Integrating Thyroid Autoantibodies With CA125 Enhances Diagnostic Precision for Ovarian Endometriosis: A Retrospective Study of 885 Patients

Yufeng Wang: conceptualization, methodology, software, data curation, investigation, formal analysis, visualization, writing – original draft. Zhiruo Qiu: conceptualization, methodology, software, data curation, validation, supervision, visualization, formal analysis, investigation, writing – original draft. Yuan Yao: conceptualization, writing – review and editing, supervision. Jie Tan: conceptualization, supervision, project administration, writing – review and editing, validation. Xuan Che: conceptualization, methodology, writing – review and editing, funding acquisition, project administration, supervision, resources.

The authors have nothing to report.

The retrospective study protocol was approved by the institutional review board of our hospital. In this study, a total of 885 female patients undergoing gynecological surgery for ovarian cysts performed between January 2019 and December 2023 were randomly included. Among these patients, 441 were diagnosed with ovarian endometriosis via postoperative pathology (case group). Four hundred and forty‐four patients were diagnosed with benign ovarian cysts (control group). The control group included 198 cases of mature cystic teratoma, 110 cases of simple cysts, 85 cases of mucinous cystadenoma, 42 cases of serous cystadenoma, 6 cases of theca cell fibroma, 1 case of sclerosing stromal tumor, 1 case of mature cystic teratoma combined with theca cell fibroma, and 1 case of mature cystic teratoma combined with mucinous cystadenoma. The clinical data of all patients, encompassing general conditions, preoperative evaluation, surgical process, and postoperative recovery, were extracted from medical records and follow‐up records. The clinical data of all patients, including general clinical characteristics, preoperative examination test results, and descriptions of surgery, were extracted from medical records and subsequent records (Table  S1 ). Inclusion criteria: (1) Patients diagnosed with ovarian cysts based on preoperative imaging, such as ultrasound or MRI. (2) Preoperative serum CA125 levels and comprehensive thyroid function tests. (3) Undergoing standardized gynecologic laparoscopic ovarian cyst removal surgery. (4) A definitive postoperative pathological diagnosis confirming the presence of ovarian cysts. Exclusion criteria: (1) Patients who underwent thyroid surgery or received medical treatment for abnormal thyroid function before the operation. (2) Other diseases associated with thyroid dysfunction, such as endometrial cancer. (3) Postoperative pathological diagnosis of ovarian borderline tumors or malignant tumors. (4) Inflammatory diseases, for instance, adenomyosis or pelvic inflammatory disease, co‐existing with ovarian cysts. (5) Patients with autoimmune diseases, hematologic disorders, severe cardiopulmonary conditions, or who are pregnant or postmenopausal. Outcome Measures: (1) Comparison of clinical characteristics: Clinical characteristic measures include patient age, age at menarche, reproductive BMI, parity, delivery, location and size of cyst, dysmenorrhea, and preoperative comorbidity. (2) Thyroid function markers and Serum CA125: Preoperative serum levels of thyroid function markers and cancer antigen 125 (CA125) were measured and compared between the two study groups. The thyroid function panel included the following markers: free triiodothyronine (FT3), free thyroxine (FT4), total triiodothyronine (T3), total thyroxine (T4), thyroid‐stimulating hormone (TSH), thyroid peroxidase antibody (TPOAb), and thyroglobulin antibody (TGAb). Serum samples for thyroid function profiling and CA125 quantification were obtained preoperatively under standardized conditions. Venous blood was drawn following ≥ 8‐h fasting within 7 days preceding surgery. To minimize hormonal variability associated with the menstrual cycle, all blood samples were collected during the non‐menstrual phase. TPOAb was considered abnormal when ≥ 5.61 IU/mL. TGAb was considered abnormal when ≥ 4.11 IU/mL. CA‐125 level was considered abnormal when > 35 U/mL. (3) Prevalence of autoimmune thyroid disease: The prevalence of autoimmune thyroid disease was compared between patients with ovarian endometriosis and those with other benign ovarian tumors. (4) Multifactorial analysis of risk factors: Logistic regression was performed to identify independent risk factors associated with ovarian endometriotic cysts. (5) Diagnostic performance of biomarkers: The diagnostic value of TgAb, TPOAb, and CA125, both individually and in combination, was evaluated in distinguishing ovarian endometriotic cysts from other benign ovarian tumors. Data were analyzed using SPSS version 25.0. Continuous variables with a normal distribution are presented as means ± standard deviations, with comparisons between two groups conducted using independent t ‐tests. For non‐normally distributed continuous variables, values are presented as medians with interquartile ranges, and group comparisons were performed using the Wilcoxon rank‐sum test. Categorical variables were analyzed using the chi‐square test or Fisher's exact test, depending on the specific data characteristics. Multivariate logistic regression analysis was employed to identify independent factors influencing the outcomes. A logistic regression model integrating multiple factors was constructed, and its diagnostic performance was evaluated using receiver operating characteristic (ROC) curve analysis. Graphs were generated with R version 4.3.2. Statistical significance was set at p  < 0.05.

