Journal of Surgical Oncology 51:231-235 (1992)

Clinicopathologic Comparisons Between Estrogen Receptor-Positive and -Negative Gastric Cancers CHEW-WUN WU, MD, FACS, SHYH-HAW TSAY, MD, TAI-JAY CHANG, BS, HWAY-MAY CHANG, BL, MAO-CHIH HSIEH, MD, WING-YIU LUI, MD, FANG-KU P’ENG, MD, FACL, A N U CHIN-WEN CHI, PhD From the Departments of Surgery (C.-W.W., T.-].C., H.-M.C., M.-C.H., W.-Y.L., F.-K.P.), Pathology (5.-H.T.), and Medical Research (C.-W.C.), Veterans General Hospital-Taipei and National Yang-Ming Medical College, Taipei, Taiwan, Republic of China

We report that a modified dextran-coated charcoal (DCC) assay, including the addition of sodium molybdate and 5% DCC stripping of endogenous hormone, detected higher estrogen receptors (ERs) than those by the conventional assay. ERs in 21 gastric adenocarcinoma were determined by the modified DCC assay; 13 patients had ERs of 2.5 to 520.2 fmol/mg protein with a mean dissociation constant of 1.9 X loP“’ M. The remaining 8 patients had no detectable amount of the receptor. There were no differences between the ER-positive and ER-negative groups in clinico-pathologic characteristics such as age, sex, tumor size, location, gross appearance, invasive depth, invasion of lymph vessel or vein of stomach wall, nodal involvement, peritoneal dissemination, liver metastasis, and curability. Histological analysis, using Japanese, Lauren’s, or Ming’s classifications, all revealed no differences between the ER-positive and -negative groups. The survival rate was identical for the 2 study groups. These results suggest that the presence or absence of ERs in gastric adenocarcinoma does not correlate to either biologic or clinicopathologic characteristics of this tumor; the role of ERs in human gastric cancer remains to be elucidated. 0 1992 Wiley-Liss, Inc. P~

KEYWORDS:dextran-coated charcoal assay, sodium molybdate, gastric cancer

INTRODUCTION Gastric cancer is a disease of later life, occurring particularly in the 6th decade, with a ma1e:female ratio of 2: 1 [ 1,2]. These data suggest that the growth of gastric cancer may be regulated by sex hormones. Sex hormones act through their specific intracellular receptors, and the occurrence of sex hormone receptors has been found in many hormone-dependent tumors such as breast, prostate and endometrial cancers (3-51. The presence of sex hormone receptors in these tumors has been used as an indicator for the selection of treatments. In addition, steroid-binding proteins have also been identified in other tumors such as brain tumors [6], lung cancer [7], and gastrointestinal cancer [8-111. The occurrence of estrogen receptors (ERs) in gastric cancer tissues was first described by Tokunaga et al. [12]. Subsequent 0 1992 Wiley-Liss, Inc.

studies showed that the ERs were positive in 1650% of cases [8-10,13-161 depending on the method of detection. In most routine receptor laboratories, ERs are determined using either biochemical assay such as dextran-coated charcoal (DCC) analysis [3], or immunohistochemical methods [ 14,161. Previous studies from our laboratory [9,10] have found a discrepancy between DCC and enzyme immunoassay (EIA); the addition of sodium molybdate to buffer and stripping of endogenous hormone of the cytosol increased the detection sensitivity of DCC assay in the same specimen. Consequently, higher

Accepted for publication July 25, 1992. Address reprint requests to Chew-Wun Wu, MD, FACS, Department of Surgery, Veterans General Hospital-Taipei, Shih-Pai, Taipei, Taiwan, 11217, Republic of China.

232

Wu et al.

levels of ER were found. These data may actually reflect the status of ERs in gastric cancer. In order to delineate the functional role of ERs, the gross and microscopic characteristics of gastric cancers with positive ER, which were measured by conventional DCC assay or immunochemical methods, have been reported [ 13,161. The purpose of the present study was to retrospectively evaluate the clinicopathologic differences between ER-positive and ER-negative gastric cancer patients whose ER were measured by the modified DCC assay. Three common histologic classifications [ 17-1 91 were used to compare our results with reports from Japan [ 12,131 and from a Western country [ 161.

