- Email: [email protected]
CA-125 levels in human uterine fluid*
FERTILITY AND STERILITY
Vol. 57, No.3, March 1992 Printed on acw.-free paper in U.S.A.
Copyright" 1992 The American Fertility Society
CA-125 levels in human uterine fluid*
Mick Abae, M.D.t Mark Gibson, M.D.:\: Jane Chapitis, M.D., Ph.D.
Daniel H. Riddick, M.D., Ph.D. John R. Brumsted, M.D.
Department of Obstetrics and Gynecology, Division of Reproductive Endocrinology and Infertility, University of Vermont College of Medicine, Burlington, Vermont
Objective: To measure uterine fluid CA-125 concentration and to determine if any menstrual cycle phase dependent changes exist in its level. Serum levels are measured for comparison. Design: CA-125 levels in uterine fluid were measured during the follicular and luteal phases of the menstrual cycle. In a sequential study, paired uterine fluid and serum samples were obtained once in both midfollicular and midluteal phases of the same menstrual cycle. Results: CA-125 in uterine fluid during the follicular phase (n = 14) ranged from 16.4 X 103 to 616.5 X 103 U/mL, and from 6.2 X 103 to 567.3 X 103 U/mL in the luteal phase (n = 11). In the paired sequential uterine fluid and serum samples, (1) the means (±SEM) CA-125 in uterine fluid were 81.5 X 103 ± 37.9 X 103 U/mL and 91.4 X 103 ± 56.8 X 103 U/mL in the midfollicular and midluteal phases, respectively (P = 0.75); (2) the CA-125 levels in serum increased in the midluteal phase (P < 0.05); and (3) compared with serum, uterine fluid levels were greater with a wider range. Conclusions: When compared with serum CA-125, uterine fluid contains high concentrations varying over a wide range without fluctuation between the follicular and luteal phases of the menstrual cycle. Fertil Steril1992;57:531-4 Key Words: CA-125, endometrium, uterine fluid
OC-125, a murine immunoglobulin G-1 monoclonal antibody, recognizes an antigenic determinant, designated CA-125, that originally was associated with an ovarian serous cystadenocarcinoma (1). CA-125 determinants are found on a high molecular weight surface glycoprotein that is expressed by a majority of nonmucinous epithelial ovarian carcinomas and a smaller fraction of malignancies of the breast, reproductive tract, and gastrointestinal tract (2). Using radioimmunoassay techniques, only Received August 14, 1991; revised and accepted November 13, 1991. * Presented in part at the 46th Annual Meeting of The American Fertility Society, Washington, D.C., October 13 to 18, 1990. t Reprint requests and present address: Mick Abae, M.D., Department of Obstetrics and Gynecology, Division of Reproductive Endocrinology and Infertility, REPSCEND Laboratories (D-5), University of Miami School of Medicine, Post Office Box 016960, Miami, Florida 331Ol. Present address: Department of Obstetrics and Gynecology, West Virginia University, Morgantown, West Virginia.
*
Vol. 57, No.3, March 1992
1% of apparently healthy menstruating women have serum CA-125 levels > 35 U/mL (3). Elevated serum levels are known to occur in 80% or more of patients with advanced ovarian carcinoma (3), in the first trimester of pregnancy (4), and in advanced cases of endometriosis (5). Studies of serum CA-125 levels during the menstrual cycle in apparently normal women demonstrated fluctuating levels, with the highest peaks occurring at the time of menses (6). Several studies have demonstrated high concentrations of CA-125 in cervical mucus (7), endometrial glands (8), decidua, chorion, and amniotic fluid (9). Jacobs et al. (10) reported on the compartmental distribution of CA-125 in the reproductive tract of healthy pregnant and nonpregnant women. Despite numerous studies, the primary site of synthesis of CA-125 in the reproductive tract remains unclear. In this study, we report the finding of high concentrations of CA-125 in uterine fluid and evaluate the hypothesis that CA-125 levels in uterine fluid fluctuate with the two phases of the menstrual cycle. Abae et al.
