Vol. 4, 2129-2134, Septenther 1998 Clinical Cancer Research 2129

Tumor Angiogenesis and Micrometastasis in Bone Marrow of

Patients with Early Gastric Cancer’

Yoshihiko Maehara,2 Shota Hasuda, Toru Abe,

Eiji Oki, Yosbihiro Kakeji, Shinji Ohno, and

Keizo Sugimachi

Department of Surgery II, Faculty of Medicine, Kyushu University,

Fukuoka 812-8582, Japan

ABSTRACT

In a subset of patients with early gastric cancer, therewere recurrences of the disease after a curative resection

had been done. Direct evidence of tumor seeding in distantorgans at the time of surgery for gastric cancer is notavailable. An immunocytochemical assay for epithelial cy-tokeratin protein may fill this gap because it is a feature ofepithelial cells that would not normally be present in bonemarrow. From 1994-1997, the bone marrow of 45 patientswith early gastric cancer was examined for tumor cells,using immunocytochemical techniques and an antibody re-

acting with cytokeratin, a component of the intracytoplas-talc network of intermediate filaments. Intratumoral mi-

crovessels were stained with anti-CD31 monoclonalantibody. Clinicopathological characteristics were deter-mined for subjects with cytokeratin-positive cells in the bonemarrow. Of these 45 patients, 9 (20.0%) had cytokeratin-positive cells in the bone marrow at the time of primarysurgery. These positive findings were not related to tumoradvance-related factors of lymph node metastasis and dis-tinct lymphatic and vascular invasion. Microvessel densityin the primary tumor exceeded 2-fold in cytokeratin-positivecells, compared with findings in negative cells (P < 0.05).

Tumor cells in bone marrow are indicative of the general

disseminative metastasis in patients with early gastric can-

cer, and the metastatic potential was closely related to an-giogenesis in the primary tumor.

INTRODUCTION

Early gastric cancer is defined as a lesion in which the

depth of invasion is limited to the mucosa or mucosa and

submucosa, regardless of whether a regional lymph node me-

tastasis is evident on histological examination, and the postop-

erative prognosis is usually favorable (1, 2). The diagnosis of

Received 12/16/97; revised 3/27/98; accepted 6/5/98.The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely toindicate this fact.I This work was supported by a Grant-in-Aid for Scientific Researchfrom the Ministry of Education, Science, Sports and Culture of Japan.2 To whom requests for reprints should be addressed. Fax: 81 92 6425482.

early gastric cancer is much more frequent in Japan because

sophisticated diagnostic techniques are now available (3).

However, even after “curative” resection of early gastric

cancer, there are recurrences in a subset of patients (4-8).

Hematogeneous metastasis is the predominant route in such

cases. It is difficult to control tumor growth in cases of a distant

metastasis in these patients, and knowledge of the stage of

subclinical tumor cell dissemination is needed to design adju-

vast treatment. Various parameters, including DNA pboidy and

oncogenes, aid in predicting the recurrence and the prognosis of

early gastric cancer (9-12). A feature common to all of these

prognostic factors is that, from the excised tumor tissues, one

attempts to extrapolate to the malignant potential of occult cells

that may possibly be present. Diagnostic techniques presently

available are not sufficiently sensitive to detect unicellular or

oligoceblular micrometastasis.

Because cytokeratin proteins are essential constituents of

the cytoskebeton of both normal and malignant epithelial cells,

they can serve as reliable markers for the epithelial origin of

cells (13). The use of a monoclonal antibody against the

cytokeratin component expressed by all tumor cells derived

from simple epithebia facilitates identification of one of l0�

epitheliab tumor cells in bone marrow. Evidence of microme-

tastasis in the bone marrow means an early relapse and a poorer

clinical outcome for patients with gastric, coborectab, mammary,

and lung cancers (14-19). Schbimok et a!. (20) found no posi-

tive cells in a large series of bone marrow aspirates from 102

patients with no evidence of malignant epitheliab disease. We

reported the presence of micrometastasis in bone marrow for

32.6% patients with gastric cancer (21). However, the presence

of micrometastasis was not focused on patients with early gas-

tric cancer.

Angiogenesis is closely involved in tumor progression and

metastasis (22, 23). Weidner et a!. (24) reported that microves-

sd density evaluated by immunostaining for endotheliab cells

was an independent prognostic indicator in breast cancer. The

level of microvessel density particularly correlated with hemato-

geneous metastasis of gastric and coborectal cancers (25, 26).

