutations of TGF-βRII or BAX were found in CI-type cancers. These results indicate that G-type cancers are likely to show more aggressive behaviors than CI-type cancers, and that O-type cancers show the intermediate characteristics of both types. However, the molecular pathogenesis of each phenotypic cancer is not associated with microsatellite alterations. HUM PATHOL 34:32-40. Copyright 2003, Elsevier Science (USA). All rights reserved.
Abbreviations:
ABC
, avidin-biotin peroxidase complex,
BAX
, bcl-2–associated X protein,
ConA
, paradoxical concanavalin A,
DGC
, differentiated-type gastric cancer,
EGC
, early gastric cancer,
HGM
, human gastric mucin,
HRFMA
, high-resolution fluorescent microsatellite analysis,
LOH
, loss of heterozygosity,
MSI
, microsatellite instability,
PBS
, phosphate-buffered saline,
PCR
, polymerase chain reaction,
TGF-β RII
, transforming growth factor-β type II receptor,
TUNEL
, terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick-end labeling
Gastric cancers are histologically divided into 2 types, the intestinal and diffuse type and the differentiated and undifferentiated type.1, 2 These 2 histologic types are considered different from the epidemiologic characteristics and genetic alterations. Most differentiated-type gastric cancers (DGCs) arise from the intestinal metaplasia and have the intestinal phenotype;1, 2, 3 however, some of the DGCs arise from gastric mucosa without intestinal metaplasia4 and are termed “gastric phenotype.”5 The latest mucin immnochemistry and immnohistochemistry methods have demonstrated that the cancers with gastric phenotype are derived from gastric foveolar epithelium and those with intestinal phenotype from complete-type intestinal metaplasia.3, 5 Gastric cancers with gastric mucin phenotype are associated with poorer outcome6 and greater malignant potential in the incipient phase of invasion and metastasis compared with those of other mucin phenotypes.7 Therefore, phenotypic classification should help improve knowledge of the biologic behavior of cancers and guide therapeutic strategy.8
Although several investigators have studied the genetic bases of gastric cancers, data on cellular phenotype are sparse. Studies by Tamura et al3, 9, 10 have addressed this issue in depth. An important factor in the rapid accumulation of genetic changes is genetic instability.11 It has thus been suggested that individuals with microsatellite instability (MSI) have a greater tendency to accumulate genetic alterations that lead to the transformation of normal cells to cancer cells.12 There have been many technical discrepancies in the conventional methods used for the analysis of genetic instabilities in the previous investigations. For example, in DNA sequencing gel, in which the polymerase chain reaction (PCR) products are electrophoresed, migration accuracy is not sufficient to enable comparison of the 2 independent lanes, and autoradiography has biased detection characteristics, preventing the intensity of bands from being estimated with high accuracy.13 These problems have influenced the results of MSI analyses carried out using these methods.14 The high-resolution fluorescent microsatellite analysis (HRFMA) system was recently developed to overcome the problems of conventional methods.13
Until now, the reason why DGCs with gastric phenotype are associated with aggressive behavior and molecular pathogenesis of each phenotypic cancer remains unclear. In sporadic colorectal cancers, the tumors with high level MSI which indicates DNA instability at >40% of microsatellite loci, are associated with a favorable clinical outcome,15, 16 as is the high rate of apoptosis in colorectal cancers.17 One study has reported the correlation between cell kinetics and prognostic impact in colorectal cancer stratified by MSI status.18 To our knowledge, however, there is no study regarding the correlation between cell kinetics and genetic instability in gastric cancers. In the present study, we evaluated cell kinetics, including cell proliferation and apoptosis, to assess the biological impact and the molecular pathogenesis of each cellular phenotype in early stage of sporadic DGCs. We also analyzed MSI, loss of heterozygosity (LOH), and frameshift mutations as genetic instabilities using the HRFMA system.