CA-125 levels in human uterine fluid
531
MATERIALS AND METHODS
This study was approved by the Committee on Human Research at the University of Vermont College of Medicine. Informed consent was obtained from all subjects. Samples of uterine fluid were collected during either the follicular (cycle days 8 to 13) or luteal (cycle days 19 to 25) phase of the menstrual cycle from 25 healthy, normally ovulatory women (ages 21 to 42 years) undergoing donor insemination or requesting tubal ligation reversal. Because of the possibility that any menstrual cycle phase dependent changes in uterine fluid CA-125 concentration for a single individual may have been obscured in a cross-sectional analysis, sequential samples of uterine fluid were obtained during the midfollicular (cycle days 8 to 11) and midluteal (cycle days 20 to 24) phases of the same menstrual cycle in 7 healthy, normally ovulatory women (ages 27 to 37 years). At the time of uterine fluid collection, serum samples were also collected in this cohort of volunteers. A modification of the technique described by Casslen (11) for collection of pure undiluted uterine fluid was used. A semisoft, 5-French Argyle pediatric feeding tube (Sherwood Medical, St. Louis, MO) was connected to a syringe and introduced into the uterine cavity under sterile conditions to a depth of approximately 7 cm. With care to avoid trauma to the endometrium, aspiration of uterine fluid was performed with application of a low-negative pressure and gentle rotation of the catheter. Clear uterine fluid, which is a yellow tinged fluid with viscosity similar to serum, was obtained in volumes ranging from 20 to 60 ~L. All samples were immediately centrifuged to provide a cell-free fraction. For centrifugation, the samples were transferred to a heparinized microcapillary tube (length 75 mm, outer diameter 1.50 mm, and inner diameter 1.10 mm) and centrifuged for 10 minutes at 13,000 X g. The supernatant was separated from the cellular sediment by breaking the glass capillary tube at the fluidcellular interface after scoring the tube with a diamond tip glass cutter. The supernatant was stored at -20°C until analysis. CA-125 measurements were made using a commercially available solid phase immunoradiometric assay (Centocor Corporation, Malvern, PA). The concentration of CA-125 was expressed as U/mL and theinterassay and intra-assay coefficients of variation were 15% and 9.3%, respectively. Paired sequential samples of uterine fluid and serum were run in the same assay. Because of high concentrations of CA-125 in uterine fluid, serial dilutions were made using phos532
Abae et aI.
CA-125 levels in human uterine fluid
phate-buffered saline (PBS) containing 1.0% bovine serum albumin (BSA) before assay. Immunoparallelism between uterine fluid CA -125 and purified assay kit standard was demonstrated using serial dilutions. To determine whether significant amounts of CA-125 were lost during the initial centrifugation, the trace residue present after removal of the supernatant was examined. The residue was diluted in 1.0 mL of PBS-BSA, vortexed for 2 minutes to resuspend, and centrifuged for 10 minutes at 13,000 X g. The supernatant was saved for CA-125 analysis. This wash procedure was repeated. CA-125 concentration in the first and second washes was considered negligible ranging from 0.3% to 0.5% of the total CA-125 present in the uterine fluid samples, indicating that the original supernatant contains nearly all ofthe assayable CA-125 in uterine fluid. Statistical Analysis
Analysis of the cross-sectional data was performed using unpaired Student's t-test. Measures of CA-125 in uterine fluid during the follicular and luteal phases within subjects were compared using Student's paired t-test. All CA-125 levels in uterine fluid were log transformed before analysis to satisfy homogeneity of variance and distribution assumptions. Means presented are geometric means. Pearson's correlation coefficient was used to examine the relationship between CA-125 concentration in the paired serum and uterine fluid samples. Probability < 0.05 was considered significant. RESULTS Cross-Sectional Study
The concentration of CA-125 in uterine fluid spanned a wide range in both the follicular and the luteal phases. Specifically, CA-125 levels ranged from 16.4 X 103 to 616.5 X 103 U/mL (mean [±SEM] = 89.3 X 103 ± 22.9 X 103 ) in the follicular phase (n = 14), whereas levels ranged from 6.2 X 103 to 567.3 X 103 U/mL (mean [±SEM] = 75.1 X 103 ± 34.5 X 103 ) in the luteal phase (n = 11) and were not significantly different. Sequential Study
In the paired samples of uterine fluid and serum from the two phases of the same menstrual cycle, the concentration of CA-125 in uterine fluid was significantly greater than the corresponding serum in all subjects (P < 0.05). Also, similar to the longitudinal study, it was again noted that uterine fluid Fertility and Sterility
CA-125 spanned over a wide range in both the midfollicular and midluteal phases (Fig. 1). The mean (±SEM) concentration of CA-125 in the sequential uterine fluid samples was 89.3 X 103 ± 37.9 X 103 U /mL in the midfollicular phase and 91.4 X 103 ± 56.8 X 103 U/mL in the midlutealphase (P = 0.75). Despite the large range of values between subjects, the paired samples from each individual tended to remain relatively constant. We did not observe any significant trends in uterine fluid CA-125 levels between the two phases of the same menstrual cycle. Values for the paired serum CA-125 levels are depicted in Figure 2. All serum CA-125 levels in both the midfollicular and midluteal phases were within the normal range of <35 U /mL. In this group of seven healthy and normally ovulatory subjects, we observed a significant increase (P < 0.05) in serum CA 125 levels in the midluteal phase as compared with the midfollicular phase of the same menstrual cycle. There was no correlation between the CA-125 concentration in uterine fluid and the paired serum levels. DISCUSSION
The data presented here describing uterine fluid CA-125 concentrations in normal ovulatory women confirm and extend the previous reports of the high levels of this glycoprotein in endometrial secretion (8). Our data show that uterine fluid CA-125 levels were significantly higher in all subjects when compared with their paired serum in both the midfollicular and midluteal phases of the same menstrual cycle. These data are also in agreement with studies that demonstrated in vitro production of CA-125 by intact endometrium (12). Thus, together with previous studies, these data further confirm
3 8
10 6
e"U?