MATERIALS AND METHODS During the past 4 years, ERs were assayed for primary adenocarcinoma of the stomach in 21 patients on whom gastrectomy was performed. No patient had previous treatments such as chemotherapy or radiotherapy. Detailed clinicopathologic comparisons were undertaken between ER-positive and ER-negative patients with gastric adenocarcinoma. All patients were regularly followed in our clinic. The minimal follow-up times were > 3 years in each group. Estrogen Receptor Assay ERs were determined as reported previously [9,10]. Briefly, samples were weighed and thawed in 6 volumes of homogenization (TEDG) buffer (10 mM Tris, 1.5 mM EDTA, 0.5 mM dithiothreitol [DTT], 20 mmolil sodium molybdate, and 10% glycerol, pH 7.4). Samples were homogenized using the polytron PCU-2- I 10 homogenizer (Brinkmann Instruments, Westbury, NY). Cytosols were prepared from the homogenates by centrifugation at 105,000 X g (40,000 rpm) for 60 minutes at 4°C in a Kontron Ultracentrifuge (Model Centrikon P2070, Kontron Analytics, Zurich, Switzerland) with a type 50 rotor. At the end of centrifugation, clear supernatants were removed. The endogenous estrogens in the cytosol were removed by the addition of 0.2 volumes of DCC solution (5% Norit A charcoal [Pfanstiehl, Waukegan, IL], 0.5% dextran T70 [Pharmacia, Uppsala, Sweden] in TEDG buffer, pH 7.4) to the cytosol. The samples were centrifuged at 1,800 X g for 10 minutes. The supernatants were incubated in the presence of increasing amounts of ‘H-17 beta-estradiol (New England Nuclear, Boston, MA, 2-100 nmolil) with or without a 500-fold excess of unlabeled ligands for 18 hours at 4°C. Free estrogen was then removed by the addition of one volume of DCC solution to 4 volumes of the cytosol. The cytosol was mixed and kept on ice for 15 minutes, and then centrifuged at 1,800 X g for 10 minutes. An aliquot of the clear supernatant was removed and counted in a liquid scitillation counter (Kontron, mR3000). Results were analyzed according to the method of Scatchard [20]. Protein

TABLE I. ER Levels and Kd in ER-Positive Gastric Cancers -~

~

Case no.

Sex

ER value

1.

F M

325.1 25.4 41.9 151.7 93.4 190. I 2.5 37.2 6.6 20.4 520.2 35.4 6.8

2. 3. 4. 5. 6.

I. 8. 9. 10.

I I. 12. 13.

M

M F M M M M M M M M

Kd

2.3 X 0.8 X 2.0 X 2.7 X 1.3 X 4.1

X

0.4 X 2.6 X 0.7 X 2.1 X 3.6 X 1.8 X 0.4 X

’

M lO-’M lO-’M lo-” M lo-’ M lo-” M lo-” M IW’M IO-”M lo-’ M 10. M lo-” M lo-‘ M 10

’

concentrations of cytosols were determined by the modified method of Lowry [21].

Pathological Classification Three common pathologic classifications were used: (1) the general rules for the gastric cancer study in surgery and pathology [ 171, (2) Lauren’s descriptive classification [ 181, and ( 3 ) Ming’s biologic classification [ 191, Statistical Analysis Statistical comparisons for significance were made using Mann-Whitney U test and the Chi-square test with Yate’s correction [22]. P values less than 0.05 were considered statistically significant. Cumulative survival rates were calculated using the Kaplan-Meier method [23]. RESULTS Estrogen Receptors ERs were detected in the cytosol of gastric adenocarcinoma removed from 13 patients. As shown in Table I , the titer of ER ranged from 2.5 to 520.2 fmol/mg protein (mean -+ SE, 112.1 +- 42.9 fmolimg protein) with dissociation constants ranging from 0.4 to 4.1 X M (mean ? SE, I .9 2 0.3 X 10-“’ M). There were no detectable amounts of ERs in the tumors taken from the remaining 8 patients. Clinicopathologic Characteristics of Gastric Adenocarinoma Clinicopathologic results of the resected gastric cancer in both groups are summarized in Table 11. The Borrmann’s pathological classification [24] of gastric cancer was used as follows: Type I, polypoid or fungating type; Type 11, circumscribed excavating type; Type 111, ulcerated and infiltrating type; Type IV, diffusely thickened type. The cancer invasion from mucosa to subserosa where the growth pattern was not infiltrative or ill-defined was classified as ps( -) [ 171. The term “ps” denoted

