- Email: [email protected]
The tumor marker CA 125 is a common constituent of normal cervical mucus
The tumor marker CA 125 is a common constituent of normal cervical mucus Henk W. A. de Bruijn, Ph.D., Ton van Beeck Calkoen-Carpay, Siemen Jager, Ph.D., Jitze M. Duk, M.D., Jan G. Aalders, M.D., and Gert Jan Fleuren, M.D. Groningen and Leiden, The Netherlands The presence of the tumor marker CA 125 was studied in the cervices of healthy women. Immunohistochemical staining of normal cervical tissue demonstrated the presence of CA 125 in the tall columnar cells of the endocervical epithelium but not in the ectocervical squamous epithelium. We measured very high levels of CA 125 in liquefied cervical mucus from women with regular menstrual cycles. At midcycle, levels ranged from 14,200 to 153,000 U/ml (n = 13) in cervical mucus, while normal levels <35 U/ml were found in the corresponding serum samples. Levels of CA 125 in cervical mucus are comparable to the high levels found in cyst fluids from ovarian tumors (median 24,600 U/ml, n = 25). When secretion of cervical mucuswas stimulated by ethinyl estradiol, equally high levels were found (7900 to 138,000 U/ml, n = 10). We conclude that the tumor marker CA 125 is synthesized and secreted by normal endocervical cells. Apparently an effective barrier exists between the endocervical mucosa and the circulation. (AM J OSSTET GVNECOL 1986;154:1088-91.)
Key words: CA 125, cervical mucus, ovarian tumor marker CA 125 is an antigenic determinant produced by nonmucinous epithelial ovarian tumors and has been proposed as a tumor marker for patients with ovarian cancer. I. 2 Its identity is defined by the interaction with the murine monoclonal antibody OC 125. 3 Increased serum levels of CA 125 were found in 90% of patients with manifest epithelial ovarian cancer, and rising or falling serum levels were associated with tumor progression and regression, respectively. Some patients with adenocarcinoma ofthe cervix or the endometrium also showed raised levels of serum CA 125.1 It was proposed that CA 125 is a differentiation antigen, shared by fetal coelomic epithelium and its derivates in the adult, which is reexpressed in neoplastic ovarian epithelial cells." We found high levels of the antigen CA 125 in normal cervical mucus from ovulatory women and confirmed its presence in the columnar cells of the normal endocervix by immunohistochemical methods.
Methods Patients. Twenty-three women who were patients of the Fertility Department of the University Hospital, Groningen, were examined. Thirteen women with nor-
From the Department of Obstetricsand Gynaecology, University H ospital, Groningen, and the Department of Pathology, University of
Leiden. Received for publication October 23, 1985; accepted January 6, 1986. Reprint requests: Dr. H. W. A. de Bruijn, Laboratory for Obstetrics and Gynaecology, University Hospital, Oostersingel59, NL- 9713 EZ Groningen, The Netherlands.
1088
mal menstrual cycles, biphasic basal body temperature curves, and normal viscoelastic properties of the cervical mucus were selected. Their ages ranged from 20 to 41 years. The cervical mucus samples were obtained at midcycle before ovulation, The mucus samples from 10 other women were obtained in the second week of a stimulated pill cycle. These women were prescribed 50 /log of ethinyl estradiol per day from the third to the fifteenth day after beginning of the menses. The cervical mucus was aspirated from the endocervix with the use of a long tuberculin syringe without needle. All cervical mucus samples investigated showed signs of good estrogenic stimulation. The cervical mucus samples were liquefied by sonication for 1 minute by an MSE Sonificator with a small tip. Debris was removed by centrifugation at 2000 X g for 5 minutes. The supernatants were stored at - 20° C. On the day of the cervical mucus collection blood samples were drawn from 19 of the 23 women. Blood samples were also collected from 13 women with chronic cervicitis. These women had a longer history of mucopurulent discharge from the external cervical os or contact bleeding. The diagnosis was confirmed by microscopic examination of colposcopically directed punch biopsies and material from endocervical scraping. As a comparison CA 125 was measured in tumor cyst fluid samples and blood samples from 32 patients with epithelial ovarian tumors. These samples were collected before any treatment was started and they were centrifuged at 2000 X g for 5 minutes. The supernatants were stored at - 20° C.