CD-3 1 is a pbatebet/endothelial cell adhesion molecule and is a

more sensitive marker for endotheliab cells than is factor VIII

antigen (27, 28). We examined micrometastasis in bone marrow

from patients with early gastric cancer, using an anti-cytokeratin

monoclonal antibody, and we determined the microvessel den-

sity in the primary tumor, using an anti-CD31 monocbonal

antibody. Bone marrow aspirates were taken immediately prior

to the initial surgery done from 1994 to 1997.

PATIENTS AND METHODS

Patients. This study included 45 unselected Japanese

patients with primary early gastric cancer, all of whom under-

went curative gastric resection in the Department of Surgery II,

Kyushu University, from 1994 to 1997. The depth of invasion

Research. on February 18, 2020. © 1998 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

2130 Gastric Cancer and Bone Marrow Micrometastasis

Table I Clinicopathological characteristics of patients with earlygastric cancer with and without cytokeratin-positive cells in the

bone marrow

Cytokeratin- Cytokeratin-negative cases positive cases

(n36) (n9) P

GenderMale 25 6 NS�Female I I 3

Age (yr) 60.9 ± 1 1 8b 56.4 ± 6#{149}8b NSTumor maximal diameter (cm) 2.81 ± 2.14” 2.52 ± 2.26” NS

Location of tumorUpper 9 2 NSMiddle 17 3Lower 10 4

HistologyDifferentiated 22 4 NSUndifferentiated 14 5

Depth of invasionMucosa 21 6 NSSubmucosa 15 3

Lymph node metastasisNegative 30 6 NSPositive 6 3

Lymphatic invasionNegative 3 1 8 NSPositive 5 1

Vascular invasionNegative 3 1 8 NS

Positive 5 1

Gastric resectionPartial 25 7 NSTotal 11 2

Lymph node dissection’Dl 2 1 NSD2andD3 34 8

a NS, not significant.

,) Mean ± SD.C Dl , complete removal of group 1 lymph node alone; D2, corn-

plete removal of groups 1 and 2 lymph nodes; and D3, completeremoval of groups 1, 2, and 3 lymph nodes.

was mucosa in 27 patients and mucosa and submucosa in 18

subjects. Standardized procedure was that a gastric resection

was done, after determining the resection line 3 cm apart from

the macroscopic edge of the localized tumor, and 6 cm for the

infiltrative tumor (29, 30). Prophylactic lymph node dissection

of more than D2 resection was carried out (3 1). Complete

excision of invaded organs was done, irrespective of the number

of sites on the organs, when there was no evidence of peritoneab

dissemination, liver metastasis, and widespread nodal involve-

ment (32). All patients were examined clinically and patholog-

ically with respect to the factors given in Table 1. Pathological

diagnosis and classification of the resected gastric cancer tissues

were made according to the General Rules for the Gastric

Cancer Study in Surgery and Pathology in Japan (33, 34).

Informed consent to participate in this study was obtained from

all patients prior to their surgery.

Bone Marrow Specimens. Details are given in our study

published previously (21). Preoperativeby, 1-2 ml of bone mar-

row aspirates from the sternum were taken in syringes contain-

ing 100 units of heparinlmb marrow, and bone marrow cells

were prepared. After density centrifugation through Ficoll-

Hypaque (400 X g for 30 mm), mononuclear cells were cob-

bected from the interphase. The cells were suspended with 0.5

ml RPMI 1640 containing 10% FCS, yielding a concentration of2 X l06/ml; the cells were then smeared on glass slides and

fixed with acetone (30 mm at 4#{176}C).For immunostaining, the

monocbonal antibody CK2 (IgGl; Boehringer Marntheim,

Mannheim, Germany) or CAM 5.2 (IgG2�, Becton Dickinson,

CA) was used at a concentration of 0.2 p�g/ml (17). This anti-

body recognizes intracellular cytokeratin component no. 1 8, an

intermediate filament representing the intracellular network of

the cytoskeleton that is expressed in simple epithelia and no-

where else. The antibody reaction was developed using the

labeled avidin-biotin technique (35), and biotin-labeled anti-

body and alkaline-phosphatase-labeled avidin were used se-

quentially. Naphthol-AS-BI-phosphate was used as a substrate

of alkaline phosphatase, and the released naphthol-AS-BI was

coupled with hexazotized new fuchsin. Endogenous phospha-

tase was inhibited by preincubation with levamisobe. Cells con-

taming cytokeratin were stained bright red. Two observers

(Y. M. and Y. K.) examined the positivity of micrometastasis of

bone marrow independently.

