© 1990 S. Karger AG, Basel 0378-7346/90/0293-0214S2.75/0

Gynecol Obstet Invest 1990;29:214-218

Clinical Evaluations of the Tumor Marker Sialyl SSEA-1 Antigen for Clinical Gynecological Disease Osamu Iwanari*, Jinya M iyako a, Yoshie Date*, Satoru Nakayama a, Satoshi Kijim a a, Masashi Moriyama a, Kenji Karinob, Jiroh Endohb, Manabu Kitao a Departments o f a Obstetrics and Gynecology, and b Laboratory Medicine, Shimane Medical University, Izumo, Japan

Key Words. Tumor marker • Sialyl SSEA-1 antigen • Various gynecologic disease Abstract. Sialyl SSEA-1 antigen (SLX) is a highly specific tumor marker composed of sugar chain antigens that have Lewis X at their terminals and bind to sialic acid. This antigen is rarely detected in normal tissues, and is present in adenocarcinoma and fetal tissues. We studied the clinical usefulness of SLX in gynecological patients and obtaind the following results. (1) The antigen was frequently positive in patients with ovarian cancer with a mean of 89.5 ± 48.3 U/ml (72.8%, 8/11) and in those with endometriosis with a mean of 39.8 ± 10.3 U/ml (75.0%, 6/8). (2) Among the gynecological malignancies, the percent positivity was low in those with cervical cancer (20.0%, 5/25), endometrial cancer (33.3%, 1/3), and cancer of the fallopian tube (33.3%, 1/3). (3) The antigen was negative in 20 with myoma uteri, 20 normal pregnant women, and 9 nonpregnant healthy women during the follicular, luteal, or menstrual phase. It was negative in 8 of 9 patients with benign ovarian cyst. False negative results were rare. (4) The SLX level was higher in the ascites than in the serum in patients with ovarian cancer and in those with benign ovarian tumors. (5) The serum SLX in patients with ovarian cancer, which was positive before tumor resection, became negative 2 weeks postoperatively. These results suggest that SLX is a tumor marker with a high specificity to adenocarcinoma of the reproductive organs.

SSEA-1 is a stage-specific antigen recognized by monoclonal antibodies to the mouse teratocarcinoma cell line F9, established by Salter and Knowles [1] in 1978. In addition to this prototype, variously modified forms of SSEA-1 antibodies were shown to be present in large amounts in human adenocarcinoma tissues [2], Among these antigens, sialyl SSEA-1 antigen (SLX), which shows high serum concentrations in patients with various adenocarcinomas, especially adenocarcinoma of the lung [3], is considered to be most useful as a tumor marker. SLX is a blood-type-related material consisting of sugar chains of varying lengths with Lewis X (X hap­ ten) on their terminals and are conjugated with sialic acid. It is rarely observed in normal adult tissues but is considered to be frequent in the tissues of adenocarci­

noma primarily of the lung and digestive organs as well as fetal tissues [4], In this study, we evaluated the clinical usefulness of SLX in the diagnosis of gynecological ma­ lignancies.

Material and Methods The study included 11 patients with ovarian cancer, 9 with benign ovarian tumors, 25 with cervical cancer (21 with squamous cell carcinoma and 4 with adenocarcinoma), 20 with myoma uteri, 8 with endometriosis, and 10 with other gynecological diseases in whom the diagnosis was histopathologically confirmed. Twenty nor­ mal pregnant Japanese women were also studied. Blood samples were taken from all the subjects before the treatment, and the ascites was collected during the surgery. Blood samples were also taken during the surgery, from 6 ovarian cancer patients. One hundred and sixty healthy nonpregnant women aged 21-59 years were also studied as the normal controls.

Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/7/2018 1:17:09 PM

Introduction

Sialyl SSEA-1 and Gynecologic Disease

215

1 •• • ,1 1

Ovarian cancer (8/11= 72.7%) Benign ovarian tumor (1 /9= 11.1%)

.u~

Cervical cancer (5/25 = 20.0%) Myoma uteri (0/20 = 0%)

|1 1

«1. i i i I I « o*



• • • • •

o

Co Adenocarcinoma)

•$?h i! !