Materials and methods  Samples Fifty DGC cases of early gastric cancers (EGCs) were randomly selected from the histopathology files of Asahikawa Medical College Hospital from 47 patients between 1998 and 2000. Written informed consent was obtained from the patients before their interviews for this study, and the ethics committee of Asahikawa Medical College approved the study. Histologic classification of the intramucosal lesions was made according to the general rules established by the Padova classification.19 EGCs were defined as cancers in which invasion was limited to the submucosal layer.20 This group comprised 13 noninvasive dysplasias (6 low-grade and 7 high-grade) and 37 submucosal invasive cancers. These tumors were treated by surgical operation (n = 32, including 27 submucosal invasive cancers) or endoscopic mucosal resection (EMR; n = 18), all of which included the adjacent normal mucosa. The specimens were fixed in 10% formalin and embedded in paraffin wax; then 4-μm consecutive sections were used for histologic examination by hematoxylin and eosin (H&E) staining and immunohistochemical staining. Analysis and classification of phenotypic expression For immunohistochemical staining, the avidin-biotin peroxidase complex (ABC) method was used for the detection of human gastric mucin (HGM) (45M1; Novocastra, Newcastle upon Tyne, UK), MUC2 (Ccp58; Novocastra) and CD10 (56C6; Novocastra). The slides were treated with the antigen-retrieval technique based on microwave oven heating in 10 mmol citrate buffer (pH 6.0). Paradoxical concanavalin A (ConA) staining to identify class III mucins in mucous neck and pyloric gland cells was done following the Katsuyama method.21 HGM staining the surface of normal gastric epithelium and class III were defined as gastric phenotype markers. MUC2 is glycoprotein expressed predominantly in goblet cells, and CD10 expression is seen at brush borders on the luminal surface of epithelial cells. These MUC2 and CD10 were defined as intestinal phenotype markers. DGCs were classified into 4 types based on the mucin phenotype according to the modified Ohmura classification9: gastric type (G type), ordinary type (O type), complete-intestinal type (CI type), and null type (N type). Briefly, the criteria for each phenotype were as follows:
•G type: HGM or ConA positive but MUC2 and CD10 negative (Fig 1)
•CI type: CD10-positive cells irrespective of staining for MUC2 that were negative for HGM and ConA (Fig 2)
•O type: gastric and intestinal phenotype markers simultaneously positively stained or positively for MUC2 alone
•N type: both gastric and intestinal phenotype markers stained negatively.
Detection of proliferation, apoptosis, and p53 overexpression Dewaxed paraffin sections were examined by the ABC method (Vector Laboratories, Burlingame, CA) method using the following primary antibodies: MIB-1 against Ki-67 antigen of proliferating cells (mouse IgG; Immunotech, Marseilles, France) and DO7 against p53 (mouse IgG diluted at 1:50; Dako, Glostrup, Denmark). The slides were treated with the antigen-retrieval technique based on microwave oven heating. Apoptotic cells in situ were detected by the terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate (dTUP) nick-end labeling (TUNEL) method described by Gavrieli et al.22 The slides were dewaxed and rehydrated through a graded alcohol series. The tissues were digested with 20 μg/mL proteinase K (Boehringer, Mannheim, Germany) for 30 minutes at 37°C. After treatment with a 2% H2O2 solution, the sections were preincubated with 100 mmol potassium cacodylate, 2 mmol cobalt chloride, 0.2 mmol dithiothreitol, pH 7.2, for 3 minutes, then incubated with the same buffer containing 0.3 U/μL terminal deoxynucleotidyl transferase (TdT) (Gibco BRL, Gaithersburg, MD) and 0.04 nmol/μL biotinylated dUTP (Boehringer, Mannheim, Germany) in a humid chamber at 37°C for 1 hour. The slides were rinsed in 30 mmol sodium citrate and 300 mmol sodium chloride for 30 minutes at room temperature and washed in phosphate-buffered saline (PBS). After blocking with 10% rabbit serum for 10 minutes and rinsing briefly in PBS, sections were incubated with ABC for 30 minutes at room temperature. Labeled cells were visualized with diaminobenzidine. The sections were then counterstained with hematoxylin. In each case, a minimum of 500 cells randomly selected from some fields in which gastric and/or intestinal mucin were positively stained were counted, and the fractions (%) of cells that showed positive nuclear staining for Ki-67 antigen and TUNEL were considered the proliferative index (PI) and apoptotic index (AI), respectively. PI and AI were counted independently by 2 physicians (N.S. and J.W.). The sections for p53 were considered positive when ≥20% tumor nuclei were stained; others were judged negative. DNA extraction Four 10-μm-thick tissue sections were serially cut from paraffin-embedded tissue blocks. DNA was extracted from the cancerous area in which cellular phenotype was expressed by immunohistochemistry, and from normal mucosa surrounding tumors. In this DNA extraction procedure, tissues were precisely microdissected under microscopic visualization, using a PixCell laser capture microdissection (LCM) system (Arcturus Engineering, Mountain View, CA) to avoid the DNA contamination of inflammatory or stromal cell nuclei and DNA of cancerous cells without overt phenotypical expressions (Fig 3).
