|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 25, No 11 (April 10), 2007: pp. 1383-1389 © 2007 American Society of Clinical Oncology. DOI: 10.1200/JCO.2006.06.7884 Serum CA-125 in Relation to Adnexal Dysplasia and Cancer in Women at Hereditary High Risk of Ovarian Cancer
From the Departments of Obstetrics and Gynecology, Clinical Epidemiology and Biostatistics, Clinical Genetics and Human Genetics, and Clinical Chemistry, VU University Medical Center, Amsterdam; and the Department of Pathology, University Medical Center Utrecht, Utrecht, the Netherlands Address reprint requests to René H.M. Verheijen, MD, PhD, Department of Obstetrics and Gynecology, VU University Medical Center, PO Box 7057, 1007 MB Amsterdam, the Netherlands; e-mail: r.verheijen{at}vumc.nl
Purpose Serum CA-125 level is commonly used as indicator for ovarian cancer recurrence. However, its value for the prediction of neoplastic lesions is unknown. The aim of this study was to investigate whether CA-125 concentrations are indicative of adnexal dysplasia and cancer in women at hereditary high risk of ovarian/tubal cancer. Patients and Methods CA-125 was obtained from 424 women at hereditary high risk of ovarian/tubal cancer attending the VU University Medical Center (Amsterdam, the Netherlands) between 1993 and 2005. Serum samples obtained at the second-to-last (n = 64) and last (n = 98) visit before surgery were tested in women who underwent adnexal surgery for diagnostic (n = 9) or prophylactic (n = 89) reasons. Serum samples obtained from 370 age-matched healthy women were used as controls.
Results Both the absolute value (P < .0001) and the serial change (P < .0001) of CA-125 were predictive for ovarian cancer (n = 8). For adnexal dysplasia (n = 23), the absolute value of CA-125 (P = .003) was predictive, but the serial change in CA-125 was not (P = .32). The odds ratio for adnexal dysplasia versus nondysplasia in the highest tertile (CA-125 levels Conclusion In patients at hereditary high risk for adnexal cancer, both the absolute value of serum CA-125 and the change in serial CA-125 are predictors for ovarian cancer. Remarkably, the absolute value of CA-125 is also predictive for adnexal dysplasia. CA-125 values should, therefore, be taken into account in the decision toward prophylactic bilateral salpingo-oophorectomy.
Ovarian carcinoma is the fourth most common cause of cancer death in women in Western countries. Currently, there is no screening instrument available to efficiently detect ovarian cancer at an early stage in the general population. Epidemiologic studies and the identification of germline mutations in the BRCA1 and BRCA2 genes led to the recognition of a hereditary predisposition for breast and/or ovarian cancer and fallopian tube carcinoma.1,2 Women at hereditary high risk are advised to undergo screening or prophylactic surgery. Gynecologic screening involves transvaginal ultrasonography (TVU) and tumor marker testing for serum CA-125.3,4 The role of CA-125 measurement for monitoring tumor response to treatment in patients with ovarian cancer and the detection of recurrences, is well established and anchored in the clinic.5 In contrast, its usefulness as a screening tool for early detection of ovarian cancer is limited by false-negative (low positivity in early-stage ovarian cancer) and false-positive (elevation of the marker in physiological and benign conditions) results.6,7 Premenopausal women have higher CA-125 serum concentrations than do postmenopausal women,8 with peak levels at the time of menstruation.9 For this reason, the usefulness of CA-125 as a screening tool in early detection of ovarian cancer is higher in postmenopausal women.10 Current evaluation relies on CA-125 levels above the cutoff level for normal women with no disease, but a rising level over time within the normal limits may also be an indication for the development of ovarian cancer.11 This study was designed, firstly, to evaluate serum CA-125 values in women at hereditary high risk of ovarian cancer compared with a control population. Secondly, we analyzed whether serum CA-125 levels obtained before adnexal surgery could be indicative of dysplasia or carcinoma in ovarian or tubal epithelium.
