Changes in CA602 and CA546 concentrations in patients during in-vitro fertilization and embryo transfer, and ovarian hyperstimulation syndrome

Transcription

1 hrep$$0301 Human Reproduction vol.12 no.3 pp , 1997 Changes in CA602 and CA546 concentrations in patients during in-vitro fertilization and embryo transfer, and ovarian hyperstimulation syndrome Kiyoshi Takamatsu 1,3, Michiko Kasuga 1, Masao Nakano 2, Yasuhiro Udagawa 1, Yasunori Yoshimura 1 and Shiro Nozawa 1 Tumour markers are also used in fields other than oncology. Like CA125, CA602 appears in relatively high concentrations in patients with non-neoplastic conditions, such as endometriosis, ascites and pelvic inflammatory disease (Nozawa et al., 1991). Both CA602 and CA125 can be used as markers for the progression of endometriosis or the effect of its treatment 1 Department of Obstetrics and Gynecology, School of Medicine, Keio University, 35 Shinanomachi, Shinjyuku-ku, Tokyo 160, Japan and 2 Department of Obstetrics and Gynecology, Saiseikai Kanagawaken Hospital, 6-6 Tomiyacho, Kanagawa-ku, Yokohama- (Barbieri et al., 1986; Suzuki et al., 1990). city, Kanagawa 221, Japan 3 To whom correspondence should be addressed Changes in CA602 and CA546, two carbohydrate-related antigens newly established in our institute, were evaluated during the course of in-vitro fertilization and embryo transfer. The concentration of CA602, an antigen with characteristics similar to those of CA125, did not change substantially until embryo transfer. In patients with clinical evidence of ovarian hyperstimulation syndrome (OHSS), a significant increase in serum concentrations of CA602 was observed 10 days after embryo transfer (P < 0.01). There were no significant changes in the patients without OHSS. A significant elevation of oestradiol concentration in OHSS patients was also observed, with the variation before oocyte retrieval being greater than that seen with CA602. The concentration of CA546, which had lower false-positive rates in benign disease, did not differ significantly in the OHSS and non-ohss patients. Our data demonstrate that serum concentrations of CA602 are increased even in early stages of OHSS, and suggest a possible role for this antigen as a marker for OHSS. Key words: CA546/CA602/in-vitro fertilization embryo transfer/ ovarian hyperstimulation syndrome/tumour markers Introduction Tumour markers are potentially useful in the diagnosis of malignant diseases. Recently we established two new tumour markers for ovarian cancers, CA602 (Nozawa et al., 1991) and CA546 (Nozawa et al., 1989). CA602, a protein antigen, showed high positive rates in patients with ovarian cancer, especially in those with serous cystadenocarcinoma. This marker exhibits characteristics similar to those of CA125 (Bast et al., 1983) and shows a relatively high false-positive rate in the presence of endometriosis (Suzuki et al., 1990; Nozawa et al., 1991). CA546, a glycosidic chain antigen, shows high positive rates in patients with mucinous cystadenocarcinoma and low false-positive rates in patients with benign diseases including endometriotic cysts (Nozawa et al., 1992a). Thus, these two markers have complementary potential for establishing the diagnosis of ovarian cancer (Nozawa et al., 1992b). The in-vitro fertilization (IVF) and embryo transfer procedure has become a common assisted reproductive technique. A serious complication of this procedure is ovarian hyperstimulation syndrome (OHSS), which is caused by the over-induction of ovulation by pharmacological means (Bettendorf and Lindner, 1987). Though the mechanism of OHSS is not clarified, the support of luteal function with injections of human chorionic gonadotrophin (HCG) is known to be one of the causes of OHSS. The discovery of indicators that could reveal the onset and the degree of severity of this syndrome would be therefore of the utmost clinical importance. A few attempts have been made to evaluate the usefulness of tumour markers in the diagnosis of OHSS, but only some studies have followed the changes in tumour markers from the beginning of ovarian stimulation (Jaeger et al., 1987; Eisermann and Collins, 1989; Lanzone et al., 1990; Zweers et al., 1990). Accordingly, in this study we measured the changes in serum concentrations of CA602 during IVF and embryo transfer to evaluate its use as a marker for the diagnosis of OHSS. Serum concentrations of CA546 were also monitored. Materials and methods Subjects A total of 17 Japanese women were enrolled in this study with informed