CA 125 in normal cervical mucus 1089
Volume 154 Number 5
,"
Fig. 1. Light microscopic picture of a frozen section of the endocervix incubated with monoclonal antibody OC 125. An intense staining of the endocervical cells is present. (Indirect peroxidase. Original magnification x 250.) MeasurementofCA 125. Levels ofCA 125 in cervical mucus, cyst fluids, and serum were assayed with a simultaneous sandwich immunoradiometric assay as described by Bast et al. ' The reagents were obt ained fr om Centocor Inc. (Malvern, Pennsylvania ). The extremely high levels of CA 125 in undiluted samples of cervical mucus gave an excess of antigen that blocked the iodine 125-labeled antibody, resulting in a strongly diminished binding. Therefore, multiple dilutions always up to 111000 were assayed for CA 125. Dilutions were made with a standard serum diluent. I Control mea surements were done by incubation of cervical mucus samples in a two-step incubation IRMA , as described by Klug et al." In this procedure the excess of anti gen was removed by washing after incubation of the sample, and a separate incubation with the labeled antibody followed. All measurements were done in duplicate. Results were expressed as units per milliliter. A sensitivity of 9 U/ml was obtained. For seru m samples an upper limit of normal was chosen at 35 U/ml. ' A control serum sample was measured in 14 different assays and a variation coefficient of 8.5 % was observed. Immunohistochemistry. Endocervical tissue was obtained from autopsy or from surgical specimens of eight patients with benign or nongynecologic disea ses. Their ages ranged from 13 to 87 years. CA 125 was demonstrated in snap-frozen cervical tissue blocks by the indirect peroxidase-peroxidase method essentially the same as described previousl y.' In brief, the following incubations were used: (I) Incubation was performed for 60 minutes with monoclonal antibod y OC 125 (gift from Oris Lapam , France). (2) The sections were incubated with horseradish peroxidase-conju-
gated rabbit antimouse seru m (Dakopatt, Denmark) for 30 minutes. (3) Development of peroxidase activity was performed with a freshl y prepared solution of3-amino9-ethylcarba zole in 0.1 moIlL acetate buffer, pH 5.0, containing 0.03% hydrogen peroxide resulting in a red reacti on product. Between incubations sections were washed three times for 10 minutes in phosphate-buffered saline , pH 7.4. Finall y, sections were counterstained with hematoxylin for 2 minutes and mounted in Kaises glycerine gelatine (Merck , Germany). Results
Immunohistochemical staining of normal cervical tissue demonstrates the presence of CA 125 in the tall columnar cells of the endocervical epithelium (in all eight tissue samples investigated ). The staining was most intense at the luminal surfaces and the apical portion of the cytoplasm (Fig. I). The ectocervical squamous epithelium was completely negative. High levels of the antigen CA 125 were found in liquified cervical mu cus from women with regular menstrual cycles. At midc ycle, the levels ranged from 14,200 to 153,000 U/ml (Table I). Women who recei ved ethinyl estradiol produced a thin and clear cervical mucus with equally high levels (7900 to 138,000 U/ml). Control measurements in the two-step incubation procedure confirmed the presence of the high CA 125 levels in the mucus and an acceptable correlation was obtained (r = 0.869 , P < 0.001) . The level of CA 125 in cervical mucus was of the same order of magnitude as that found in cyst fluid samples derived from benign and malignant epithelial ovarian tumors (Table II).
1090 de Bruijn et al.
May, 1986
Am J Obstet Gynecol
Table I. Midcycle CA 125 levels in cervical mucus and serum from women with regular menstrual cycles
-,---.-No.
Natural cycle I
2 3 4 5 6 7 8 9 10 II 12 13 Mean
Cervical mucus (Ulml)
66,500 27,800 II 6,000 117,000 75,700 108,000 153,000 104,000 14,200 16,800 73,700 59,300 26,400 73,700 ± 44,200
18 31 30 23 21 21 20 -* 20 23
Table II. Median levels of CA 125
Cervical mucus from preovulatory women Cyst fluids from epithelial ovarian tumors Serum from patients with epithelial ovarian cancer before treatment Serum from healthy women Serum from patients with chronic cervicitis
~l
Median CA 125 (Ulml)
Range
23
65,500
25
24,600
32
424
11-4,480
19
<35
13-35
13
<35
10-69
7,900-153,000 770-2,140,000
Ethinyl estradiol stimulated
14 15 16 17 18 19 20 21 22 23 Mean
109,000 84,600 109,000 138,000 30,000 7,900 60,800 II,400 16,600 21,700 58,900 ± 48,100
26 22 24 23 22 35 30 26 13 23
*No serum sample available.