233

ER in Gastric Cancer TABLE 11. ER Status and Clinicopathologic Characteristics in Patients With Gastric Cancer

100

ER status Positive No. cases Mean age (yr) Sex Men Women Tumor maximal diameter (cm) Location Upper portion Lower portion Gross type Borrmann I , 11 Borrmann 111, IV Histological depth of cancer cell invasion PS( - )"

PS(+)

Cancer cell invasion of stomach wall Lymph vessels Present Absent Veins Present Absent Nodal involvement Present Absent Peritoneal dissemination Absent Liver metastasis Present Absent Curability Curative Noncurative

Negative ..

13 65.1

8 63.4

11 2 6.6

7 1 6.4

P value

-

80

-

~

,..

NS

60 -

NS

40 -

ER-positive (n=13)

ER-negative(n=B)

NS

4 9

2 6

6 7

4

2 I1

2 6

11

6

2

2

0 13

3 5

NS

9 4

6 2

NS

13

8

...

NS

2o

t

0

4

1

0

12

8

9

7

4

1

300

600

900

1200

NS Time in Days

NS

NS

NS

NS

Prognostic serosal invasion. NS = not significant.

prognostic serosal invasion; ps( -1 patients had better prognosis than those of the remainder, i.e., ps(+) patients. There were no substantial differences between the ERpositive and ER-negative groups in clinical data such as sex or age. There were also no substantial differences between the ER-positive and ER-negative groups in pathologic data such as tumor size, location, gross appearance, histologic depth of invasion, nodal involvement, peritoneal dissemination, or liver metastasis. There were 2 Borrmann type IV in the 13 ER-positive groups and 2 in the 8 ER-negative groups. Statistically, there was no difference. There was no cancer cell invasion to the vein(s) of the stomach wall in the ER-positive group, whereas 3 in 8 (38%) of the ER-negative group had invasion to the vein, but the difference was not statistically significant. There

Fig. I . Cumulative survival curves in patients with ER-positive gastric cancer and those with ER-negative gastric cancer.

were 11 cases with cancer cell invasion to the lymph vessels of the stomach wall in the ER-positive group as compared to 6 in 8 (75%) in the ER-negative group. There was also no statistically significant difference between the 2 groups. The curative resection [ 171 rates were 69% in the ERpositive group and 88% in the ER-negative group. There was no operative mortality in either group. The survival rate was identical for the 2 groups (Fig. I ) .

Japanese Histologic Classification of Gastric Cancer Table I11 shows the histological classifications of our tumors by 3 different systems. Using the General Rules for Gastric Cancer Study in Japan 1171, there were 9 undifferentiated (7 poorly differentiated, 2 mucinous) and 4 differentiated (3 moderately differentiated, 1 papillary) cancer tissues in the ER-positive group. There were 5 undifferentiated (3 poorly differentiated, 1 signet-ring cell, 1 mucinous), and 3 differentiated ( I well differentiated, 2 moderately differentiated) in the ER-negative group. No statistically significant difference between these 2 groups was noted. The growth pattern [ 171 of cancer tissue is classified into 3 types according to stromal reaction. The cancer with a small number of stromal cells is classified as medullary type, and that with a large number as scirrhous type. The intermediate type is that between the medullary and scirrhous types [ 171. There were 4 scirrhous types in the ER-positive group and 1 in the ER-negative group. Statistically, there was no significant difference between these two groups. The intra- and extramural growth pattern 1171 of the cancer is classified as follows: a,the cancer lesion shows expansive growth and exhibits a distinct border from the surrounding tissue; y, the cancer lesion shows infiltrative

Wu et al.