Study Population In total, 424 women at hereditary high risk for breast/adnexal cancer were attending the Familial Cancer Clinic of the VU University Medical Center (Amsterdam, the Netherlands) for annual gynecologic screening between 1993 and 2005. Clinical geneticists were responsible for a detailed family pedigree. Pathology reports of cancer cases in the family were reviewed and revised by a pathologist of the VU University Medical Center. BRCA1/2 mutations were detected by a combination of protein truncation test (PTT), denaturing gradient gel electrophoresis (DGGE), and direct DNA sequencing (DS).12 If aberrations were detected, DNA sequencing was used to elucidate the exact mutation. Additionally, we performed multiplex ligation-dependent probe amplification (MLPA) to screen BRCA1 for exon deletions or duplication.13 Women with a proven BRCA1/2 mutation in their family but themselves tested negatively were excluded from analysis (n = 33), as were three women with a personal history of ovarian cancer. Control serum samples (n = 370) were obtained from a serum bank containing samples from a cohort of 1,000 female healthy volunteers who agreed to undergo ultrasonography of the pelvis and abdomen to exclude ovarian cancer. Control samples were age matched (± 1 year) to the study population. Because no information on the menstrual history of controls was available, we defined controls 52 years and older as postmenopausal, and otherwise as premenopausal. The characteristics of women at first visit (n = 388) and controls are summarized in Table 1. Menopausal status was noted in the medical chart during every visit at the clinic. Three hundred eighty-eight women underwent a total of 1,376 screening visits that included both CA-125 and TVU. The woman-years of screening were 1,152. Eight cases of ovarian cancer were identified, which gives an annual incidence of 0.007 for ovarian cancer in this population. Diagnostic adnexal surgery was performed in nine women, of whom four (44%) were BRCA1/2 mutation carriers. Indication to undergo diagnostic surgery was elevation of CA-125 levels (n = 3), an abnormal TVU (n = 1), and both CA-125 and TVU abnormalities (n = 5; Table 2). Prophylactic bilateral salpingo-oophorectomy (pBSO) was performed in 89 women, of whom 60 (67%) were BRCA1/2 mutation carriers. Adnexal tissue pathology is summarized in Table 3. Dysplasia of the ovaries or fallopian tubes was defined according to the criteria of Fox14,15 as a multilayering of (nonciliated) cells with atypical hyperchromatic nuclei, and mitoses that are often above the basal layer.16-18 One gynecologic pathologist performed all reviews (P.v.D.).
The research protocol was approved by the Scientific Committees of the Department of Pathology and the VU University Medical Center Research Institute for Cancer and Immunology.
Serum Samples
CA-125 Assay
Statistical Analysis
CA-125 Levels in Women at High Risk and in Controls CA-125 levels are listed in Table 4. The 95th percentiles of CA-125 levels for women at hereditary high risk and for controls were 29 U/mL and 36 U/mL, respectively.
CA-125 was higher (P < .0001) in premenopausal women at high risk (median, 12 U/mL; range, 2 to 173 U/mL) than in postmenopausal women at high risk (median, 8 U/mL; range, 2 to 43 U/mL), and also higher (P < .0001) in premenopausal controls (median, 11 U/mL; range, 2 to 56 U/mL) than in postmenopausal controls (median, 9 U/mL; range, 3 to 22 U/mL). At first visit, CA-125 was above the 35 U/mL cutoff level recommended by the assay manufacturer in 11 women (2%) at high risk (median, 41 U/mL; range, 36 to 173 U/mL), and in 19 (5%) controls (median, 42 U/mL; range, 36 to 56). Nine of these women at high risk were premenopausal. Two of the women had a fibroid uterus apparent at TVU ultrasound, one woman was diagnosed with breast cancer (pT2N0M0) at the time of elevated CA-125 (173 U/mL), two women were pregnant, and TVU was normal in six women, including the two postmenopausal women. As for controls, all 19 women were premenopausal. CA-125 was analyzed in relation to age, divided into categories of 5 years, in women at high risk and controls (Fig 1). In agreement with a previous study,8 a descending trend of CA-125 was seen both in women at high risk and in controls after 40 years of age.