Immunohistochemistry of CD31. Tumors were cob-

lected and fixed in 10% formalin. Sections of 5-p.m thickness

from paraffin-embedded blocks were deparaffinized in xybene

and rehydrated in a graded series ofethanol. After quenching the

endogenous peroxidase activity in methanol containing 0.3%

(v/v) hydrogen peroxidase for 30 mm, sections were pretreated

with 12.5 mg of protease type XXIV (Sigma Chemical Co., St.

Louis, MO)/100 ml PBS (pH 7.4) for 15 mm at 37#{176}C.Nonspe-

cific binding was blocked by treatment with 10% (v/v) normal

goat serum for in PBS for 15 mm. Primary mouse anti-CD31

antibody (1:50 at 4#{176}C;DAKO Corp., Carpinteria, CA) was

applied to the sections, and the preparations were incubated

overnight in a moist chamber (36, 37). After washing in PBS,

biotinylated goat anti-mouse immunogbobulin G (Vector Labo-

ratories, Burlingame, CA) diluted 1:200 was applied, followed

by incubation for 30 mm at room temperature. After a thorough

washing in PBS, peroxidase-conjugated streptoavidin (DAKO

LSAB kit; DAKO) was applied, and the preparations were

incubated for 30 mm. Peroxidase labeling was developed with

diaminobenzidine and hydrogen peroxidase, and nuclear coun-

terstaining was performed with Mayer’s hernatoxylin solution

(Fig. 1). Specificity of binding for all antibodies was examined

by applying nonimmune sera instead of specific antibodies.

Microvessel Counting. Evaluation of microvesseb den-

sity (as a continuous variable) was determined, using the mod-

ified technique of Weidner et a!. (24). The entire tumor section

was systematically scanned at X 100 to search for areas of the

most intense neovascularization; these were identified as having

the highest density of red-brown staining, CD31-positive cells

or cell clusters. These neovascular “hotspots” were included into

the counts only if they were adjacent to tumor tissue. Whenever

a highly vascubarized area was evident at X 100, individual

microvessebs were counted on a single X 200 field (1 mm2) in

this area. Any red-brown-staining endothelial cell or endothelial

cell clusters, clearly separated from adjacent microvessels, were

regarded as a single, countable microvessel. Neither vessel

lumens nor RBCs were used to define a microvesseb. Immu-

nopositive macrophages and plasma cells were excluded on

Research. on February 18, 2020. © 1998 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

l� a� � �� �.�.

�

Clinical Cancer Research 2131

Fig. 1 Immunohistochemicalstaining of endothelial cellswith antibody against CD-31.X200.

morphological grounds. Results were expressed as the mean of

values in five fields.

Postoperative Chemotherapy. All of the patients with

cytokeratin-positive cells were treated with postoperative adju-

vant chemotherapy. An iv. injection of 10 mg of mitomycin C

(Kyowa Hakko Co., Japan) was given on the day of operation,

and fluorinated pyrimidine UFT (Taiho Pharmaceutical Co.,

Japan; Ref. 38) p.o. at a daily dose of 400 mg was started 2

weeks after the operation and was continued for 1 year.

Statistical Analysis. The BMDP Statistical Package pro-

gram (BMDP, Los Angeles, CA) for the IBM 3090 mainframe

computer was used for all analyses (39). The BMDP P4F and

P3S programs were used for the x2 test and the Mann-Whitney

test to compare data on patients with and without cytokeratin-

positive cells in their bone marrow. The bevel of significance

was P < 0.05.

RESULTS

Cytokeratin-positive cells were present in 9 of 45 (20.0%)

patients with early gastric cancer, the depth of penetration being

the mucosa or mucosa and submucosa. Alkaline phosphatase-

stained cells in cytocentrifuge preparations varied from a single

cell to a cluster of 10 cells, as determined histologically.

Cytokeratin-positive cells were confirmed to be cancer cells,

determined using Papanicolaou staining. Seeding of cancer cells

had already occurred in these cases, even though extensive

lymph node dissection had been done and a surgically “cura-

tive” operation was carried out.