A total of 885 patients were included in the study, with 441 diagnosed with ovarian endometriosis based on postoperative pathology (case group) and 444 diagnosed with other benign ovarian tumors (control group). No significant differences were observed between the two groups in terms of age, age at menarche, body mass index (BMI), reproductive history, maximum cyst diameter, or the prevalence of hypertension and diabetes ( p  > 0.05). The clinical characteristics of the patients are presented in Table  1 . Clinical characteristics between ovarian endometriosis and control groups. The incidence of dysmenorrhea was significantly higher in the ovarian endometriosis group compared to the control group (49.7% vs. 15.3%, p  < 0.001). Additionally, the proportion of bilateral ovarian cysts was significantly higher in patients with ovarian endometriosis than in the control group (26.1% vs. 8.8%, p  < 0.001). Furthermore, the prevalence of endometrial polyps and uterine fibroids was significantly higher in patients with ovarian endometriosis compared to the control group (21.9% vs. 15.9%, p  = 0.015; 35.6% vs. 26.2%, p  = 0.003, respectively). Ovarian endometriosis is associated with higher rates of dysmenorrhea, bilateral ovarian cysts, endometrial polyps, and uterine fibroids. We compared biomarker levels between the case group (ovarian endometriosis patients) and the control group. The serum levels of CA125, thyroid function markers (FT3, FT4, T3, T4, TSH), and thyroid autoantibodies (TPOAb, TGAb) were compared between ovarian endometriosis patients and controls. The ovarian endometriosis group demonstrated significantly higher incidence rates of abnormal TPOAb (56.9% vs. 6.5%; p  < 0.001), TGAb (68.5% vs. 12.4%; p  < 0.001), and CA125 (75.3% vs. 10.8%; p  < 0.001) compared to the control group, indicating these biomarkers' potential association with endometriotic cyst susceptibility. Complete comparative data are presented in Table  2 and Figure  1 . Comparison of ovarian endometriosis and control groups of thyroid function indicators and abnormal situations of CA125. Note: Data are presented as number (percentage) for categorical variables. Chi‐square test was used for categorical variables. Abbreviations: CA125, carbohydrate antigen 125; FT3, free triiodothyronine; FT4, free thyroxine; T3, triiodothyronine; T4, thyroxine; TGAb, thyroglobulin antibody; TPOAb, thyroid peroxidase antibody; TSH, thyroid‐stimulating hormone. TPOAb was considered abnormal when ≥ 5.61 IU/mL. TGAb was considered abnormal when ≥ 4.11 IU/mL. CA‐125 level was considered abnormal when > 35 U/mL. FT3 was considered abnormal when  6.01 pmol/L. FT4 was considered abnormal when  19.24 pmol/L. T3 was considered abnormal when  2.33 nmol/L. T4 was considered abnormal when  150.84 nmol/L. TSH was considered abnormal when  4.94 μIU/mL. Fisher's exact test was applied when expected cell counts were < 5. Thyroid function indicators and CA125 levels in the two groups *** p  < 0.01; ns, no statistical significance. Normal reference value: Free triiodothyronine (FT3): 2.43–6.01 pmol/L; Free thyroxine (FT4): 9.1–19.24 pmol/L; Thyroid‐stimulating hormone (TSH): 0.35–4.94 μIU/mL; Total triiodothyronine (T3): 0.98–2.33 nmol/L; Total thyroxine (T4): 62.68–150.84 nmol/L; Thyroid peroxidase antibody (TPOAb) < 5.61 IU/mL; Thyroglobulin antibody (TGAb) < 4.11 IU/mL; Cancer antigen 125 (CA125): 0–35 U/mL. In contrast, no statistically significant intergroup differences were observed in thyroid hormone profiles, including FT3, T3, FT4, T4, and TSH ( p  > 0.05), suggesting these parameters are not directly linked to ovarian endometriosis pathogenesis. These results demonstrate that elevated serum levels of CA125, TPOAb, and TGAb hold significant diagnostic value in distinguishing ovarian endometriosis from non‐endometriosis populations, with aberrant expression of these biomarkers strongly correlating with disease presence. We further investigated the relationship between ovarian endometriosis and autoimmune thyroid disorders. Autoimmune thyroid disease comprises organ‐specific disorders characterized by antibodies against thyroid antigens. These conditions feature elevated thyroid autoantibodies (TPOAb, TGAb, and TRAb), which are diagnostic regardless of whether thyroid hormone abnormalities (FT3, FT4, or