Endometriosis is a gynecologic condition in women of reproductive age that primarily produces infertility and pain and is defined as the presence of viable endometrial glands and stroma outside the uterine cavity, mainly on the pelvic peritoneum but also on the ovaries, the rectovaginal septum, and, more rarely, in other sites.1 The pathogenesis of endometriosis is probably multifactorial and involves the loss of differentiation capacity of the endometriotic cells,2 cell-adhesion molecules for adhesion of endometrium to the peritoneum, neoangiogenesis, the peritoneal fluid, and the immune system.1 Endometriomas are invaginations of ovarian surface epithelium containing endometrial tissue, commonly referred to as ovarian cysts lined by endometrial tissue.1 Ovarian endometrioma is a prevalent gynecologic disorder still lacking specific serum markers.
In the present study, we measured plasma levels of urocortin, a 40-amino acid peptide belonging to the corticotropin-releasing hormone (CRH) family,3 in women affected by endometriomas to assess its diagnostic performance in distinguishing endometriomas from other benign ovarian cysts. The differential diagnosis between endometriomas and other benign ovarian cysts is important in current clinical practice to assist decisions about medical and surgical therapy. Ultrasonography and CA 125 measurements are the usual methods of noninvasive diagnosis of endometrioma, but they do not achieve optimal sensitivity and specificity.4 We choose to study urocortin, because it is expressed by the human endometrium,5 and it is supposed to affect several aspects of both endometriosis and endometrium physiology.6 Indeed, urocortin 1) induces endometrial cell differentiation,7 2) stimulates the expression of cell-adhesion molecules,8 3) acts on the immune system,9 and finally, 4) influences vascular endothelial tone10 through its endothelial receptors.11
As a final point, in the present study we also investigated the localization of immunoreactive urocortin in ovarian endometriotic tissue by immunohistochemistry and urocortin levels in peritoneal fluid and cystic content of endometriomas to verify the hypothesis that urocortin circulating into bloodstream originates from endometriotic lesions.
MATERIALS AND METHODS
Written informed consent was obtained from all patients before inclusion in the study, which was approved by the University of Siena Human Investigation Committee. The study sample consisted of women who underwent laparoscopic excision of ovarian cysts (n=80) and were enrolled prospectively between March 2004 and January 2006 at the University of Siena academic hospital. After laparoscopy and histopathology, they were classified into two groups: one group included 40 consecutive patients diagnosed as having stage III or IV endometriosis according to the American Society for Reproductive Medicine revised classification of endometriosis.12 Of these, 28 had only ovarian endometrioma, and 12 had peritoneal endometriosis associated with ovarian endometrioma (Table 1). The second group included consecutive patients with benign, nonendometriotic ovarian cysts (n=40), including serous (n=11) and mucinous (n=12) cystadenomas, dermoid cysts (n=6), and hemorrhagic corpora lutea cysts (n=11).
Blood samples were drawn from a peripheral vein with a polypropylene syringe and a butterfly needle, immediately before anesthesia for laparoscopy, and then transferred to chilled tubes containing ethylenediaminetetraacetic acid (10 mg/mL blood) and to dry tubes where blood was allowed to clot at room temperature. The tubes were centrifuged immediately at 4°C (3,000g for 10 minutes). All plasma and serum samples were kept at –80°C until assay.
In a representative subset of patients with endometriosis (n=12) with the same clinical and demographic characteristics of the whole endometriosis group, peritoneal fluid was obtained by aspiration of cul de sac fluid at the time of laparoscopy, while cystic fluid was collected by needle aspiration. All fluid samples were centrifuged at 400g for 10 minutes, and aliquots of the supernatants were stored at –20°C until urocortin assay. In addition, ovarian endometriotic tissue was collected and fixed by immersion on 10% buffered formalin, embedded in paraffin, and routinely processed to further immunohistochemistry.
Urocortin levels were measured using previously published methodology,13 except for the delayed addition of tracer to improve assay sensitivity. Briefly, duplicate 100-μL aliquots of plasma extract or human urocortin-(1–40) standard were mixed with 100 μL assay buffer containing rat urocortin-(1–40) antiserum at a 1:2,100 dilution and incubated for 40 hours at 4°C. One hundred microliters of buffer containing approximately 25,000 cpm 125I-labeled human urocortin-(1–40) were then added, and the tubes were incubated for an additional 6 hours before the addition of preprecipitated sheep antirabbit second antibody, as previously described.13 The specificity of the urocortin antiserum had been checked by measuring the cross-reactivity of peptides with sequence homology in the urocortin assay, ie, human CRH-(1–20) and human CRH-(1–41) (Peninsula Laboratories, St. Helen’s Merseyside, UK), human urocortin II (stresscopin related peptide-[6–43] NH2), and human urocortin III (stresscopin-[3–40] NH2; Phoenix Pharmaceuticals Inc, Belmont, CA) as well as with adrenocorticotropic hormone, sauvagine, and urotensin 1 (Sigma-Aldrich Corporation, St. Louis, MO) and thyroglobulin. None of these molecules displayed significant cross-reactivity, even at a high concentration (1 mg/mL).