234

TABLE 111. Histopathological Data According to 3 Common Classifications Data Japanese classification Histologic pattcrn of cancer tissue Differentiated Undifferentiated Amount of interstitial connective tissue Intermediate type Scirrhous type Intra- and extramural growth pattern of cancel 01.

ER(+) (n = 13)

ER(-) (n = 8)

4 9

3 5

8 5

I

6 7

4

8

4 4

1

P value

NS

NS

P

Y Lauren’s classification Diffuse type Intestinal type Ming’s classification Infiltrating type Expanding type

NS

=

5

4

NS NS

not significant.

growth with an ill-defined border; p, the growth pattern of the cancer lesion is intermediate between a and y. There were 8 y-growth patterns in the ER-positive group and 3 in the ER-negative group. Statistically, there was also no significant difference between the 2 groups.

Lauren’s and Ming’s Classifications The classification by Lauren [ 181, who divided gastric cancers into intestinal and diffuse types, was proposed, based upon histogenesis. Intestinal type cancer is described as a glandular tumor resembling colonic carcinoma, and diffuse type cancer is composed of solitary or small clusters of cells without forming glands. The intestinal type was found in 7 cases in the ER-positive group as compared to 4 in the ER-negative group (Table 111). There was no statistically significant difference. Ming’s classification [ 191 divided gastric cancers into expanding and infiltrative types, which were meaningful to express the biological potential as the mode of growth and invasions. The expanding type was found in 5 cases in the ER-positive group as compared to 4 in the ERnegative group (Table 111). There were also no statistically significant differences between the two groups. DISCUSSION In the present study, we retrospectively evaluated the characteristics of ER-positive and ER-negative gastric adenocarcinoma in terms of various clinicopathologic parameters. No substantial differences were found in clinical data between the 2 groups of patients. No significant

rences could be observed in detailed histologic data between these 2 groups. The survival rate was also similar. The cutoff limits for ERs have been defined at greater than 3 fmol/mg protein by Tokunaga et al. [l 11. However, the survival rate of the present study was similar in both ER-positive and -negative patients, even when the cutoff limit was raised to 20 fmol/mg protein (data not shown). Thus the present study considered all detectable ER as ER-positive. This was in agreement with Sica et al. [81. A previous study in our laboratory found that the antibody binding assay detected lower ER than those by DCC assay 1101. Moreover, the immunohistological methods may not detect the true high affinity binding to type I receptor, but rather lower affinity binding to types I1 and 111 receptors, which do not represent specific binding 1251. Consequently, results from Harrison et al. [ 161 and Yozozaki et al. [26] using immunohistological methods showed that ER-negative patients had a significant survival advantage over the ER-positive patients. These data were at variance with our observations (Fig. I ) . However, our results were from limited cases and short-term follow-up. Large series and long-term follow study are needed in the future. Studies in Japan claimed that ER-positive gastric cancers were characterized grossly as Borrmann type IV and microscopically as diffuse type with a scirrhous growth pattern [ 13,261 and were composed of poorly differentiated cancer cells [26]. The study of Harrison et al. [ 161 in England noted that sex, the Lauren pathologic type, and cancer cell grade all failed to reach independent significance as prognostic factors in a smear. Our study also failed to find any specific features for the ER-positive cases (Tables 11, 111). Recent studies suggest that ER may have important functions in gastric tissues [9,10,27-291. However, its role in gastric cancer remains obscure. Conflicting reports [30-32] from in vivo and in vitro studies indicate that estrogen can have either a stimulatory or inhibitory effect on gastric cancer cell growth. Clinical trials with the antiestrogen agent tamoxifen for the treatment of gastric cancer [33-351 also obtained controversial results. Research during the past few years has revealed a correlation between hormone dependency and oncogene (c-H-ras) expression in rat and human mammary carcinomas [36,37]. In contrast, amplification of c-erbB-2 gene has been demonstrated in breast and stomach carcinomas [38,39]. The amplification of the gene correlated with earlier relapse and shorter survivals [40,41]. The expression of its oncoprotein appeared to be a marker of poor short-term prognosis [42]. Recently, estrogen was found to regulate c-erbB-2 expression in breast cancer cell lines 1431. These data seems to suggest that ER have moved from the role of prognostic guides and guides to therapy

ER in Gastric Cancer

into a new role, that of potential pathogenic factors in oncogenesis.

