Tumor Markers in Ovarian Malignancies

Tumor Markers in Ovarian Malignancies KAREN EAGLE, JONATHAN A. LEDERMANN Department of Oncology, University College London Medical School, London, United Kingdom Key Words. Ovarian cancer Tumor markers CA125 ABSTRACT Epithelial ovarian cancer is the most common ovarian malignancy. CA125, the glycoprotein defined by the antibody OC 125, is the most important clinical marker for the diagnosis, treatment and follow-up of epithelial ovarian cancer. However, like most tumor markers, it is neither wholly specific nor sensitive for the disease. We discuss how CA125 in combination with other tests can be used in the differential diagnosis of pelvic masses and as part of the investigations for cancer screening. CA125 is an important indicator of response to treatment, guiding therapeutic decisions and identifying those patients whose response to INTRODUCTION Malignant epithelial ovarian tumors account for 90% of all malignancies of the ovary and are the fourth most common cause of tumor-related death in women. The empirical lifetime risk of developing ovarian cancer is 1:70 [1], and most women present with advanced disease (FIGO stage III or stage IV), which is rarely curable. Tumor-associated antigens released into the circulation have been described in many diseases. Ideally, a tumor marker should be able to detect subclinical disease (i.e., screening), useful in monitoring the response to treatment, and to identify early recurrence so that further treatment can be instituted. Furthermore, the release of circulating tumor antigen provides an identifiable surface target on the tumor cell that might be used for in vivo diagnosis or antigendirected therapy. No serum tumor marker, with the possible exception of human β chorionic gonadotrophin, meets all these criteria. Nevertheless, measurement of many serum tumor markers has been incorporated into clinical practice. This review will focus on CA125, the most clinically applicable tumor marker for ovarian cancer, and will briefly describe other tumor markers and possible future applications of tumor markers. Bast and colleagues in 1981 first described CA125, a 200 kd glycoprotein recognized by the murine monoclonal antibody OC 125 as a marker for epithelial malignancies [2]. chemotherapy and survival is short. CA125 has recently been shown to correlate well with response and can be used to define relapse. Thus, it can be used as a surrogate endpoint in the assessment of new therapeutic modalities as well as in reducing the need for tumor imaging. At the moment, the other tumor markers for non-germ-cell neoplasms of the ovary are clinically less important than CA125, but their role alone or in association with CA125 is the subject of intense study as the search for ideal tumor markers to identify early disease, prognosis, and relapse continues. The Oncologist 1997;2:324-329 A raised level of antigen was detectable in the serum of 82% of women with epithelial ovarian cancer but in only 1% of healthy blood donors [3]. Epithelial ovarian cancers with low or normal levels of CA125 are usually mucinous tumors. The antigen is not specific to ovarian cancer as raised serum levels may also be found in 29% of other cancers (lung, breast, pancreas, and colorectum) and in 6% of women with nonmalignant conditions such as cirrhosis with ascites, acute pancreatitis, ovarian cysts, endometriosis, and pelvic inflammatory disease. Screening and Diagnosis As early-stage ovarian cancer carries a much more favorable prognosis, there is an urgent need to identify subclinical disease. A satisfactory method of screening subclinical disease for ovarian cancer is needed (Table 1). Serological markers are theoretically an ideal approach but none have 100% specificity and sensitivity. In a retrospective study of stored sera from the JANUS serum bank in Norway, raised CA125 levels were found in half of the samples collected within 18 months of the diagnosis of ovarian cancer, and approximately 25% of samples had raised serum CA125 levels within 60 months preceding the diagnosis of ovarian cancer [4]. In a review of the literature, Jacobs and Bast [5] found that about 50% of patients with stage I disease had elevated levels of CA125. However, even in the Correspondence: Jonathan A. Ledermann, M.D., Department of Oncology, University College London Medical School, 91 Riding House Street, London, W1P 8BT, United Kingdom. Telephone: +44-171-380-9430; Fax: +44-171-436-2956; e-mail: j.ledermann@ucl.ac.uk Accepted for publication July 30, 1997. AlphaMed Press 1083-7159/97/$5.00/0 The Oncologist 1997;2:324-329