Analysis of microsatellite instability and loss of heterozygosity by high-resolution fluorescent microsatellite analysis We examined 9 microsatellite loci on chromosomes for MSI and LOH using the following 9 microsatellite markers (2 mononucleotide, 7 dinucleotide repeat): 2p (BAT26), 2p (D2S123), 3p (D3S1067), 4q (BAT25), 5q (D5S346, D5S409), 13q (D13S153), 17p (D17S520), and 18q (D18S34). One primer of each primer pair was fluorescent-labeled at the 5' end. PCR was carried out in a 10 μL reaction volume containing 100 ng genomic DNA, 1X PCR buffer (PerkinElmer, Foster City, CA), 200 μmol/L of deoxynucleoside triphosphate, 600 μmol/L of each primer, and 1.5 U of AmpliTaq Gold polymerase (PerkinElmer). The MgCl2 concentration was 1.5 mmol/L. The following PCR cycles were used for amplification: 95°C for 10 minutes, then 30 cycles of 95°C for 45 seconds, 55°C for 1 minute, 72°C for 30 seconds. PCR products were evaluated with an ABI Prism 310 Genetic analyzer (PerkinElmer), based on capillary electrophoresis, and automated sizing of the alleles by GeneScan (Applied Biosystems, Foster City, CA) for MSI and LOH. MSI was defined as positive when unequivocal extra peak bands in tumor DNA that differed by a multiple of 2 base pairs in dinucleotide markers or 1 base pair in mononucleotide markers from DNA in normal mucosa were observed, and was also characterized by the appearance of drastic additional alleles in the tumor DNA. The former type of MSI was considered a minor pattern (Fig 4A) and the latter type a major pattern (Fig 4B) according to Tokunaga's classification.13 On the basis of these criteria, tumors were considered to be of the major pattern if they exhibited at least 1 major pattern of MSI in each loci examined, and tumors were considered to be of the minor pattern if they had none of the major patterns of MSI, only minor patterns. Tumors were defined as MSI-H when unstable loci were observed in >30% and MSI-L when unstable loci were observed in <30%. 23 MSS was defined if no unstable loci were found. LOH was determined to be positive when the allelic ratio (AR; ie, T1:T2/N1:N2) was <0.7 ( Fig 4C), 24 as Kobayashi et al used in their gastric cancer study. 25 Briefly, T1 and N1 are the highest peak areas of the shorter allele for cancer and normal mucosa sample, and T2 and N2 are the highest peak areas of the longer allele. In cases in which AR was >1.0, the ratio was inverted (1/AR) to obtain results in the range of 0 to 1. Tumors exhibiting MSI at a given locus were not evaluated for LOH. Frameshift mutation analysis To detect frameshift mutations in MSI-positive tumors, the 2 repetitive mononucleotide sequences in coding exons, the poly (A)10 tract of transforming growth factor-β type II receptor (TGF-β RII) and the poly (G)8 tract of bcl-2-associated X protein (BAX), were amplified. PCR was performed for 32 cycles at 94°C for 1 minute, 55°C for 1 minute, and 72°C for 1 minute. Statistical analysis Statistical comparisons were performed by the Mann-Whitney U test between 2 independent groups and by the X2 test between 2 proportions. Statistical significance was defined as P <0.05. Results Of the 50 gastric cancers evaluated, 7 did not express any cellular phenotype (N type). Of the remaining 43 cases, 11 were G type (26%), 20 were O type (47%), and 12 were CI type (28%). Tumor sizes (mean ± standard deviation) were 2.4 ± 1.4 cm for G type, 2.7 ± 2.0 cm for O type, and 2.0 ± 1.4 cm CI type, differences that are not statistically significant. Cell kinetics in cellular phenotype With regard to the relationship between cell kinetics and cellular phenotype, PIs and AIs are listed in Table 1.
| | |  | | G type (n = 11) | O type (n = 20) | CI type (n = 12) | |
|---|
| PI (%; mean ± SE) | 44.1 ± 4.4 | 45.9 ± 3.4 | 35.2 ± 3.5 | |
| AI (%; mean ± SE) | 1.5 ± .4*,† | 4.0 ± .7* | 4.1 ± .8† | |
| AI/PI ratio | .04 ± .02§,∥ | .10 ± .02§ | .13 ± .03∥ | |
| *P < .05 and †P < .01; Comparison of 2 groups, §P < .05 and ∥P < .005. | | | | |
AI were significantly lower in G-type cancers than in O-type and CI-type cancers ( P = 0.01 and P = 0.006, respectively), but PI was not. Tumor progression should be analyzed on the balance between proliferation and cell loss. 26 Thus we calculated the value of AI against PI (AI/PI ratio) to provide an indication of net cell production. 27, 28 The AI/PI ratio (mean ± standard error) was 0.04 ± 0.02 for G type, 0.10 ± 0.02 for O type, and 0.13 ± 0.03 for CI type, significantly lower in G type than in O and CI types ( P = 0.02 and P = 0.001, respectively). Microsatellite instability and loss of heterozygosity status The results of genetic instabilities are presented in Table 2.
