Experimental immunologyNon-invasive diagnosis of endometriosis based on a combined analysis of four plasma biomarkers

Purpose: The objective of the current study was to evaluate whether the analysis of different proin - flammatory and angiogenesis-regulating cytokines in a well-defined patient population can be accurate for the diagnosis of endometriosis at different stages.

For this prospective study, plasma samples were collected from women undergoing laparoscopic surgery for subfertility. The stage of endometriosis was confirmed during laparoscopy.

Plasma levels of interleukin 6 (IL-6), carbohydrate antigen (CA)-125 were significantly higher in women with endometriosis compared with controls (46.70 ±27.08 vs. 27.08 ±16.29, p < 0.05; 49.39 ±28.32 vs. 9.78 ±3.87, respectively). In women with minimal-mild endometriosis, only the plas - ma level of IL-6 was increased (56.45 ±23.21 vs. 27.08 ±16.29, p < 0.05). In women with moderate- severe endometriosis, plasma levels of IL-6 and CA-125 were significantly increased, and those of Epo and tumor necrosis factor α (TNF- α) were significantly decreased (44.53 ±31.07 vs. 27.08 ±16.29, p < 0.05; 57.84 ±61.44 vs. 9.78 ±3.87, p < 0.05; 26.60 ±10.19 vs. 30.32 ±7.94, p < 0.05; 58.61 ±7.93 vs. 65.40 ±9.86, p < 0.05, respectively). The area under curve (AUC) for Epo, TNF- α, IL-6 and CA-125 were 0.280, 0.322, 0.729 and 0.864, respectively.

The results of our study show that progression of endometriosis is associated with the elevated level of serum IL-6. Clearly, larger prospective studies are required to determine the diagnos - tic potential of measuring circulating inflammatory cytokine levels like IL-6 in endometriosis. Key words: endometriosis, IL-6, CA-125, Epo, TNF- α. (Centr Eur J Immunol 2013; 38 (2): 154-158)

This perplexing disease affects 6-10% of women in the reproductive age (ages of 12-80 years) [1]. Laparoscopy is accepted as the “gold” standard for the diagnosis of endometriosis. The extent of endometriosis is generally staged by the American Society for Reproductive Medicine (ASRM) scoring system, which categorizes minimal and mild endometriosis as stage I and stage II, and moderate and severe endometriosis as stage III and stage IV [2]. A non-invasive diagnostic test for endometriosis might abolish the need for surgery. However, such a noninvasive diagnostic test is currently unavailable [3]. In the current practice, unexpected preoperative findings potentially can lead to under-treatment or unnecessary surgery. Accurate non-invasive diagnostic techniques are clearly necessary to manage women with endometriosis more effectively. In addition, a survey [4], which was taken by 7025 women with endometriosis, revealed that 65% of the women with endometriosis were initially diagnosed with another con - dition, which led to an average delay of 8 years between the start of symptoms and the actual diagnosis of endometriosis [5, 6]. DOI: 10.5114/ceji.2013.35204 Central European Journal of Immunology 2013; 38(2) 155 The immune deviations in endometriosis, including increased local production of some cytokines as well as ele - vated autoantibody production of local and systemic immu - nity,suggest that endometriosismay be an autoimmunedis - order [7-11]. The objectiv e of the current study was to evaluate whether the analysis of different proinfl ammatory and angiogenesis-regulating cytokines in a well-defined patient population can be accurate for the diagnosis of endometrio - sis at different stages.

