Distinct endometriosis involvement confers divergent oncologic outcomes in ovarian clear cell carcinoma

Objective To evaluate the clinicopathologic characteristics and survival outcomes of ovarian clear cell carcinoma (OCCC) patients with different endometriosis statuses.

This retrospective study included OCCC patients diagnosed between 2012 and 2021, classified into three groups based on the Sampson and Scott criteria: Without (no endometriosis), Arising (OCCC arising from endometriosis), and Coexisting (OCCC coexisting with endometriosis). Clinical and pathological characteristics were compared across groups, and survival outcomes were analyzed using Kaplan–Meier methods. Prognostic factors for progression-free survival (PFS) and overall survival (OS) were identified through univariate and multivariate analyses.

Among 242 patients, 53.7% were in the Without group, 29.3% in the Arising group, and 16.9% in the Coexisting group. The Arising group had the highest prevalence of early FIGO stage disease (91.6%) compared to the Coexisting (75.6%, p = 0.041) and Without (67.7%, p = 0.000) groups. Lymph-node metastasis was significantly lower in the Arising group (2.8%) than in the Coexisting (19.5%, p = 0.010) and Without (10%, p = 0.011) groups. Notably, the Arising group demonstrated unique atypical endometriosis features. In univariate analysis, the presence of endometriosis (either arising from or coexisting with endometriosis) was associated with improved PFS (p = 0.004 and p = 0.009, respectively); however, multivariate analysis confirms only coexisting with endometriosis as an independent factor (HR: 0.11, 95% CI: 0.01–0.84). For OS, the Arising group demonstrated the most significant benefit, with a 5-year OS of 92.4% compared to the Coexist- ing group (83.9%, p = 0.293) and the Without group (62.6%, p = 0.023). Multivariate analysis identified only FIGO stage (HR: 5.89, 95% CI: 2.06–16.82) as an independent prognostic factor for OS, while endometriosis did not reach statistical significance (HR: 0.62, 95% CI: 0.26–1.53).

Classifying OCCC with endometriosis statuses reveals distinct prognostic patterns. Coexisting with endome- triosis positively impacts PFS, while the Arising subgroup shows the most significant OS benefit but may be confounded with other factors.

Clinicopathologic · Sampson and Scott criteria · Atypical endometriosis · Survival outcomes · Prognostic factors First author: Jie Deng. * Tianjin Yi [email protected] 1 Department of Obstetrics and Gynecology, Key Laboratory of Birth Defects and Related Diseases of Women and Children of Ministry of Education, West China Second University Hospital, Sichuan University, Chengdu 610041, People’s Republic of China 2 The Third People’s Hospital of Xindu District, Chengdu 610041, People’s Republic of China 3 West China School of Medicine, Sichuan University, Chengdu 610041, People’s Republic of China 4 Department of Pathology, West China Second University Hospital, Sichuan University, Chengdu 610041, People’s Republic of China 274 Archives of Gynecology and Obstetrics (2025) 312:273–286 What does this study add to the clinical work? Applying the Sampson and Scott criteria to classify ovarian clear cell carcinoma patients by types of endometriosis involvement reveals distinct clinico- pathologic characteristics and survival outcomes. Endometriosis presence, particularly in the coex- isting subgroup, is an independent favorable fac- tor for progression-free survival, while the arising subgroup shows improved 5-year overall survival, emphasizing the prognostic value of detailed endo- metriosis classification in patient management.

Endometriosis is a chronic, estrogen-dependent, inflamma- tory disorder characterized by ectopic endometrial glands and stroma, affecting approximately 10% of reproductive- aged women [1 ]. Despite its benign nature, the chronic inflammation and immune dysregulation in endometriosis may increase the risk of infertility and cancer [2 , 3]. Epi- demiological evidence strongly links endometriosis with epithelial ovarian cancer (EOC), with particularly ovarian clear cell carcinoma (OCCC; OR = 3.05) and ovarian endo- metrioid carcinoma (OR = 2.04) [4]. OCCC, a histological subtype of EOC, is distinct in its clinical presentation, histo- pathology, and genetics [5]. Although relatively uncommon, OCCC accounts for approximately 5% to 12% of all EOC cases and up to 30% in Asian populations [6]. Although early stage OCCC patients have favorable outcomes, advanced or recurrent disease responds poorly to chemotherapy, resulting in worse prognoses compared to high-grade serous carci- noma [7, 8]. This highlights the urgent need for improved etiological understanding and prognostic markers. Endometriosis has been reported in 25–58% of OCCC cases [9 –11]. Some studies suggest better prognoses in endometriosis-associated OCCC, potentially due to younger age and earlier disease stage at diagnosis [12–14]. However, other studies have failed to demonstrate significant survival differences between these groups [15, 16] or to establish endometriosis as an independent prognostic factor [17–19]. These inconsistencies leave the prognostic role of endome- triosis in OCCC patients open to debate. Immunohistochemistry (IHC) plays a crucial role in accurately diagnosing OCCC