MSI was observed in 7 of G type (64%), 15 (75%) of O type (75%), and 6 of CI type (50%). The incidence of MSI-H, MSI-L, and MSS was not statistically significant among the 3 cellular phenotypes. As for MSI pattern, major type and minor type were seen in 6 (86%) and 1 (14%) of G type, in 11 (73%) and 4 (27%) of O type, and in 0 (0%) and 6 (100%) of CI type, respectively. The major type pattern of MSI showed higher incidence in G and O types than in CI type (P = 0.002). No correlation between LOH and cellular phenotype was found. Genetic alterations of D17S520 (17p) were observed in 17 of 41 (41%) informative cases, composed of 15 MSI and 2 LOH. Genetic instabilities at this locus (17p) in informative cases were observed in 4 G type (36%), 11 O type (58%), and 2 CI type (18%). Thus the rate of genetic instabilities on D17S520 was significantly lower in CI type than in O type (P = 0.03) (Table 3).
| | | | | G type (n = 11) | O type (n = 20) | CI type (n = 12) | P | |
|---|
| MSI | 7 (64%) | 15 (75%) | 6 (50%) | NS | |
|  MSI-H | 3 | 7 | 1 | | |
| MSI-L | 4 | 8 | 5 | | |
| MSI pattern | | | | | |
| Major pattern | 6 | 11 | 0 | .002*,† | |
| Minor pattern | 1 | 4 | 6 | | |
| LOH | 5 (45%) | 6 (30%) | 6 (50%) | NS | |
| Frame shift mutations | | | | | |
| TGF-βRII§ | 2 (29%) | 3 (20%) | 0 (0%) | NS | |
| BAX§ | 1 (14%) | 2 (13%) | 0 (0%) | NS | |
| *G type versus CI type and †O type versus CI type. §Parentheses in TGF-βRII and BAX indicate the percentage in the cases with MSI. | | | | |
| | | | | G type | O type | CI type | P | |
|---|
| Genetic instabilities of 17p | | | | | |
| Present | 4 | 11 | 2 | .03* | |
| Absent | 7 | 8 | 9 | | |
| p53 overexpression | | | | | |
| Present | 6 | 11 | 8 | NS | |
| Absent | 5 | 9 | 4 | | |
| *O type versus CI type. | | | | |
Target gene analysis Of the tumors with MSI, the frame shift mutations of cancer-related genes TGF-βRII and BAX were found in 2 (29%) and 1 (14%) of G-type cancers and in 3 (20%) and 2 (13%) of O-type cancers, respectively. However, no tumor mutated for TGF-βRII and BAX was found in CI-type cancers (Table 2). All of the tumors with the frame shift mutations were noted in MSI-H status. p53 immunohistochemistry p53 Immunohistochemistry was performed to analyze p53 expression for an association with genetic alterations at locus D17S520. Immunostaining with p53 antibody (DO7) was positive in 6 (55%) of G type, 11 (56%) of O type, and 8 (67%) of CI type (Table 3). No significant difference in incidence of p53 overexpression was found among the 3 cellular phenotypes. Further, there was no correlation between 17p chromosomal instabilities (D17S520) and p53 overexpression. Twenty-four cases had p53 overexpression by immunostaining and informative at D17S520l; however, only 8 (33%) of these showed genetic instabilities. Furthermore, 17 cases were immunonegative and informative, 9 (53%) of which showed genetic instabilities. Genetic alterations and cell kinetics MSI status was not associated with cell kinetics such as PI, AI, and AI/PI. Furthermore, AI was not associated with the frame shift mutations of TGF-βRII and BAX or with the genetic alterations and overexpression of p53 (data not shown). Discussion The present study clearly demonstrates that the amount of naturally occurring apoptosis against cell proliferation was significantly lower in G-type cancers than in O-type and CI-type cancers. Tumor progression should be considered in the context of both proliferative activity and cell loss.26 The AI/PI ratio is reportedly significantly lower in undifferentiated-type gastric cancers than in well-differentiated type cancers, indicating the more invasive, extensive, and rapidly growing nature of poorly differentiated cancers.27, 28 Thus our results probably mean that G-type cancers reflect their more aggressive nature when compared with other cellular phenotypic cancers in early stages of DGC. We recently observed a few important characteristics of gastric cancers with gastric phenotype: 1. The frequency of complex-type gastric cancer, comprised predominantly of DGCs with partly undifferentiated type carcinomas, tended to be higher in G-type cancers than in other phenotypes. 