Endometriosis (6/8 = 75.0%)

•! {"• i• i



Pregnancy (0/20 = 0%)

tim * , ~ •f* •

Others (3/10 = 30.0%)

! “T •;

| !

f



----------------1— i--------------------- »--------------------- ,--------------------O (positive No./n = %)

The SLX concentration was measured using an FH-6 Ohtsuka radioimmunoassay kit. This assay system had a sensitivity of 5 U/ml. The antigen was regarded as positive when the concentra­ tion exceeded 38 U/ml.

Results In the 160 healthy controls aged 21-59 years, the serum SLX level was 21.9 ± 4.0 U/ml (mean ± SD) and exceeded 38 U/ml in 3 (1.9%); 40 U/ml (38-year-old), 41 U/ml (44-year-old) and 44 U/ml (49-year-old). Fig­ ure 1 shows the serum SLX levels in patients with gyne­ cological disease. Serum SLX was positive in 8 (72.7%) of the 11 patients with ovarian cancer. Six of these eight had stage III lesions (serous cystadenocarcinoma in 3, clear cell adenocarcinoma in 2, and mucinous cystadenocarci­ noma in 1), and the remaining 2 had stage Ic lesions. Of the 3 SLX-negative patients, 1 had stage Ic (mucinous cystadenocarcinoma) and 2 had stage III lesions (clear cell adenocarcinoma and endometrioid adenocarcinoma in 1 each). Eight of the nine patients with benign ovarian tumors (5 with serous cystadenoma, 1 with mucinous cystadenoma and 2 with benign cystic teratoma) were negative, and 1 with benign cystic teratoma was positive (43 U/ml). Five (20.0%) of the twenty-five patients with cervical cancer were SLX-positive and 3 (60.0%) of these 5 had an endocervical type adenocarcinoma (stage lb in

38

SO

100

150

200

SLX, U/ml

2 and stage Illb in 1) and the other 2, a squamous cell carcinoma (stage IV). The 20 patients with myoma uteri were all negative. Of the 8 patients with endometriosis, 6 (75.0%) were positive. The antigen was positive in 3 of the 10 patients with other gynecological diseases, namely in 1 of the 3 with endometrial cancer, and in 1 with Krukenberg’s tumor. The 20 normal pregnant women (8-41 weeks of gestation) were all negative. Figure 2 shows the SLX levels in the serum and ascites of patients with ovarian tumors. In the 6 patients with ovarian cancer, the mean SLX level was 80.5 ± 37.5 U/ml in the serum and 154.3 ± 82.8 U/ml in the ascites, with the latter being significantly higher (0.02 < p < 0.05) than the former. In the 3 with benign ovarian tumors, the mean SLX level was 25.0 ± 4.1 U/ml in the serum and 38.0 ± 4.3 U/ml in the ascites, the latter also being significantly higher (0.02 < p < 0.05). However, the SLX level in the ascites was 50 U/ml or less in patients with benign tumors but 100 U/ml or above in all patients with malignant tumors. Figure 3 show serum SLX levels in 6 ovarian cancer patients pre- and postsurgery. The antigen became nega­ tive in all these patients 2 weeks postoperatively. All these patients were treated intraoperatively with anticancer agents. Two had residual tumors 2 cm in diameter. Figure 4 shows the serum SLX levels in 9 healthy women during the follicular, luteal, and menstrual phase. The results were consistently negative.

Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/7/2018 1:17:09 PM

Fig. 1. Serum SLX levels in women with ovarian cancer (n = 9), uterine cervical cancer (n = 25), myoma uteri (n = 20), endometrio­ sis (n = 8), pregnancy (n = 20) and others (endometrial cancers, 3 can­ cers of the tube, 3 hydatidiform moles and 1 Krukenberg’s tumor).

|

216

Iwanari/Miyako/Date/Nakayama/Kijima/Moriyama/Karino/Endoh/Kitao

Ope.