and distinguishing it from other ovarian or metastatic carcinomas [20, 21]. OCCC typically demonstrates high expression of Paired box gene 8 (PAX-8) (approximately 95%), a reliable marker for ovarian origin, along with strong positivity for Hepato - cyte nuclear factor-1 beta (HNF-1β) (92–100%) and Napsin A (83–100%), making these markers particularly useful for accurate diagnosis. Cytokeratin 7 (CK7) is consistently positive in nearly all cases (close to 100%), whereas Cytokeratin 20 (CK20) is usually negative, effec- tively differentiating OCCC from metastatic colorectal carcinoma, which typically expresses CK20. Wilms' tumor 1 (WT1) negativity (0–5% positivity) distinguishes OCCC from high-grade serous carcinoma, which strongly expresses WT1. Estrogen receptor (ER) and progesterone receptor (PR) are predominantly negative in OCCC, with ER positivity reported in approximately 10% of cases and PR positivity even less frequently (approximately 5%). The p53 protein usually exhibits wild-type expres- sion patterns in OCCC; however, aberrant p53 staining occurs in approximately 10–24% of cases. In contrast, endometriosis tissues exhibit significantly elevated estro- gen receptor beta (ER-β)—over 100-fold higher compared to normal endometrial tissue—and reduced PR expres- sion [22]. These distinct hormonal receptor patterns in endometriosis likely contribute to its pathogenesis and resistance to treatment. Since Sampson's first description of malignancy aris- ing from endometriosis in 1925 [23]and Scott's subse- quent histological criteria [24], studies have broadly categorized OCCC into cases with or without endome- triosis. In prior studies, OCCC patients were typically categorized into two broad groups: those with and without endometriosis. In these studies, endometriosis-associated OCCC included the following conditions [14, 24–26]: (1) histological identification of ovarian cancer and endo- metriosis in the same ovary, with evidence of malignant transition from endometriosis [e.g., atypical endometrio - sis (A-EMS)]; (2) endometriosis in one ovary and ovarian cancer in the contralateral ovary; and (3) coincidental identification of ovarian cancer in any ovary or pelvic endometriosis. The first condition adheres to the Samp- son and Scott’s criteria and is defined as “OCCC aris - ing from endometriosis”, while the latter two conditions represent the concurrent existence of endometriosis and OCCC in the same patient (“OCCC coexisting with endo - metriosis”). Recent histological evidence suggests that “Arising” and “Coexisting” cases may represent different stages along a malignant continuum [27]. To better understand this relationship, we refined the classification of OCCC into three distinct groups: OCCC arising from endometriosis (“Arising”), OCCC coex - isting with endometriosis (“Coexisting”), and OCCC without endometriosis ("Without"). To date, no studies 275Archives of Gynecology and Obstetrics (2025) 312:273–286 have simultaneously analyzed the histological and clini- cal characteristics and outcomes of these three groups. Our study aims to elucidate how distinct endometriosis involvement patterns influence clinicopathological fea- tures and long-term oncological outcomes in OCCC.

Study design and patient population This retrospective study analyzed consecutive patients newly diagnosed with OCCC and treated at our institu- tion from February 2012 to February 2021. Clinical data were systematically collected and reviewed for all eligible patients. Demographic, clinical, and pathological data col- lected included age at diagnosis, height, weight, age at menarche, FIGO stage, serum tumor markers, tumor grade, histological findings, and treatment details. Inclusion and exclusion criteria Inclusion criteria were as follows: (1) Patients who under - went comprehensive staging and debulking surgery at our institution. (2) OCCC diagnosis confirmed by histological slide reexamination by two experienced pathologists. (3) Complete clinicopathological data and follow-up informa- tion available. Exclusion criteria included patients with histologically confirmed invasive epithelial ovarian cancers other than OCCC, cases of OCCC mixed with other histology sub- types, treated with fertility preservation surgery and those with incomplete clinical data. Histopathological assessment OCCC and endometriosis diagnoses were confirmed by two gynecological pathologists. Endometriosis was identified based on the presence of ectopic endometrial glands and stroma. According to Sampson and Scott’s criteria [23, 24], patients were categorized into three groups: (1) Coexisting group: Patients with histologically confirmed endometriosis concurrent with OCCC (in the same or contralateral ovary or other sites) but without evidence of malignant transition. (2) Arising group: To accurately define OCCC arising from endometriosis, cases in the Arising group were required to meet four criteria: A. Histopathological analysis confirmed the presence of both benign and neoplastic endometrial tissues within the tumors. B. Histological findings were compatible with an endometrial origin. C. Comprehensive clinical and imaging assessments ensured the exclusion of other primary tumor sites. D. A morphologic demonstra- tion of a continuum between benign and malignant epithe- lium [with atypical endometriosis (A-EMS) identified as the precancerous lesion [28]]. (3) Without group: Patients with OCCC and no endometriosis on final histological examina- tion or prior history of endometriosis. Prognostic analysis The primary endpoints were overall survival (OS) and progression-free survival (PFS) of OCCC patients, strati- fied by endometriosis involvement. Subgroup analysis was performed according to the Coexisting, Arising, and Without groups. Chemotherapy resistance was defined as tumor pro- gression or recurrence within six months of the last chemo- therapy. OS was defined as the time from initial treatment to death or last follow-up, while PFS was defined as the time from initial treatment to tumor recurrence, determined by histopathological evidence or new imaging. Surgical procedures All patients underwent surgical treatment at our institution following FIGO staging guidelines. Comprehensive stag - ing surgery was conducted for stage I–II, while debulking surgery was performed for stage III–IV cases. Main pro- cedures included total hysterectomy, bilateral salpingo- oophorectomy, pelvic and para-aortic lymphadenectomy or lymph-node sampling, omentum resection, and peritoneal biopsies as necessary. R0 was defined as no residual tumor or complete remission of the tumor, based on a combination of clinical and pathological findings. Chemotherapy and maintenance therapy First-line chemotherapy consisted of platinum-based regi- mens combined with paclitaxel, including TC (paclitaxel/ carboplatin) and TP (paclitaxel/cisplatin), administered intravenously or intraperitoneally. Maintenance therapy with anti-angiogenic agents, such as bevacizumab, was offered based on treatment protocols and availability. Platinum resistance was defined based on the widely accepted clini- cal criterion: disease progression within 6 months following the completion of platinum-based chemotherapy. Follow‑up Patients were followed up from the date of surgical or patho- logical diagnosis through outpatient visits, re-admission for 276 Archives of Gynecology and Obstetrics (2025) 312:273–286 treatment, or telephone follow-ups. There was no loss to follow-up, and the cutoff was September 2021. Follow-up included patient-reported complaints, pelvic exams, CA125 levels, and imaging (ultrasound, CT, MRI, or PET-CT, as needed). Statistical analysis Statistical analyses were performed using R version 3.6.2. Categorical variables were summarized as frequencies and percentages, while continuous variables were expressed as means ± standard deviations or medians with interquartile ranges, as appropriate. For comparisons between two groups, the Chi-square test was used to analyze categorical variables, ensuring appropriate statistical evaluation of relationships. To compare variables among the three groups (Coexisting, Arising, and Without groups), the Chi-square test was also applied. Kaplan–Meier method estimated survival curves for OS and PFS, compared with log-rank tests (GraphPad Prism version 9.0 for macOS, San Diego, California USA). Survival analysis was conducted using the Cox proportional hazards model (CoxPHFitter from lifelines python package) to evaluate factors influencing PFS and OS. Hazard ratios (HRs) with corresponding 95% confidence intervals (CIs) were calculated to quantify the strength of associations. Sta- tistical significance was set at p < 0.05 (two-tailed).

We included a total of 242 eligible patients following the selection workflow detailed in Fig.  1. These patients were further classified into three groups using the Sampson and Scott criteria: Without group (n = 130, 53.7%), Arising group (n = 71, 29.3%), and Coexisting group (n = 41, 16.9%). Patients characteristics Patients’ characteristics were summarized in Table  1. Patients in the Coexisting and Arising groups were younger (mean ages: 45.7 ± 8.9 years and 47.6 ± 8.6 years, respec- tively), compared to the Without group (52.3 ± 9.3 years; p < 0.000 and p = 0.012). More patients in the Coexisting and Arising groups were pre-menopausal (p < 0.000 and p = 0.005). The Arising group had a higher proportion of patients with a BMI < 24.0 (74.6%) than the Coexisting group (48.8%, p = 0.01) and a higher percentage of patients with fewer than two gestations compared to the Without group (42.2% vs. 26.1%, p = 0.029). Early FIGO stage (I + II) was more prevalent in the Arising group (91.6%) compared to the Coexisting (75.6%, p = 0.041) and Without (67.7%, p = 0.000) groups. Lymph-node metastasis was least com- mon in the Arising group (2.8%) compared to Coexisting (19.5%, p = 0.010) and Without (10%, p = 0.011) groups. There were no significant differences in preoperative CA125 Fig. 1 The flowchart of study selection and classification of OCCC patients based on Sampson and Scott criteria. A total of 275 patients diagnosed with primary ovarian clear cell carcinoma (OCCC) from 2012 to 2021 were initially identified. After excluding 30 patients who underwent fertility-preserving surgery and 3 patients with insuf- ficient data, 242 eligible patients were included. These patients were classified using the Sampson and Scott criteria into three groups: the Without group (patients with OCCC without endometriosis, n = 130), the Arising group (patients with OCCC arising from endometriosis, n = 71), and the Coexisting group (patients with OCCC coexisting with endometriosis, n = 41) 277Archives of Gynecology and Obstetrics (2025) 312:273–286 levels, ascites or peritoneal lavage fluid metastasis, or throm- bosis across groups. Patterns of care Patterns of care were outlined in Table  2. Most patients underwent laparotomy, though laparoscopy was more frequent in the Arising group than in the Without group (40.9% vs. 23.9%, p = 0.018). Staging surgery was per - formed most frequently in the Arising group (91.6%), compared to the Coexisting (75.6%) and Without (67.7%) groups (p < 0.001). Lymphadenectomy was performed in 97.6% of Coexisting and 94.4% of Arising patients, com- pared to 83.8% in the Without group (p = 0.043). Para- aortic lymphadenectomy was also more common in the Coexisting (65.9%) and Arising (66.2%) groups than in the Without group (49.2%, p = 0.031). While there were no significant differences in the number of lymph nodes resected, a higher percentage of patients in the Coexisting (80.5%) and Arising (81.7%) groups had ≥ 20 lymph nodes removed compared to the Without group (63.9%) ( p = 0.011). The percentage of patients with no residual tumor (R0) was significantly higher in the Arising group (97.2%) compared to the Without group (83.1%, p = 0.007). Regarding chemotherapy, TC (paclitaxel + carboplatin) remained the most commonly used regimen across all groups, with no significant differences (p = 0.652). How- ever, chemotherapy with Bevacizumab was significantly less frequent in the Arising group (8.5%) compared to the Without group (21.5%, p = 0.030). Most patients in the Arising group (91.6%) received ≤ 6 cycles of chemother - apy, significantly more than in the Without group (66.4%, p < 0.001). Table 1 Clinicopathologic characteristics of ovarian clear cell carcinoma patients with different endometriosis involvement *Lymph-node metastasis was diagnosed through pathological examination Characteristics Without Coexisting Arising p value n = 130 % n = 41 % n = 71 % Three-group comparison Without vs. Coexisting Without vs. Arising Coexisting vs. Arising Age (years, mean ± SD) 52.3 ± 9.2 45.7 ± 8.8 47.6 ± 8.6   < 50 48 36.9 29 70.7 40 56.3 < 0.001 < 0.001 0.012 0.190   ≥ 50 82 63.1 12 29.3 31 43.7 Menopause  No 44 33.9 27 65.9 39 54.9 < 0.001 < 0.001 0.005 0.351  Yes 86 66.2 14 34.2 32 45.1 BMI (kg/m2, mean ± SD) 23.0 ± 2.9 23.0 ± 2.9 23.3 ± 4.6   < 24.0 82 63.1 20 48.8 53 74.6 0.022 0.140 0.130 0.010   ≥ 24.0 48 36.9 21 51.2 18 25.4 Gestation   ≥ 2 96 73.9 28 68.3 41 57.8 0.064 0.621 0.029 1.000   < 2 34 26.1 13 31.7 30 42.2 FIGO stage  I + II 88 67.7 31 75.6 65 91.6 < 0.001 0.436 0.000 0.041  III + IV 42 32.3 10 24.4 6 8.4 Present with thrombosis  Yes 28 21.5 5 12.2 7 9.9 0.074 0.274 0.058 0.946  No 102 78.5 36 87.8 64 90.1 Preoperative CA125(U/ml)   < 35 47 36.2 14 34.2 33 46.5 0.289 0.981 0.203 0.292   ≥ 35 71 54.6 23 56.1 32 45.1 Ascites or peritoneal lavage fluid  Positive 24 18.5 5 12.2 12 16.9 0.682 0.548 0.951 0.653  Negative 100 76.9 33 80.5 56 78.9 Lymph-node status*  Negative 96 73.8 32 78 65 91.6 0.018 0.031 0.011 0.010  Positive 13 10 8 19.5 2 2.8 278 Archives of Gynecology and Obstetrics (2025) 312:273–286 Table 2 Patterns of care and prognosis of ovarian clear cell carcinoma patients with different endometriosis involvement PFS progression-free survival, OS overall survival Variables Without Coexisting Arising p value n = 130 % n = 41 % n = 71 % Three-group comparison Without vs. Coexisting Without vs. Arising Coexisting vs. Arising Surgical approach  Laparotomy 99 76.2 31 75.6 42 59.2 0.031 1 0.018 0.119  Laparoscopy 31 23.9 10 24.4 29 40.9 Surgical procedure  Staging surgery 88 67.7 31 75.6 65 91.6 < 0.001 0.444 < 0.001 0.041  Total hysterectomy 119 91.5 40 97.6 69 97.2 0.157 0.334 0.209 1.000  Bilateral salpingo- oophorectomy 130 100.0 41 100.0 71 100.0 1.000 1.000 1.000 1.000  Omentum resection 118 90.8 40 97.6 68 95.8 0.146 0.274 0.313 1.000  Pelvic lymphadenectomy 82 63.1 30 73.2 55 77.5 0.089 0.319 0.053 0.778  Para-aortic lymphad- enectomy 64 49.2 27 65.9 47 66.2 0.031 0.093 0.030 1.000  Debulking surgery 42 32.3 10 24.4 6 8.5 < 0.001 0.444 < 0.001 0.041 Lymphadenectomy  Yes 109 83.8 40 97.6 67 94.4 0.012 0.043 0.053 0.754  No 21 16.2 1 2.4 4 5.6 No. of lymph nodes resected   R0 22 16.9 5 12.2 2 2.8  Chemothrapy  TC (paclitaxel + carbo- platin) 59 45.4 22 53.7 34 47.9 0.652 0.455 0.848 0.695  TP (paclitaxel + cis- platin) 8 6.2 1 2.4 1 1.4 0.227 0.598 0.231 1.000  TC/P(paclitaxel + cispl- atin/carboplatin) 45 34.6 12 29.3 21 29.6 0.694 0.658 0.569 1.000  With bevacizumab 28 21.5 5 12.2 6 8.5 0.041 0.273 0.030 0.755 Number of chemotherapy   ≤ 6 87 66.4 30 73.2 65 91.6 0.001 0.013 6 43 32.9 11 26.8 6 8.5  Follow-up: median (range) (months) 23.3 (0.9–110) 25.3 (0.4–111.9) 21.5 (1.1–95.0)  5-year PFS rate (%) 50.5 91.8 85.3 < 0.005 0.006 0.006 0.731  5-year OS rate (%) 62.6 83.9 92.4 0.01 0.17 0.023 0.293 279Archives of Gynecology and Obstetrics (2025) 312:273–286 Pathological characteristics Figure  2 illustrates the pathological characteristics of OCCC with different status. The Arising group displayed carcinoma with continuous atypical endometriosis (A-EMS), marked by moderate-to-severe cytologic atypia and a crowded or micropapillary appearance. IHC analysis (Table  3) showed