2. Differentiation of cancer cells was caused in G-type cancers with reduction of cell–cell and cell–extracellular matrix adhesion molecule such as syndecan-1, but not in other phenotypic cancers. These results led us to hypothesize that characteristics of DGCs with gastric phenotype might be similar to those of undifferentiated type carcinoma. From a cell kinetic standpoint, the present study supports this hypothesis as expected. Regarding the molecular features of undifferentiated-type carcinomas, it has been recently reported that MSI and LOH are less common in early stage of cancers, and epigenetic inactivation of E-cadherin via promotor hypermethylation may be an early critical event in the development of undifferentiated-type cancers.29 However, we did not find any difference in the prevalence of MSI and LOH among the 3 cellular phenotypes of DGCs. This indicates that G-type cancers show a different molecular pathogenesis than undifferentiated-type cancers. Previous reports suggest that the incidence of MSI is significantly higher in G-type cancers than in CI-type and O-type cancers, but the incidence of LOH is low in G-type cancers.9, 10 However, we were not able to confirm such a tendency. There are several possible explanations for these differences. First, it can be explained by the method of DNA extraction. The laser capture microdissection that we used in this study allows procurement of relatively pure tumor cell populations from the complex heterogeneous cell mixtures.30 Therefore, the specificity of genetic alterations in DNA extracted selectively from the area where each cellular phenotype expressed is higher than that in the hand-microdissected samples.31 Second, in assessing tumors for the presence of MSI and LOH, many investigators differ in terms of which and how many loci should be analyzed. Third, the method of analysis of MSI and LOH affects the outcome. Using conventional methods, the electrophoreitic profiles of PCR products may not always be reproducible.13, 32 Moreover, assessment of MSI (especially minor pattern) using an autoradiograph is difficult, as reported previously.13, 32, 33 Nonspecific stutter bands, seen as small fragments, can make autoradiographs difficult to interpret and quantitate. The HFRMA assay used in the present study allows for more accurate assessment of phenomena such as stutter or allele shift patterns.13 Therefore, it might be impossible to detect a minor-type pattern in MSI with conventional radioactive PCR. Interestingly, if only major pattern was considered in MSI, which is readily diagnosed even with conventional autoradiography, the incidence of MSI in G-type, O-type, and CI-type cancers changed to 55% (6 of 11), 55% (11 of 20) and 0% (0 of 12), respectively. These results are similar to those reported by Tamura et al9, 10 using the conventional method. Taking these results together, HFRMA, which detects objectively and precisely both minor type and major type for MSI, is superior to the conventional method. The genetic basis for differences in MSI between major type and minor type remains unclear. Oki et al34 stated that the major pattern in MSI had not been found in various types of human mismatch repair gene-deficient cell lines; however, this type of alteration was seen in patients who are clinically at higher risk for cancers.13 In the present study, the incidence of the major type pattern of MSI was significantly higher in G- and O-type than in CI-type cancers. These results strongly support the observation that cancers with gastric phenotype have greater malignant potential than those with intestinal phenotype.6, 7 Furthermore, we found that the incidence of genetic instability at 17p (p53) locus was the lowest in CI-type cancers. Commonly, p53 mutations are considered bad prognostic markers in the gastric cancers. On the basis of this finding, CI-type cancers are considered to have less-aggressive behavior than the other phenotypic cancers. MSI status is significantly correlated with cell kinetics (ie, AI, PI and AI/PI) in the sporadic colorectal cancers.18 However, we failed to detect any such correlation in DGCs. In addition, no correlation between AI and p53 overexpression and genetic instabilities at 17p locus was found. There are reports that AI differs in tumors with genetic mutation and tumors with protein overexpression of p53,35, 36 and a lack of correlation between mutation of the p53 gene and overexpression of the p53 protein has been reported in some studies of gastric cancers.37, 38 Apoptosis is regulated by various oncogenes and suppressor genes, and occurs in both a p53-dependent and p53-independent manner in gastric cancers.28 The present data therefore suggest that apoptosis of DGCs is not enhanced by a functional loss of p53, which is derived from genetic alterations at 17p and detected by immunohistochemical overexpression. Most cancers with MSI harbor frame-shift mutations in coding mononucleotide repeat sequences in cancer-related genes. Mutations of these target genes, such as TGF-βRII and BAX, are considered to affect