325 Tumor Markers in Ovarian Malignancies Table 1. Screening for ovarian cancer A. Identify preclinical/early stage I disease Prognosis similar to clinical stage I disease? B. High prevalence Who should be screened? age determined? high-risk groups? C. Validity High sensitivity and specificity? Positive predictive value 10% (i.e., 1 out of 10 diagnostic laparotomies positive)? D. Screening test Simple, acceptable, inexpensive? presence of an ultrasonographically defined mass, serum CA125 cannot reliably distinguish between a malignant or benign mass. CA125 levels were raised above 35 U/ml in 78% of women with malignant masses, but also in 22% of those with benign masses [6]. The predictive value of CA125 measurement in postmenopausal women is a little greater, and using a cut-off of 65 U/ml, the false positive rate was about 8% [7]. Population screening with ultrasonography alone has not proved to be a cost-effective means of detecting ovarian cancer. However, the sensitivity and specificity of this investigation can be increased by transvaginal ultrasonography and transvaginal color doppler imaging. Serum CA125 measurement in healthy women has been used as a means of selecting women for ultrasonography. This increases the specificity of examination, but the predictive value of screening is about 10% [8]. At this level, a significant number of surgical explorations would be performed for nonmalignant ovarian pathology. Furthermore, it is not clear how often patients should have examinations repeated, as some of the patients with normal CA125 at screening subsequently developed ovarian cancer on follow-up [9]. Currently, the combination of CA125 and transvaginal color doppler studies is likely to be the most successful screening tool, particularly if applied to women with a strong family history of ovarian cancer, as they have a higher risk of developing the disease. The role of screening in the general population is unclear but is now being tested in a large randomized trial in the UK using the criteria illustrated in Table 2. The inclusion of other tumor markers may further increase the specificity of screening. Einhorn et al. evaluated CA125 concentrations together with those of CA15-3 and TAG-72 in 219 patients undergoing diagnostic laparotomy for pelvic masses. They found that the three tumor markers increased the specificity for detecting ovarian cancer but reduced the sensitivity of the CA125 assay [10]. Similar observations were made by Soper et al., who showed that by combining CA15-3 and TAG- 72 measurements with CA125, the specificity for detecting ovarian cancer rose from 83% to 98%, but the sensitivity was reduced from 88% to 73% [11]. Several other serological markers have been described in association with ovarian cancer. None have the same level of sensitivity and specificity as CA125. As some have a higher specificity, they may be used in combination with CA125 although the sensitivity would be sacrificed. In stage I disease where serum CA125 has a low sensitivity, there may be scope to increase the sensitivity of screening by using a combination of tumor markers. Preliminary reports with CA125, OVX1, and M-CSF are encouraging; 98% of sera from 46 patients with stage I disease had elevated levels of one of these serological markers [12]. A D-dimer of CA125 has recently been analyzed in 56 patients with epithelial ovarian cancer and 65 women with benign ovarian disease. All women in whom the CA125 level was >65 U/ml and the D-dimer level was >416 ng/ml had ovarian cancer (specificity and positive predictive value of 100% with a sensitivity of 73%) [13]. Monitoring Response to Therapy The use of tumor markers to monitor response to treatment is particularly helpful in ovarian cancer where there is often a lack of clinically or radiologically measurable disease. A reduction in the serum CA125 level correlates well with clinical response. Failure of CA125 to fall with chemotherapy indicates drug resistance and identifies a need to change treatment (Fig. 1). Formal CA125 response criteria have been formulated (Table 3) [14]. These parameters were tested on 629 of 989 patients considered assessable for response according to serum CA125 levels in three trials (North Thames Ovary Trials 3 and 4, Gynecologic Oncology Group [GOG] protocol #97). The tumor response rate was 66% in patients assessable by CA125 in GOG #97 and 62% by the GOG-defined response rate. The response rate in patients where CA125 was not assessable was 67%. The specificity was very high and the sensitivity was 68% [14]. It should be noted that removal of ascites will interfere with the serum CA125 level. Relapse It has been accepted for a long time that a rise in CA125 into the abnormal range is highly predictive for relapse [3, 15]. However, the lead time to clinical relapse is variable and a clearer definition of relapse is needed if CA125 measurement is to be used as a definition of clinical progression, particularly in the context of a clinical trial. In a recent analysis of 255 patients from a North Thames Ovary Trial, progressive disease was defined as a doubling of the CA125 from the upper limit of normal (30 U/ml cut-off in this study). The sensitivity was 85.9% and the specificity was 91.3%, giving a positive predictive value of 94.8%. If a confirmatory elevated CA125 was obtained, the false positive rate fell to less than 2%, with only a small fall in the sensitivity. The median lead time to clinical progression was 63 days [16].