~
10 5
I
-<
U
"Ci
.;;
~ 10' Q)
.:: ·C
.... Q)
~
103~
_______________________________
Follicular
Luteal
Figure 1 Concentration ofCA-125, measured by RIA, in uterine fluid obtained during the midfollicular and midluteal phases of the same menstrual cycle in seven women. Symbols represent the same individual subjects represented in Figure 2. No significant difference was noted (P = 0.75). Vol. 57, No.3, March 1992
20
-::J E
18 16
...........
::J
14
LO
12
N
I
U
~
::J 0:::
10 8 6 4
•
W
Vl
2 0
Follicular
Luteal
Figure 2 Concentration of CA-125, measured by RIA, in serum obtained during the midfollicular and midluteal phases of the same menstrual cycle in seven women. Symbols represent the same individual subjects represented in Figure 1. A significant difference between follicular and luteal samples was detected (P < 0.05).
that CA-125 is an excretory product of the endometrium (8, 13). We noted wide ranges of uterine fluid CA-125 during two phases of the menstrual cycle: midfollicular, 38-fold; midluteal, 91-fold. A similar though narrower 30-fold range of CA-125 characterized proliferative and secretory endometrial tissue (10). Contiguous reproductive tract fluid, cervical mucus at midcycle, contained an ll-fold range of CA-125 (7). For comparison, we and others (10) found that serum CA-125 levels during the menstrual cycle vary 4- to ll-fold. Although the data presented demonstrate very high concentrations of CA-125 in uterine fluid, it does not directly shed light on the origin of serum CA-125. The independence of cycle-dependent changes in serum CA-125 levels and uterine fluid levels seems to argue against a role for uterine fluid CA-125 in accounting for serum levels. However, because of a lack of understanding of the dynamics of human uterine fluid, CA-125 analysis based on concentrations alone may be insufficient. We have previously demonstrated in women that the preovulatory ovarian follicular contribution to the circulating levels of CA-125 is negligible by evaluating its concentration in follicular fluid and also by observing no CA-125 production in vitro by either luteinized granulosa cells obtained during in vitro fertilization or by midluteal corpus luteum (unpublished observation). Other investigators have also excluded a possible ovarian origin of circulating levels of CA-125 in normal women by demonstrating similar levels in the blood draining the active and inactive ovaries compared with peripheral venous blood (unpublished observation). Evidence of an endometrial origin for CA-125 is from immunohisAbae et al.
CA-125 levels in human uterine fluid
533
tochemical studies (13), tissue content studies (10), and in vitro endometrial production of CA-125 (12). An analysis ofCA-125 content in the endometrial and circulating compartments may provide an explanation for the circulating levels. For this analysis, the endometrial luminal compartment was calculated by the sum: (1) endometrial glandular luminal volume, a space not accessible during uterine fluid sampling and (2) uterine fluid volume. Total endometrial volume was estimated by the previously proposed formula (V = ! 7r X r2 X h, where rand h are the radius and height of the functional endometrium, respectively) (14). Using the values r = 2.3 mm (midfollicular) and 2.95 mm (midluteal) (15), and h = 35 mm (16), the total endometrial volume is 194 ~L (midfollicular) and 319 ~L (midluteal). The estimated proportion of endometrial volume that is glandular lumen is 3% midfollicular and 18% midluteal (17); hence, the glandular luminal volumes are 5.8 ~L and 57.4 ~L, respectively. Summing the glandular luminal volumes with their respective uterine fluid volumes (105 ~L midfollicular and 40 ~L midluteal) (18), the resultant endometrial luminal compartments are 110.8 ~L and 97.4 ~L, respectively. Using the mean concentrations of uterine fluid CA-125 from the present study, total endometrialluminal compartment CA-125 content in the midfollicular and midluteal phases are 9.0 X 103 and 8.9 X 103 U, respectively. Similar estimation of the total circulating CA-125 content can be derived. Using 2,700 cc plasma volume (19), representing 20% to 25% of total extracellular fluid, total circulating CA-125 content is 16.6 X 103 U in the midfollicular phase and 28.6 X 103 U in the midluteal phase. Although the data presented here indicate approximately a 1:5,000 concentration gradient of CA-125 between the peripheral circulation and the endometrial lumen, the total contents of CA-125 in these two fluid compartments may indicate that the secretion of CA-125 by the endometrium is bidirectional, intraluminal, and intravascular. Further studies are needed to better define the physiological mechanisms associated with the absorption of this large glycoprotein and to determine its regulation by the ovarian steroids and, most intriguing, its possible role within the normal female reproductive tract.