TSH) are present [ 9 ]. Comprehensive assessment of thyroid parameters (FT3, FT4, T3, T4, TSH) and autoantibodies (TPOAb, TGAb) demonstrated that patients with ovarian endometriosis exhibited a substantially higher incidence of concurrent autoimmune thyroid disease compared to controls (51% vs. 15%, p  < 0.001). See Figure  2 . This marked association suggests potential shared immunological mechanisms underlying both conditions and highlights the importance of thyroid evaluation in patients presenting with ovarian endometriosis. Prevalence of autoimmune thyroid diseases in ovarian endometriosis versus control groups. Subsequently, multivariate logistic regression was performed on variables demonstrating statistical significance in univariate analysis (CA125, TPOAb, TGAb). After adjusting for clinical confounders—including dysmenorrhea, endometrial polyps, uterine fibroids, and lesion location—the model identified elevated CA125, TPOAb, and TGAb levels as robust predictors of ovarian endometriosis: TPOAb: OR = 2.63 (95% CI: 1.46–4.77, p  = 0.001), TGAb: OR = 4.03 (95% CI: 2.48–6.60, p  < 0.001), CA125: OR = 18.4 (95% CI: 12.5–27.6, p  < 0.001). These associations persisted after additional adjustment for non‐normally distributed covariates: TPOAb: OR = 2.59 (95% CI: 1.41–4.81, p  = 0.002), TGAb: OR = 3.33 (95% CI: 2.01–5.57, p  < 0.001), CA125: OR = 15.4 (95% CI: 10.3–23.4, p  < 0.001). Sensitivity analyses accounting for distributional assumptions confirmed the stability of these biomarkers as statistically significant discriminators between cases and controls ( p  < 0.05). Collectively, these results establish CA125, TPOAb, and TGAb as independent risk factors for ovarian endometriosis (see Table  3 ). Multivariate analysis of biomarkers as independent risk factors for ovarian endometriosis. Abbreviations: CA125, carbohydrate antigen 125; OR, odds ratio; TGAb, thyroglobulin antibody; TPOAb, thyroid peroxidase antibody. Correct dysmenorrhea, endometrial polyps, uterine fibroids, and the location of cysts. To evaluate the diagnostic utility of preoperative TGAb, TPOAb, and CA125 levels in identifying ovarian endometriotic cysts, receiver operating characteristic (ROC) curve analyses were performed. The area under the curve (AUC) values for TGAb, TPOAb, and CA125 were 0.793 (95% CI: 0.763–0.824), 0.736 (95% CI: 0.702–0.769), and 0.895 (95% CI: 0.873–0.916), respectively (all p  < 0.001), indicating significant discriminatory capacity for all three biomarkers. Upon optimization of the ROC curves using Youden's index, the optimal diagnostic cutoff values were established as follows: TGAb: 4.985 IU/mL, TPOAb: 5.735 IU/mL, and CA125: 28.885 U/mL. These threshold values represent the optimal balance between sensitivity and specificity for each marker in the differential diagnosis of ovarian endometriotic cysts. Complete diagnostic performance metrics are presented in Table  3 . Finally, we conducted multivariate logistic regression analysis to evaluate the diagnostic performance of combined serum markers CA125, TPOAb, and TgAb for ovarian pathological detection. Receiver operating characteristic (ROC) curves were generated to assess diagnostic efficacy. The combined diagnostic model incorporating all three biomarkers (CA125, TPOAb, and TgAb) demonstrated superior diagnostic performance, yielding the highest area under the curve (AUC) of 0.924 (95% CI: 0.906–0.942, p  < 0.05). This combined approach achieved a sensitivity of 88.9% and specificity of 84.5% (Figure  3 , Table  4 ). ROC curves for TPOAb, TGAb, CA125, and combined testing. Analysis of the diagnostic value of TgAb, TPOAb, and CA125 alone and in combination for ovarian endometriosis. Abbreviations: 95% CI, 95% confidence interval; AUC, area under the curve; CA125, carbohydrate antigen 125; TGAb, thyroglobulin antibody; TPOAb, thyroid peroxidase antibody. Furthermore, in the differential diagnosis of ovarian endometriotic cysts from other benign ovarian cysts, the combined model exhibited significantly enhanced sensitivity compared to any individual marker alone ( p  < 0.01). This finding suggests that the integration of thyroid autoantibodies (TPOAb and TGAb) with the conventional ovarian cancer marker CA125 substantially improves diagnostic accuracy for distinguishing specific ovarian pathologies.