Eagle, Ledermann 326 CA125 kµ/l 100,000 10,000 1,000 100 10 carboplatin 1 2 CA125 Response to chemotherapy 3 4 5 Taxol ECF Figure 1. The patient relapsed four times between 1992 and 1996, and on each occasion except the last responded to carboplatin. Eventually, carboplatin resistance developed and a change to paclitaxel failed to produce a marker or clinical response. She developed intestinal obstruction and was treated with intermittent epirubicin and cisplatin and continuous infusional 5-fluorouracil (ECF). She achieved a good clinical and CA125 response. 12/24/92 10/20/93 3/15/94 8/3/94 Table 2. Screening strategy Serum CA125 + OVX1 Ultrasound (transabdominal and transvaginal) Transvaginal color doppler imaging Laparotomy 2/14/95 Date Measurements of CA125 are frequently taken after the completion of chemotherapy, outside clinical trials. From the results of Rustin et al. the predictive power of CA125 followup certainly reduces the need for regular abdominopelvic scans [16]. However, while normal levels of CA125 are reassuring for the patient and her doctor, their measurement often evokes a period of anxiety. Furthermore, there are no clear guidelines to follow during the period between CA125 relapse and clinical progression of disease. We do not know whether the institution of second-line therapy at the time of CA125 (subclinical) relapse is preferable to waiting until clinical relapse has occurred. The Medical Research Council (MRC) Gynaecological Cancer Working Party and the European Organisation for Research and Treatment of Cancer (EORTC) Gynaecological Cancer Cooperative Group have recently started randomized trials to determine whether there is any benefit in survival and quality of life from the early introduction of chemotherapy based on CA125 relapse. Although these trials address important scientific and health-economic issues, recruitment could be slow as clinicians and patients may feel uncomfortable about not knowing the results of tests that have been taken. 9/12/95 5/21/96 10/8/96 2/17/97 Table 3. CA125-determined response 50% Response 50% decrease in level after two samples Confirmed by fourth sample 75% Response Serial reduction of CA125 level by 75% over three consecutive samples In approximately 20% of patients, serum CA125 levels are not elevated. The majority of these patients have mucinous tumors. In such cases, other tumor markers, such as carcinoembryonic antigen or TAG-72, which are more often elevated in mucinous tumors, may be used. The clinician will need to depend more on CT imaging in cases where serological markers are not detectable. Prognostic Marker It would be helpful if reliable prognostic indicators for survival could be determined before treatment. This is particularly true for stage I disease where there is doubt about the need for adjuvant therapy. For more advanced tumors, the value of prognostic markers is less clear as there are fewer options for therapy at the moment. Preoperative CA125 does not appear to be an independent risk factor for survival, although a multivariate analysis of 201 patients with stage I disease concluded that the preoperative CA125 level was the most powerful prognostic factor for survival [17]. In addition, it does reflect a larger tumor burden and more advanced stage [18]. Serum CA125 levels may rise shortly after surgery and even in stage