consent. They were being treated for infertility, especially bilateral tubal occlusion, in the outpatient clinic of Saiseikai Kanagawaken Hospital (Yokohama-city, Kanagawa, Japan). All patients desired the IVF and embryo transfer procedure. Five of the 17 patients were excluded because of endometriosis. Finally, 12 patients aged years were registered. A hormonal examination, including a gonadotrophin-releasing hormone (GnRH) and a thyrotrophin-releasing hormone test, before the IVF and embryo transfer procedure showed no abnormal hormonal conditions. IVF and embryo transfer procedure The IVF and embryo transfer procedure was performed as described previously (Konishi et al., 1993). Individualized ovarian stimulation was performed with pure follicle stimulating hormone (FSH) human menopausal gonadotrophin (HMG) HCG in combination with GnRH analogue. Briefly, while endogenous gonadotrophins were being suppressed with GnRH analogue (Suprecure; Hoechst, Frankfurt, European Society for Human Reproduction and Embryology 441

2 K.Takamatsu et al. Figure 1. CA602 concentration during in-vitro fertilization and embryo transfer. Points of blood sample collection were: (i) during the previous cycle (not in the menstrual period); (ii) when the leading follicle became 15 mm in diameter; (iii) at oocyte retrieval; (iv) at embryo transfer; and (v) 10 days after embryo transfer. (d) Cases with a positive pregnancy test 14 days after oocyte retrieval. n number of patients; OHSS ovarian hyperstimulation syndrome. Statistical analysis Data are presented as the means SEM. A statistical analysis was performed using paired or non-paired Student s t-tests. The post-hoc test was Scheffé s F-test. Differences were considered to be statistic- ally significant if the P value was All analyses were performed using standard computer software for statistical analysis (StatView- J, version 4.02; Abacus Concept Inc., Berkeley, CA, USA). Germany), pure FSH (Fertinom P; Serono Japan, Tokyo, Japan) was administered for 4 days following the administration of HMG (Humegon; Sankyo, Tokyo, Japan). When the diameter of the leading follicle was 17 mm, IU HCG (HCG Mochida; Mochida, Tokyo, Japan) were injected. Oocytes were retrieved transvaginally under ultrasonographic guidance 36 h after the injection. After 48 h, fertilized eggs were transferred to the uterine cavity. After the embryo transfer, HCG was injected every 2 days on five occasions to support the luteal phase. A pregnancy test was performed on day 14 after oocyte retrieval. Blood samples were collected as follows: (i) during the previous cycle (not in the menstrual period) for the purposes of excluding women with apparent endometriosis; (ii) when the leading follicle grew to 15 mm in diameter; (iii) at oocyte retrieval; (iv) at embryo transfer; and (v) 10 days after embryo transfer. Assays for tumour markers and hormones Serum concentrations of CA602 and CA546 were measured with kits from Mochida Co. Ltd, which were based on the double-determinant enzyme immunoassay (Nozawa et al., 1992b,c). The threshold values for distinguishing women with ovarian cancer from healthy women were set at 63 U/ml for CA602 and 12 U/ml for CA546. The 442 Figure 2. CA546 concentration during in-vitro fertilization and embryo transfer. Points of blood sample collection were as for Figure 1. (d) Cases with a positive pregnancy test 14 days after oocyte retrieval. See Figure 1 for abbreviations. interassay coefficients of variation were % for CA602 and % for CA546. The serum concentration of oestradiol was measured with semiautomatic equipment (SR-1; Biochem Immunosystems, Tokyo, Japan). The complete blood count and total protein concentration in serum were also measured. To detect OHSS after embryo transfer, patients were checked using transvaginal ultrasonography and questioned about their condition. According to the classification of OHSS by the World Health Organization (WHO Scientific Group, 1973), patients with swelling of the ovaries to 5 cm in diameter were categorized as having OHSS. Massive ascites and complaints of lower abdominal pain were also defined as OHSS. Results Of the 12 women who underwent pharmacological induction of ovulation, five (41.6%) developed OHSS. All had mild or