not only carcinogenesis, but also cell proliferation and apoptosis.39, 40 Frame-shift mutations of TGF-βRII and BAX were not associated with cell kinetics. It is thought that TGF-βRII and BAX gene mutations may be involved in the progressive stage of gastric carcinogenesis.41, 42 Interestingly, no mutations of TGF-βRII and BAX were found in CI-type cancers, and all of the G-type and O-type cancers with the frame-shift mutations had MSI-H status. Our present study and the previous study have revealed that gastric cancers with gastric phenotype have greater malignant potential than those with intestinal phenotype. In EGCs with gastric phenotype, endoscopic mucosal resection (EMR) or minimal invasive surgery should be considered as curative therapy. The likelihood of complete curative therapy may decline with metastasis or recurrence, if EMR is performed to those cancers with gastric phenotype.8 However, no significant difference in clinicopathologic factors among the 3 cellular phenotypes was seen on evaluation with hematoxylin and eosin staining (data not shown), in agreement with the results of Kabashima et al.8 Therefore, we are in agreement with their conclusion that the phenotypic classification by immunohistochemistry is considered useful in evaluating malignant potential of cancers. In conclusion, this study is the first to show that cell kinetics and mutation patterns of MSI clearly differ between G-type and CI-type cancers, strongly suggesting that G-type cancers grow more aggressively than CI-type cancers. Based on results from the HRFMA system, MSI is not associated with the pathogenesis of these phenotypic cancers. The result of this investigation do not support the previous evidence;9, 10 G-type cancers were related to the “mutator” pathway, characterized by MSI, and CI-type cancers followed the “suppressor” pathway, characterized by p53 alterations. In the present study, the number of cancers analyzed is small, particularly considering that 3 different categories of cellular phenotypes are compared. Thus further investigations are needed with a larger sample size for differentiated-type gastric cancers to more clearly characterize the relationship between cellular phenotype and cell kinetic status and genetic alterations. However, we believe that the data presented in this paper based on a pilot study is very interesting and will generate interest in extending the studies. Hereafter we should attempt to investigate both genetic and epigenetic alterations by using other genetic markers to clarify the mechanisms involved in the development of each cellular phenotypic cancer. Acknowledgements We thank Dr. Gopala Kovvali for helpful suggestions and Dr. Zafar K. Mirza for comments on the manuscript. References 1.
1
Lauren P.
The two histological main types of gastric carcinoma: diffuse and so-called intestinal type carcinoma. An attempt at a histo-clinical classification.
Acta Pathol Microbiol Scand. 1965;64:31–49. MEDLINE 2.
2
Nakamura K, Sugano H, Takagi K.
Carcinoma of the stomach in incipient phase: Its histogenesis and histological appearances.
Jpn J Cancer Res. 1968;59:251–258. 3.
3
Endoh Y, Tamura G, Motoyama T, et al.
Well-differentiated adenocarcinoma mimicking complete-type intestinal metaplasia in the stomach.
Hum Pathol. 1999;30:826–832. Abstract |
Full-Text PDF (13355 KB)
|
CrossRef
4.
4
Sasaki I, Yao T, Nawata H, et al.
Minute gastric carcinoma of differentiated type with special reference to the significance of intestinal metaplasia, proliferative zone, and p53 protein during tumor development.
Cancer. 1999;85:1719–1729. 5.
5
Tatematsu M, Ichinose M, Miki K, et al.
Gastric and intestinal phenotypic expression of human stomach cancers as revealed by pepsinogen immunohistochemistry and mucin histochemistry.
Acta Pathol Jpn. 1990;40:494–504. MEDLINE 6.
6
Tajima Y, Shimoda T, Nakanishi Y, et al.
Gastric and intestinal phenotypic marker expression in gastric carcinomas and its prognostic significance: Immunohistochemical analysis of 136 lesions.
Oncology. 2001;61:212–220. 7.
7
Koseki K, Takizawa T, Koike M, et al.
Distinction of differentiated-type early gastric carcinoma with gastric type mucin expression.
Cancer. 2000;89:724–732. 8.
8
Kabashima A, Yao T, Sugimachi K, et al.
Relationship between biologic behavior and phenotypic expression in intramucosal gastric carcinomas.
Hum Pathol. 2002;33:80–86. Abstract | Full Text |
Full-Text PDF (295 KB)
|
CrossRef
9.
9
Ohmura K, Tamura G, Endoh Y, et al.
Microsatellite alterations in differentiated-type adenocarcinomas and precancerous lesions of the stomach with special reference to cellular phenotype.