In serum, levels of CA 125 were below the limit of 35 Vlml in all 19 women investigated at the time of cervical mucus aspiration (Table I). Only one of 13 patients with chronic cervicitis demonstrated an elevated CA 125 level of 65 Vlml (Table II). In a control experiment we confirmed the presence of elevated CA 125 levels in preoperative serum samples from patients with epithelial ovarian cancer; in 30 of 32 patients levels >35 Vlml were seen.
Comment Tumor antigens are used as markers in clinical oncology mainly for immunohistochemical diagnosis and monitoring of the effect of therapy. These antigens often occur in normal cells and are, so far, not tumor specific. In the search for specific antigens from ovarian cancer a monoclonal antibody OC 125 was developed that reacted with a cell line from an ovarian cystadenocarcinorna." The undefined antigen was termed CA 125, and to date CA 125 has been characterized only as a high molecular weight glycoprotein. I Our study demonstrates the presence of CA 125 in cervical mucus and we conclude that CA 125 is a secretory product of normal cells and not a tumor-specific antigen. The pro-
duction site of the CA 125 is, at least in part, the columnar endocervical epithelium, since the presence of CA 125 in the epithelial cells could be demonstrated immunohistochemically. The columnar epithelium of the endocervical canal and its diverging crypts are known to produce and secrete glycoproteins. These cells show secretory granules and irregular vacuolization characteristic of the mucin cell type. They stain deeply red with the periodic acid-Schiff method." It cannot be excluded that some of the CA 125 in the cervical mucus originates from other parts of the female genital tract. Presence ofCA 125 in cervical mucus due to deposition by semen could be excluded since the antigen could not be detected in seminal plasma or on spermatozoa (unpublished observations). Despite the high levels in cervical mucus, only small amounts of CA 125 could be detected in serum from normal women. Apparently this is because an effective barrier exists between the endocervical mucosa and the circulation. We investigated serum CA 125 levels in patients with chronic cervicitis because in endocervical biopsy specimens subepithelial infiltration of lymphocytes and plasma cells is often present. This condition may enhance transfer of CA 125 into the circulation. One of the 13 patients studied showed a raised level. Niloff et al." were unable to find elevated serum levels in women with vaginitis or cervicitis. However, they used 65 Vlml as the upper limit of normal. Although CA 125 is not tumor specific, it may be quite useful for monitoring the response to therapy and early detection of tumor recurrence in patients with ovarian epithelial cancer. The specificity of raised serum levels of CA 125 as an indicator of malignancy remains to be assessed. The presence of large quantities of CA 125 in the normal female genital tract makes it worthwhile to study the functional properties of this antigen.
Volume 154 Number 5
CA 125 in normal cervical mucus
REFERENCES 1. Bast RC, Klug TL, John ES, et al. A radioimmunoassay using a monoclonal antibody to monitor the course of epithelial ovarian cancer. N Engl J Med 1983;309:883-7. 2. Channey PA, Moore M, Wilkinson PM, James RD. Ovarian cancer antigen CA 125: a prospective clinical assessment of its role as a tumour marker. Br J Cancer 1984;50: 765-9. 3. Bast RC, Feeny 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. 4. Niloff JM, Klug TL, Schaetzl E, Zurawski VR, Knapp RC, Bast RC. Elevation of serum CA 125 in carcinomas of the fallopian tube, endometrium, and endocervix. AM J OBSTET CYNECOL 1984; 148: 1057-8. 5. Kabawat SE, Bast RC, Bhan AK, Welch WR, Knapp RC, Colvin RB. Tissue distribution of a coelomic-epithelium
6.
7.
8.