For this prospective study, plasma samples were col - lected from women undergoing the laparoscopic surgery due to infertility at the gynecology clinic of the Department of Obstetrics and Gynecology, Istanbul University School of Medicin e (Istanbul, Turkey). The study had received approval from the Ethics Committee of the Istanbul Uni - versity School of Medicine and written consent from the participants before its initiation. Blood (4 ml) was drawn on the day of the surgery, centrifuged at 1000 rpm for 15 min, labeled and stored at –80°C till analysis. The stage of endometriosis was confirmed during laparoscopy accord - ing to the revised American Fertility Society classification [12]. For each patient, relevant clinical information (age, stage of endometriosis, current medication and, number and type of previous operations) was recorded. The first comparison was of controls (22 patients in which endometriosis was excluded laparoscopically by an experienc ed gynecologic surgeon) versus all stages of endometriosis (33 patients). Endometriosis patients were then divided into three groups according to the stage of endometriosis: controls (33 patients; no endometriosis), minimal-mild endometriosis (16 patients; stages I-II) and moderate-severe endometriosis (17 patients; stages III-IV). Exclusioncriteria were as follows: women (1) who were on hormonal medication; (2) who underwent an operation within 6 months; (3) who had other pelvic inflammatory disease. Selection and measurement of biomarkers After a widespread literature search, four plasma bio - markers were selected due to reported significant differ - ences in the plasma concentrations between women with and without endometriosis.Erythropoietin(Epo), interleukin (IL)-6, tumor necrosis factor α (TNF- α) and carbohydrate antigen (CA)-125, are suggested to play a role in the devel - opment of endometriosis due to either their angiogenic potential or as autocrine/paracrine factors or by encourag - ing vascularization or survival and proliferation of ectopic endometrial cells [13-17]. Plasma concentrations of Epo, IL-6 and TNF- α were determined by using commercially available ELISA kits (R&D Systems Inc., Minneapolis, MN, USA) according to the manufacturer’s instructions. Plasma concentrations of the CA-125 level were measured using Microparticle Enzyme Immunoassay (MEIA) Abbott AxSYM instrument (Abbott Diagnostics, Abbott Park, IL, USA). Inter-assay coefficient of variation for Epo, IL-6, TNF- α and CA-125 was < 10, 6.4, 3.5 and < 10%, respectively.Intra-assay coef - ficient for Epo, IL-6, TNF- α and CA-125 was 5.9, 4.2, 1.8 and < 10%, respectively. Statistical analysis Data were expressed as mean ±SD. Statistical analysis was performed using the Student’s t-test as appropriate, whereas means were compared among groups by a one-way analysis of variance (ANOVA). When homogeneity and normality of the samples were not appropriate, non-para - metric statistical methods (Kruskal-Wallisand Mann-Whit - ney tests) were applied for comparison. If a significant over - all difference was found in a one-way ANOVA or a Kruskal-Wallis test, post hoc Tukey or multiple compar - ison test were performed to identify any significant differ - ences among individual groups. Receiver operating char - acteristi c (ROC) curves were generated and confidence intervals for areas under ROC curves were calculated. Sta - tistical significancewas calculated using the Statistical Pack - age for Social Sciences (SPSS Inc., Chicago, IL, USA) and was considered to be P < 0.05.