Hum Pathol. 2000;31:1031–1035. Abstract |
Full-Text PDF (224 KB)
|
CrossRef
10.
10
Endoh Y, Sakata K, Tamura G, et al.
Cellular phenotypes of differentiated-type adenocarcinomas and precancerous lesions of the stomach are dependent on the genetic pathways.
J Pathol. 2000;191:257–263. MEDLINE |
CrossRef
11.
11
Loeb LA.
Mutator phenotype may be required for multistage carcinogenesis.
Cancer Res. 1991;51:3075–3079. MEDLINE 12.
12
Lee HS, Lee BL, Kim SH, et al.
Microsatellite instability in synchronous gastric carcinomas.
Int J Cancer. 2001;91:619–624. MEDLINE |
CrossRef
13.
13
Tokunaga E, Oki E, Oda S, et al.
Frequency of microsatellite instability in breast cancer determined by high-resolution fluorescent microsatellite analysis.
Oncology. 2000;59:44–49. 14.
14
Oda S, Oki E, Maehara Y, et al.
Precise assessment of microsatellite instability using high-resolution fluorescent microsatellite analysis.
Nucleic Acids Res. 1997;25:3415–3420. MEDLINE |
CrossRef
15.
15
Thibodeau SN, Bren G, Schaid D.
Microsatellite instability in cancer of the proximal colon.
Science. 1993;260:816–819. MEDLINE 16.
16
Lothe RA, Peltomaki P, Meling GI, et al.
Genomic instability in colorectal cancer: Relationship to clinicopathological variables and family history.
Cancer Res. 1993;53:5849–5852. MEDLINE 17.
17
Michael-Robinson JM, Biemer-Hüttmann AE, Purdie DM, et al.
Tumour-infiltrating lymphocytes and apoptosis are independent features in colorectal cancer stratified according to microsatellite instability status.
Gut. 2001;48:360–366. MEDLINE |
CrossRef
18.
18
Michael-Robinson JM, Reid LE, Purdie DM, et al.
Proliferation, apoptosis, and survival in high-level microsatellite instability sporadic colorectal cancer.
Clin Cancer Res. 2001;7:2347–2356. MEDLINE 19.
19
Rugge M, Correa P, Dixon MF, et al.
Gastric dysplasia: The Padova international classification.
Am J Surg Pathol. 2000;24:167–176. MEDLINE |
CrossRef
20.
20
Japanese Research Society for Gastric Cancer .
The General Rules for Gastric Cancer Study.
ed 12. Tokyo, Japan: , Kinbara; 1993;. 21.
21
Katsuyama T, Spicer SS.
Histochemical differentiation of complex carbohydrates with variants of the concanavarin A horseradish peroxidase method.
J Histochem Cytochem. 1978;26:223–250. 22.
22
Gavrieli Y, Sherman Y, Ben-Sasson SA.
Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation.
J Cell Biol. 1992;119:493–501. MEDLINE |
CrossRef
23.
23
Boland CR, Thibodeau SN, Hamilton SR, et al.
A national cancer institute workshop on microsatellite instability for cancer detection and familial predisposition: Development of intestinal criteria for the determination of microsatellite instability in colorectal cancer.
Cancer Res. 1998;58:5248–5257. MEDLINE 24.
24
Goessl C, Heicappell H, Münker R, et al.
Microsatellite analysis of plasma DNA from patients with clear cell renal carcinoma.
Cancer Res. 1998;58:4728–4732. MEDLINE 25.
25
Kobayashi K, Okamoto T, Takayama S, et al.
Genetic instability in intestinal metaplasia is a frequent event leading to well-differentiated early adenocarcinoma of the stomach.
Eur J Cancer. 2000;36:1113–1119. Abstract | Full Text |
Full-Text PDF (205 KB)
|
CrossRef
26.
26
Wyllie AH.
The biology of cell death in tumors.
Anticancer Res. 1996;5:131–136. MEDLINE 27.
27
Shinohara T, Ohshima K, Murayama H, et al.
Apoptosis and proliferation in gastric carcinoma: The association with histological type.
Histopathology. 1996;29:123–129. MEDLINE 28.
28
Ikeda M, Shomori K, Endo K, et al.
Frequent occurrence of apoptosis in an early event in the oncogenesis of human gastric carcinoma.
Virchows Arch. 1998;432:43–47. MEDLINE |
CrossRef
29.
29
Tamura G, Sato K, Akiyama S, et al.
Molecular characterization of undifferentiated-type gastric carcinoma.
Lab Invest. 2001;81:593–598. MEDLINE 30.