9.
related antigen recognized by the monoclonal antibody OC 125. Int J Cynecol Pathol 1983;2:275-85. Klug TL, Bast RC, Niloff JM, Knapp RC, Zurawski VR. Monoclonal antibody immunoradiometric assay for an antigenic determinant (CA 125) associated with human epithelial ovarian carcinomas. Cancer Res 1984;44:1048-53. Nap M, Ten Hoor KA, Fleuren CJ. Crossreactivity with normal antigens in commercial anti-CEA sera, used for immunohistology. The need for tissue controls and absorptions. AmJ Clin PathoI1983;79:25-31. Friedrich ER. The normal morphology and ultrastructure of the cervix. In: Blandau RJ, Moghissi K, eds. The biology of the cervix. Chicago: The University of Chicago Press, 1973:79-102. Niloff JM, Knapp RC, Schaetzl E, Reynolds C, Bast RC. CA 125 antigen levels in obstetric and gynecologic patients. Obstet Cynecol 1984;64:703-7.
Human chorionic gonadotropin in early normal and pathologic pregnancy Discordant levels in peripheral maternal blood and blood from the uterine and abdominal cavities Johan Amt Steier, M.D., Roar Sandvei, M.D., and Ole L. Myking, M.D. Bergen, Norway Human chorionic gonadotropin was assayed in 25 cases after first-trimester induced abortion, in 45 cases of spontaneous abortion in the first trimester, and in 27 cases of ectopic pregnancy. Blood was obtained from an antecubital vein and from the uterine cavity. In the cases of ectopic pregnancy blood was also obtained from the abdominal cavity. In the group of induced abortion the human chorionic gonadotropin levels in peripheral maternal blood did not differ significantly from the levels in the uterine cavity. In the groups of spontaneous abortion and ectopic pregnancy the human chorionic gonadotropin levels were significantly higher in blood from the uterine cavity and the abdominal cavity, respectively. In four cases (three with spontaneous abortion and one with an ectopic pregnancy) human chorionic gonadotropin was not detectable in peripheral maternal blood, while it was found in blood from the uterine and abdominal cavities. (AMJ OBSTETGYNECOL 1986;154:1091-4.)
Key words: Human chorionic gonadotropin, ectopic pregnancy, spontaneous abortion, induced abortion Human chorionic gonadotropin (hCG) is produced in large quantities by the syncytiotrophoblast. By use of sensitive radioimmunoassays it is detectable in peripheral maternal serum as early as 6 to 8 days after conception. 1 In early normal pregnancy the hCG levels
From the Department of Obstetrics and Gynecologyand the Hormone Laboratory, University of Bergen. Received for publication November 5, 1985; acceptedJanuary 22,
1986. Reprint requests: Dr. A. Steier, Department of Obstetrics and Gynecology, University of Bergen, N-5016 Haukeland Hospital, Norway.
rise rapidly, with a doubling time of 1.7 to 2 days, reaching a peak after 60 to 80 days." In pathologic conditions such as ectopic pregnancy and spontaneous abortion the levels are found to be lower than in normal gestation of the same duration." Some pathologic cases are on record in which no trace of hCG in peripheral blood was found, although histologic examination of the surgical specimens confirmed pregnancy. Devitalized trophoblast with no circulating hCG has been suggested as the most likely explanation" We have observed pathologic pregnancies in which blood from the uterine cavity and blood from the peritoneal cavity have con-
1091
Key words: CA 125, cervical mucus, ovarian tumor marker CA 125 is an antigenic determinant produced by nonmucinous epithelial ovarian tumors and has been proposed as a tumor marker for patients with ovarian cancer. I. 2 Its identity is defined by the interaction with the murine monoclonal antibody OC 125. 3 Increased serum levels of CA 125 were found in 90% of patients with manifest epithelial ovarian cancer, and rising or falling serum levels were associated with tumor progression and regression, respectively. Some patients with adenocarcinoma ofthe cervix or the endometrium also showed raised levels of serum CA 125.1 It was proposed that CA 125 is a differentiation antigen, shared by fetal coelomic epithelium and its derivates in the adult, which is reexpressed in neoplastic ovarian epithelial cells." We found high levels of the antigen CA 125 in normal cervical mucus from ovulatory women and confirmed its presence in the columnar cells of the normal endocervix by immunohistochemical methods.
Methods Patients. Twenty-three women who were patients of the Fertility Department of the University Hospital, Groningen, were examined. Thirteen women with nor-
From the Department of Obstetricsand Gynaecology, University H ospital, Groningen, and the Department of Pathology, University of
Leiden. Received for publication October 23, 1985; accepted January 6, 1986. Reprint requests: Dr. H. W. A. de Bruijn, Laboratory for Obstetrics and Gynaecology, University Hospital, Oostersingel59, NL- 9713 EZ Groningen, The Netherlands.