Acknowledgment. We thank Vincent R. Zurawski, Jr., Ph.D., (Centocor, Malvern, PAl for the generous gift of the CA-125 RIA kits.
534
Abae et al.
CA -125 levels in human uterine fluid
REFERENCES 1. Bast RC Jr, Feeney M, Lazarus H, Nadler LM, Colvin RB, Knapp RC. Reactivity of a monoclonal antibody with human ovarian carcinoma. J Clin Invest 1981;68:1331-7. 2. Jacobs I, Bast RC. The CA-125 tumour-associated antigen: a review of the literature. Hum Reprod 1989;4:1-12. 3. Bast RC, Klug TL, St. John E, Niloff JM, Lazarus H, Berkowitz RS, et al. A radioimmunoassay using monoclonal antibody to monitor the course of epithelial ovarian cancer. N Engl J Med 1983;309:883-7. 4. Brumsted JR, Nakajima ST, Badger G, Riddick DH, Gibson M. Serum concentration of CA-125 during the first trimester of normal and abnormal pregnancies. J Reprod Med 1990;35: 499-502. 5. Barbieri RL, Niloff JM, Bast RC Jr, Schaetzl E, Kistner RW, Knapp RC. Elevated serum concentrations of CA-125 in patients with advanced endometriosis. Fertil SteriI1986;45: 630-4. 6. Pittaway DE, Fayez JA. Serum CA-125 levels increase during menses. Am J Obstet GynecoI1987;156:75-6. 7. de Bruijn HWA, Calkoen-carpay T, Jager S, Duk JM, Aalder J G, Flerren GJ. The tumor marker CA -125 is a common constituent of normal cervical mucus. Am J Obstet Gynecol 1986;154:1088-91. 8. Weintraub J, BischoffP, TsenfL, Redard M, Vassilakoss P. CA-125 is an excretory product of endometrial glands. BioI Reprod 1990;42:721-6. 9. O'Brien TJ, Hardin JW, Bannon GA, Norris JS, Quirk JG. CA-125 antigen in human amniotic fluid and fetal membranes. Am J Obstet Gynecol 1986;155:50-5. 10. Jacobs IJ, Fay TN, Stabile I, Bridges J, Oram DHO, Grudzinkas JG. The distribution of CA-125 in the reproductive tract of pregnant and non-pregnant women. Br J Obstet GynecoI1988;95:1190-4. 11. Casslen B. Proteinases and proteinase inhibitors in uterine fluid, with special reference to IUD-users. Acta Obstet Gynecol Scand 1981;98(Suppl):1-38. 12. Brumsted JR, McBean JH, Deaton JL, Gibson M. CA-125 secretion by luteal endometrium in vitro. Hum Reprod 1990;5: 682-4. 13. Kabawat SE, Bast RC Jr, Welch WR, Knapp RC, Colvin RB. Immunopathologic characterization of a monoclonal antibody that recognizes common surface antigens of human ovarian tumors of serous, andometroid, and clear cell type. Am J PathoI1983;79:98-104. 14. Giorlandino C, Gleicher N, Cinzia N, Vizzone A, Gentili P, Taramanni C. The sonographic picture of endometrium in spontaneous and induced cycles. Fertil SterilI987;47:508-11. 15. Lenz S, Lindenberg S. Ultrasonic evaluation of endometrial growth in women with normal cycles during spontaneous and stimulated cycles. Hum Reprod 1990;5:377-81. 16. Muckle CWo Clinical anatomy of the uterus, fallopian tubes, and ovaries. In: Sciarra JJ, editor. Obstetrics and gynecology. Philadelphia: Harper and Row, 1987:1. 17. Johannisson E, Landgren B-M, Rohr HP, Diczfalusy E. Endometrial morphology and peripheral hormone levels in women with regular menstrual cycle. Fertil Steril 1987;48: 401-8. 18. Clemetson CAB, Kim JK, de Jesus TPS, MallikaJjuneswara YR, Wilds JH. Human uterine fluid potassium and the menstrua! cycle. J Obstet Gynaecol Br Commonw 1973;80:553-61. 19. Harrison RA. Intravenous fluids and acid-base balance. In: Dripps RD, Eckenhoff JE, Vandam LD, editors. Introduction to anesthesia the principles of safe practice. 7th ed. Philadelphia: Saunders, 1988:261.