In a patient with mucosal gastric cancer, the tumor meas-

ured only 0.9 X 0.6 cm, was a macroscopically depressed type

(lIc), localized in the angle of the stomach, and was limited to

the mucosa (Fig. 2, A and B). In this patient, a small cluster of

cytokeratin-positive cells was present in bone marrow aspirates

(Fig. 2C).

Micrometasta.sis in the Bone Marrow and Clinicopath-

obogical Factors. Positive findings of micrometastasis in the

bone marrow of patients with early gastric cancer did not

depend on gender, age, tumor size, tissue differentiation or

location ofthe tumor (Table 1). The depth of penetration was the

mucosa and mucosa and submucosa, and the tumor size was

smaller. The rates of lymph node metastasis and distinct lym-

phatic and vascular involvements were low, and these events

were not rebated to the presence of micrometastasis. There were

evidently no cytokeratin-positive micrometastatic cells in the

lymph nodes.

Micrometastasis in the Bone Marrow and MicrovesselDensity. Microvesseb counts determined by anti-CD3 1 anti-

body in primary gastric cancer tissues were 13.5 ± 6.0 for the

cytokeratin-negative patients and 27.9 ± 8.3 for the cytokeratin-

positive ones, with a statistical significance of P < 0.05

(Table 2). Thus, the presence of micrometastasis in the bone

marrow showed a close relation to tumor angiogenesis.

DISCUSSION

Even after curative resection, there can be recurrences of

the cancer in patients with early gastric malignancy (4-8). The

most frequent mode of recurrence is by a hematogeneous

spread, and tumor cells have already disseminated to distant

organs at the time of surgery. Availability of a highly sensitive

method would aid in predicting metastatic potential and clinical

outcome, and more effective treatments could be designed.

Clinicopathological factors of age of the patient, tissue differ-

entiation, growth pattern of pen A type, and nodal involvement

were reported to be prognostic for the occurrence of recurrences

Research. on February 18, 2020. © 1998 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

A

�a�i.�,, � �

� �t�IY�J .3,

, �:.�

CIsp”

2132 Gastric Cancer and Bone Marrow Micrometastasis

Fig. 2 Resected tissue specimen and microphotographs from a patientwith mucosal gastric cancer. Macroscopic size of the tumor was 0.6 X0.9 cm and the depressed type (A). Cancer cells were limited in the depthof the gastric mucosa (B; X 18). High-power photomicrograph of bonemarrow specimen reveals a small cluster of malignant cells (C; X 1000).

in early gastric cancer ( 1 , 7, 40). Gastric cancer markers may

provide prognostic information independent of and complemen-

tary to conventional parameters, including growth potential,

oncogenes, tumor-suppressor genes, and DNA flow cytometry,

as well as other growth factors (9 -1 2, 4 1 ). The common char-

acteristic of these prognostic factors is that they correlate a

property of the primary tumor with the subsequent outcome.

The method we have described here relates to aspects of

the actual behavior of the tumor, microscopic dissemination of

cancer cells in the bone marrow. There are reports of microme-

tastasis in the bone marrow of patients with gastric, coborectal,

Table 2 Microvessel density of early gastric cancer with andcytokeratin-positive cells in the bone marrow

without

Cytokeratin- Cytokeratin-negative cases positive cases

Factor (n = 36) (n 9) P

Microvessel density 13.5 ± 6ff’ 27.9 ± 8.3” 0.05

a Mean ± SD.

breast, prostate, or lung cancers (13-19, 42). Evidence of mi-

crometastasis means an early relapse, and the clinical outcome

for these patients can be predicted. We also found cytokeratin-

positive cells in bone marrow of our patients with gastric cancer,

and the micrometastasis in the bone marrow did not correlate

with p53 overexpression and proliferating activity of the tumor

(21).

Micrometastasis was noted in 20.0% of our patients with

early gastric cancer; thus, seeding of cancer cells can occur even

in the early stages of the cancer. In these cases, small tumors are

just an early manifestation of systemic disease, and metastases

has already occurred. However, the cancer cells were single to

at most 10, and a few cells were present in bone marrow;

however, vascular involvement in primary tumor or lymph node

metastasis is difficult to identify.