CA-125 Levels Before Surgery Pathology results of prophylactic adnexal surgery (n = 89) are included in Tables 2 and 3. Dysplasia of the ovarian epithelium was found in eight cases, and dysplasia of the fallopian tube epithelium in 17 cases (three women had dysplasia both in the ovaries and in the fallopian tubes). No relation was found between BRCA1/2 mutations and dysplasia: 70% of women with dysplasia (n = 16 of 23) were BRCA1/2 mutation carriers, and 65% of women with no lesions (n = 41 of 63) were proven mutation carriers. One ovarian and two fallopian tube carcinomas were found at pBSO. CA-125 in women with adnexal dysplasia was within the normal range (median, 14 U/mL; range, 6 to 80 U/mL) but significantly higher (P = .002) compared with women with no adnexal abnormalities (median, 10 U/mL; range, 4 to 51 U/mL). CA-125 in the three women with occult cancer was also within the normal range (median, 15 U/mL; range, 11 to 20 U/mL). Sensitivity, specificity, and positive predictive value of CA-125 levels 14 U/mL or higher for adnexal dysplasia or occult adnexal cancer were 54%, 75%, and 47%, respectively. Nine women at hereditary high risk underwent diagnostic adnexal surgery because of an elevated CA-125 level during screening (n = 2), an abnormal TVU (n = 1), or a combination of both (n = 6). In five women (three BRCA1 mutation carriers), an invasive ovarian cancer was diagnosed; one with International Federation of Gynecology and Obstetrics (FIGO) stage I, one with FIGO stage II, two with FIGO stage III, and one with FIGO stage IV disease (Table 2). Furthermore, one woman (BRCA2 mutation carrier) had a borderline ovarian tumor FIGO stage I, and a Brenner tumor was diagnosed in one woman. One woman had dysplasia of the fallopian tube, and one woman had no abnormalities. A higher level of CA-125 (P = .03) was found in the five women with ovarian cancer detected at gynecologic screening followed by diagnostic surgery (median, 394 U/mL; range, 159 to 1,693 U/mL) than in the three women with occult adnexal malignancies diagnosed during prophylactic surgery (median, 15 U/mL; range, 11 to 20 U/mL). In the entire group of women who underwent surgery (n = 98), age (P = .05) and the absolute value of CA-125 (P < .0001) were significant predictors for ovarian cancer (eight of 98 women). Women diagnosed with ovarian cancer at surgery had a median age of 52 years (range, 45 to 59 years), and women with no cancer had a median age of 46 years (range, 31 to 67 years). No significant effects were found for mutation status (P = .74), menopausal status (P = .11), or interval between the time of surgery and the time of CA-125 testing (P = .59). Figure 2 shows the relationship between specificity and sensitivity for the absolute value of CA-125 without a fixed cutoff level of CA-125, representing an area under the curve of 0.874. The median time between CA-125 measurement and surgery for the eight women with ovarian cancer was 1 month (range, 0 to 7 months) and for the others was 8 months (range, 0 to 12 months). The change of CA-125 between the two visits preceding surgery could be measured in 64 women, including five cases of ovarian cancer, with a median time between the two measurements of 12 months (range, 0.6 to 59 months). This time interval was not significant (P = .73). The median time between the two serum samples for ovarian cancer cases alone was 12 months (range, 8.4 to 18 months). The change in their serial CA-125 values was strongly predictive for ovarian cancer (P < .0001). A model including both the absolute CA-125 value and the change in serial CA-125 values could not be estimated because of colinearity between the absolute CA-125 value and the change in CA-125 values. However, the model based on the change in serial CA-125 values had a superior fit (Cox and Snell R2 = 0.42) compared with the model based on the absolute CA-125 values (Cox and Snell R2 = 0.21).19 The superior fit becomes apparent from Figure 3, where women without and with ovarian cancer are perfectly separated.