Controls and endometriosis patients were matched for age (controls: 26.86 ±9.13, endometriosis patients: 31.24 ±7.24, p > 0.05). The plasma levels of IL-6, CA-125 were significantly higher, whereas the plasma levels of Epo and TNF- α were non-significantly decreased, in women with endometriosis compared with controls (IL-6: 46.70 ±27.08 vs. 27.08 ±16.29, p < 0.05; CA-125: 49.39 ±28.32 vs. 9.78 ±3.87, p 0.05; TNF- α: 60.22 ±9.11 vs. 65.40 ±9.86, p > 0.05, respectively) (Table 1). Table1. Comparisonof serum erythropoietin(Epo), interleukin (IL)-6, tumor necrosis factor α (TNF- α) and carbohydrate anti - gen (CA)-125values in women with and without endometriosis Endometriosis Controls p patients (n = 22) (n = 33) IL-6, mIU/ml 46.70 ±27.08 27.08 ±16.29 0.016 CA-125, U/ml 49.39 ±28.32 9.78 ±3.87 0.0001 TNF- α, pg/ml 26.26 ±9.31 65.40 ±9.86 0.051 Epo, mIU/ml 60.22 ±9.11 30.32 ±7.94 0.099 Note: Student’s t-test was used to evaluate statistically significant differences between the values. Non-invasive diagnosis of endometriosis based on a combined analysis of four plasma biomarkers Central European Journal of Immunology 2013; 38(2) 156 In women with minimal-mild endometriosis, the only significant difference was the increased plasma level of IL-6 in comparison to the controls (IL-6: 56.45 ±23.21 vs. 27.08 ±16.29, p < 0.05). In women with moderate-severe endometriosis, plasma levels of IL-6 and CA-125 were sig - nificantly increased, and those of Epo and TNF- α were sig - nificantly decreased, when compared with controls (IL-6: 44.53 ±31.07 vs. 27.08 ±16.29, p < 0.05; CA-125: 57.84 ±61.44 vs. 9.78 ±3.87, p < 0.05; Epo: 26.60 ±10.19 vs. 30.32 ±7.94, p < 0.05; TNF- α: 58.61 ±7.93 vs. 65.40 ±9.86, p < 0.05, respectively) (Table 2). Receiver operating characteris tic curve analysis was used to examine the diagnostic test performance of Epo, TNF- α, IL-6 and CA-125. The area under curve (AUC) for Epo, TNF- α, IL-6 and CA-125 were 0.280, 0.322, 0.729 and 0.864, respectively (Fig. 1).

The results of this study show that minimal-mild endometriosis patients displayed higher serum IL-6 levels, while moderate -severe endometriosis patients displayed higher serum Epo, TNF- α, IL-6 and CA-125 levels than other patients. Increased serum IL-6 levels observed in patients with minimal-mild endometriosis in our study are in line with findings of some reports [18-20] but depart from others that found no value for serum IL-6 in the diagnos is of endometriosis at any stage [21, 22] and from yet others that reported slightly elevated serum IL-6 levels in controls [23-25]. Additionally, it is worthwhile to note that serum IL-6 was the only marker that was constantly elevated in all our comparison groups (endometriosis group, minimal- mild endometriosis group and moderate-severeendometrio - sis group). Inflammatory markers such as TNF- α and Epo were slightly higher in the controls than in the endometrios is group. This observation may be explained by the possibil - ity that controls with adhesions, etc. had increased plasma concentra tions of inflammatory cytokines. Comparable TNF- α [20, 23-25] levels were previously reported in women with and without endometriosis. However, other investigators reported elevated levels of TNF- α and Epo in endometriosis patients compared with controls [13, 18, 26]. These discrepancies may be explained by differences in the study design (i.e. different inclusion criteria and different Table2. Comparison of serum erythropoietin (Epo), interleukin (IL)-6, tumor necrosis factor α (TNF- α) and carbohydrate anti - gen (CA)-125 values in women with and without endometriosis in relation to the stage of the disease Stage I-II endometriosis Stage III-IV endometriosis Controls (n = 22) p patients ( n = 16) patients ( n = 17) Mean ±SD Median Min-Max Mean ±SD Median Min-Max Mean ±SD Median Min-Max IL-6, mIU/ml 56.45 ±23.21 29.2 18.9-205.2 44.53 ±31.07 b 33.9 18.7-150.0 27.08 ±16.29 20.5 18.7-92.0 0.014 CA-125, U/ml 11.39 ±4.81 19.0 6.12-19.9 57.84 ±61.44 a,b 39.8 8.7-309.3 9.78 ±3.87 15.9 3.56-18.47 0.000 TNF- α, pg/ml 67.48 ±11.29 a 35.4 54.9-79.0 58.61 ±7.93 b 21.6 48.0-75.3 65.40 ±9.86 33.9 49.8-80.2 0.013 Epo, mIU/ml 24.75 ±3.39 25.8 19.1-28.0 26.60 ±10.19 b 22.6 18.0-60.0 30.32 ±7.94 35.3 21.1-46.5 0.021 ap < 0.05: stage I-II vs. stage III-IV bp < 0.05: stage III-IV vs. controls Note: Kruskal-Wallis test was used to calculate the overall p-values; Mann-Whitney test was used to perform the pairwise comparisons. 1.0 0.8 0.6 0.4 0.2 0.0 0.0 0.2 0.4 0.6 0.8 1.0 specificity sensitivity Epo IL-6 TNF- α CA-125