Urocortin levels were measured in a blinded fashion in a single assay. The assay had a sensitivity of approximately 50 pg/mL, with coefficient of variation of 8%.
Serum CA 125 concentration was assessed by Cobas Core CA 125 enzyme-immunoassay analysis kit (Roche, Basel, Switzerland). The sensitivity of the assay was less than 1 unit/L, and the intra- and interassay variations were less then 5.6% and 7.8%, respectively.
Formalin-fixed, paraffin-embedded specimens were cut into 4-μm slices and were stained by immunohistochemistry by using the avidin-biotin-peroxidase method. All samples and controls were processed together. After exposure to 1% H2O2 in methanol to block endogenous peroxidase, sections were treated with normal goat serum for 30 minutes to suppress nonspecific binding. The specific rabbit polyclonal antiserum raised against human urocortin (kindly donated by Dr. Wylie Vale, Salk Institute, LA Jolla, CA) was diluted 1:500 in phosphate-buffered saline and applied on the slides for 12 hours at 4°C.
Sections were then treated with biotinylated goat anti-rabbit immunoglobulin G for 30 minutes at room temperature and incubated with the avidin-biotin-peroxidase complex (Oncogene Research Products, San Diego, CA) for 30 minutes. Peroxidase activity was visualized by exposing the sections for 3 minutes to 1 mg/mL 3,3′-diaminobenzidine tetrahydrochloride (Sigma Chemical Company, St. Louis, MO) in phosphate-buffered saline containing 0.3% H2O2.
The sections were then counterstained with hematoxylin. Negative controls were obtained by using the antibody preadsorbed with the corresponding peptide at the concentration of 20 mcg/mL of diluted antibody.
Data distribution was analyzed with the Kolmogorov-Smirnov test. The concentrations of CA 125 were normally distributed and therefore expressed as mean±standard error, and compared by unpaired Student t test. Urocortin concentrations departed from normality and were expressed as medians and compared with Mann-Whitney test (two groups) or Friedman analysis of variance followed by Dunn test (three compartments with matched samples). Statistical significance was set at P<.05.
Sensitivity, specificity, and likelihood ratios were calculated with exact 95% confidence intervals. Receiver operating characteristic curves were obtained with their respective areas under the curves±standard error and compared by the method of Hanley and McNeil14 using Analyse-It software package (Analyse-It Software, Ltd., Leeds, England).15 The sample size was calculated to estimate the sensitivity and specificity of plasma urocortin levels in the diagnosis of endometrioma compared other benign ovarian cysts, considering expected sensitivity and specificity of 0.80, minimal acceptable lower 95% confidence limit of 0.55, and power (1-β)=0.95.16
Urocortin was measurable in all samples evaluated. Plasma urocortin levels were twice as high in the women with endometrioma (median 49 pg/mL, interquartile range 41–63 pg/mL) as in the control group (19 pg/mL, 15–23 pg/mL, P<.001, Fig. 1). As expected, CA 125 concentrations were also higher in the group with endometrioma (50±4 units/L) than in controls (24±2 units/L, P<.001, Fig. 1).
In the patients with ovarian endometrioma, urocortin concentrations were higher in the peritoneal fluid than in plasma and still higher in cyst fluid (Fig. 2). In detail, the levels of urocortin in the cystic content (188 pg/mL, 135–218 pg/mL) were markedly higher than in peritoneal fluid (125 pg/mL, 108–140 pg/mL, P<.05) and reached more than three times the plasma concentration (P<.001, Fig. 2), suggesting a local source. To further support this hypothesis, we found strong and diffuse urocortin immunostaining in the epithelial cells of the endometriotic glands (Fig. 3A, B). In the periglandular endometrial stroma, urocortin was localized mainly in the vascular walls, whereas stromal and inflammatory cells were seldom positive (Fig. 3C).