3 CA602 and CA546 concentrations during IVF and embryo transfer the OHSS group, the serum concentrations of CA602 were significantly increased 10 days after embryo transfer. At this point, the mean serum concentration of CA602 in the OHSS group was significantly higher than that in the non-ohss group (Table I). The CA546 concentration did not change substantially during the course of IVF and embryo transfer (Figure 2). No difference in the CA546 concentration was found between the OHSS and the non-ohss groups. The concentrations of oestradiol before oocyte retrieval were pg/ml in the OHSS group and pg/ml in the non-ohss group; the difference was not statistically significant. However, 10 days after embryo transfer, the concentration of oestradiol in the OHSS group ( pg/ml) was significantly higher than that in the non- OHSS group ( pg/ml) (Figure 3). Discussion CA602 and CA546 are newly established carbohydrate-related tumour markers in our institute, which are used for the diagnosis of ovarian cancers. To evaluate the possibility of tumour markers also acting as markers for OHSS, we measured the serum concentrations of CA602 and CA546 at five points during IVF and embryo transfer. Although no change in the CA602 concentration was observed until embryo transfer in the OHSS group, there was an immediate and sharp increase 10 days after embryo transfer. At the time of oocyte retrieval, a needle is inserted into the ovary; the bleeding that occurs could lead to stimulation of the peritoneum. However, in the non-ohss group, there was no change in the course of the Figure 3. Oestradiol concentrations during in-vitro fertilization and procedure of the study. In the OHSS group, three out of five embryo transfer. Points of blood sample collection were as for patients became pregnant. This might suggest that the elevation Figure 1. (d) Cases with a positive pregnancy test 14 days after oocyte retrieval. See Figure 1 for abbreviations. of CA602 was due not to OHSS but to pregnancy. It has been reported that CA602 and CA125 concentrations do increase in the first trimester of pregnancy (Halila et al., 1986; Nozawa moderate stage OHSS, and no severe haemoconcentration was et al., 1991). However, there were two patients in the OHSS observed. There was no significant difference in complete group who were not pregnant but who showed an increase in blood count or total protein concentration during the course CA602, and one pregnant patient without OHSS but with no of the study in both the OHSS and non-ohss groups. The increase in CA602. These results suggested that the elevation number of follicles observed before oocyte retrieval was 8.0 of CA602 in the OHSS group could indeed be explained 1.3 in the OHSS group and in the non-ohss by OHSS. group, and this was a significant difference (P 0.01). The All cases of OHSS in this study were never more than at numbers of oocytes retrieved and fertilized were the grade II or moderate stage (Rabau et al., 1967; Schenker and respectively in the OHSS group, and and Weinstein, 1978; Bettendorf and Lindner, 1987; Blankstein and respectively in the non-ohss group. Both et al., 1987). Even at this early stage, the serum concentrations differences were statistically significant (P 0.01 and P of CA602 were increased significantly compared with those 0.01 respectively). At most, three fertilized oocytes were in the non-ohss group. In contrast, CA546 showed no transferred. Four pregnancies occurred in the study subjects. significant change, confirming the low rate of false-positives The course of the three pregnancies in the OHSS group was for this marker. uneventful. The one pregnancy in the non-ohss group ended During the time-course of the experiment, changes in the in a chemical abortion. concentration of CA602 were not large enough to detect OHSS The mean concentrations of CA602 in the OHSS group in its early stages, such as before embryo transfer. With were higher than those in the non-ohss group at all sampling reference to CA125, the concentration of this antigen has been points through to embryo transfer, but at no time was there a reported to increase in patients with OHSS (Jaeger et al., statistically significant difference. During stimulation of the 1987). Scarpellini and Scarpellini (1992) also found that ovaries and until embryo transfer, the CA602 concentrations CA125 was increased 8 days after ovulation. The elevation of in both the OHSS and the non-ohss groups were steady, CA125 was detected in the mid-luteal phase (Oezaksit et al., almost below the threshold value of 63 U/ml (Figure 1). In 1993). From these data, it seems that tumour markers may not 443