Fertility and Sterility
Vol. 57, No.3, March 1992 Printed on acw.-free paper in U.S.A.
Copyright" 1992 The American Fertility Society
CA-125 levels in human uterine fluid*
Mick Abae, M.D.t Mark Gibson, M.D.:\: Jane Chapitis, M.D., Ph.D.
Daniel H. Riddick, M.D., Ph.D. John R. Brumsted, M.D.
Department of Obstetrics and Gynecology, Division of Reproductive Endocrinology and Infertility, University of Vermont College of Medicine, Burlington, Vermont
Objective: To measure uterine fluid CA-125 concentration and to determine if any menstrual cycle phase dependent changes exist in its level. Serum levels are measured for comparison. Design: CA-125 levels in uterine fluid were measured during the follicular and luteal phases of the menstrual cycle. In a sequential study, paired uterine fluid and serum samples were obtained once in both midfollicular and midluteal phases of the same menstrual cycle. Results: CA-125 in uterine fluid during the follicular phase (n = 14) ranged from 16.4 X 103 to 616.5 X 103 U/mL, and from 6.2 X 103 to 567.3 X 103 U/mL in the luteal phase (n = 11). In the paired sequential uterine fluid and serum samples, (1) the means (±SEM) CA-125 in uterine fluid were 81.5 X 103 ± 37.9 X 103 U/mL and 91.4 X 103 ± 56.8 X 103 U/mL in the midfollicular and midluteal phases, respectively (P = 0.75); (2) the CA-125 levels in serum increased in the midluteal phase (P < 0.05); and (3) compared with serum, uterine fluid levels were greater with a wider range. Conclusions: When compared with serum CA-125, uterine fluid contains high concentrations varying over a wide range without fluctuation between the follicular and luteal phases of the menstrual cycle. Fertil Steril1992;57:531-4 Key Words: CA-125, endometrium, uterine fluid
OC-125, a murine immunoglobulin G-1 monoclonal antibody, recognizes an antigenic determinant, designated CA-125, that originally was associated with an ovarian serous cystadenocarcinoma (1). CA-125 determinants are found on a high molecular weight surface glycoprotein that is expressed by a majority of nonmucinous epithelial ovarian carcinomas and a smaller fraction of malignancies of the breast, reproductive tract, and gastrointestinal tract (2). Using radioimmunoassay techniques, only Received August 14, 1991; revised and accepted November 13, 1991. * Presented in part at the 46th Annual Meeting of The American Fertility Society, Washington, D.C., October 13 to 18, 1990. t Reprint requests and present address: Mick Abae, M.D., Department of Obstetrics and Gynecology, Division of Reproductive Endocrinology and Infertility, REPSCEND Laboratories (D-5), University of Miami School of Medicine, Post Office Box 016960, Miami, Florida 331Ol. Present address: Department of Obstetrics and Gynecology, West Virginia University, Morgantown, West Virginia.
*
Vol. 57, No.3, March 1992
1% of apparently healthy menstruating women have serum CA-125 levels > 35 U/mL (3). Elevated serum levels are known to occur in 80% or more of patients with advanced ovarian carcinoma (3), in the first trimester of pregnancy (4), and in advanced cases of endometriosis (5). Studies of serum CA-125 levels during the menstrual cycle in apparently normal women demonstrated fluctuating levels, with the highest peaks occurring at the time of menses (6). Several studies have demonstrated high concentrations of CA-125 in cervical mucus (7), endometrial glands (8), decidua, chorion, and amniotic fluid (9). Jacobs et al. (10) reported on the compartmental distribution of CA-125 in the reproductive tract of healthy pregnant and nonpregnant women. Despite numerous studies, the primary site of synthesis of CA-125 in the reproductive tract remains unclear. In this study, we report the finding of high concentrations of CA-125 in uterine fluid and evaluate the hypothesis that CA-125 levels in uterine fluid fluctuate with the two phases of the menstrual cycle. Abae et al.