consistent expression patterns across groups: HNF-1β was positive in nearly all cases (98.6%), and Napsin A in 84.9%. CK7 was expressed universally, and WT1 was largely nega- tive (positive in 21.1%). The expression rates of ER and PR were low (18.2% and 9.5% positivity, respectively). Wild- type p53 was positive in 77.7% of cases, while PAX-8 was positive in 89.3%. CK20 was negative in nearly all cases (2% positive). Survival analysis The median follow-up was 31.09  months (range: 6–112 months). Among 242 patients, there were 45 recur - rences: 27.69% in the Without group (36/130), 7.32% in Coexisting (3/41), and 8.45% in Arising (6/71). Mortality rates were 23.08% (30/130) in Without, 12.20% (5/41) in Coexisting, and 5.63% (4/71) in Arising. The 5-year PFS rates were significantly higher in Coexisting (91.8%) and Arising (85.3%) compared to Without (50.5%, p = 0.006) (Fig.  3A). The 5-year OS rate was highest in Arising (92.4%), which was significantly greater than in the Without group (62.6%, p = 0.023) but not significantly different from Coexisting (83.9%, p = 0.293) (Fig. 3B). Prognostic factors of PFS in univariate and multivariate analyses Univariate and multivariate analyses for PFS (Table 4) iden- tified several factors as significant prognostic indicators. In univariate analysis, significant factors included FIGO stage (p < 0.001), no residual tumor (R0) (p = 0.029), platinum resistance (p = 0.017), lymph-node involvement (p = 0.036), ascites or peritoneal lavage fluid metastasis (p = 0.003), and the presence of endometriosis (p < 0.001). Subgroup analyses revealed that tumors coexisting with ( p = 0.009) and arising from (p = 0.004) endometriosis were correlated with longer PFS. In multivariate analysis, the presence of endometriosis remained an independent favorable factor for PFS (HR: 0.24, 95% CI: 0.09–0.66, p = 0.006). Specifically, tumors coexisting with endometriosis (HR: 0.11, 95% CI: 0.01–0.84, p = 0.033) were associated with significantly better PFS outcomes. Tumors arising from endometrio- sis showed a trend toward improved PFS (HR: 0.34, 95% CI: 0.11–1.05, p = 0.061), though not reaching statistical significance. Prognostic factors of OS in univariate and multivariate analyses Univariate and multivariate analyses for OS (Table  5) iden- tified several factors as significant indicators. In univariate analysis, significant factors included FIGO stage (p < 0.001), residual tumor status (R0) (p < 0.001), platinum resistance (p < 0.001), lymph-node involvement (p = 0.008), ascites or peritoneal lavage fluid metastasis (p < 0.001), thrombosis (p < 0.005), and the presence of endometriosis (p = 0.006). Subgroup analyses revealed that only carcinoma arising from endometriosis (p = 0.012) correlated significantly with improved OS. In multivariate analysis, FIGO stage (HR: 5.89, 95% CI: 2.06–16.82, p < 0.001) was identified as an independ- ent factor affecting OS, while residual tumor status (HR: 1.82, 95% CI: 0.78–4.29, p = 0.168), platinum resistance (HR: 2.39, 95% CI: 0.86–6.67, p = 0.096), and endometriosis presence (HR: 0.62, 95% CI: 0.26–1.53, p = 0.302) did not reach statisti- cal significance.

Unlike prior studies that primarily categorized patients based solely on the presence or absence of endometriosis, our research employs the Sampson and Scott criteria [23, 24] to rigorously classify endometriosis subtypes. Nearly half (46.3%) had cancer associated with endometriosis, with one-third (29.3%) having cancer arising from endo- metriosis and a smaller yet significant proportion (16.9%) having cancer coexisting with endometriosis. Understanding the role of endometriosis in OCCC pro- gression is crucial for guiding treatment and predicting prognosis. A key question remains whether OCCC coexist- ing with endometriosis behaves more like OCCC arising from endometriosis or OCCC without endometriosis. In our cohort, we observed significant clinical and prognos - tic differences between these groups. Both the Arising and Coexisting groups were younger at diagnosis, more likely pre-menopausal, received less intense chemotherapy, and had higher 5-year PFS rates than the Without group, consist- ent with studies demonstrating favorable prognosis in OCCC with endometriosis [10, 12, 13, 18, 19, 29]. Notably, only the Arising group demonstrated significantly higher 5-year OS rates, with no significant OS difference observed between the Coexisting and Without groups. To evaluate whether differences in treatment approaches influenced survival outcomes, we systematically analyzed surgical and chemotherapeutic interventions across the three groups (Table  2). Our analysis revealed both consist- encies and variations that help contextualize the observed survival patterns. Notably, five key treatment variables showed no significant differences across groups: total 280 Archives of Gynecology and Obstetrics (2025) 312:273–286 hysterectomy (91.5–97.6%, p = 0.157), bilateral salpingo- oophorectomy (100% across groups, p = 1.000), omentum resection (90.8–97.6%, p = 0.146), pelvic lymphadenec- tomy (63.1–77.5%, p = 0.089), and chemotherapy regimens, including TC/P (78.9–86.2%, p = 0.227–0.694). This uni- formity in radical cytoreduction and first-line chemotherapy suggests comparable baseline treatment intensity across groups, minimizing potential