4 K.Takamatsu et al. Table I. CA602 concentrations (U/ml) during in-vitro fertilization and embryo transfer at various sampling times Group Sampling point a OHSS b c d e Non-OHSS f OHSS ovarian hyperstimulation syndrome. a Points of blood sample collection: (i) during the previous cycle (not in the menstrual period); (ii) when the leading follicle became 15 mm in diameter; (iii) at oocyte retrieval; (iv) at embryo transfer; and (v) 10 days after embryo transfer. b e P 0.01 compared with point 5. f P 0.03 compared with the corresponding sampling point in the OHSS group. be suitable predictors of OHSS, but may be indicators of concentrations of CA602. There was a case in our study, for OHSS. Compared with CA125, the great advantage of CA602 example, in which the oestradiol concentration was high, is higher sensitivity. The detection limits of both markers were ~2800 pg/ml, before oocyte retrieval, but the CA602 concentration almost the same since it was known that 258 CA602 units was in the normal range, 52 U/ml. In the patients excluded were equal to 108 CA125 units (Nozawa et al., 1991). The from this study because of endometriosis, high oestradiol gradient of the standard curve (units absorbance curve) for concentrations before oocyte retrieval were not related to the CA602 kit is steeper than that for the CA125 kit. Thus, CA602 concentrations (data not shown). CA602 can detect a smaller change with an at least 2-fold From the economic point of view, the analysis of tumour greater sensitivity than CA125. Recently, a second-generation markers is a relatively high cost procedure. However, this is CA125 assay was introduced into clinical use. However, it balanced by high sensitivity, and much progress has been was reported that the ability of this assay to differentiate made towards establishing simpler and cheaper kits for measuring between patients with and without endometriosis is small, CA602. especially in the lower range (Hornstein et al., 1995). From In conclusion, our preliminary results suggest that CA602 these points of view, CA602 is especially useful when small may serve as an indicator of OHSS. Further prospective changes in a narrow range are considered, as in OHSS. investigations of large numbers of patients will reveal its The origin of CA602 in OHSS remains to be clarified. One usefulness. hypothesis is that the gonadotrophic stimulus leads to ovarian enlargement, which modifies the surface of the ovary and also increases steroid production, which in turn increases the Acknowledgements thickness of the endometrium; the tumour marker then pours The authors are indebted to Dr Shin-ichi Yoshimura, Dr Jun-ichi into the circulatory stream (Scarpellini and Scarpellini, 1992). Kobayashi and Dr Yasuhiro Konishi for their support and suggestions, Some investigators have concluded that the high CA125 and to Ms Hiromi Enomoto and Ms Miho Murata for their excellent technical assistance. We also thank Mochida Co. Ltd for generously concentration in OHSS patients could be related to a peritoneal providing the kits for measuring CA602 and CA546. contribution (Oezaksit et al., 1993). On the other hand, CA125 concentrations in serum from ovarian cancer patients were not correlated with the size of the ovarian tumour (Fleuren et al., References 1987). In animal models, isolation of both ovaries from the Amit, A., Yaron, Y., Yovel, I. et al. (1993) Repeated aspiration of ovarian peritoneal cavity did not prevent ascites formation in the follicles and early corpus luteum cysts in an in-vitro fertilization programme reduces the risk of ovarian hyperstimulation syndrome in high responders. presence of OHSS (Yarali et al., 1993). With respect to Hum. Reprod., 8, peritoneal stimulation, repeated aspiration of ovarian follicles Barbieri, R.L., Niloff, J.M., Bast, R.C., Jr et al. (1986) Elevated serum and early corpus luteum cysts has been found to reduce the concentration of CA125 in patients with advanced endometriosis. Fertil. Steril., 15, risk of OHSS (Amit et al., 1993). In endometriosis, some Bast, R.C., Jr, Klug, T.L., St John, E. et al. (1983) A radioimmunoassay using investigators focused on the angiogenic factor released by the a monoclonal antibody to monitor the course of epithelial ovarian cancer. ovary (McLaren et al., 1996). This factor may also be released N. Engl. J. Med., 309, in OHSS and alter the permeability of the peritoneum and Bettendorf, G. and Lindner, C. (1987) The ovarian hyperstimulation syndrome. Horm. Metab. Res., 19, ovary, resulting in the increased concentrations of CA602. Blankstein, J., Shalev, J., Saadon, T. et al. (1987) Ovarian hyperstimulation Among current biochemical analyses, oestradiol