CA-125 levels in human uterine fluid
531
MATERIALS AND METHODS
This study was approved by the Committee on Human Research at the University of Vermont College of Medicine. Informed consent was obtained from all subjects. Samples of uterine fluid were collected during either the follicular (cycle days 8 to 13) or luteal (cycle days 19 to 25) phase of the menstrual cycle from 25 healthy, normally ovulatory women (ages 21 to 42 years) undergoing donor insemination or requesting tubal ligation reversal. Because of the possibility that any menstrual cycle phase dependent changes in uterine fluid CA-125 concentration for a single individual may have been obscured in a cross-sectional analysis, sequential samples of uterine fluid were obtained during the midfollicular (cycle days 8 to 11) and midluteal (cycle days 20 to 24) phases of the same menstrual cycle in 7 healthy, normally ovulatory women (ages 27 to 37 years). At the time of uterine fluid collection, serum samples were also collected in this cohort of volunteers. A modification of the technique described by Casslen (11) for collection of pure undiluted uterine fluid was used. A semisoft, 5-French Argyle pediatric feeding tube (Sherwood Medical, St. Louis, MO) was connected to a syringe and introduced into the uterine cavity under sterile conditions to a depth of approximately 7 cm. With care to avoid trauma to the endometrium, aspiration of uterine fluid was performed with application of a low-negative pressure and gentle rotation of the catheter. Clear uterine fluid, which is a yellow tinged fluid with viscosity similar to serum, was obtained in volumes ranging from 20 to 60 ~L. All samples were immediately centrifuged to provide a cell-free fraction. For centrifugation, the samples were transferred to a heparinized microcapillary tube (length 75 mm, outer diameter 1.50 mm, and inner diameter 1.10 mm) and centrifuged for 10 minutes at 13,000 X g. The supernatant was separated from the cellular sediment by breaking the glass capillary tube at the fluidcellular interface after scoring the tube with a diamond tip glass cutter. The supernatant was stored at -20°C until analysis. CA-125 measurements were made using a commercially available solid phase immunoradiometric assay (Centocor Corporation, Malvern, PA). The concentration of CA-125 was expressed as U/mL and theinterassay and intra-assay coefficients of variation were 15% and 9.3%, respectively. Paired sequential samples of uterine fluid and serum were run in the same assay. Because of high concentrations of CA-125 in uterine fluid, serial dilutions were made using phos532
Abae et aI.
CA-125 levels in human uterine fluid
phate-buffered saline (PBS) containing 1.0% bovine serum albumin (BSA) before assay. Immunoparallelism between uterine fluid CA -125 and purified assay kit standard was demonstrated using serial dilutions. To determine whether significant amounts of CA-125 were lost during the initial centrifugation, the trace residue present after removal of the supernatant was examined. The residue was diluted in 1.0 mL of PBS-BSA, vortexed for 2 minutes to resuspend, and centrifuged for 10 minutes at 13,000 X g. The supernatant was saved for CA-125 analysis. This wash procedure was repeated. CA-125 concentration in the first and second washes was considered negligible ranging from 0.3% to 0.5% of the total CA-125 present in the uterine fluid samples, indicating that the original supernatant contains nearly all ofthe assayable CA-125 in uterine fluid. Statistical Analysis
Analysis of the cross-sectional data was performed using unpaired Student's t-test. Measures of CA-125 in uterine fluid during the follicular and luteal phases within subjects were compared using Student's paired t-test. All CA-125 levels in uterine fluid were log transformed before analysis to satisfy homogeneity of variance and distribution assumptions. Means presented are geometric means. Pearson's correlation coefficient was used to examine the relationship between CA-125 concentration in the paired serum and uterine fluid samples. Probability < 0.05 was considered significant. RESULTS Cross-Sectional Study
The concentration of CA-125 in uterine fluid spanned a wide range in both the follicular and the luteal phases. Specifically, CA-125 levels ranged from 16.4 X 103 to 616.5 X 103 U/mL (mean [±SEM] = 89.3 X 103 ± 22.9 X 103 ) in the follicular phase (n = 14), whereas levels ranged from 6.2 X 103 to 567.3 X 103 U/mL (mean [±SEM] = 75.1 X 103 ± 34.5 X 103 ) in the luteal phase (n = 11) and were not significantly different. Sequential Study
In the paired samples of uterine fluid and serum from the two phases of the same menstrual cycle, the concentration of CA-125 in uterine fluid was significantly greater than the corresponding serum in all subjects (P < 0.05). Also, similar to the longitudinal study, it was again noted that uterine fluid Fertility and Sterility
CA-125 spanned over a wide range in both the midfollicular and midluteal phases (Fig. 1). The mean (±SEM) concentration of CA-125 in the sequential uterine fluid samples was 89.3 X 103 ± 37.9 X 103 U /mL in the midfollicular phase and 91.4 X 103 ± 56.8 X 103 U/mL in the midlutealphase (P = 0.75). Despite the large range of values between subjects, the paired samples from each individual tended to remain relatively constant. We did not observe any significant trends in uterine fluid CA-125 levels between the two phases of the same menstrual cycle. Values for the paired serum CA-125 levels are depicted in Figure 2. All serum CA-125 levels in both the midfollicular and midluteal phases were within the normal range of <35 U /mL. In this group of seven healthy and normally ovulatory subjects, we observed a significant increase (P < 0.05) in serum CA 125 levels in the midluteal phase as compared with the midfollicular phase of the same menstrual cycle. There was no correlation between the CA-125 concentration in uterine fluid and the paired serum levels. DISCUSSION
The data presented here describing uterine fluid CA-125 concentrations in normal ovulatory women confirm and extend the previous reports of the high levels of this glycoprotein in endometrial secretion (8). Our data show that uterine fluid CA-125 levels were significantly higher in all subjects when compared with their paired serum in both the midfollicular and midluteal phases of the same menstrual cycle. These data are also in agreement with studies that demonstrated in vitro production of CA-125 by intact endometrium (12). Thus, together with previous studies, these data further confirm
3 8
10 6
e"U?