In the group of women who underwent prophylactic surgery (n = 89), the absolute value of CA-125 was a significant (P = .003) predictor for the presence of dysplasia in the adnexal tissue. No significant effects were found for age, DNA status, menopausal status, or interval between the time of surgery and the time of CA-125 testing (P = .44, .69, .36 and .37, respectively). The change of CA-125 between the two serial visits preceding surgery could be measured in 56 women, and median time between the two measurements was 11 months (range, 0.6 to 18 months) for women diagnosed with dysplasia and 12 months (range, 2.8 to 59 months) for women with no lesions. Neither the change in CA-125 levels (P = .32) nor the interval between the two serum samples (P = .10) was predictive for dysplasia. Table 5 shows dysplasia and cancer prevalence in prophylactically removed adnexal tissue separately for CA-125 tertile groups. The odds ratio for adnexal dysplasia versus nondysplasia in the highest tertile (CA-125 levels  14 U/mL) compared with the lowest tertile (CA-125 < 10 U/mL) was 6 (95% CI, 1.32 to 36.66). Cancer prevalence was low.
CA-125 serum levels are related to ovarian function and decrease with age.8 The association between elevated CA-125 levels and ovarian cancer is stronger for postmenopause than premenopause.10,20,21 This factor should also be considered when screening for ovarian cancer over an extended period of time.11,22 Data available in the literature on CA-125 serial serum measurements refer only to postmenopausal women in the general population.23 The majority of women at hereditary high risk, however, enter a screening program for ovarian cancer already in premenopause. CA-125 was significantly lower in postmenopausal than in premenopausal women, and CA-125 levels decreased with increasing age in both high-risk women and controls. Therefore, as in the general population, in postmenopause, a value within the normal range does not exclude an abnormality in the adnexal tissue. The majority of CA-125 levels at first visit that were above the cutoff were measured in premenopausal women. This could explain the elevated CA-125 in the six women who had no abnormalities. Because CA-125 values may vary with the commercial test used, it is mandatory to use one and the same test throughout screening of individual women.24 In our study, CA-125 values obtained at two consecutive visits and measured with the same test were sufficient to predict ovarian cancer. In agreement with Skates et al,25 we found that the difference in CA-125 levels between two consecutive visits had a higher predictive value for ovarian cancer than did the absolute value of a single measurement. In our study, no ovarian cancer was diagnosed in women with initial levels of CA-125 above the cutoff of 35 U/mL. On the other hand, women with an increase of CA-125 over time were more often diagnosed with ovarian cancer. The median interval of 1 year between two measurements of serum CA-125, which is in agreement with screening guidelines may have an impact on early diagnosis of ovarian cancer. Three of the five women diagnosed with ovarian cancer at surgery had CA-125 levels that were well above the cutoff at the planned annual screening visit. Two women presented with abdominal pain 3 and 4 months before the planned annual screening visit, with CA-125 levels of 1,693 and 394 U/mL, respectively. CA-125 levels in all these women were normal 1 and a half years before. Ovarian cancer can be predicted by both the absolute serum CA-125 level and the change in CA-125 levels between two consecutive visits. However, even with yearly screening visits, ovarian cancer is often diagnosed at an advanced stage, with the consequent high mortality rate. An upward trend in CA-125, measured every 3 months and plotted on a curve included in the clinical chart, may contribute to an earlier diagnosis of adnexal cancer. Even though the exact significance of dysplasia in relation to the developments of ovarian and fallopian tube cancer is not yet well defined, prediction of dysplasia leading to prophylactic surgery could be of importance for the prevention of ovarian cancer. Remarkably, dysplasia could be predicted on the basis of the absolute value of CA-125, which ranked higher in women diagnosed with dysplasia than in women with no adnexal lesions. The odds ratio for dysplasia was 6 for women with CA-125 of 14 U/mL or higher, suggesting that this cutoff value may be taken as an additional parameter toward decision making in women considering pBSO. Because differences in CA-125 are marginal, CA-125 values above this cutoff and still within the normal range are not in themselves mandatory for pBSO in women hesitant to undergo the procedure. In conclusion, CA-125 levels do not behave differently in women at hereditary high risk compared with controls, and are subject to the same changes caused by age and menopause. Both the absolute value of serum CA-125 and the change in serial CA-125 are predictors for ovarian cancer. Remarkably, for the presence of dysplasia, the absolute value of CA-125 is the best predictor and should, therefore, be taken into account as an additional parameter for considering pBSO, irrespective of age, menopausal status, or mutation status.
