Diagnostic Value of Serum Cancer Antigen 125, Carcinoembryonic Antigen, Cancer Antigen 19-9, Anti Müllerian Hormone, White Blood Cell Count, Platelet Count, and Neutrophil to Lymphocyte Ratio in Endometriosis: A Retrospective Study

Endometriosis is a benign disorder, which is dependent on estrogen hormone levels. The main feature of this disease is the travelling of endometrial tissue out of the uterine cavity and into the lower abdominal cavity, along with inflammatory responses ( 1 - 3 ). The endometriosis prevalence in women of reproductive age ranges from 2 to 10%, which can increase to 30-50% in females with infertility or persistent pelvic pain ( 2 , 4 - 6 ). Clinical manifestations of endometriosis include dysmenorrhea, dyspareunia, chronic pelvic pain, irregular menstrual cycles, pain during urination and defecation, blood in urine and feces, and chronic fatigue ( 2 , 5 ). Additionally, endometriosis is related to an augmented risk of obstetric complications, including infertility and spontaneous abortion ( 7 ). The precise pathogenesis of endometriosis is currently unknown ( 8 , 9 ). However, based on the results of some studies, some genetic and environmental factors, such as nutrition, body mass index (BMI), menarche age,and number of cesarean deliveries, may influence the probability of endometriosis onset ( 10 , 11 ). The gold standard diagnostic method for endometriosis is direct observation of endometriosis foci during laparoscopic surgery and histopathological examination of obtained tissue samples ( 12 ). Numerous studies have been conducted to discover non-invasive diagnostic measurements for endometriosis, including hormones, chemokines, angiogens, cytokines, inflammatory and autoimmune factors, imaging methods, and direct examination of uterine endometrial tissue. However, there is still no non-invasive method with acceptable specificity and sensitivity for accurate diagnosis of endometriosis before surgery ( 13 , 14 ). Cancer antigen 125 or CA-125 is a protein produced in coelomic epithelial cells present in the endometrium, ovary, fallopian tubes, and peritoneum. CA-125 is important in the diagnosis of ovarian cancers ( 15 ). According to some studies, the level of CA-125 increases in endometriosis patients. Nevertheless, its level varies during the menstrual phases and between different types of endometriosis. Therefore, the role of this biomarker in the diagnosis and screening of endometriosis, especially in mild to moderate types (grades 1 to 3), has not yet been proven and requires more investigation ( 16 , 17 ). An increase in the neutrophil to lymphocyte ratio (NLR) indicates a systemic inflammation that can be seen in various diseases ( 18 ). The use of NLR as both a diagnostic factor and a tool for monitoring treatment has been extensively studied. Its diagnostic efficacy has been demonstrated across a range of inflammatory medical conditions, including neonatal sepsis, diabetic peripheral neuropathy, Parkinson’s disease, allergic disorders, and sarcoidosis ( 19 - 23 ). Considering that endometriosis is an inflammatory disorder, higher values of NLR are expected in this disease. Nonetheless, previous research has not accurately recognized the role of this factor in the diagnosis or screening of endometriosis patients, thus, further studies are required ( 16 , 24 , 25 ). Several studies have investigated the diagnostic performance of carcinoembryonic antigen (CEA), anti Mullerian hormone (AMH), cancer antigen 19-9 (CA 19-9), and platelet and WBC count in endometriosis. However, the exact diagnostic cut-off value of these factors, as well as CA-125 and NLR, are not identified and currently, the available data are inconclusive ( 26 - 29 ). Since endometriosis frequently affects young women, offering more accessible and non-invasive diagnostic methods leads to earlier detection and treatment, preventing serious related complications, such as infertility. Regarding the importance of this issue and the limited available evidence in this field, particularly in the population of Iranian women, this study was carried out with the objective of investigating the role of CA-125, CA 19-9, CEA, WBC count, PLT count, AMH, and NLR in endometriosis diagnosis.