1088
mal menstrual cycles, biphasic basal body temperature curves, and normal viscoelastic properties of the cervical mucus were selected. Their ages ranged from 20 to 41 years. The cervical mucus samples were obtained at midcycle before ovulation, The mucus samples from 10 other women were obtained in the second week of a stimulated pill cycle. These women were prescribed 50 /log of ethinyl estradiol per day from the third to the fifteenth day after beginning of the menses. The cervical mucus was aspirated from the endocervix with the use of a long tuberculin syringe without needle. All cervical mucus samples investigated showed signs of good estrogenic stimulation. The cervical mucus samples were liquefied by sonication for 1 minute by an MSE Sonificator with a small tip. Debris was removed by centrifugation at 2000 X g for 5 minutes. The supernatants were stored at - 20° C. On the day of the cervical mucus collection blood samples were drawn from 19 of the 23 women. Blood samples were also collected from 13 women with chronic cervicitis. These women had a longer history of mucopurulent discharge from the external cervical os or contact bleeding. The diagnosis was confirmed by microscopic examination of colposcopically directed punch biopsies and material from endocervical scraping. As a comparison CA 125 was measured in tumor cyst fluid samples and blood samples from 32 patients with epithelial ovarian tumors. These samples were collected before any treatment was started and they were centrifuged at 2000 X g for 5 minutes. The supernatants were stored at - 20° C.
CA 125 in normal cervical mucus 1089
Volume 154 Number 5
,"
Fig. 1. Light microscopic picture of a frozen section of the endocervix incubated with monoclonal antibody OC 125. An intense staining of the endocervical cells is present. (Indirect peroxidase. Original magnification x 250.) MeasurementofCA 125. Levels ofCA 125 in cervical mucus, cyst fluids, and serum were assayed with a simultaneous sandwich immunoradiometric assay as described by Bast et al. ' The reagents were obt ained fr om Centocor Inc. (Malvern, Pennsylvania ). The extremely high levels of CA 125 in undiluted samples of cervical mucus gave an excess of antigen that blocked the iodine 125-labeled antibody, resulting in a strongly diminished binding. Therefore, multiple dilutions always up to 111000 were assayed for CA 125. Dilutions were made with a standard serum diluent. I Control mea surements were done by incubation of cervical mucus samples in a two-step incubation IRMA , as described by Klug et al." In this procedure the excess of anti gen was removed by washing after incubation of the sample, and a separate incubation with the labeled antibody followed. All measurements were done in duplicate. Results were expressed as units per milliliter. A sensitivity of 9 U/ml was obtained. For seru m samples an upper limit of normal was chosen at 35 U/ml. ' A control serum sample was measured in 14 different assays and a variation coefficient of 8.5 % was observed. Immunohistochemistry. Endocervical tissue was obtained from autopsy or from surgical specimens of eight patients with benign or nongynecologic disea ses. Their ages ranged from 13 to 87 years. CA 125 was demonstrated in snap-frozen cervical tissue blocks by the indirect peroxidase-peroxidase method essentially the same as described previousl y.' In brief, the following incubations were used: (I) Incubation was performed for 60 minutes with monoclonal antibod y OC 125 (gift from Oris Lapam , France). (2) The sections were incubated with horseradish peroxidase-conju-
gated rabbit antimouse seru m (Dakopatt, Denmark) for 30 minutes. (3) Development of peroxidase activity was performed with a freshl y prepared solution of3-amino9-ethylcarba zole in 0.1 moIlL acetate buffer, pH 5.0, containing 0.03% hydrogen peroxide resulting in a red reacti on product. Between incubations sections were washed three times for 10 minutes in phosphate-buffered saline , pH 7.4. Finall y, sections were counterstained with hematoxylin for 2 minutes and mounted in Kaises glycerine gelatine (Merck , Germany). Results
Immunohistochemical staining of normal cervical tissue demonstrates the presence of CA 125 in the tall columnar cells of the endocervical epithelium (in all eight tissue samples investigated ). The staining was most intense at the luminal surfaces and the apical portion of the cytoplasm (Fig. I). The ectocervical squamous epithelium was completely negative. High levels of the antigen CA 125 were found in liquified cervical mu cus from women with regular menstrual cycles. At midc ycle, the levels ranged from 14,200 to 153,000 U/ml (Table I). Women who recei ved ethinyl estradiol produced a thin and clear cervical mucus with equally high levels (7900 to 138,000 U/ml). Control measurements in the two-step incubation procedure confirmed the presence of the high CA 125 levels in the mucus and an acceptable correlation was obtained (r = 0.869 , P < 0.001) . The level of CA 125 in cervical mucus was of the same order of magnitude as that found in cyst fluid samples derived from benign and malignant epithelial ovarian tumors (Table II).