327 Tumor Markers in Ovarian Malignancies I disease may not return to normal until four weeks after laparotomy. A raised level of CA125 after this time has prognostic value as the patient has persistent disease and does not have a true stage I tumor. In a collaborative study from the Gynaecological Tumour Marker Group in Germany, it was reported that the postoperative value of CA125 was a predictor for survival. All patients with less than 2 cm residual disease who had a CA125 >65 U/ml died within 42 months, whereas 48% of those whose level was less than this were alive at six years [19]. The rate of fall of CA125 in response to chemotherapy, particularly during the early period of treatment, does provide prognostic information. Mogensen reported that if the titer of CA125 was 10 U/ml after three cycles of treatment, the median survival was 60 months. If it was 100 U/ml, the median survival was seven months [20]. Similar findings have been reported by others [21-24]. The MRC Working Party on Gynaecological Cancer analyzed 248 patients from 11 centers and found that the absolute value of the third serum CA125 level was the most important factor for predicting progressive disease at 12 months, with the addition of residual bulk disease at the end of initial surgery slightly improving the predictive power; however, there was still a false positive rate of 20% [25]. Second-look procedures are now uncommonly performed, but the prognosis is significantly worse in patients with raised CA125 levels at the time of surgery. CA125 is generally considered to be an insensitive predictor of laparotomy findings [26], although others have suggested that it is an independent prognostic factor [27]. It has been suggested that the sensitivity of predicting disease can be increased by combining the measurement of CA125 with the OVX1 assay, which detects an epitope on a high molecular weight mucin molecule. The prediction of a positive second-look laparotomy could be increased from 35% with CA125 alone to 56% [28]. Other Tumor Markers The use of other tumor markers in epithelial ovarian cancer has been briefly discussed above and listed in Table 4. None so far are being used as frequently as CA125 measurement in clinical practice. Levels of sialyl Tn, an antigen of the core region of mucin oligosaccharide, are raised in the preoperative serum of almost half the patients with ovarian cancer, and raised levels are an adverse prognostic determinant [29]. Various cytokines, including M-CSF, GM-CSF, IL-1, IL-6, and TNF-α are produced by ovarian cancer cells. Both M-CSF and its receptor fms can be expressed by ovarian cancer cells, and the levels of fms have been demonstrated to correlate with both advanced histological grade and clinical stage and are therefore associated with a poor clinical outcome [30, 31]. IL-6 is produced by ovarian cancer cells and can be isolated from the ascitic fluid and serum Table 4. Serological tumor markers for ovarian cancer CA125 Carcinoembryonic antigen CA15-3 TAG-72 HMFG2 (Human milk fat globule) PLAP (Placental alkaline phosphatase) NB70/K Urinary gonadotrophin core fragment peptide OVX1 CA19-9 LASA (Lipid-associated sialic acid) Tissue peptide antigen Table 5. Tumor markers in nonepithelial ovarian cancer Tumor marker α-fetoprotein (AFP) human β-chorionic gonadotrophin (β-hcg) Placental alkaline phosphatase (PALP) Lactate dehydrogenase (LDH) Estradiol Inhibin Tumor type Germ cell tumors Dysgerminomas Stromal tumors (including granulosa cell tumors) Granulosa cell tumors of patients. In one study, there was a correlation between an elevation of the serum level of IL-6 and disease extent with a raised IL-6 in 76% of patients with macroscopic disease, but only 13% in those with microscopic disease and 17% in healthy controls [32]. The IL-6 produced by such tumor cells, however, is not distinguishable from that produced by cells of the immune system, making it a nonspecific marker [33]. Furthermore, cytokine production may not arise from tumor cells. IL-10, a cytokine with various immunoinhibitory functions, was raised in ascites from nearly all women