~
10 5
I
-<
U
"Ci
.;;
~ 10' Q)
.:: ·C
.... Q)
~
103~
_______________________________
Follicular
Luteal
Figure 1 Concentration ofCA-125, measured by RIA, in uterine fluid obtained during the midfollicular and midluteal phases of the same menstrual cycle in seven women. Symbols represent the same individual subjects represented in Figure 2. No significant difference was noted (P = 0.75). Vol. 57, No.3, March 1992
20
-::J E
18 16
...........
::J
14
LO
12
N
I
U
~
::J 0:::
10 8 6 4
•
W
Vl
2 0
Follicular
Luteal
Figure 2 Concentration of CA-125, measured by RIA, in serum obtained during the midfollicular and midluteal phases of the same menstrual cycle in seven women. Symbols represent the same individual subjects represented in Figure 1. A significant difference between follicular and luteal samples was detected (P < 0.05).
that CA-125 is an excretory product of the endometrium (8, 13). We noted wide ranges of uterine fluid CA-125 during two phases of the menstrual cycle: midfollicular, 38-fold; midluteal, 91-fold. A similar though narrower 30-fold range of CA-125 characterized proliferative and secretory endometrial tissue (10). Contiguous reproductive tract fluid, cervical mucus at midcycle, contained an ll-fold range of CA-125 (7). For comparison, we and others (10) found that serum CA-125 levels during the menstrual cycle vary 4- to ll-fold. Although the data presented demonstrate very high concentrations of CA-125 in uterine fluid, it does not directly shed light on the origin of serum CA-125. The independence of cycle-dependent changes in serum CA-125 levels and uterine fluid levels seems to argue against a role for uterine fluid CA-125 in accounting for serum levels. However, because of a lack of understanding of the dynamics of human uterine fluid, CA-125 analysis based on concentrations alone may be insufficient. We have previously demonstrated in women that the preovulatory ovarian follicular contribution to the circulating levels of CA-125 is negligible by evaluating its concentration in follicular fluid and also by observing no CA-125 production in vitro by either luteinized granulosa cells obtained during in vitro fertilization or by midluteal corpus luteum (unpublished observation). Other investigators have also excluded a possible ovarian origin of circulating levels of CA-125 in normal women by demonstrating similar levels in the blood draining the active and inactive ovaries compared with peripheral venous blood (unpublished observation). Evidence of an endometrial origin for CA-125 is from immunohisAbae et al.
CA-125 levels in human uterine fluid
533
tochemical studies (13), tissue content studies (10), and in vitro endometrial production of CA-125 (12). An analysis ofCA-125 content in the endometrial and circulating compartments may provide an explanation for the circulating levels. For this analysis, the endometrial luminal compartment was calculated by the sum: (1) endometrial glandular luminal volume, a space not accessible during uterine fluid sampling and (2) uterine fluid volume. Total endometrial volume was estimated by the previously proposed formula (V = ! 7r X r2 X h, where rand h are the radius and height of the functional endometrium, respectively) (14). Using the values r = 2.3 mm (midfollicular) and 2.95 mm (midluteal) (15), and h = 35 mm (16), the total endometrial volume is 194 ~L (midfollicular) and 319 ~L (midluteal). The estimated proportion of endometrial volume that is glandular lumen is 3% midfollicular and 18% midluteal (17); hence, the glandular luminal volumes are 5.8 ~L and 57.4 ~L, respectively. Summing the glandular luminal volumes with their respective uterine fluid volumes (105 ~L midfollicular and 40 ~L midluteal) (18), the resultant endometrial luminal compartments are 110.8 ~L and 97.4 ~L, respectively. Using the mean concentrations of uterine fluid CA-125 from the present study, total endometrialluminal compartment CA-125 content in the midfollicular and midluteal phases are 9.0 X 103 and 8.9 X 103 U, respectively. Similar estimation of the total circulating CA-125 content can be derived. Using 2,700 cc plasma volume (19), representing 20% to 25% of total extracellular fluid, total circulating CA-125 content is 16.6 X 103 U in the midfollicular phase and 28.6 X 103 U in the midluteal phase. Although the data presented here indicate approximately a 1:5,000 concentration gradient of CA-125 between the peripheral circulation and the endometrial lumen, the total contents of CA-125 in these two fluid compartments may indicate that the secretion of CA-125 by the endometrium is bidirectional, intraluminal, and intravascular. Further studies are needed to better define the physiological mechanisms associated with the absorption of this large glycoprotein and to determine its regulation by the ovarian steroids and, most intriguing, its possible role within the normal female reproductive tract.