confounding effects related to treatment variability. However, significant differences were observed in the Arising group, which demonstrated higher rates of laparos- copy (40.9%, p = 0.031), staging surgery (91.6%, p < 0.001), para-aortic lymphadenectomy (66.2%, p = 0.031), and com- plete tumor resection (R0) (97.2%, p = 0.013), while under- going less debulking surgery (8.5%, p < 0.001) and receiving bevacizumab less frequently (8.5%, p = 0.041). This distinct surgical paradigm—characterized by minimally invasive techniques, systematic nodal evaluation, and optimal cytore- duction—likely contributed to the improved OS observed in the Arising group. The higher rate of complete tumor resec- tion (97.2% R0) in the Arising group may have reduced the need for adjuvant biologics, as reflected in the lower rates of debulking surgery and bevacizumab use. Consistent with landmark studies in ovarian cancer [30], R0 has been identi- fied as a pivotal determinant of survival. To further explore potential confounding factors that may influence the results, we analyzed the impact of sur - gical approach (laparotomy or laparoscopy), lymph-node status, and residual tumor status (R0 or > R0) on PFS and OS using univariate and multivariate analyses (Table  4 and Table  5). In univariate analysis, R0 was a significant factor for both PFS and OS, because it was analyzed in isolation. Fig. 2 Pathological character- istics of OCCC with different endometriosis involvement. (A) Hematoxylin and eosin (HE) stained section of ovarian clear cell carcinoma (OCCC) arising from endometriosis. The yellow arrow indicates a region of clear cell carcinoma, characterized by its typical clear cytoplasm and distinct cell borders. The gray arrow points to con- tiguous atypical endometriosis (A-EMS), which shows cellular atypia, suggesting its role as a precursor lesion to malig- nancy. Higher magnification views of these areas highlight the transition between atypical endometriosis and invasive carcinoma. Additionally, a distant endometriotic lesion is shown. (B) HE and immuno- histochemical (IHC) staining of OCCC specimens categorized based on their association with endometriosis: without, aris- ing from, or coexisting with endometriosis. The first column shows HE staining of carcinoma tissue, displaying the typical polyhedral clear cells of OCCC, separated by delicate fibrovas- cular or hyalinized stroma. The other columns show consistent IHC staining patterns across all three subtypes for HNF-1β (nuclear staining), Napsin A (cytoplasmic staining), and CK-7 (membranous/cytoplasmic staining) 281Archives of Gynecology and Obstetrics (2025) 312:273–286 This indicates that achieving complete tumor resection (R0) is associated with better survival outcomes when considered independently. However, in multivariate analysis, R0 was not a significant factor. Its lack of significance suggests that its effect is confounded or mediated by other variables, such as FIGO stage, platinum resistance, lymph-node status, or others. These findings underscore the importance of consid- ering the broader clinicopathological context when inter - preting the results. The improved outcomes in the Arising group likely reflect the combined benefits of optimal cytore- duction and other biological characteristics, rather than the effect of R0 alone. These findings challenge the prevailing notion that all OCCC cases associated with endometriosis exhibit uniform clinical characteristics and prognoses. A more nuanced understanding of the role of endometriosis in OCCC is essential for guiding personalized treatment strate- gies and improving patient outcomes. Previous studies have often treated OCCC “arising from” and “coexisting with” endometriosis as one entity, endome- triosis-associated ovarian cancer (EAOC) [ 11, 13, 18, 25, 26], despite unclear pathological distinctions. According to Sampson and Scott criteria, “arising from” cases exhibit malignant transformation, with A-EMS as the precancer - ous lesion for OCCC [ 31]. A-EMS is rare, found in 1.7% to 4.4% of all endometriotic lesions [31– 33], though pre- sent in 12–35% of ovarian endometriosis cases [34]. In our study, all “arising from” cases showed continuous A-EMS, accounting for nearly 30% of all OCCC cases. The World Health Organization defines A-EMS by atypical epithelium resembling endometrial neoplasia or cyst lining changes with stratification and cytologic atypia [35]. Apart from A-EMS, no significant histologic and immunophenotypic differences were observed in the carcinoma between the Coexisting, Arising, and Without groups. A-EMS is associated with increased malignant progres- sion risk [ 31, 32], frequently positive for ER and PR, with wild-type p53 staining and a low proliferation index [28, 36]. This differs from OCCC, which is often hormone receptor- negative and may exhibit abnormal p53/Ki67 expression. Molecularly, both OCCC and A-EMS lack ARID1A expres- sion, but ARID1A remains intact in distant endometriotic [37]. HNF-1β is expressed in OCCC but not in contiguous or distant endometriosis, and estrogen receptor is expressed in both contiguous and distant endometriosis but not in OCCC, supporting the progression from benign to malignant. The association between endometriosis and OCCC has been increasingly recognized, with