Figure 4 shows the receiver operating characteristic curves of urocortin and CA 125 for the diagnosis of ovarian endometrioma compared with other benign ovarian cysts. Urocortin, with an area under the receiver operating characteristic curve=0.961±0.021, was more effective than CA 125 in discriminating the women with endometrioma (P=.016, Table 2).14,17 Whereas urocortin detected 88% of the cases of endometrioma with 90% specificity, CA 125 detected only 65% of the cases with the same specificity. Accordingly, the positive and negative likelihood ratios for urocortin were 8.8 and 0.14, respectively (Table 2).
The subgroup of women with only endometrioma and absence of peritoneal implants still had high plasma urocortin concentrations (49 pg/mL, 41–62 pg/mL, P<.001 compared with control group), which did not differ from the subgroup with endometrioma plus peritoneal implants (Fig. 2B). Elevated plasma urocortin levels (above 33 pg/mL, that corresponds to 90% specificity) were found in 25 of 28 women with isolated endometrioma (sensitivity 89%, 95% confidence interval 73–93%).
In the present study we found that circulating levels of urocortin are higher in women affected by endometriomas than in women with other benign ovarian cysts. The finding that urocortin concentrations are higher in the cystic content than in peritoneal fluid and plasma of patients with endometriomas and the peptide localization shown by immunohistochemistry in epithelial cells of the endometriotic glands together suggest that urocortin is expressed and secreted by the endometriotic tissue and therefore open new questions about the putative role of urocortin in the pathogenesis of endometriosis. In this regard, urocortin has been shown to influence key events implicated in the development of endometriosis, such as endometrial growth and differentiation,7 endometrial adhesion,8 immune system modulation,9 and angiogenesis.10
A potential role of urocortin in endometrial cell cycle regulation may be hypothesized because CRH receptor activation inhibits the proliferation of endometrial cells derived from a tumor cell line,18 and the expression of urocortin is reduced in endometrial cancer.19 Therefore, we may speculate that urocortin secretion is part of the host response against ectopic endometrium implantation. This may be accomplished not only by growth inhibition but also by stimulating stromal cell decidualization.7 This is important because a typical phenotype shown by the endometrium of women with endometriosis is the impaired decidualization, which makes this tissue prone to implant and survive outside the uterus.2
Along with impaired decidualization, endometrial cells of women with endometriosis are acquainted with a higher adhesion capacity that favors their implantation on the peritoneal or ovarian surface,20 and they also face an aberrant immune response at the host site that contributes to their ectopic implantation and survival.8 Urocortin has been shown to stimulate the secretion of matrix metalloproteinase-9 from human placental cells,21 and whether it affects the same adhesion molecule in endometrial cells is still unknown, but this would certainly help to explain urocortin’s role in the pathogenesis of endometriosis. As regards the immune system modulation, urocortin is released by lymphocytes and stimulates mast cell and natural killer cell activation, thereby promoting an early proinflammatory response.9 These proimmune effects of urocortin, however, may be unable to improve the immune response to endometriotic implants. In fact, endometriotic lesions are richly infiltrated by immune cells but remain resistant to autoimmune clearing.8,22
Urocortin inhibits angiogenesis and promotes vascular smooth muscle relaxation through its agonistic action on its type 2 receptor (namely, CRHR2). The antiangiogenic and vasodilatory actions mediated by such a receptor are physiologically relevant, as illustrated by CRHR2-deficient mice, which are hypertensive and have increased systemic vascular resistance.23 It is thus plausible that urocortin acts on endometrial vessels, especially during the secretory phase of menstrual cycle, in a way similar to its action on other vascular beds, ie, as a tonic stimulus to limit neovascularization and maintain vascular quiescence.10 Ovarian endometriomas are poorly vascularized tumors24 and the local accumulation of urocortin, as shown by the present data, may possibly help in limiting local angiogenesis. During the initial phase of endometriosis development, however, when angiogenesis is required to render implants viable,25 urocortin might function as a protective mechanism against disease progression.
Whatever the putative role of urocortin in endometriosis, the clinical significance of the higher urocortin plasma levels in women with ovarian endometrioma compared with other benign ovarian cysts warrants consideration. The differential diagnosis between these two benign ovarian conditions is important in current practice to assist decisions about medical and surgical therapy. Ultrasonography and CA 125 measurements are the usual methods of noninvasive diagnosis of endometrioma, but they do not achieve optimal sensitivity and specificity.4 A meta-analysis of 23 studies involving 2,866 patients has come to the conclusion that CA 125 detects only half of the cases of endometriosis stage III/IV with appropriate specificity.26 Other noninvasive methods like transvaginal ultrasonography and pelvic examination achieve high sensitivity for ovarian endometriosis,27 but their false-positive rate is still concerning. On the contrary, our data suggest that urocortin may be a new marker of ovarian endometrioma because its diagnostic performance outweighs that of CA 125.