concentra- syndrome: prediction by number and size of preovulatory ovarian follicles. tions in serum or urine are only used for the evaluation of Fertil. Steril., 47, OHSS (Schenker and Weinstein, 1978). In this study, oestradiol Eisermann, J. and Collins, J.L. (1989) Enzyme immunoassay determination of CA125 in serum: minimal endometriosis after ovarian hyperstimulation. concentrations in the OHSS group were significantly higher Fertil. Steril., 51, than those in the non-ohss group at 10 days after embryo Fleuren, G.J., Nap, M., Aalders, J.G. et al. (1987) Explanation of the limited transfer. The oestradiol concentration is, however, thought to correlation between tumor CA 125 content and serum CA 125 antigen levels in patients with ovarian tumors. Cancer, 60, be associated with the number of follicles (MacDougall et al., Halila, H., Stenman, U.H. and Seppala, M. (1986) Ovarian cancer antigen 1993), and the serum oestradiol concentrations before oocyte CA125 levels in pelvic inflammatory disease and pregnancy. Cancer, 57, retrieval exhibited a relatively wide range compared with the

5 CA602 and CA546 concentrations during IVF and embryo transfer Hornstein, M.D., Harlow, B.L., Thomas, P.P. et al. (1995) Use of a new CA125 assay in the diagnosis of endometriosis. Hum. Reprod., 10, Jaeger, W., Diedrich, K. and Wildt, L. (1987) Elevated levels of CA-125 in serum of patients suffering from ovarian hyperstimulation syndrome. Fertil. Steril., 48, Konishi, Y., Yoshimura, S., Hosaka, M. et al. (1993) A study on improving the pregnancy rate in in vitro fertilization and embryo transfer. J. Fertil. Implant., 10, (in Japanese). Lanzone, A., Fulghesu, A.M., Guida, C. et al. (1990) Serum CA125 levels do not depend on ovarian steroidogenesis. Fertil. Steril., 54, MacDougall, M.J., Tan, S.L., Balen, A. et al. (1993) A controlled study comparing patients with and without polycystic ovaries undergoing in-vitro fertilization. Hum. Reprod., 8, McLaren, J., Prentice, A., Charnock-Jones, D.S. et al. (1996) Vascular endothelial growth factor is produced by peritoneal fluid macrophages in endometriosis and is regulated by ovarian steroids. J. Clin. Invest., 98, Nozawa, S., Yajima, M., Kojima, K. et al. (1989) Tumor-associated mucintype glycoprotein (CA54/61) defined by two monoclonal antibodies (MA54 and MA61) in ovarian cancers. Cancer Res., 49, Nozawa, S., Yajima, M., Sasaki, H. et al. (1991) A new CA125-like antigen (CA602) recognized by two monoclonal antibodies against a newly established ovarian clear cell carcinoma cell line (RMG-II). Jpn. J. Cancer Res., 82, Nozawa, S., Aoki, D., Yajima, M. et al. (1992a) CA54/61 as a marker for epithelial ovarian cancer. Cancer Res., 52, Nozawa, S., Udagawa, Y., Sasaki, H. et al. (1992b) Studies of clinical usefulness of new tumor markers of ovarian cancer, CA54/61 and CA602 CA602 assay reagent kit performance its normal value and correlations with other tumor markers. Jpn. J. Cancer Chemother., 19, (in Japanese). Nozawa, S., Udagawa, Y., Sasaki, H. et al. (1992c) Studies of preclinical and clinical usefulness of new tumor markers of ovarian carcinoma, CA54/61 and CA Evaluation of the diagnostic accuracy of CA54/61, a study of the normal range of its values and its correlation with other tumor markers. Jpn. J. Cancer Chemother., 19, (in Japanese). Oezaksit, G., Turhan, N.O., Oral, H. et al. (1993) Relationship between serum CA 125 levels, endometrial thickness and corpus luteum function in different stages of ovarian activity. J. Endocrinol. Invest., 16, Rabau, E., Serr, D.M., David, A. et al. (1967) Human menopausal gonadotropins for anovulation and sterility. Am. J. Obstet. Gynecol., 96, Scarpellini, L. and Scarpellini, F. (1992) CA-125 and ovarian hyperstimulation. Acta Eur. Fertil., 23, Schenker, J.G. and Weinstein, D. (1978) Ovarian hyperstimulation syndrome: a current survey. Fertil. Steril., 30, Suzuki, M., Sekiguchi, K., Ohwada, M. et al. (1990) Clinical value of a new serum tumor marker CA602 in ovarian cancers. J. Jpn. Soc. Cancer Ther., 25, WHO Scientific Group (1973) Agents stimulating gonadal function in the human. Report of a WHO Scientific Group, Geneva, Switzerland. Yarali, H., Fleige-Zahradka, B.G., Yuen, B.H. et al. (1993) The ascites in the ovarian hyperstimulation syndrome does not originate from the ovary. Fertil. Steril., 59, Zweers, A., De Boever, J., Serreyn, R. et al. (1990) Correlation between peripheral CA125 levels and ovarian activity. Fertil. Steril., 54, Received on August 29, 1996; accepted on December 6,