The authors indicated no potential conflicts of interest.
Conception and design: Brenda B.J. Hermsen, Silvia von Mensdorff-Pouilly, Johannes Berkhof, Peter Kenemans, René H.M. Verheijen Administrative support: Brenda B.J. Hermsen Provision of study materials or patients: Brenda B.J. Hermsen, Silvia von Mensdorff-Pouilly, Fred H. Menko, Johan J.P. Gille, Paul J. van Diest, Marinus A. Blankenstein, René H.M. Verheijen Collection and assembly of data: Brenda B.J. Hermsen, Johannes Berkhof, Marinus A. Blankenstein, René H.M. Verheijen Data analysis and interpretation: Brenda B.J. Hermsen, Silvia von Mensdorff-Pouilly, Johannes Berkhof, Fred H. Menko, Paul J. van Diest, Marinus A. Blankenstein, Peter Kenemans, René H.M. Verheijen Manuscript writing: Brenda B.J. Hermsen, Silvia von Mensdorff-Pouilly, Johannes Berkhof, Fred H. Menko, Paul J. van Diest, Marinus A. Blankenstein, Peter Kenemans, René H.M. Verheijen Final approval of manuscript: Brenda B.J. Hermsen, Silvia von Mensdorff-Pouilly, Johannes Berkhof, Fred H. Menko, Johan J.P. Gille, Paul J. van Diest, Marinus A. Blankenstein, Peter Kenemans, René H.M. Verheijen
Supported by Biocare Foundation Grant No. 02-22. Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Zweemer RP, van Diest PJ, Verheijen RH, et al: Molecular evidence linking primary cancer of the fallopian tube to BRCA1 germline mutations. Gynecol Oncol 76:45-50, 2000[CrossRef][Medline] 2. Piek JM, Dorsman JC, Zweemer RP: Women harboring BRCA1/2 germline mutations are at risk for breast and female adnexal carcinoma. Int J Gynecol Pathol 22:315-316, 2003[CrossRef][Medline] 3. Karlan BY, Raffel LJ, Crvenkovic G, et al: A multidisciplinary approach to the early detection of ovarian carcinoma: Rationale, protocol design, and early results. Am J Obstet Gynecol 169:494-501, 1993[Medline] 4. Laframboise S, Nedelcu R, Murphy J, et al: Use of CA-125 and ultrasound in high-risk women. Int J Gynecol Cancer 12:86-91, 2002[CrossRef][Medline] 5. Verheijen RH, Mensdorff-Pouilly S, van Kamp GJ, et al: CA 125: Fundamental and clinical aspects. Semin Cancer Biol 9:117-124, 1999[CrossRef][Medline] 6. Yedema CA, Kenemans P: Serum CA-125 screening for ovarian cancer, in HIJ Wildschut, CP Weiner, TJ Peters (eds): When to Screen in Obstetrics and Gynecology (ed 2). New York, NY, Saunders, 2006, pp 629-648 7. Stirling D, Evans DG, Pichert G, et al: Screening for familial ovarian cancer: Failure of current protocols to detect ovarian cancer at an early stage according to the International Federation of Gynecology and Obstetrics system. J Clin Oncol 23:5588-5596, 2005 8. Bon GG, Kenemans P, Verstraeten R, et al: Serum tumor marker immunoassays in gynecologic oncology: Establishment of reference values. Am J Obstet Gynecol 174:107-114, 1996[CrossRef][Medline] 9. Bon GG, Kenemans P, Dekker JJ, et al: Fluctuations in CA 125 and CA 15-3 serum concentrations during spontaneous ovulatory cycles. Hum Reprod 14:566-570, 1999 