In recent years, emerging evidence indicates that endometriosis has been increasingly linked to systemic autoimmune diseases, including systemic lupus erythematosus, rheumatoid arthritis, and multiple sclerosis, with shared mechanisms such as chronic inflammation, immune dysregulation, and disturbances in local and systemic immune environments [ 10 , 11 , 12 , 13 ]. Autoimmune thyroid disease, particularly Hashimoto's thyroiditis and thyroid autoantibodies, also shows a consistently higher prevalence in women with endometriosis [ 14 ], likely mediated by common inflammatory pathways and genetic susceptibility [ 2 , 15 ]. In this study, we demonstrate that ovarian endometriosis is strongly associated with elevated serum CA125, TPOAb, and TGAb, which act as independent risk factors and diagnostic markers. These findings reinforce the role of thyroid autoimmunity in endometriosis pathogenesis while providing clinical value through a novel multimarker diagnostic model that may improve early detection and patient management. The elevated TPOAb and TGAb levels observed in ovarian endometriosis patients (56.9% and 68.5%, respectively) support the hypothesis that thyroid autoantibodies contribute to immune dysregulation, enabling ectopic endometrial cell survival. Mechanistically, TPOAb/TGAb may impair natural killer cell cytotoxicity and promote inflammatory cytokine release, fostering a permissive microenvironment for endometrial cell adhesion and angiogenesis [ 16 , 17 ]. This aligns with prior studies showing autoimmune thyroid disease (AITD) prevalence in endometriosis cohorts [ 18 ], though our data uniquely quantify thyroid autoantibodies as independent predictors (TPOAb: OR = 2.63; TGAb: OR = 4.03) after adjusting for confounders. CA125 remains the cornerstone biomarker for endometriosis, yet its limited specificity underscores the need for complementary markers. Our ROC analysis confirmed CA125's diagnostic value (AUC = 0.895) at a cutoff of 28.885 U/mL (sensitivity 82.5%, specificity 86.6%), consistent with prior reports [ 19 , 20 ]. However, integrating TPOAb and TGAb significantly improved diagnostic performance (AUC = 0.924, sensitivity 88.9%, specificity 84.5%), addressing a critical gap in noninvasive differentiation of endometriomas from other benign ovarian tumors. The co‐occurrence of autoimmune thyroid disease and ovarian endometriosis (51% vs. 15% in controls) further implicates shared immunological pathways. Thyroid autoantibodies may exacerbate pelvic inflammation via molecular mimicry or cross‐reactivity with endometrial antigens [ 21 , 22 ], and further mechanistic studies are warranted. Clinically, preoperative detection of TPOAb > 5.735 IU/mL or TGAb > 4.985 IU/mL—particularly alongside elevated CA125—should raise suspicion for endometriosis, aiding surgical decision‐making in radiologically ambiguous cases. This study's retrospective single‐center design limits generalizability, and our findings specifically apply to ovarian endometriosis. Future multicenter prospective studies should be proposed, such as including more types of endometriosis patients and ovarian endometriosis phenotypes (e.g., rASRM stages, adhesion severity). Furthermore, we suggest a potential overlap in immune dysregulation, which aligns with emerging evidence linking endometriosis to systemic autoimmunity, and more mechanistic studies (e.g., in vitro models, cytokine profiling) are needed to clarify whether thyroid autoimmunity drives endometriosis pathogenesis. Our research found that the synergistic diagnostic value of CA125, TPOAb, and TGAb provides a practical tool for reducing diagnostic delays and guiding clinical management. These findings support the incorporation of thyroid autoimmunity screening into standard diagnostic procedures to optimize risk stratification and personalized management pathways for symptomatic patients. Meanwhile, thyroid autoimmunity is a previously underrecognized influencing factor for ovarian endometriosis. In‐depth mechanistic studies are needed to understand the complex interactions between autoimmune diseases and endometriosis.