Tanigawa et al. (25, 26) reported that tumors that devel-

oped hematogeneous metastasis after surgery had significantly

higher microvessel density than did tumors with rebated perito-

neal metastasis and nonmetastatic tumors. Vascubarization is

usually required for tumor cells to enter the blood circulation

(23). Newly formed tumor vessels are devoid of or backing in

smooth muscle, are tortuous and sinusoidal, have increased

vascular length and diameter, have incomplete endothebial cell

lining and basement membrane, and are prone to spontaneous

hemorrhage and/or thrombosis, thus enabling tumor cells to

enter circulating systems (24, 43). In such cases, cancer cells

released from the primary site would be transported to the bone

marrow. Because the presence of micrometastatic cells in the

bone marrow was closely related to angiogenesis in the primary

tumor, detection of micrometastasis in the bone marrow does

have clinical significance when attempting to evaluate the he-

matogeneous metastatic potential of gastric cancer.

A higher sensitivity test to detect micrometastasis in the

bone marrow can be achieved by making use of the reverse

transcription-PCR (44-46). However, this method needs further

standardization, PCR primers specific for each RNA tool have

to be designed, the optimal number of PCR cycles has to be

defined, as do best cutoff levels, and so on (18, 47).

Overt bone or skeleton metastases are rare in patients with

gastric cancer; however, bone marrow is more often involved

than expected based on clinical findings (17). The apparent

discrepancy between clinically rare bone metastases and the

marrow micrometastases frequently detected by immunocyto-

chemistry can be explained by a reduced proliferative behavior

of the cells and often invoked state of dormancy (20, 42). The

capacity of tumor cells to proliferate in the bone marrow and to

manifest metastasis depends on the microenvironment. Jauch et

a!. ( 1 8) reported that positive bone marrow aspirations are a

surrogate marker of general tumor-cell dissemination or mini-

Research. on February 18, 2020. © 1998 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

Clinical Cancer Research 2133

mal residual disease, rather than the start of metastatic growth in

the skeletal system. The survival time was shorter in the cyto-

keratin-positive group than in negative group in cases of gastric

cancer (16-18). Therefore, these patients may be at a higher risk

for complications arising from peritoneal dissemination or liver

metastasis with no manifest clinical metastasis in the bone

marrow (2, 48, 49).

Because cytokeratin-positive cells were present in the bone

marrow of our patients with early gastric cancer, these cells can

serve as valid indicators of the metastatic activity of these

cancers. Patients presenting with disseminated cytokeratin-

positive cells at the time of primary surgery can be followed to

detect any distant metastasis. Therefore, such patients may be

good candidates for postoperative adjuvant trials, even when a

curative resection is done for patients with early gastric cancer.

ACKNOWLEDGMENTS

We thank M. Ohara for comments and K. Miyammoto and J.

Tsuchihashi for technical assistance.

REFERENCES

I. Machare, Y., Okuyama, T., Oshiro, T., Baba, H., Ansi, H., Akazawa,K., and Sugimachi, K. Early carcinoma of the stomach. Surg. Gynecol.Obstet., 177: 593-597, 1993.

2. Macham, Y., Emi, Y., Baba, H., Adachi, Y., Akazawa, K., Ichiyoshi,Y., and Sugimachi, K. Recurrences and related characteristics of gastriccancer. Br. J. Cancer, 74: 975-978, 1996.

3. Macham, Y., Oshiro, T., Oiwa, H., 0th, 5., Baba, H., Akazawa, K.,and Sugimachi, K. Gastric cancer in patients over 70 years of age. Br. J.Surg., 82: 102-105, 1995.

4. Ichiyoshi, Y., Toda, T., Minamisono, Y., Nagasaki, S., Yakeishi, Y.,and Sugimachi, K. Recurrence in early gastric cancer. Surgery, 107:489-495, 1990.

5. Macham, Y., Orita, H., Okuyama, T., Moriguchi, S., Tsujitani, S.,Korenaga, D., and Sugimachi, K. Predictors oflymph node metastasis inearly gastric cancer. Br. J. Surg., 79: 245-247, 1992.

6. Orita, H., Matsusaka, T., Wakasugi, K., Kume, K., Fujinaga, Y.,Fuchigami, T., and Iwashita, A. Clinicopathologic evaluation of recur-rence in early gastric cancer. Surg. Today, 22: 19-23, 1992.

7. Sano, T., Sasako, M., Kinoshita, T., and Maruyama, K. Recurrenceof early gastric cancer. Follow-up of 1475 patients and review of theJapanese literature. Cancer (Phila.), 72: 3 174-3 178, 1993.

8. Shiozawa, N., Kodama, M., Chida, T., Arakawa, A., Tur, G. E., andKoyama, K. Recurrent death among early gastric cancer patients: 20-years’ experience. Hepato-Gastroenterol., 41: 244-247, 1994.