with ovarian cancer [34], but it does not appear to be produced by tumor cells. Further Applications of Tumor Markers The antigens recognized by monoclonal antibodies are found on tumor cells and provide an opportunity to target antibodies to tumors in vivo, either for diagnosis or therapy. Diagnostic radioimmunolocalization studies have been performed with antibodies to CA125, TAG-72, and HMFG2 antigens and the monoclonal antibody 791T/36 [35-39]. Tumors, both primary or recurrent, can be imaged preoperatively, but this technique is still in its infancy. However, radiolabeled B72.3 antibody (Oncoscint) has a product license and is the

Eagle, Ledermann 328 antibody most commonly used in clinical practice. It is now possible to produce smaller molecularly engineered antibody fragments with high affinity for their antigen. Engineered molecules are likely to increase significantly the sensitivity and specificity of radioimmunodetection. Radioimmunotargeting has also been applied to ovarian cancer. Preliminary results of intraperitoneal radiolabeled antibody therapy, particularly in the adjuvant setting, are encouraging [40, 41]. Whether these radiolabeled antibodies produce their effect as an adjuvant by radiation or alteration of the host s immunity through the idiotypic network is unclear. There is now a randomized trial in the UK comparing intraperitoneal radiolabeled HMFG2 antibody to no treatment after negative second-look laparotomy. Nonepithelial Ovarian Cancer A detailed discussion of tumor markers in ovarian germ cell tumors is beyond the scope of this review and is outlined in Table 5. α-fetoprotein and human β chorionic gonadotrophin are probably the best known tumor markers in clinical practice and are invaluable in the diagnosis, treatment, and follow-up of ovarian germ cell tumors. Serum placental alkaline phosphatase and lactate dehydrogenase are also sometimes helpful as markers of dysgerminoma. Stromal tumors comprise approximately 10% of ovarian cancer. Traditionally, stromal tumors produce estradiol, and this has been used as a biochemical tumor marker. Granulosa cell tumors a subgroup of stromal tumors, causing approximately 2% of ovarian malignancies have been demonstrated to produce both estradiol and inhibin [42]. REFERENCES 1 Herbst AL. The epidemiology of ovarian carcinoma and the current status of tumour markers to detect disease. (Review). Am J Obstet Gynecol 1994;170:1099-1107. 2 Bast RC, Feeney M, Lazarus H et al. Reactivity of a monoclonal antibody with a human ovarian carcinoma. J Clin Invest 1981;68:1311-1337. 3 Bast RC, Klug TL, St. John E et al. A radioimmunoassay using a monoclonal antibody to monitor the course of epithelial ovarian cancer. N Engl J Med 1983;309:883-887. 4 Zurawski R, Orjaster H, Andersen A et al. Elevated serum CA125 levels prior to diagnosis of ovarian neoplasia: relevance for early detection of ovarian cancer. Int J Cancer 1988;42:677-680. 5 Jacobs I, Bast RC. The CA125 tumour-associated antigen: a review of the literature. Hum Reprod 1989;4:1-12. 6 Vasilev SA, Schaerth JB, Campeau J et al. Serum CA125 levels in preoperative evaluation of pelvic masses. Obstet Gynecol 1988;71:751-756. 