line Fig. 1. Receiver operating characteristic (ROC) curve for the identification of endometriosis for serum erythropoietin (Epo), interleukin (IL)-6, tumor necrosis factor α (TNF- α) and car - bohydrate antigen (CA)-125 Erbil Yagmur et al. Central European Journal of Immunology 2013; 38(2) 157 menstrual cycle phases) as well as genetic diversity. It is important to note that although CA-125 is the most exten - sively investigated and used biomarker of endometriosis [27], it is well established that CA-125 levels lack diag - nostic power as a single biomarker of endometriosis [28]. Regarding the role of cytokines in the pathogenesis of endometriosis, our findings are consistent with previous reports [8-10, 29, 30], which state that endometriosis is asso - ciated with the elevatedlevel of serum IL-6 and the elevated level of this cytokine correlates with a more advanced stage of the disease. However, we were unable to reveal any sta - tistically significant differences in the case of the other test - ed cytokines (Epo and TNF- α), which is also consistent with some studies [29, 30].

Establishing a non-invasive diagnostic test for endometriosis can have a radical impact on the patients’ quality of life and the accuracy of the treatment by poten - tially reducing the number of unnecessary laparoscopies. The results of our study show that progress ion of endometriosis is associated with the elevated level of serum IL-6. Undoubtedly, larger well-designed prospective stud - ies are urgently needed to determine the diagnostic poten - tial of cytokines like IL-6 in endometriosis. We certify that there is no conflict of interest with any financial organization regarding the material discussed in this manuscript.