ACKNOWLEDGMENT by a grant from ~ ~ This study was Science Council of the Republic of China (NSC 8 1-0412B075-48). REFERENCES I . Fujimoto 1: Role of population based on cancer registry. Japanese Journal of Cancer Clinics 23:557-582, 1977. 2. Wynne GG: The sex ratio in gastric cancer and hypothetical considerations relative to etiology. Br J Cancer 22: 164-171, 1968. 3. Liu TJ, Chi CW, Wu LH: The estrogen receptor and the progesterone receptor in Chinese breast cancer. J Formosan Med Assoc 83~45-58, 1984. 4. Walsh PC, Hicks LL: Characterization and measurement of androgen receptors in human prostatic tissue. Prog Clin Biol Res 3 3 5 4 3 , 1979. 5 . Smith DC, Prentice R, Thompson DJ, et al: Association of exogenous estrogen and endometrial carcinoma. N Engl J Med 293:1164-1167, 1975. 6. Lee LS, Chi CW: Chang TJ, et al: Steroid hormone receptors in meningiomas of Chinese patients. Neurosurgery 25:541-545, 1989. 7. Chaudhuri PK, Thomas PA, Walker MJ, et al: Steroid receptors in human lung cancer cytosols. Cancer Lett 16:327-332, 1982. 8. Sica V , Nola E, Contieri E, et al: Estradiol and progesterone receptors in malignant gastrointestinal tumors. Cancer Res 44:467&4674, 1984. 9. Wu CW, Chi CW, Chang TJ, et al: Sex hormone receptors in gastric cancer. Cancer 65: 139&1400, 1990. 10. Wu CW, Chang HM, Kao HL, et al: The nontransformed progesterone and estrogen receptors in gastric cancer. Gastroenterology 102:1639-1 646, 1992. 11. P’eng FK, Lui WY, Chang TJ, et al: Glucocorticoid receptors in hepatocellular carcinoma and adjacent liver tissue. Cancer 62:2 134-2138, 1988. 12. Tokunaga A, Kojima N, Andoh T, et al: Hormone receptors in gastric cancer. Eur J Cancer Clin Oncol 19:687-689, 1983. 13. Tokunaga A, Nishi K, Matsukura N, et al: Estrogen and progesterone receptors in gastric cancer. Cancer 57: 1376-1379, 1986. 14. Uehara Y, Takahashi T, Kojima 0, et al: Peroxidase-antiperoxidase staining for estrogen and progesterone in scirrhous type of gastric cancer: Possible existence of the estrogen receptor. Jpn J Surg 16:245-149, 1986. 15. Kojima 0, Takahashi T, Kawakami S , et al: Localization of estrogen receptors in gastric cancer using immunohistochemical staining of monoclonal antibody. Cancer 67:240 1-2406, 1991 . 16. Harrison JD, Jones JA, Ellis 10, Morris DL: Oestrogen receptor D5 antibody is an independent negative prognostic factor in gastric cancer. Br J Surg 78:334-336, 1991. 17. Japanese Research Society for Gastric Cancer: The general rules for the gastric cancer study in surgery and pathology. Jpn J Surg I1:127-145, 1981. 18. Lauren P: The two histological main types of gastric carcinoma. Diffuse and so-called intestinal type carcinoma. An attempt at a histoclinical classification. Acta Path Microbiol Scand 64:3 1 4 9 , 1965. 19. Ming SC: Gastric carcinoma: A pathological claasification. Cancer 39:2475-1485, 1977. 20. Scatchard G: The attraction of proteins for small molecules and ions. Ann NY Acad Sci 51:660-672, 1949. 21. Hartree EF: Determination of protein: A modification of the Lowry method that gives a linear photometric response. Anal Biochem 48:422-427, 1972.