The authors have nothing to report.

Endometriosis, a chronic inflammatory condition affecting 10%–15% of reproductive‐aged women, is characterized by the ectopic implantation of endometrial‐like tissue and is associated with pelvic pain, infertility, and diminished quality of life [ 1 , 2 ]. Despite its clinical prevalence, the pathogenesis remains incompletely understood. While Sampson's retrograde menstruation theory posits that retrograde menstrual flow facilitates pelvic implantation, this mechanism alone cannot explain why only 10% of women with retrograde menstruation develop endometriosis or the occurrence of extra‐pelvic lesions [ 3 , 4 ]. Current diagnostic strategies rely heavily on serum carbohydrate antigen 125 (CA125), yet its limited specificity underscores the need for complementary biomarkers. Emerging evidence underscores a multifactorial etiology involving aberrant immunomodulation, hormonal dysregulation, and autoimmune dysfunctions [ 5 , 6 ]. Notably, thyroid hormones modulate estrogen metabolism in endometrial cells, promoting proliferation and survival of ectopic lesions, while thyroid autoantibodies (e.g., thyroid peroxidase antibody [TPOAb], thyroglobulin antibody [TGAb]) may exacerbate inflammatory responses critical to endometriosis progression [ 7 , 8 ]. However, existing studies lack rigorous validation in large populations and fail to evaluate the synergistic diagnostic utility of combining thyroid autoantibodies with CA125. Furthermore, epidemiological data linking thyroid autoimmunity to ovarian endometriotic cysts remain observational, underpowered, and insufficient to establish causal relationships. Critical gaps persist in understanding immune‐endocrine crosstalk in endometriosis pathogenesis and its translation to clinical practice. Prior research predominantly examines isolated biomarkers, neglecting combinatorial models that address diagnostic heterogeneity. No studies have systematically quantified the independent and additive predictive value of TPOAb/TGAb alongside CA125 through multivariate modeling or optimized diagnostic thresholds via receiver operating characteristic (ROC) analysis. This retrospective case–control study analyzed 885 patients (441 ovarian endometriosis cases, 444 benign ovarian tumor controls) to (1) compare clinical profiles, (2) evaluate thyroid autoantibodies (TPOAb/TGAb), and (3) establish a composite diagnostic model integrating CA125, TPOAb, and TGAb. Using multivariate regression and ROC curve optimization, we provide evidence for integrating thyroid immune evaluation into endometriosis management and advancing precision diagnostics.

The authors declare no conflicts of interest.

Table S1: Patient characteristics by study group.