molecular studies shedding light on the mechanisms driving this progression [38]. Endometriosis creates a pro-tumorigenic microenvi- ronment characterized by cyclic hemorrhage-induced iron overload and reactive oxygen species (ROS) accumulation [39]. Chronic inflammation upregulates pro-inflammatory cytokines (e.g., IL-6 and TNF-α) and activates the NF-κB/ Table 3 The expression of IHC markers in ovarian clear cell carcinoma patients with different endometriosis involvement Comparisons: a. Without vs. Coexisting; b. Without vs. Arising; c. Coexisting vs. Arising IHC immunohistochemistry, HNF1β hepatocyte nuclear factor-1 beta (HNF1β), NA not available Markers Total Positive (%) Negative (%) Without Coexisting Arising Comparisons Positive (%) Negative (%) Positive (%) Negative (%) Positive (%) Negative (%) a b c HNF1β 72 71 (98.6) 1 (1.4) 38 (97.4) 1 (2.6) 12 (100) 0 (0) 21 (100) 0 (0) 1 1 NA WT1 194 41 (21.1) 153 (78.8) 21 (20.2) 83 (79.8) 7 (23.3) 23 (76.7) 13 (21.7) 47 (78.3) 0.906 0.981 1 ER 192 35 (18.2) 157 (81.8) 16 (16.3) 82 (83.7) 4 (14.3) 24 (85.7) 15 (22.7) 51 (77.3) 1 0.981 1 PR 189 18 (9.5) 171 (90.5) 12 (12.6) 83 (87.4) 1 (3.6) 27 (96.4) 5 (7.6) 61 (92.4) 0.307 0.410 0.515 Napsin A 185 157 (84.9) 28 (15.1) 75 (81.5) 17 (18.5) 30 (96.8) 1 (3.2) 52 (83.9) 10 (16.1) 0.074 0.873 0.140 CK7 187 187 (100) 0 (0) 99 (100) 0 (0) 28 (100) 0 (0) 60 (100) 0 (0) 1 1 1 CK20 149 3 (2.0) 146 (98.0) 3 (3.8) 76 (96.2) 0 (0) 21 (100) 0 (0) 49 (100) 0.852 0.436 NA P53 202 157 (77.7) 45 (22.3) 85 (79.4) 22 (20.6) 27 (84.4) 5 (15.6) 45 (71.4) 18 (28.6) 0.715 0.316 0.255 PAX-8 28 25 (89.3) 3 (10.7) 16 (84.2) 3 (15.8) 2 (100) 0 (0) 7 (100) 0 (0) 1 0.670 NA 282 Archives of Gynecology and Obstetrics (2025) 312:273–286 COX-2/PGE2 signaling pathway [40], which not only induces DNA damage but also suppresses mismatch repair (MMR) mechanisms, leading to genomic instability [41]. Over time, oxidative stress drives somatic mutations in key regulatory genes, including ARID1A (15–62% of OCCC cases) and PIK3CA (31.5–55%), which are hallmarks of malignant transformation [42, 43]. Loss of ARID1A disrupts the SWI/SNF chromatin remodeling complex, impairing DNA repair and promoting unchecked prolif- eration in endometriosis-derived lesions. Concurrently, PIK3CA mutations constitutively activate the PI3K/AKT/ mTOR pathway, enhancing cell survival and chemore- sistance through downstream effectors like eEF1A2 and PPP1R14B [44]. The dichotomy between “Arising” and “Coexisting” OCCC subgroups reflects distinct molecular trajectories. Despite their shared clinical characteristics, the Coexist- ing group does not experience the same OS benefit as the Arising group, suggesting key molecular differences that influence disease progression and treatment response. In the Arising group, OCCC evolves directly from endome- triosis, implying a continuous clonal progression. This may lead to a more homogeneous tumor population with spe- cific molecular vulnerabilities, potentially making it more responsive to treatment. In contrast, the Coexisting group comprises separate endometriosis and OCCC lesions, sug- gesting independent tumor origins or parallel evolution, which could lead to greater intratumoral heterogeneity. This heterogeneity may contribute to treatment resistance and a less favorable OS. Additionally, the Coexisting group may include cases that have already undergone malignant trans- formation from atypical endometriosis (A-EMS) and cases originating from other pathological subtypes not directly associated with endometriosis. By applying the Sampson and Scott criteria, our study enhances understanding of the pathological and clinical con- nection between endometriosis and OCCC, showing that dif- ferent types of endometriosis involvement may contribute to varied outcomes. Recognizing these distinctions is essential for elucidating the clinical behavior and prognosis of OCCC Fig. 3 Kaplan–Meier survival curves for Coexisting, Aris- ing, and Without groups. (A) Overall Survival (OS) and (B) Progression-Free Survival (PFS) for patients in each group. The log-rank test showed significant differences in sur- vival among the three groups for both OS (p = 0.0102) and PFS (p = 0.0002). Specifically, patients in the Arising group had a more favorable 5-year OS than those in the Without group (92.4% vs. 62.6%, p = 0.023). The Arising and Coexisting groups also had a more favora- ble 5-year PFS than the Without group (85.3% and 91.8%, respectively, p = 0.006). There was no significant difference in the 5-year OS and PFS between the Arising and Coexisting groups (p = 0.293 and p = 0.731, respectively) 283Archives of Gynecology and Obstetrics (2025) 312:273–286 associated with endometriosis, potentially guiding tailor surveillance and treatment decisions. Future studies should investigate the biological mechanisms underlying this dis - crepancy, particularly differences in the tumor microenviron- ment, immune landscape, and metabolic reprogramming. A deeper molecular characterization of these subgroups could identify novel therapeutic targets tailored to each distinct evolutionary pathway. However, our study has limitations. As a retrospective analysis from a single specialized center, selection bias may impact the generalizability of our findings. One limitation of our study