7 Malkasian GD, Knapp R, Lavin P et al. Preoperative evaluation of serum CA 125 levels in premenopausal and postmenopausal Approximately 30% of granulosa cell tumors and most extraovarian recurrences do not produce estradiol. Inhibin is a polypeptide hormone produced by the granulosa cells of the ovary and inhibits follicle-stimulating hormone secretion by the anterior pituitary gland. It is a glycoprotein consisting of two subunits β and α. Measurement of inhibin has been restricted by a lack of sensitivity and cross-reactivity with the active dimeric form and inactive α subunits. A new radioimmunoassay has now been developed which recognizes both subunits of inhibin and is more sensitive. Serum levels of inhibin have been demonstrated to correlate closely with clinical disease and, like CA125, can predict relapse sometime before it becomes symptomatic [43]. CONCLUSIONS Serological markers provide a means of monitoring tumor activity at many stages of the disease diagnosis, therapy, and relapse. However, it is important that they are used appropriately and their significance is understood. Knowledge about raised levels of CA125 often raises questions as well as answers; we need to be able to make use of the information available. Early knowledge about relapse does not necessarily help outcome, as better therapies are needed. Progress in therapy is likely to come from a combination of better drugs and a greater understanding of the biology of the disease. Study of serological and tumor-related surface markers needs to continue. Markers for ovarian cancer, and, in particular, CA125, have led the way for epithelial tumors and provide a valuable model for further studies. patients with pelvic masses: discrimination of benign from malignant disease. Am J Obstet Gynecol 1988;159:341-346. 8 Jacobs I, Bridges J, Reynolds C et al. Multimodal approach to screening for ovarian cancer. Lancet 1988;1:268-271. 9 Jacobs I, Prys Davies A, Bridges J et al. Prevalence screening for ovarian cancer in postmenopausal women by CA125 measurement and ultrasonography. BMJ 1993;306:1030-1034. 10 Einhorn N, Knapp R, Zurawski V. CA125 assay used in conjunction with CA 15-3 and TAG-72 assays for discrimination between malignant and non-malignant diseases of the ovary. Acta Oncol 1989;28:655-657. 11 Soper JT, Hunter V, Daly L et al. Pre-operative serum tumour associated antigen levels in women with pelvic masses. Gynecol Oncol 1990;75:249-254. 12 Woolas RP, Xu FJ, Jacobs IJ et al. Elevation of multiple serum markers in stage I ovarian cancer. J Natl Cancer Inst 1993;85:1748-1751. 13 Gadduchi A, Baicchi U, Marrai R et al. Preoperative evaluation of D-dimer and CA 125 in differentiating benign from malignant ovarian masses. Gynecol Oncol 1996;60:197-202.

329 Tumor Markers in Ovarian Malignancies 14 Rustin GJ, Nelstrop AE, McClean P et al. Defining response of ovarian carcinoma to initial chemotherapy according to serum CA 125. J Clin Oncol 1996;14:1545-1551. 15 van der Burg MEL, Lammes FB, Verweij J. The role of CA 125 in the early diagnosis of progressive disease in ovarian cancer. Ann Oncol 1990;1:301-302. 16 Rustin GJS, Nelstrop AE, Tuxen MK et al. Defining progression of ovarian carcinoma during follow-up according to CA125: a North Thames Ovary Group Study. Ann Oncol 1996;7:361-364. 17 Nagele F, Petru E, Medl M et al. Preoperative CA 125: an independent prognostic factor in patients with stage I epithelial ovarian cancer. Obstet Gynecol 1995;86:259-264. 18 Cruikshank DG, Fullerton WT, Klopper A. The clinical significance of pre-operative serum CA 125 in ovarian cancer. Br J Obstet Gynaecol 1987;94:692-695. 19 Möbus V, Kreienberg R, Crombach G et al. Evaluation of CA 125 as a prognostic and predictive factor in ovarian cancer. Journal of Tumour Markers 1988;3:251-258. 20 Mogensen O. Prognostic value of CA 125 in advanced ovarian cancer. Gynecol Oncol 1992;44:207-212. 21 Redman CWE, Blackledge GR, Kelly K et al. Early serum CA 125 response and outcome in epithelial ovarian cancer. Eur J Cancer 1990;26:593-596. 22 Fisken J, Leonard RCF, Stewart M et al. The prognostic value of early CA 125 serum assay in epithelial ovarian cancer. Br J Cancer 1993;68:140-145. 23 Ron IG, Inbar M, Gelernter I et al. Use of CA 125 response to predict survival parameters of patients with advanced ovarian carcinoma. Acta Obstet Gynecol Scand 1994;73:658-662. 