The total number of patients in the case and control groups was 230 and 116, respectively. Specifically, the control group consisted of 27 cases of serous cystadenomas, 26 patients with dermoid cyst, 15 cases of paratubal cysts, 12 cases of corpus luteum cysts, 11 patients with mucinous cystadenomas, 10 cases of follicular cysts, one individual with a Müllerian anomaly, one transmasculine individual who underwent total hysterectomy, and 13 healthy individuals. The mean age of individuals was 34.08 ± 7.83 years for the case group and 31.31 ± 11.97 for the control group. The mean BMI of the individuals in the control and case groups were 24.88 ± 4.22 and 24.97 ± 3.98 kg/m 2 , respectively. The most common manifestation among the patients of the case group was a persistent ovarian cyst (68.3%), followed by dysmenorrhea (63%), chronic pelvic pain (58.3%), dyspareunia (36.5%), abnormal uterine bleeding (AUB) (23.9%), gastrointestinal symptoms (22.59%), and urinary symptoms (7.8%). The most frequent manifestation in the control group was abdominal pain (57.8%), followed by incidental finding (22.5%), AUB (16.4%), and dysmenorrhea (3.4%). The period from the onset of symptoms to endometriosis diagnosis was 30.79 ± 39.04 (range: 0-240) months. Among the patients with endometriosis, 12.3% had a history of abortion, 24.7% had a history of infertility, and 6.2% experienced both conditions. Table 1 presents the mean and 95% CI of CA-125, CEA, CA 19-9, NLR, WBC count, platelet count, and AMH in both the case and control groups. The factors CA-125, CA 19-9, NLR, and WBC were significantly higher in the case group. The mean and 95% CI of CA-125, CEA, CA 19-9, NLR, WBC count, platelet count, and AMH in case and control groups CI; Confidence interval, CA-125; Cancer antigen 125, CEA; Carcinoembryonic antigen, NLR; Neutrophil to lymphocyte ratio, WBC; White blood cell, AMH; Anti Müllerian hormone, and *; The P<0.05 were considered significant. The diagnostic performance of CA-125, CEA, CA 19-9, NLR, WBC count, platelet count, AMH, and the combination of CA-125 and NLR were evaluated for the detection of endometriosis by analyzing the ROC curve ( Table 2 ). Diagnostic performance of CA-125, CEA, CA 19-9, NLR, WBC count, platelet count, AMH, and the combination of CA-125 and NLR using ROC curve analysis CA-125; Cancer antigen 125, CA 19-9; Cancer antigen 19-9, CEA; Carcinoembryonic antigen, NLR; Neutrophil to lymphocyte ratio, WBC; White blood cell, AMH; Anti Müllerian hormone, ROC; Receiver operating curve, AUC; Area under the curve, CI; Confidence interval, and *; The P<0.05 were considered significant. Diagnostic performance of CA-125 (purple curve), NLR (blue curve), and the combination of CA-125 and NLR (green curve). CA-125; Cancer antigen 125, NLR; Neutrophil to lymphocyte ratio, and ROC; Receiver operating curve. The combination of CA 125 and NLR showed the best diagnostic performance with an AUC of 0.903 (95% CI: 0.869-0.937). The AUC of the combination of CA 125 and NLR was greater than the AUC for each of CA-125 or NLR alone. The AUC for CA-125 was 0.896 (95% CI: 0.862-0.930), and the most appropriate cut-off value was 12.7 IU/mL with sensitivity of 93.9% and specificity of 60.9%. We also assessed the utility of CA-125 for endometriosis detection at the value of 35 (sensitivity=61.4%, specificity=98.3%). The AUC for NLR was 0.699 (95% CI: 0.637-0.760), and the ideal cut-off value was 1.5 with sensitivity of 83.4% and specificity of 52.6% ( Fig .1 ). The AUC for CA 19-9 was 0.692 (95% CI: 0.614-0.769). The ROC curve analysis identified the value of 13.2 IU/mL as the optimal cut-off, balancing both sensitivity (73%) and specificity (65%).