Besides the discrimination of endometriomas from other benign ovarian cysts, future investigation should clarify whether urocortin also detects nonovarian endometriosis, as well as low-stage disease. Those are conditions for which a serum marker is even more necessary, due to the poor performance of noninvasive methods.2,26 Furthermore, the clinical use of urocortin as a diagnostic tool in reproductive-age women will need additional studies to determine if the peptide concentration is appreciably affected by circadian rhythm, menstrual cycle, exercise, stress, and concurrent diseases. Finally, to put this new marker in the context of standard care, it should be considered that most ovarian cysts of benign appearance are primarily treated with medical therapy before a definitive diagnosis is established. The measurement of plasma urocortin at this moment will hopefully assist the clinician to estimate the probability that the patient has an endometriotic cyst. Therefore, studies including a large number of unselected women with adnexal masses should determine the positive and negative predictive values of plasma urocortin measurement, alone and in combination with other markers, in the diagnosis of endometrioma.
Another important research avenue is to define the presence of urocortin in other normal and abnormal states. Recent experimental data show that the liver and kidney are potential sources of circulating urocortin in healthy animals.28 To date, plasma urocortin levels have been found increased in men and women with heart failure29 and in pregnant women with hypertensive disorders30 or postterm gestation.31 Urocortin is expressed in adrenal tumors,32 but data are still lacking on its possible production and secretion by other neoplastic tissues. It would be particularly important to investigate whether urocortin levels are altered in women with ovarian cancer because, if it is not, then urocortin could be more specific than CA125 in distinguishing between malignant and benign conditions.
In conclusion, the immunolocalization of urocortin and its higher levels in the cystic content than in peritoneal fluid and plasma suggest that urocortin may be secreted by the endometriotic tissue. Urocortin is a sensitive and specific marker for the differential diagnosis of endometrioma compared with other benign ovarian cysts.
1. Giudice LC, Kao LC. Endometriosis. Lancet 2004;364:1789–99.
2. Klemmt PA, Carver JG, Kennedy SH, Koninckx PR, Mardon HJ. Stromal cells from endometriotic lesions and endometrium from women with endometriosis have reduced decidualization capacity. Fertil Steril 2006;85:564–72.
3. Vaughan J, Donaldson C, Bittencourt J, Perrin MH, Lewis K, Sutton S, et al. Urocortin, a mammalian neuropeptide related to fish urotensin I and to corticotropin-releasing factor. Nature 1995;378:287–92.
4. Guerriero S, Mais V, Ajossa S, Paoletti AM, Angiolucci M, Melis GB. Transvaginal ultrasonography combined with CA-125 plasma levels in the diagnosis of endometrioma. Fertil Steril 1996;65:293–8.
5. Florio P, Arcuri F, Ciarmela P, Runci Y, Romagnoli R, Cintorino M, et al. Identification of urocortin mRNA and peptide in the human endometrium. J Endocrinol 2002;173:R9–14.
6. Florio P, Vale W, Petraglia F. Urocortins in human reproduction. Peptides 2004;25:1751–7.
7. Torricelli M, De Falco G, Florio P, Rossi M, Leucci E, Vigano P, et al. Secretory endometrium highly expresses urocortin messenger RNA and peptide: possible role in the decidualization process. Hum Reprod 2007;22:92–6.
8. Li W, Challis JR. Corticotropin-releasing hormone and urocortin induce secretion of matrix metalloproteinase-9 (MMP-9) without change in tissue inhibitors of MMP-1 by cultured cells from human placenta and fetal membranes. J Clin Endocrinol Metab 2005;90:6569–74.
9. Baigent SM. Peripheral corticotropin-releasing hormone and urocortin in the control of the immune response. Peptides 2001;22:809–20.
10. Bale TL, Giordano FJ, Vale WW. A new role for corticotropin-releasing factor receptor-2: suppression of vascularization. Trends Cardiovasc Med 2003;13:68–71.
11. Simoncini T, Apa R, Reis FM, Miceli F, Stomati M, Driul L, et al. Human umbilical vein endothelial cells: a new source and potential target for corticotropin-releasing factor. J Clin Endocrinol Metab 1999;84:2802–6.