30
Bonner RF, Emmert-Buck M, Cole K, et al.
Laser capture microdissection: Molecular analysis of tissue.
Science. 1997;278:1481–1483. MEDLINE |
CrossRef
31.
31
Dillon D, Zheng K, Costa J.
Rapid, efficient genotyping of clinical tumor samples by laser-capture microdissection/PCR/SSCP.
Exp Mol Pathol. 2001;70:195–200. MEDLINE |
CrossRef
32.
32
Cawkwell L, Bell SM, Lewis FA, et al.
Rapid detection of allele loss in colorectal tumors using microsatellites and fluoresent DNA technology.
Br J Cancer. 1993;67:1262–1267. MEDLINE 33.
33
Cawkwell L, Li D, Lewis FA, et al.
Microsatellite instability in colorectal cancer: Improved assessment using fluorescent polymerase chain reaction.
Gastroenterology. 1995;109:465–471. Abstract |
Full-Text PDF (624 KB)
|
CrossRef
34.
34
Oki E, Oda S, Maehara Y, et al.
Mutated gene-specific phenotype of dinucleotide repeat instability in human colorectal carcinoma cells deficient in DNA mismatch repair.
Oncogene. 1999;18:2143–2147. MEDLINE |
CrossRef
35.
35
Simms LA, Radford-Smith G, Biden KG, et al.
Reciprocal relationship between the tumor suppressors p53 and BAX in primary colorectal cancers.
Oncogene. 1998;17:2003–2008. MEDLINE 36.
36
Biden KG, Simms LA, Cummings M, et al.
Expression of Bcl-2 protein is decreased in colorectal adenocarcinomas with microsatellite instability.
Oncogene. 1999;18:1245–1249. MEDLINE |
CrossRef
37.
37
Soong R, Robbins PD, Dix BR, et al.
Concordance between p53 protein overexpression and gene mutation in large series of common human carcinogenesis.
Hum Pathol. 1996;27:1050–1055. Abstract |
Full-Text PDF (752 KB)
|
CrossRef
38.
38
Keller G, Rudelius M, Vogelsang H, et al.
Microsatellite instability and loss of heterozygosity in gastric carcinoma in comparison to family history.
Am J Pathol. 1998;152:1281–1289. MEDLINE 39.
39
Rampino N, Yamamoto H, Ionov Y, et al.
Somatic frameshift mutations in the BAX gene in colon cancers of the microsatellite mutator phenotype.
Science. 1997;275:967–969. MEDLINE |
CrossRef
40.
40
Markowitz S, Wang J, Myeroff L, et al.
Inactivation of the type II TGF-beta receptor in colon cancer cells with microsatellite instability.
Science. 1995;268:1276–1277. MEDLINE 41.
41
Wu MS, Lee CW, Shun CT, et al.
Clinicopathological significance of altered loci of replication error and microsatellite instability–associated mutations in gastric cancer.
Cancer Res. 1998;58:1494–1497. MEDLINE 42.
42
Oliveira C, Seruca R, Seixas M, et al.
The clinicopathological features of gastric carcinomas with microsatellite instability may be mediated by mutations of different “target genes”: A study of the TGF-βRII, IGFII-R, and BAX genes.
Am J Pathol. 1998;153:1211–1219. MEDLINE Third Department of Internal Medicine, Asahikawa Medical College, Asahikawa, Japan ☆ Address correspondence and reprint requests to Jiro Watari, MD, Third Department of Internal Medicine, Asahikawa Medical College, 2-1-1-1 Midorigaoka-Higashi, Asahikawa, Hokkaido 078-8510, Japan. PII: S0046-8177(02)29702-6 doi:10.1053/hupa.2003.2 © 2003 Elsevier Science (USA). All rights reserved. | |
|
FULL TEXT29702-6/journalimage?img=PIIS0046817702297026.gr1.lrg.gif&fig=&kwhquery=&issn=0046-8177&ishighres=false&allhighres=false&free=yes','journalimage');)
29702-6/journalimage?img=PIIS0046817702297026.gr2.lrg.gif&fig=&kwhquery=&issn=0046-8177&ishighres=false&allhighres=false&free=yes','journalimage');)
29702-6/journalimage?img=PIIS0046817702297026.gr3.lrg.gif&fig=&kwhquery=&issn=0046-8177&ishighres=false&allhighres=false&free=yes','journalimage');)
29702-6/journalimage?img=PIIS0046817702297026.gr4.lrg.gif&fig=fig4&kwhquery=&issn=0046-8177&ishighres=false&allhighres=false&free=yes','journalimage');)