Based on the findings of the current study, the average values of CA 19-9, CA-125, WBC count, and NLR were significantly greater in the case group compared to the control group. Consistent with the findings of the present study, a recent meta-analysis demonstrated significantly greater NLR in endometriosis patients compared to both patients with other benign tumors and healthy subjects ( 18 ). Several studies revealed that CA 19-9 and CA-125 levels, and WBC count were remarkably higher in endometriosis patients than in patients with other benign conditions ( 30 - 33 ). However, the results of a study suggested that the level of CA-125 was not significantly different in endometriosis patients compared to patients with benign ovarian cysts ( 27 ). Contrary to our study, in which the platelet count did not differ between the control and case groups, a study discovered a higher level of platelet count in endometriosis patients ( 34 ). In the current study, AUC (0.896) and the most ideal cut-off value (12.7) were calculated (specificity=60.9%, sensitivity=93.9%). We also examined both specificity (98.3%) and sensitivity (61.4%) at the cut-off value of 35 for CA-125. The cut-off value equal to 35 was initially recognized for the detection of epithelial ovarian cancer. While this value is commonly used in the literature as the cut-off for CA-125 in diagnosing endometriosis, numerous studies have reported varying optimal cut-off levels for endometriosis diagnosis ( 26 ). In our findings, the AUC for the NLR was 0.699 and 1.5 was found to be the best cut-off value (specificity=52.6%, sensitivity=83.4%). However, a study conducted in South Korea suggested 2.01 as the optimal cut-off for NLR (specificity=60.1%, sensitivity=59.7%). In that study, the AUC for the NLR was 0.615 ( 33 ). Similar to NLR, other markers, such as platelet to lymphocyte ratio (PLR) and monocyte to lymphocyte ratio (MLR), were found to be higher in endometriosis patients compared with normal females ( 17 ). Our results revealed the CA 19-9 sensitivity (73%) and specificity (65%) at the cut-off of 13.2 IU/mL. However, the most favorable cut-off of this biomarker ranges from 7.5 to 37.0 in the literature ( 26 ). The highest diagnostic performance was achieved by combining both CA-125 and NLR values, resulting in an AUC of 0.903. Consistent with our findings, several studies have found that combined CA-125 and NLR leads to a greater diagnostic performance than each marker alone, when detecting endometriosis ( 33 , 35 ). In a similar study conducted in Turkey, it was found that the average CA-125 values, NLR, and combined biomarkers were remarkably greater in endometriosis patients. The optimal CA-125 cut-off point was 23.7 (specificity=81%, sensitivity=75%). For NLR, a cut-off value of 1.89 resulted in 70% sensitivity and 74% specificity. The best cut-off value for combined biomarkers was 41.0 (sensitivity=80%, specificity=82%). Additionally, positive associations between NLR and CA-125, and also between NLR and endometriosis score were discovered ( 31 ). A study was carried out in 2021 with the aim of differentiating benign adnexal cysts from endometrioma in adolescents and pre-menopausal women ( 29 ). Based on the findings of this study, females with ovarian endometriomas had significantly greater levels of CA-125, platelet count, average serum PLR, and plateletcrit. However, the findings of that study showed that hematological markers do not have the ability to differentiate ovarian endometrioma from other benign cysts in any of the two age groups of young and older women ( 29 ).

Currently, there is no simple non-invasive diagnostic method, accurate enough for a definite diagnosis of endometriosis. Different studies have measured various biomarkers to improve endometriosis diagnosis and screening. However, the optimal cut-off values vary in the literature for each marker. Our study showed that the measurement of NLR and CA-125 may serve as convenient diagnostic methods. However, the combination of CA-125 and NLR had higher specificity and sensitivity. More studies are necessary to discover the optimal cutoff points for each biomarker, and to determine the most appropriate combination, particularly with regard to the combined value of NLR and CA-125.

We retrospectively investigated the recorded data of 346 patients who underwent elective surgery from March 2016 to February 2023 at the obstetrics and gynecology department of Shohada Tajrish Hospital, Tehran, Iran. This study enrolled all patients who had complete medical data, including the report of pathological evaluation after the surgery, as well as the values of complete blood count (CBC), CA-125, CA 19-9, CEA, and AMH. We excluded the patients who had received any treatment for endometriosis or were experiencing any gynecological conditions (such as uterine leiomyoma, adenomyosis, uterine polyp, endometrial hyperplasia, pelvic inflammatory disease, and malignant ovarian tumors). Additionally, we excluded pregnant or breastfeeding patients and those with a history of chronic diseases (including infectious, autoimmune, and endocrine diseases), and malignancies. We have extracted socio-demographic, clinical, and paraclinical data of all included individuals from the hospital data system as follows: age, occupation, weight, height, BMI, signs and symptoms, past obstetric history (gravidity, parity, abortion, and infertility), menstrual phase, levels of AMH, CA-125, CEA, CA 19-9, WBC and platelet count, NLR, histopathological report, and the interval from the symptom onset to diagnosis. The case group consisted of 230 patients with pathological confirmation of endometriosis. The control group included 116 women with pathological confirmation of benign tumors of the ovary or other benign conditions such as Müllerian anomalies. In this study, data analysis was done using SPSS statistical software (version 21, IBM Corp., USA). We used range, mean ± standard deviation (SD), 95% confidence intervals (95% CI), and range for the quantitative variables. Qualitative variables were described using the frequency (%). We verified the normality of quantitative variables by Kolmogorov-Smirnov analysis. To compare the means between case and control groups, an independent sample t test was used for normally distributed data, while the Mann-Whitney U test was performed for nonparametric data. The area under the curve (AUC) was calculated for each hematological marker (NLR, WBC and platelet count, CA-125, AMH, CEA, CA 19-9, and the combination of CA-125 and NLR) using the receiver operating curve (ROC) analysis. A statistically significant P value was set at <0.05. An ethical approval was taken prior to data collection from the Ethical Committee of Shahid Beheshti University of Medical Sciences (IR.SBMU.RETECH.REC.1402.131). Patients’ confidentiality was maintained throughout the study. Patients’ consent was obtained after the objectives and steps of the study were clearly explained to them.