1090 de Bruijn et al.
May, 1986
Am J Obstet Gynecol
Table I. Midcycle CA 125 levels in cervical mucus and serum from women with regular menstrual cycles
-,---.-No.
Natural cycle I
2 3 4 5 6 7 8 9 10 II 12 13 Mean
Cervical mucus (Ulml)
66,500 27,800 II 6,000 117,000 75,700 108,000 153,000 104,000 14,200 16,800 73,700 59,300 26,400 73,700 ± 44,200
18 31 30 23 21 21 20 -* 20 23
Table II. Median levels of CA 125
Cervical mucus from preovulatory women Cyst fluids from epithelial ovarian tumors Serum from patients with epithelial ovarian cancer before treatment Serum from healthy women Serum from patients with chronic cervicitis
~l
Median CA 125 (Ulml)
Range
23
65,500
25
24,600
32
424
11-4,480
19
<35
13-35
13
<35
10-69
7,900-153,000 770-2,140,000
Ethinyl estradiol stimulated
14 15 16 17 18 19 20 21 22 23 Mean
109,000 84,600 109,000 138,000 30,000 7,900 60,800 II,400 16,600 21,700 58,900 ± 48,100
26 22 24 23 22 35 30 26 13 23
*No serum sample available.
In serum, levels of CA 125 were below the limit of 35 Vlml in all 19 women investigated at the time of cervical mucus aspiration (Table I). Only one of 13 patients with chronic cervicitis demonstrated an elevated CA 125 level of 65 Vlml (Table II). In a control experiment we confirmed the presence of elevated CA 125 levels in preoperative serum samples from patients with epithelial ovarian cancer; in 30 of 32 patients levels >35 Vlml were seen.
Comment Tumor antigens are used as markers in clinical oncology mainly for immunohistochemical diagnosis and monitoring of the effect of therapy. These antigens often occur in normal cells and are, so far, not tumor specific. In the search for specific antigens from ovarian cancer a monoclonal antibody OC 125 was developed that reacted with a cell line from an ovarian cystadenocarcinorna." The undefined antigen was termed CA 125, and to date CA 125 has been characterized only as a high molecular weight glycoprotein. I Our study demonstrates the presence of CA 125 in cervical mucus and we conclude that CA 125 is a secretory product of normal cells and not a tumor-specific antigen. The pro-
duction site of the CA 125 is, at least in part, the columnar endocervical epithelium, since the presence of CA 125 in the epithelial cells could be demonstrated immunohistochemically. The columnar epithelium of the endocervical canal and its diverging crypts are known to produce and secrete glycoproteins. These cells show secretory granules and irregular vacuolization characteristic of the mucin cell type. They stain deeply red with the periodic acid-Schiff method." It cannot be excluded that some of the CA 125 in the cervical mucus originates from other parts of the female genital tract. Presence ofCA 125 in cervical mucus due to deposition by semen could be excluded since the antigen could not be detected in seminal plasma or on spermatozoa (unpublished observations). Despite the high levels in cervical mucus, only small amounts of CA 125 could be detected in serum from normal women. Apparently this is because an effective barrier exists between the endocervical mucosa and the circulation. We investigated serum CA 125 levels in patients with chronic cervicitis because in endocervical biopsy specimens subepithelial infiltration of lymphocytes and plasma cells is often present. This condition may enhance transfer of CA 125 into the circulation. One of the 13 patients studied showed a raised level. Niloff et al." were unable to find elevated serum levels in women with vaginitis or cervicitis. However, they used 65 Vlml as the upper limit of normal. Although CA 125 is not tumor specific, it may be quite useful for monitoring the response to therapy and early detection of tumor recurrence in patients with ovarian epithelial cancer. The specificity of raised serum levels of CA 125 as an indicator of malignancy remains to be assessed. The presence of large quantities of CA 125 in the normal female genital tract makes it worthwhile to study the functional properties of this antigen.