9. Korenaga, D., Saito, A., Baba, H., Watanabe, A., Okamura, T.,Macham, Y., and Sugimachi, K. Cytophotometrically determined DNAcontent, mitotic activity, and lymph node metastasis in clinical gastriccancer. Surgery, 107: 262-267, 1990.

10. Yonemura, Y., Ninomiya, I., Ohoyama, S., Fushida, S., Kimura, H.,Tsugawa, K., Kamata, T., Yamaguchi, A., Miyazaki, I., Endou, Y.,Tanaka, M., and Sasaki, T. Correlation of c-erbB-2 protein expressionand lymph node status in early gastric cancer. Oncology, 49: 363-367,1992.

11. Kakeji, Y., Korenaga, D., Tsujitani, S., Baba, H., Anal, H.,Macham, Y., and Sugimachi, K. Gastric cancer with p53 overexpressionhas high potential for metastasising to lymph nodes. Br. J. Cancer,67: 589-593, 1993.

12. Joypaul, B. V., Hopwood, D., Newman, E. L., Qureshi, S., Grant,A., Ogston, S. A., Lane, D. P., and Cuschieri, A. The prognosticsignificance of the accumulation of p53 tumour-suppressor gene proteinin gastric adenocarcinoma. Br. J. Cancer, 69: 943-946, 1994.

13. Schlirnok, G., and RiethmUller, G. Detection, characterization andtumorigenicity of disseminated tumor cells in human bone marrow.Semin. Cancer Biol., 1: 207-2015, 1990.

14. Mansi, J. L., Berger, U., Easton, D., McDonnell, T., Redding,W. H., Gazet, J-C., McKinna, A., Powles, T. J., and Coombes, R. C.Micrornetastases in bone marrow in patients with primary breast cancer:evaluation as an early predictor of bone metastases. Br. Med. J., 295:1093-1096, 1987.

15. Cote, R. J., Rosen, P. P., Lesser, M. L., Old, L. J., and Osborne,M. P. Prediction of early relapse in patients with operable breast cancerby detection of occult bone marrow micrometastases. J. Clin. Oncol., 9:

1749-1756, 1991.

16. Schlimok, G., Funke, I., Pantel, K., Strobel, F., Lindemann, F.,Wiue, J., and Riethmtlller, G. Micrometastatic tumour cells in bonemarrow of patients with gastric cancer: Methodological aspects ofdetection and prognostic significance. Eur. J. Cancer, 11: 1461-1465,

1991.

17. Lindemann, F., Schlimok, G., Dirschedl, P., Witte, J., andRiethmtiller, G. Prognostic significance of micrometastatic tumourcells in bone marrow of coborectal cancer patients. Lancet, 340: 685-689, 1992.

18. Jauch, K-W., Heiss, M. M., Gruetzner, U., Funke, I., Pantel, K.,Babic, R., Eissner, H-i., Riethm#{252}ller,G., and Schildberg, F-W. Frog-nostic significance of bone marrow micrometastases in patients withgastric cancer. J. Clin. Oncol., 14: 1810-1817, 1996.

19. Cote, R. J., Beanie, E. i., Chaiwun, B., Shi, S-R., Harvey, I., Chen,S-C., Sherrod, A. E., Groshen, S., and Taylor, C. R. Detection of occultbone marrow micrometastases in patients with operable lung carcinoma.Ann. Surg., 222: 415-425, 1995.

20. Schlimok, G., Funke, I., Bock, B., Schweiberer, B., Witte, i., andRiethmUller, G. Epithelial tumor cells in bone marrow of patients withcoborectal cancer: immunocytochemical detection, phenotypic charac-terization, and prognostic significance. J. Clin. Oncol., 8: 83 1-837,1990.

21 . Maehara, Y., Yamarnoto, M., Oda, S., Baba, H., Kusumoto, T.,Ohno, S., Ichiyoshi, Y., and Sugimachi, K. Cytokeratin-positive cells inbone marrow for identifying distant micrometastasis of gastric cancer.Br. J. Cancer, 73: 83-87, 1996.

22. Folkman, J., Watson, K., Ingber, D., and Hanahan, D. Induction ofangiogenesis during the transition from hyperplasia to neoplasia. Nature(Lond.), 339: 58-61, 1989.