Postoperative

Discussion FH-6, a monoclonal antibody prepared by Hakomori et al. [2], specifically recognizes SLX, a sugar chain anti­ gen with a known structure [5, 6]. It is considered to react with various cancer cell lines and fetal tissues but has a weak reactivity with normal cells [7]. For this rea­ son, SLX seems to be a promising carcinoembryonic antigen with a high specificity for malignant tissues, especially adenocarcinoma [8], As there have been few

reports on application in clinical gynecology, we carried out related studies. Imura et al. [9] suggested a serum SLX concentration of 38 U/ml as a cutoff value, on the basis of evaluation in 1,105 individuals of both sexes. Our findings that the serum SLX levels exceeded 38 U/ml in only 3 (1.9%) of our 160 healthy nonpregnant women supports the appro­ priateness of this values for screening of gynecological malignancies. Moreover, unlike the serum SLX level appears to be unaffected by the menstrual cycle, because

Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/7/2018 1:17:09 PM

Fig. 2. Serum and ascitic fluid SLX levels in patients with malig­ nant ovarian cancer (n = 6; •) and benign ovarian tumor (n =3; o). Fig. 3. Serum SLX levels before and after resection of ovarian cancer (n = 6). Ope. = operation. Fig. 4. Serum SLX levels at the follicular and luteal phases and during menses in normal women (n = 9).

217

Sialyl SSEA-1 and Gynecologic Disease

marker. For example, CA 125, which is considered to be highly sensitive to ovarian tumors, was reported by Don­ ald et al. [11] and Takahashi et al. [12] to be increased also during menstruation and in patients with endome­ triosis. Simultaneous evaluation of serum CA 125 and SLX levels will thus likely exclude false positive findings linked to menstruation and/or endometriosis. In patients with ovarian tumors, the ascitic SLX level was significantly higher than the serum SLX level in both patients with malignant disease and those with benign diseases, but the level in the ascites was 100 U/ml or above in the former group while it was 50 U/ml or below in the latter group. Also, in patients with malig­ nant ovarian tumors, the ascitic SLX level tended to be higher with advance in the disease. These findings sug­ gest that the ascitic SLX level is useful for clinical eval­ uation of a possible malignancy. The serum of all 6 patients with ovarian cancer became negative for SLX 2 weeks after the tumor resec­ tion. This suggests that SLX disappears rapidly from the body, that the serum level is relatively unaffected by inflammation related to surgery and that secretion is supressed by anticancer agents. Further investigation on the relationship of response to those of other tumor markers, especially CA 125, is considered to contribute to an increase in the clinical usefulness of SLX for diag­ nosing gynecological disease.

Acknowledgment We thank M. Ohara for pertinent comments.

References 1 Salter D, Knowles BB: Monoclonal antibody defining a stage specific mouse embryonic antigen (SSEA-1). Proc Natl Acad Sci USA 1978;75:5565-5568. 2 Hakomori S, Nudelman E, Levery SB, et al: Novel fucolipids accumulating in human adenocarcinoma. I. Glycolipids with dior trifucosylatcd type 2 chain. J Biol Chcm 1984;259:672— 4680. 3 Kannagi R, Fukushi Y, Tachikawa T, et al: Quantitative and qualitative characterization of human cancer-associated serum glycoprotein antigens expressing fucosyl or sialyl-fucosyl type 2 chain polyactosamine. Cancer Res 1986;46:2619-2626. 4 Abe K, McKibbin JM, Hakomori S: The monoclonal antibody directed to difucosylated type 2 chain (Fuc al-» 2 Gal (31->4 [Fuc a l —>3] GlcNAc; (3—>R, Y determinant). J Biol Chem 1983; 258:11793-11797. 5 Fukushi Y, Kudclman E, Levery SB, et al: Novel fucolipids accumulating in human cancer. III. A hybridoma antibody (FH-6) defining a human cancer-associated difucogangliosidc

Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/7/2018 1:17:09 PM

it was 38 U/ml or less in 20 normal pregnant women and 9 healthy nonpregnant women during the follicular, lu­ teal, or menstrual phases. The percent positivity of the antigen was high in patients with ovarian cancer and endometriosis, being 72.7 and 75.0%, respectively, but low in those with benign ovarian tumors, cervical cancer, and myoma uteri, being 11.1,20.0, and 0%, respectively. Concerning malignancies of the ovary, endometrium, and fallopian tube, the antigen was positive in 13 (65%) of the 20 patients with adenocarcinoma, as compared with only 2 (9.5%) of the 20 patients with squamous cell carcinoma. From these results, the serum SLX has a high specificity for gynecological adenocarcinoma. The percent positivity for the antigen was high in our ovarian cancer patients (72.7%), but was higher than the values reported by Imura et al. [8] or Inouc et al. [10] (69%). Since it tended to increase as the lesion was more advanced, and was 66.6% for stage I lesions as compared with 75.0% for stage III lesions, the high percent positiv­ ity in our study may be explained by the large proportion of stage III patients among the studied subjects. We also examined patients with endometriosis in whom the percent positivity was high. SLX was positive in 5 of 7 patients, with its serum level being between 38 and 50 U/ml, in all these 5 patients. Since Inoue et al. [10] observed localization SLX in glandular epithelium and luminal surface of the glands in the basal layer, this increase in the percent positivity for serum SLX may be due to the abnormal proliferation of the endometrial glands associated with endometriosis. Because the serum SLX level was 50 U/ml or less in all SLX-positive patients with endometriosis or benign ovarian tumors, the serum SLX levels from 38 to 50 U/ml was considered to be a transitional range and that of 51 U/ml or above indicative of a malignancy. Adenocarcinoma was noted in 12 ( 92. 3 % ) of the 13 can­ cer patients with a serum SLX levels of 51 U/ml or above. Of the 10 patients with a serum SLX level of the transitional zone, 6 (60.0%) had endometriosis, 2 had cervical cancer, 1 had benign ovarian tumor, 1 had endo­ metrial cancer. The possibility of adenocarcinoma is high when the serum SLX level is 51 U/ml or above, and a diagnosis of endometriosis is more likely when the SLX level is in the transitional range. Based on these criteria, the specificity of the serum SLX level to adeno­ carcinoma is considered to be improved. In addition, false positive results may be eliminated by evaluation SLX with other tumor markers, thus fur­ ther enhancing the clinical usefulness of this tumor

6

7

8

9

10

Iwanari/Miyako/Date/Nakayama/Kijima/Moriyama/Karino/Endoh/Kitao

(IV3NeuAcV3III3Fuc2nLC3). J Biol Chem 1984;259:1051110517. Kannagi R, Fukushi Y, Tachikawa T, et al: Quantitative and qualitative characterization of human cancer-associated serum glycoprotein antigens expressing fucosyl or sialylfucosyl type 2 chain polyactosamine. Cancer Res 1986;46:2619-2623. Fukushi Y, Kannagi R, Hakomori S, et al: Location and distri­ bution of difucoganglioside (IV'NeuAcV'III'FucjnLct) in nor­ mal and tumor tissues defined by its monoclonal antibody FH-6. Cancer Res 1985;45:371 1-3715. Imura FI, Endoh J, Ohkura H, et al: Laboratory and clinical evaluations of the measurement of the new tumor marker sialyl SSEA-1. II. Measurement of its serum levels in patients with various malignant and benign diseases. Gan Karyo 1987; 14: 1322-1331. Imura H, Endoh J, Ohkura H, et al: Laboratory and clinical evaluations of the measurement of the new tumor marker sialyl SSEA-1. 1 Study of the normal value and conditions of measure­ ment. Gan Karyo 1987;14:1315-1321. Inoue M, Shimizu C, Sasagawa T, et al: The clinical value of sialyl SSEA-1 antigen in patients with gynecologic tumors. Acta Obstet Gynecol Jap 1987;39:2120-2124.

11 Donald E, Pittaway M, Fayez JA: Serum CA-125 antigen levels increase during menses. Am J Obstet Gynecol 1987; 156:75— 76. 12 Takahashi K, Kijima S, Yoshino K, et al: Differential diagnosis between leiomyomata uteri and adenomyosis using CA 125 as a new tumor marker of ovarian carcinoma. Acta Obstet Gynecol Jap 1987;37:591-595.

Received: April 14, 1989 Accepted: September 15, 1989 Osamu Iwanari, MD Department of Obstetrics and Gynecology Shimane Medical University Izumo 693 (Japan)

Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/7/2018 1:17:09 PM

218

Clinical evaluations of the tumor marker sialyl SSEA-1 antigen for clinical gynecological disease.

Sialyl SSEA-1 antigen (SLX) is a highly specific tumor marker composed of sugar chain antigens that have Lewis X at their terminals and bind to sialic...
783KB Sizes 0 Downloads 0 Views