Acknowledgment. We thank Vincent R. Zurawski, Jr., Ph.D., (Centocor, Malvern, PAl for the generous gift of the CA-125 RIA kits.
534
Abae et al.
CA -125 levels in human uterine fluid
REFERENCES 1. Bast RC Jr, Feeney M, Lazarus H, Nadler LM, Colvin RB, Knapp RC. Reactivity of a monoclonal antibody with human ovarian carcinoma. J Clin Invest 1981;68:1331-7. 2. Jacobs I, Bast RC. The CA-125 tumour-associated antigen: a review of the literature. Hum Reprod 1989;4:1-12. 3. Bast RC, Klug TL, St. John E, Niloff JM, Lazarus H, Berkowitz RS, et al. A radioimmunoassay using monoclonal antibody to monitor the course of epithelial ovarian cancer. N Engl J Med 1983;309:883-7. 4. Brumsted JR, Nakajima ST, Badger G, Riddick DH, Gibson M. Serum concentration of CA-125 during the first trimester of normal and abnormal pregnancies. J Reprod Med 1990;35: 499-502. 5. Barbieri RL, Niloff JM, Bast RC Jr, Schaetzl E, Kistner RW, Knapp RC. Elevated serum concentrations of CA-125 in patients with advanced endometriosis. Fertil SteriI1986;45: 630-4. 6. Pittaway DE, Fayez JA. Serum CA-125 levels increase during menses. Am J Obstet GynecoI1987;156:75-6. 7. de Bruijn HWA, Calkoen-carpay T, Jager S, Duk JM, Aalder J G, Flerren GJ. The tumor marker CA -125 is a common constituent of normal cervical mucus. Am J Obstet Gynecol 1986;154:1088-91. 8. Weintraub J, BischoffP, TsenfL, Redard M, Vassilakoss P. CA-125 is an excretory product of endometrial glands. BioI Reprod 1990;42:721-6. 9. O'Brien TJ, Hardin JW, Bannon GA, Norris JS, Quirk JG. CA-125 antigen in human amniotic fluid and fetal membranes. Am J Obstet Gynecol 1986;155:50-5. 10. Jacobs IJ, Fay TN, Stabile I, Bridges J, Oram DHO, Grudzinkas JG. The distribution of CA-125 in the reproductive tract of pregnant and non-pregnant women. Br J Obstet GynecoI1988;95:1190-4. 11. Casslen B. Proteinases and proteinase inhibitors in uterine fluid, with special reference to IUD-users. Acta Obstet Gynecol Scand 1981;98(Suppl):1-38. 12. Brumsted JR, McBean JH, Deaton JL, Gibson M. CA-125 secretion by luteal endometrium in vitro. Hum Reprod 1990;5: 682-4. 13. Kabawat SE, Bast RC Jr, Welch WR, Knapp RC, Colvin RB. Immunopathologic characterization of a monoclonal antibody that recognizes common surface antigens of human ovarian tumors of serous, andometroid, and clear cell type. Am J PathoI1983;79:98-104. 14. Giorlandino C, Gleicher N, Cinzia N, Vizzone A, Gentili P, Taramanni C. The sonographic picture of endometrium in spontaneous and induced cycles. Fertil SterilI987;47:508-11. 15. Lenz S, Lindenberg S. Ultrasonic evaluation of endometrial growth in women with normal cycles during spontaneous and stimulated cycles. Hum Reprod 1990;5:377-81. 16. Muckle CWo Clinical anatomy of the uterus, fallopian tubes, and ovaries. In: Sciarra JJ, editor. Obstetrics and gynecology. Philadelphia: Harper and Row, 1987:1. 17. Johannisson E, Landgren B-M, Rohr HP, Diczfalusy E. Endometrial morphology and peripheral hormone levels in women with regular menstrual cycle. Fertil Steril 1987;48: 401-8. 18. Clemetson CAB, Kim JK, de Jesus TPS, MallikaJjuneswara YR, Wilds JH. Human uterine fluid potassium and the menstrua! cycle. J Obstet Gynaecol Br Commonw 1973;80:553-61. 19. Harrison RA. Intravenous fluids and acid-base balance. In: Dripps RD, Eckenhoff JE, Vandam LD, editors. Introduction to anesthesia the principles of safe practice. 7th ed. Philadelphia: Saunders, 1988:261.
Fertility and Sterility