23. Folkman, I. What is the evidence that tumors are angiogenesisdependent? J. Natl. Cancer Res., 82: 4-6, 1990.

24. Weidner, N., Folkman, J., Pozza, F., Bevilaqoua, P., Allred, E. N.,Moore, D. H., Mcli, S., and Gasparmni, G. Tumor angiogenesis. A newsignificant and independent prognostic factor in early-stage breast car-cinoma. I. Nail. Cancer Inst., 84: 1875-1887, 1992.

25. Tanigawa, N., Amaya, H., Matsumura, M., Shimomatsuya, T.,Horiuchi, T., Muraoka, R., and lid, M. Extent of tumor vasculanzationcorrelates with prognosis and hematogenous metastasis in gastric car-cinomas. Cancer Res., 56: 2671-2676, 1996.

26. Tanigawa, N., Amaya, H., Matsumura, M., Lu, C., Kitaoka, A.,Matsuyama, K., and Muraoka, R. Tumor angiogenesis and mode ofmetastasis in patients with colorectal cancer. Cancer Res., 57: 1043-1046, 1997.

27. Toi, M., Kashitani, J., and Tominaga, T. Tumor angiogenesis is anindependent prognostic indicator in primary breast carcinoma. mt. I.Cancer, 55: 371-374, 1993.

28. Horak, E. R., Leek, R., Klenk, N., Lejeune, S., Smith, K., Stuart, N.,Greenall, M., Stepniewska, K., and Harris, A. L. Angiogenesis, assessedby platelet/endothelial cell adhesion molecule antibodies, as indicator ofnode metastases and survival in breast cancer. Lancet, 340: 1120-1124,

1992.

29. Kawasaki, S. A clinicopathological study on upward intramuralextension of cancer of the stomach. Fukuoka Acta Medica, 66: 1-23,

1975 (in Japanese with English Abstract).

Research. on February 18, 2020. © 1998 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

2134 Gastric Cancer and Bone Marrow Micrometastasis

30. Bozzetti, F., Bonfanti, G., Bufalino, R., Menotti, V., Persano, S.,Andreola, S., Doci, R., and Gennari, L. Adequacy of margins of resec-tion in gastrectomy for cancer. Ann. Surg., 196: 685-690, 1982.

31. Machare, Y., Okuyama, T., Moriguchi, S., Orita, H., Kusumoto, H.,Korenaga, D.. and Sugimachi, K. Prophylactic lymph node dissection inpatients with advanced gastric cancer promotes increased survival time.Cancer (Phila.), 70: 392-395, 1992.32. Korenaga, D., Okamura, T., Babe, H., Saito, A., and Sugimachi, K.

Results of resection of gastric cancer extending to adjacent organs. Br. J.Surg., 75: 12-15, 1988.

33. Japanese Research Society for Gastric Cancer. The General Rulesfor the Gastric Cancer Study in Surgery and Pathology. Part I. ClinicalClassification. Jpn. J. Surg., 11: 127-139, 1981. Part II. Histologicalclassification of gastric cancer. Jpn. J. Surg., 11: 140-145, 1981.

34. Japanese Research Society for Gastric Cancer (ed.), Japanese Cbas-sification of Gastric Carcinoma. Tokyo: Kanehara and Co., Ltd., 1995.

35. Guesdon, J-L., Ternynck, T., and Avrameas, S. The use of avidin-

biotin interaction in immunoenzymatic techniques. J. Histochem. Cyto-chem., 27: 1131-1139, 1979.

36. Bossi, P., Viale, G., Lee, A. K. C., Alfano, R., Coggi, G., andBosari, S. Angiogenesis in colorectal tumors: microvessel quantificationin adenomas and carcinomas with climcopathobogical correlations. Can-cer Res., 55: 5049-5053, 1995.

37. Vermeulen, P. B., Verhoeven, D., Fierens, H., Hubens, G.,Goovaerts, G., Van Marck, E., Dc Bruijn, E. A., Oosterom, A. T., andDirix, L. Y. Microvessel quantification in primary colorectal carcinoma:an immunohistochemical study. Br. J. Cancer, 71: 340-343, 1995.

38. Sugimachi, K., Macham, Y., Ogawa, M., Kakegawa, T., andTomita, M. Dose intensity of uracil and tegalur in postoperative chem-otherapy for patients with poorly differentiated gastric cancer. CancerChemother. Pharmacol., 40: 233-238, 1997.