10. Jacobs I, Davies AP, Bridges J, et al: Prevalence screening for ovarian cancer in postmenopausal women by CA 125 measurement and ultrasonography. BMJ 306:1030-1034, 1993 11. Skates SJ, Menon U, MacDonald N, et al: Calculation of the risk of ovarian cancer from serial CA-125 values for preclinical detection in postmenopausal women. J Clin Oncol 21:206-210, 2003 (suppl)[CrossRef][Medline] 12. Sevilla C, Moatti JP, Julian-Reynier C, et al: Testing for BRCA1 mutations: A cost-effectiveness analysis. Eur J Hum Genet 10:599-606, 2002[CrossRef][Medline] 13. Hogervorst FB, Nederlof PM, Gille JJP, et al: Large genomic deletions and duplications in the BRCA1 gene identified by a novel quantitative method. Cancer Res 63:1449-1453, 2003 14. Fox H, Wells M: Pathology of early malignant change in the ovary. Int J Gynecol Pathol: 12:153-155, 1993[Medline] 15. Fox H, Wells M: Obstetrical and Gynecological Pathology (ed 4). New York, NY, Churchill Livingstone, 1995 16. Schlosshauer PW, Cohen CJ, Penault-Llorca F, et al: Prophylactic oophorectomy: A morphologic and immunohistochemical study. Cancer 98:2599-2606, 2003[CrossRef][Medline] 17. Piek JM, van Diest PJ, Zweemer RP, et al: Dysplastic changes in prophylactically removed fallopian tubes of women predisposed to developing ovarian cancer. J Pathol 195:451-456, 2001[CrossRef][Medline] 18. Hermsen BB, van Diest PJ, Berkhof J, et al: Low prevalence of (pre)malignant lesions in the breast and high prevalence in the ovary and fallopian tube in women at hereditary high risk of breast and ovarian cancer. Int J Cancer 119:1412-1418, 2006[CrossRef][Medline] 19. Cox DR, Snell EJ: The Analysis of Binary Data (ed 2). London, United Kingdom, Champman and Hall, 1989, pp 208-209 20. Jeyarajah AR, Ind TE, Skates S, et al: Serum CA125 elevation and risk of clinical detection of cancer in asymptomatic postmenopausal women. Cancer 85:2068-2072, 1999[Medline] 21. Menon U, Talaat A, Jeyarajah AR, et al: Ultrasound assessment of ovarian cancer risk in postmenopausal women with CA125 elevation. Br J Cancer 80:1644-1647, 1999[CrossRef][Medline] 22. Jacobs IJ, Menon U: Progress and challenges in screening for early detection of ovarian cancer. Mol Cell Proteomics 3:355-366, 2004 23. Jacobs IJ, Skates SJ, MacDonald N, et al: Screening for ovarian cancer: A pilot randomised controlled trial. Lancet 353:1207-1210, 1999[CrossRef][Medline] 24. Davelaar EM, Schutter EM, Mensdorff-Pouilly S, et al: Clinical and technical evaluation of the ACS: OV serum assay and comparison with three other CA125-detecting assays. Ann Clin Biochem 40:663-673, 2003[CrossRef][Medline] 25. Skates SJ, Xu FJ, Yu YH, et al: Toward an optimal algorithm for ovarian cancer screening with longitudinal tumor markers. Cancer 76:2004-2010, 1995 (suppl)[CrossRef][Medline] Submitted April 8, 2006; accepted January 31, 2007.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2007 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