12. American Fertility Society. Revised American Fertility Society classification of endometriosis: 1985. Fertil Steril 1985;43:351–2.
13. Florio P, Torricelli M, Galleri L, De Falco G, Leucci E, Calonaci G, et al. High fetal urocortin levels at term and preterm labor. J Clin Endocrinol Metab 2005;90:5361–5.
14. Hanley JA, McNeil BJ. A method of comparing the areas under receiver operating characteristics curves derived from the same cases. Radiology 1983;148:839–43.
15. Stephan C, Wesseling S, Schink T, Jung K. Comparison of eight computer programs for receiver-operating characteristic analysis. Clin Chem 2003;49:433–9.
16. Flahault A, Cadilhac M, Thomas G. Sample size calculation should be performed for design accuracy in diagnostic test studies. J Clin Epidemiol 2005;859–62.
17. Khan KS, Khan SF, Nwosu CR, Arnott N, Chien PF. Misleading authors’ inferences in obstetric diagnostic test literature. Am J Obstet Gynecol 1999;181:112–5.
18. Graziani G, Tentori L, Portarena I, Barbarino M, Tringali G, Pozzoli G, et al. CRH inhibits cell growth of human endometrial adenocarcinoma cells via CRH-receptor 1-mediated activation of cAMP-PKA pathway. Endocrinology 2002;143:807–13.
19. Florio P, De Falco G, Leucci E, Torricelli M, Torres PB, Toti P, et al. Urocortin expression is downregulated in human endometrial carcinoma. J Endocrinol 2006;190:99–105.
20. Collette T, Maheux R, Mailloux J, Akoum A. Increased expression of matrix metalloproteinase-9 in the eutopic endometrial tissue of women with endometriosis. Hum Reprod 2006;21:3059–67.
21. Ness RB, Modugno F. Endometriosis as a model for inflammation-hormone interactions in ovarian and breast cancers. Eur J Cancer 2006;42:691–703.
22. Kempuraj D, Papadopoulou N, Stanford EJ, Christodoulou S, Madhappan B, Sant GR, et al. Increased numbers of activated mast cells in endometriosis lesions positive for corticotropin-releasing hormone and urocortin. Am J Reprod Immunol 2004;52:267–75.
23. Bale TL, Giordano FJ, Hickey RP, Huang Y, Nath AK, Peterson KL, et al. Corticotropin-releasing factor receptor 2 is a tonic suppressor of vascularization. Proc Natl Acad Sci U S A 2002;99:7734–39.
24. Guerriero S, Ajossa S, Mais V, Risalvato A, Lai MP, Melis GB. The diagnosis of endometriomas using colour Doppler energy imaging. Hum Reprod 1998;13:1691–95.
25. Groothuis PG, Nap AW, Winterhager E, Grummer R. Vascular development in endometriosis. Angiogenesis 2005;8:147–56.
26. Mol BW, Bayram N, Lijmer JG, Wiegerinck MA, Bongers MY, van der Veen F, et al. The performance of CA-125 measurement in the detection of endometriosis: a meta-analysis. Fertil Steril 1998;70:1101–8.
27. Eskenazi B, Warner M, Bonsignore L, Olive D, Samuels S, Vercellini P. Validation study of nonsurgical diagnosis of endometriosis. Fertil Steril 2001;76:929–35.
28. Charles CJ, Rademaker MT, Richards AM, Yandle TG. Plasma urocortin 1 in sheep: regional sampling and effects of experimental heart failure. Peptides 2006;27:1801–5.
29. Ng LL, Loke IW, O’Brien RJ, Squire IB, Davies JE. Plasma urocortin in human systolic heart failure. Clin Sci (Lond) 2004;106:383–8.
30. Florio P, Torricelli M, De Falco G, Leucci E, Giovannelli A, Gazzolo D, et al. High maternal and fetal plasma urocortin levels in pregnancies complicated by hypertension. J Hypertens 2006;24:1831–40.
31. Torricelli M, Ignacchiti E, Giovannelli A, Merola A, Scarpetti E, Calonaci G, et al. Maternal plasma corticotrophin-releasing factor and urocortin levels in post-term pregnancies. Eur J Endocrinol 2006;154:281–5.
© 2007 The American College of Obstetricians and Gynecologists
32. Fukuda T, Takahashi K, Suzuki T, Saruta M, Watanabe M, Nakata T, et al. Urocortin 1, urocortin 3/stresscopin, and corticotropin-releasing factor receptors in human adrenal and its disorders. J Clin Endocrinol Metab 2005;90:4671–8.