1. Giudice LC, Kao LC (2004): Endometr iosis. Lancet 364: 1789-1799. 2. American College of Obstetricians and Gynecologists (2010): Practicebulletin no. 114: managementof endometriosis. Obstet Gynecol 116: 223-236. 3. Somigliana E, Vercellini P,Vigan o` P,et al. (2010): Non-inva - sive diagnosis of endometriosis: the goal or own goal? Hum Reprod 25: 1863-1868. 4. European Endometriosis Alliance. Endometriosis. April 2006: www. endometriosis.org. 5. Zondervan KT, Yudkin PL, Vessey MP, et al. (1999): Preva - lence and incidence of chronic pelvic pain in primary care: evidence from a national general practice database. Br J Obstet Gynaecol 106: 1149-1155. 6. Ballard KD, Lowton K, Wright JT (2006): What’s the delay? A qualitative study of women’s experience of reaching a diag - nosis of endometriosis. Fertil Steril 85: 1296-1301. 7. Gazvani R, Templeton A (2002): Peritoneal environment, cytokines and angiogenesis in the pathophysiology of endo- metriosis. Reproduction 123: 217-226. 8. Oral E, Olive DL, Arici A (1996): The peritoneal environment in endometriosis. Hum Reprod Update 2: 385-398. 9. Harada T, Iwabe T, Terakawa N (2001): Role of cytokines in endometriosis. Fertil Steril 76: 1-10. 10. Matarese G, De Placido G, Nikas Y ,Alviggi C (2003): Patho - genesis of endometriosis: natural immunity dysfunction or autoimmune disease? Trends Mol Med 9: 223-228. 11. Ulukus M, Arici A (2005): Immunology of endometriosi s. Minerva Ginecol 57: 237-248. 12. American Society of Reproductive Medicine (1997): Revised American Society for Reproductive Medicine classification of endometriosis: 1996. Fertil Steril 67: 817-821. 13. Matsuzaki S, Murakami T, Uehara S, et al. (2001): Erythro - poietin concentrations are elevated in the peritoneal fluid of women with endometriosis. Hum Reprod 16: 945-948. 14. Mihalyi A, Kyama CM, Simsa P, et al. (2005): Role of immunolog ic factors in the development of endometriosis: indications for treatment strategies. Therapy 2: 623-639. 15. Kyama CM, Overbergh L, Debrock S, et al. (2006): Increased peritonealand endometrial gene expression of biologicallyrel - evant cytokines and growth factors during menstrual phase in women with endometriosis. Fertil Steril 85: 1667-1675. 16. Debrock S, De Strooper B, Vander Perre S, et al. (2006): Tumour necrosis factor-alpha, interleukin-6 and interleukin- 8 do not promote adhesion of human endometrial epithelial cells to mesothelial cells in a quantitative in vitro model. Hum Reprod 21: 605-609. 17. Kyama CM, Mihalyi A, Simsa P, et al. (2008): Non-steroidal targets in the diagnosis and treatment of endometriosis. Curr Med Chem 15: 1006-1017. 18. Bedaiwy MA, Falcone T, Sharma RK, et al. (2002): Predic - tion of endometri osis with serum and periton eal markers: a prospective controlled trial. Hum Reprod 17: 426-431. 19. Martínez S, Garrido N, Coperias JL, et al. (2007): Serum inter - leukin-6 levels are elevated in women with minimal-mild endometriosis. Hum Reprod 22: 836-842. 20. Othman Eel-D, Hornung D, Salem HT, et al. (2008): Serum cytokines as biomarkers for nonsurgical prediction of endometrio sis. Eur J Obstet Gynecol Reprod Biol 137: 240-246. 21. D’Hooghe TM, Debrock S (2002): Endometriosis, retrograde menstruation and peritoneal inflammation in women and in baboons. Hum Reprod Update 8: 84-88. 22. Somigliana E, VVigan o` P,Tirelli AS, et al. (2004): Use of the concomitant serum dosage of CA 125, CA 19-9 and inter - leukin-6 to detect the presence of endometriosis. Results from a series of reproductive age women undergoing laparoscopic surgery for benign gynaecological conditions. Hum Reprod 19: 1871-1876. 23. Kalu E, Sumar N, Giannopoulos T, et al. (2007): Cytokine profiles in serum and peritoneal fluid from infertile women with and without endometriosis. J Obstet Gynaecol Res 33: 490-495. 24. Socolov R, Butureanu S, Angioni S, et al. (2011): The value of serological markers in the diagnosis and prognosis of endometriosis: a prospective case-control study. Eur J Obstet Gynecol Reprod Biol 154: 215-217. 25. V odolazkaia A, El-Aalamat Y ,Popovic D, et al. (2012): Eval - uation of a panel of 28 biomarkers for the non-invasive diag - nosis of endometriosis. Hum Reprod 27: 2698-2711. 26. Xavier P ,Belo L, Beires J, et al. (2006): Serum levels of VEGF and TNF-a and their association with C-reactive protein in patientswith endometriosis.ArchGynecolObstet 273:227-231. 27. Gupta S, Agarwal A, Sekhon L, et al. (2006): Serum and peri - toneal abnormalities in endometriosis: potential use as diag - nostic markers. Minerva Ginecol 58: 527-551. Non-invasive diagnosis of endometriosis based on a combined analysis of four plasma biomarkers Central European Journal of Immunology 2013; 38(2) 158 28. Kennedy S, Bergqvist A, Chapron C, et al.; on behalf of the ESHRE Special Interest Group for Endometriosis Endometri - um Guideline Development Group (2005): ESHRE guideline for the diagnosis and treatment of endometriosis. Hum Reprod 20: 2698-2704. 29. Milewski £, Barcz E, Dziunycz P, et al. (2008): Association of leptin with inflammatory cytokines and lymphocyte sub - populations in peritoneal fluid of patients with endometriosis. J Reprod Immunol 79: 111-117. 30. Dziunycz P,Milewski £, Radomski D, et al. (2009): Elevated ghrelin levels in the peritoneal fluid of patients with endometriosis: associations with vascular endothelial growth factor (VEGF) and inflammatory cytokines. Fertil Steril 92: 1844-1849. Erbil Yagmur et al.