Volume 154 Number 5
CA 125 in normal cervical mucus
REFERENCES 1. Bast RC, Klug TL, John ES, et al. A radioimmunoassay using a monoclonal antibody to monitor the course of epithelial ovarian cancer. N Engl J Med 1983;309:883-7. 2. Channey PA, Moore M, Wilkinson PM, James RD. Ovarian cancer antigen CA 125: a prospective clinical assessment of its role as a tumour marker. Br J Cancer 1984;50: 765-9. 3. Bast RC, Feeny 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. 4. Niloff JM, Klug TL, Schaetzl E, Zurawski VR, Knapp RC, Bast RC. Elevation of serum CA 125 in carcinomas of the fallopian tube, endometrium, and endocervix. AM J OBSTET CYNECOL 1984; 148: 1057-8. 5. Kabawat SE, Bast RC, Bhan AK, Welch WR, Knapp RC, Colvin RB. Tissue distribution of a coelomic-epithelium
6.
7.
8.
9.
related antigen recognized by the monoclonal antibody OC 125. Int J Cynecol Pathol 1983;2:275-85. Klug TL, Bast RC, Niloff JM, Knapp RC, Zurawski VR. Monoclonal antibody immunoradiometric assay for an antigenic determinant (CA 125) associated with human epithelial ovarian carcinomas. Cancer Res 1984;44:1048-53. Nap M, Ten Hoor KA, Fleuren CJ. Crossreactivity with normal antigens in commercial anti-CEA sera, used for immunohistology. The need for tissue controls and absorptions. AmJ Clin PathoI1983;79:25-31. Friedrich ER. The normal morphology and ultrastructure of the cervix. In: Blandau RJ, Moghissi K, eds. The biology of the cervix. Chicago: The University of Chicago Press, 1973:79-102. Niloff JM, Knapp RC, Schaetzl E, Reynolds C, Bast RC. CA 125 antigen levels in obstetric and gynecologic patients. Obstet Cynecol 1984;64:703-7.
Human chorionic gonadotropin in early normal and pathologic pregnancy Discordant levels in peripheral maternal blood and blood from the uterine and abdominal cavities Johan Amt Steier, M.D., Roar Sandvei, M.D., and Ole L. Myking, M.D. Bergen, Norway Human chorionic gonadotropin was assayed in 25 cases after first-trimester induced abortion, in 45 cases of spontaneous abortion in the first trimester, and in 27 cases of ectopic pregnancy. Blood was obtained from an antecubital vein and from the uterine cavity. In the cases of ectopic pregnancy blood was also obtained from the abdominal cavity. In the group of induced abortion the human chorionic gonadotropin levels in peripheral maternal blood did not differ significantly from the levels in the uterine cavity. In the groups of spontaneous abortion and ectopic pregnancy the human chorionic gonadotropin levels were significantly higher in blood from the uterine cavity and the abdominal cavity, respectively. In four cases (three with spontaneous abortion and one with an ectopic pregnancy) human chorionic gonadotropin was not detectable in peripheral maternal blood, while it was found in blood from the uterine and abdominal cavities. (AMJ OBSTETGYNECOL 1986;154:1091-4.)
Key words: Human chorionic gonadotropin, ectopic pregnancy, spontaneous abortion, induced abortion Human chorionic gonadotropin (hCG) is produced in large quantities by the syncytiotrophoblast. By use of sensitive radioimmunoassays it is detectable in peripheral maternal serum as early as 6 to 8 days after conception. 1 In early normal pregnancy the hCG levels
From the Department of Obstetrics and Gynecologyand the Hormone Laboratory, University of Bergen. Received for publication November 5, 1985; acceptedJanuary 22,
1986. Reprint requests: Dr. A. Steier, Department of Obstetrics and Gynecology, University of Bergen, N-5016 Haukeland Hospital, Norway.
rise rapidly, with a doubling time of 1.7 to 2 days, reaching a peak after 60 to 80 days." In pathologic conditions such as ectopic pregnancy and spontaneous abortion the levels are found to be lower than in normal gestation of the same duration." Some pathologic cases are on record in which no trace of hCG in peripheral blood was found, although histologic examination of the surgical specimens confirmed pregnancy. Devitalized trophoblast with no circulating hCG has been suggested as the most likely explanation" We have observed pathologic pregnancies in which blood from the uterine cavity and blood from the peritoneal cavity have con-
1091