39. Dixon, W. I. (ed.), BMDP Statistical Software. Berkeley, CA:University of California Press, 1988.

40. Oiwa, H., Macham, Y., Ohno, S., Sakaguchi, Y., Ichiyoshi, Y., andSugimachi, K. Growth pattern and p53 overexpression in patients withearly gastric cancer. Cancer (Phila.), 75: 1454-1459, 1995.

41. Kakeji, Y., Macham, Y., Orita, H., Emi, Y., Ichiyoshi, Y.,Korenaga, D., and Sugimachi, K. Argyrophilic nucleolar organizerregion in endoscopically obtained biopsy tissue: a useful predictor ofnodal metastasis and prognosis in carcinoma of the stomach. J. Am.Coil. Surgeons, 182: 482-487, 1996.

42. Wood, D. P., Jr., Banks, E. R., Humphreys, S., McRoberts, J. W.,and Rangnekar, V. M. Identification ofbone marrow micrometastases in

patients with prostate cancer. Cancer (Phila.), 74: 2533-2540, 1994.

43. Dewhirst, M. W. Angiogenesis and blood flow in solid tumors. In:B. A. Teicher (ed.), Drug Resistance in Oncology, pp. 3-24. New York:Marcel Dekker, Inc., 1993.

44. Datta, Y. H., Adams, P. T., Drobyski, W. R., Ethier, S. P., Terry,V. H., and Roth, M. S. Sensitive detection of occult breast cancer by thereverse-transcriptase polymerase chain reaction. J. Cbin. Oncob., 12:

475-482, 1994.

45. Moreno, J. G., Croce, C. M., Fischer, R., Monne, M., Vihko, P.,Mulholland, S. G., and Gomella, L. G. Detection of hematogenousmicrometastasis in patients with prostate cancer. Cancer Res., 52:

6110-6112, 1992.

46. Noguchi, S., Aihara, T., Nakamori, S. Motomura, K., Inaji, H.,

Imaoka, S., and Koyama, H. The detection of breast carcinoma micro-metastases in axillary lymph nodes by means of reverse transcriptase-polymerase chain reaction. Cancer (Phila.), 74: 1595-1600, 1994.

47. Schoenfeld, A., Luqmani, Y., Smith, D., O’Reilly, S., Shousha, S.,Sinnett, H. D., and Coombes, R. C. Detection of breast cancer micro-metastases in axilbary lymph nodes by using pobymerase chain reaction.Cancer Res., 54: 2986-2990, 1994.

48. Machare, Y., Moriguchi, S., Kakeji, Y., Kohnoe, S., Korenaga, D.,Haraguchi, M., and Sugirnachi, K. Pertinent risk factors and gastric

carcinoma with synchronous peritoneal dissemination or liver metasta-sis. Surgery, 110: 820-823, 1991.

49. Moriguchi, S., Machare, Y., Korenaga, D., Sugimachi, K., andNose, Y. Risk factors which predict pattern of recurrence after curativesurgery for patients with advanced gastric cancer. Surg. Oncol., 1:341-346, 1992.

50. Okamura, T., Korenaga, D., Babe, H., Saito, A., and Sugimachi, K.Postoperative adjuvant chemotherapy inhibits early recurrence of earlygastric carcinoma. Cancer Chemother. Pharmacol., 23: 319-322, 1989.

51. Macham, Y., Moriguchi, S., Yoshida, M., Takahashi, I., Korenaga,D., Sugimachi, K. Splenectomy does not correlate with length of sur-vival in patients undergoing curative total gastrectomy for gastric car-cinoma. Cancer (Phila.), 67: 3006-3009, 1991.

52. Maeda, K., c:hung, Y-S., Takatsuka, S., Ogawa, Y., Sawada, T.,Yamashita, Y., Onoda, N., Kato, Y., Nina, A., Arimoto, Y., Kondo, Y.,and Sowa, M. Tumor angiogenesis as a predictor of recurrence in gastric

carcinoma. J. Clin. Oncol., 13: 477-481, 1995.

Research. on February 18, 2020. © 1998 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

1998;4:2129-2134. Clin Cancer Res   Y Maehara, S Hasuda, T Abe, et al.   patients with early gastric cancer.Tumor angiogenesis and micrometastasis in bone marrow of

  Updated version

  http://clincancerres.aacrjournals.org/content/4/9/2129

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/4/9/2129To request permission to re-use all or part of this article, use this link

Research. on February 18, 2020. © 1998 American Association for Cancerclincancerres.aacrjournals.org Downloaded from