24 Gadduchi A, Zola P, Landoni F et al. Serum half-life of CA 125 during early chemotherapy as an independent prognostic variable for patients with advanced epithelial ovarian cancer: results of a multicentric Italian study. Gynecol Oncol 1995;58:42-47. 25 Fayers PM, Rustin GJS, Wood R et al. The prognostic value of serum CA 125 in patients with advanced ovarian carcinoma: an analysis of 573 patients by the Medical Research Council Working Party on Gynaecological Cancer. Int J Gynecol Cancer 1993;3:285-292. 26 Khoo S-K, Hurst T, Webb M et al. Predictive value of serial CA125 antigen levels in ovarian cancer evaluated by secondlook laparotomy. Eur J Cancer Clin Oncol 1987;23:765-771. 27 Markar A, Kristensen GB, Bormer OP et al. CA 125 measured before second-look laparotomy is an independent prognostic factor for survival in patients with epithelial ovarian cancer. Gynecol Oncol 1992;45:323-328. 28 Xu F-J, Yu Y-H, Daly L et al. OVX1 radioimmunoassay complements CA-125 for predicting the presence of residual ovarian carcinoma at second-look surgical surveillance procedures. J Clin Oncol 1993;11:1506-1510. 29 Kobayashi H, Terao T, Kawashima Y. Serum sialyl Tn as an independent predictor of poor prognosis in patients with epithelial ovarian cancer. J Clin Oncol 1992;10:95-101. 30 Kacinski BM, Carter D, Mittal K et al. Ovarian adenocarcinomas express fms-complementary transcripts and fms antigen, often with co-expression of CSF-1. Am J Pathol 1990;137:135-147. 31 Kacinski BM. CSF-1 and its receptor on ovarian and other gynaecological neoplasms. In: Sharp F, Mason WP, Creason W, eds. Ovarian Cancer 2: Biology, Diagnosis and Management. London: Chapman and Hall, 1992:93-99. 32 Berek JS, Chung C, Kaldi K et al. Serum IL-6 levels correlate with disease status in epithelial ovarian cancer patients. Am J Obstet Gynecol 1991;164:1038-1043. 33 Watson J, Sensintaffar JL, Berek JS et al. Constitutive production of interleukin 6 by ovarian cancer cell lines and by primary ovarian tumour cultures. Cancer Res 1990;50:6959-6965. 34 Gottlieb WH, Abrams JS, Watson JM et al. Presence of IL-10 in the ascites of patients with ovarian and other intra-abdominal cancers. Cytokine 1992;4:385-390. 35 Granowska M, Britton KE, Shepherd JH et al. A prospective study of 123-I labeled monoclonal antibody imaging in ovarian cancer. J Clin Oncol 1986;4:730-736. 36 Powell MC, Perkins AC, Pimm MV et al. Diagnostic imaging of gynecologic tumors with the monoclonal antibody 791T/36. Am J Obstet Gynecol 1987;157:28-34. 37 Perkins AC, Powell MC, Wastie ML et al. A prospective evaluation of OC 125 and magnetic resonance imaging in patients with ovarian cancer. Eur J Nucl Med 1989;16:311-316. 38 Gallup DG. Multicenter clinical trial of 111-In-Cyt-103 in patients with ovarian cancer. In: Maguire RT, Van Nostrand D, eds. Diagnosis of Colorectal and Ovarian Cancer. New York: Marcel Dekker, Inc., 1992:111-124. 39 Kalofonos HP, Kosmas C, Hird V et al. Targeting of tumors with murine and reshaped human monoclonal antibodies against placental alkaline phosphatase: immunolocalisation, pharmacokinetics and immune response. Eur J Cancer 1994;30A:1842-1850. 40 Stewart JS, Hird V, Snook D et al. Intraperitoneal yttrium-90- labeled monoclonal antibody in ovarian cancer. J Clin Oncol 1990;8:1941-1950. 41 Hird V, Snook D, Dhokia B et al. Adjuvant therapy of ovarian cancer with radioactive monoclonal antibody. Br J Cancer 1993;68:403-406. 42 Lappohn RE, Burger HG, Bouma J et al. Inhibin as a marker for granulosa cell tumors. N Engl J Med 1989;321:790-793. 43 Cooke I, O Brien M, Charnock FM et al. Inhibin as a marker for ovarian cancer. Br J Cancer 1995;71:1046-1050.