235

22. Snedecor GW, Cochran WG (eds): “Statistical Methods.” Ames: Iowa State University Press, 1967, pp 128-129. 23. Kaplan EC, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457460, 1958. 24. Borrmann R: Geschwulste des Magens und Duodenums. in Henke F,~Lubarschi 0 (eds): ~“Handbuchuder speziallen pathologkchen ~ ~ l Anatomie und histologie.” Berlin: Springer Verlag, vol lV/I, 1926. DD 865-1095, 25. Cham& GC, Mercer WD, McGuire WL: Are histochemical methods for estrogen receptor valid’? J Histochem Cytochem 281792-798, 1980. 26. Yozozaki H, Takekura N , Takanashi A, et al: Estrogen receptors in gastric adenocarcinoma: A retrospectrive immunohistochemical analysis. Virchows Archiv [A] 41 3:297-302, 1988. 27. Adeniyi KO: Gastric acid secretion and parietal cell mass: Effect of sex hormones. Gastroenterology 101:6&69, 1991. 28. Nishi K, Tokunaga A, Shimizu Y, et al: Immunohistochemical study of intracellular estradiol in human gastric cancer. Cancer 5911328-1332, 1987. 29. Luqmani Y, Bennett C, Paterson 1, et al: Expression of the pS2 gene in normal, benign and neoplastic human stomach. Int J Cancer 44:806-8 12, 1989. 30. Furukawa H, lwanaga T , Koyama H, Taniguchi H: Effect of sex hormone on carcinogenesis in the stomach of rats. Cancer Res 42:s 181-51 82, 1982. 31. Nohga K , Katsurnata T, lshiwata K, et al: Experimental and clinical study on chempendocrine therapy for gastric cancer. J Jpn Surg Soc 88:1109-1112, 1987. 32. Harrison JD, Watson S , Morris DL: The effect of sex hormones and tamoxifen on the growth of human gastric and colorectal cancer cell lines. Cancer 63:2148-2151, 1989. 33. Kitaoka H: Sex hormone dependency and endocrine therapy for patients with diffuse carcinoma of the stomach. Jpn J Cancer Chemother 10:2453-1460, 1983. 34. Kojima 0, Takahashi T: Endocrine therapy for scirrhous carcinoma of the stomach. Jpn J Cancer Chemother 13:25261531, 1986. 35. Harrison JD, Morris DL, Ellis 10, et al: The effect of tamoxifen and estrogen receptor status on survival in gastric carcinoma. Cancer 64:1007-1010, 1989. 36. Huang FL, Cho-Chung YS: Hormone-regulated expression of cellular ras oncogene in mammary carcinomas in rats. Biochem Biophys Res Commun 123:141-147, 1984. 37. Debortoli ME, Abou-Issa H, Haley BE, Cho-Chung YS: Amplified expression of p21 ras protein in hormone-dependent mammary carcinomas of humans and rodents. Biochem Biophys Res Commun I27:699-706, 1985. 38. Tal M, Wetzler M, Josefberg Z, et al: Sporadic amplification of the HER2/neu protooncogene in adenocarcinomas of various tissues. Cancer Res 48:1517-1520, 1988. 39. Slamon DJ, Clark GM, Wong SG, et al: Human breast cancer: correlation of relapse and survival with amplification of the HER2/neu oncogene. Science 235:177-182, 1987. 40. Walker RA, Gullick WJ, Varley JM: An evaluation of immunoreactivity for c-erbB-2 protein as a marker of poor short-term prognosis in breast cancer. Br J Cancer 60:42M29, 1989. 41. Falck VG, Gullick WJ: C-erbB-2 oncogene product staining in gastric adenocarcinoma. An immunohistochemical study. J Pathol 159:107-111, 1989. 42. Yonemura Y, Ninomiya I , Ohoyama S , et al: Expression of c-erbB-2 oncoproteion in gastric carcinoma. Immunoreactivity for c-erbB-2 protein is an independent indicator of poor short-term prognosis in patients with gastric carcinoma. Cancer 67:29142918, 1991. 43. Warri AM, Laine AM, Majasuo KE, et al: Estrogen suppression of erbB2 expression is associated with increased growth rate of ZR-75- I Human breast cancer cells in vitro and in nude mice. Int J Cancer 49:6 16-623, 199I .