is the potential underdiagnosis of endometriosis, as routine Douglas pouch peritoneal biopsy was not univer- sally performed. Future prospective studies may consider systematic peritoneal biopsies to classify endometriosis status more accurately. Additionally, while we identified clinical differences among subgroups, further studies on bio- logical mechanisms, such as microenvironmental or immune factors driving these prognostic differences, should be taken into consideration. Multi-center prospective cohorts will be crucial for validating our findings and uncovering molecu- lar pathways distinguishing OCCC subtypes associated with endometriosis. Table 4 Factors associated with progression-free survival in patients with OCCC Risk factors N Univariate analysis Multivariate analysis HR 95% CI  p value  Adjusted HR 95% CI p value Age (years)   < 50 117 1 (Ref) – –   ≥ 50 125 1.35 0.59–1.47 0.303 BMI (kg/m2)   < 24.0 155 1 (Ref) – –   ≥ 24.0 87 0.97 0.38—1.86 0.417 Surgery approach  Laparotomy 172 1 (Ref) – –  Laparoscopy 70 0.89 0.45–1.74 0.642 FIGO stage  I–II 184 1 (Ref) – – 1 (Ref) – –  III–IV 58 3.52 1.91–6.50 R0 29  2.19  1.08–4.42  0.029  1.42 0.58–3.48   0.444 Resistant to platinum  No 214 1 (Ref) – – 1 (Ref) – –  Yes 16 3.25 1.23–8.55 0.017 0.84 0.19–3.67 0.82 Lymph nodes  Negative 192 1 (Ref) – – 1 (Ref) – –  Positive 23 2.41 1.06–5.47 0.036 1.88 0.51–6.95 0.342 Ascites or peritoneal lavage fluid  Negative 189 1 (Ref) – – 1 (Ref) – –  Positive 41 2.86 1.44–5.68 0.003 2.49 0.87–7.13 0.089 Present with thrombosis  No 202 1 (Ref) – –  Yes 40 1.01 0.98–3.77 0.081 Endometriosis  No 130 1 (Ref) – – 1 (Ref) – –  Yes 112 0.25 0.12–0.52 < 0.001 0.24 0.09–0.66 0.006 Endometriosis involvement  Without 130 1 (Ref) – – 1 (Ref) – –  Coexisting 41 0.21 0.06–0.68 0.009 0.11 0.01–0.84 0.033  Arising 71 0.28 0.12–0.67 0.004 0.34 0.11–1.05 0.061 284 Archives of Gynecology and Obstetrics (2025) 312:273–286 Supplementary Information The online version contains supplemen- tary material available at https:// doi. org/ 10. 1007/ s00404- 025- 08025-3. Author contributions Project development: JD and TY; data collection: JD and JL; data analysis: LX and TY; manuscript writing: JD and TY; manuscript editing: JD, LX, JL and TY. All authors approved the final submitted draft. Funding This work was supported by National Natural Science Foun- dation of China (Grant No. 82002756). Data availability The patients’ clinicopathological data used to gener- ate the results are presented in Supplemental Table. Declarations Conflicts of interest The authors declare no competing interests. Ethics approval This study was performed in line with the principles of the Declaration of Helsinki. Since the retrospective study utilized anonymous patient information, the Research Ethics Committee of West China Second Hospital, Sichuan University, confirmed that no ethical approval was required. Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and repro- duction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party

in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If

is not included in the article’s Creative Commons licence and Table 5 Factors associated with overall survival in patients with OCCC Risk factors N Univariate analysis Multivariate analysis HR 95% CI p value Adjusted HR 95% CI p value Age (years)   < 50 117 1 (Ref) – –   ≥ 50 125 1.26 0.60–2.68 0.401 BMI (kg/m2)   < 24.0 155 1 (Ref) – –   ≥ 24.0 87 0.88 0.38–1.93 0.211 Surgery approach  Laparotomy 172 1 (Ref) – –  Laparoscopy 70 0.41 0.13–1.02 0.112 FIGO stage  I–II 184 1 (Ref) – – 1 (Ref) – –  III–IV 58 7.12 3.71–13.59 < 0.001 5.89 2.06–16.82 R0 29  3.68  1.88–7.22  < 0.001  1.82  0.78–4.29  0.168 Resistant to platinum  No 214 1 (Ref) – – 1 (Ref) – –  Yes 16 14.46 6.69–31.24 < 0.001 2.39 0.86–6.67 0.096 Lymph nodes  Negative 192 1 (Ref) – – 1 (Ref) – –  Positive 23 3.05 1.33–6.99 0.008 0.87 0.32–2.40 0.789 Ascites or peritoneal lavage fluid  Negative 189 1 (Ref) – – 1 (Ref) – –  Positive 41 5.11 2.67–9.73 < 0.001 2.18 0.84–5.67 0.111 Present with thrombosis  No 202 1 (Ref) – – 1 (Ref) – –  Yes 40 3.6383 1.62–8.08 < 0.005 2.544 0.83–7.27 0.105 Endometriosis  No 130 1 (Ref) – – 1 (Ref) – –  Yes 112 0.35 0.17—0.74 0.006 0.62 0.26–1.53 0.302 Endometriosis involvement status  Without 130 1 (Ref) – – 1 (Ref) – –  Coexisting 41 0.48 0.18–1.22 0.125 0.62 0.22–1.96 0.415  Arising 71 0.26 0.09–0.75 0.012 0.63 0.21–1.98 0.429 285Archives of Gynecology and Obstetrics (2025) 312:273–286 your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by- nc- nd/4. 0/.

1. Burney RO, Giudice LC (2012) Pathogenesis and pathophysiology of endometriosis. Fertil Steril 98(511–519):20120720. https:// doi. org/ 10. 1016/j. fertn stert. 2012. 06. 029 2. Kvaskoff M, Mu F, Terry KL et al (2015) Endometriosis: a high- risk population for major chronic diseases? Hum Reprod Update 21(500–516):20150311. https:// doi. org/ 10. 1093/ humupd/ dmv013 3. Wischmann T, Ditzen B (2024) Endometriosis: patient-doctor communication and psychological counselling. Arch Gynecol Obstet 309(599–610):20231205. https:// doi. org/ 10. 1007/ s00404- 023- 07292-2 4. Pearce CL, Templeman C, Rossing MA et al (2012) Association between endometriosis and risk of histological subtypes of ovar - ian cancer: a pooled analysis of case-control studies. Lancet Oncol 13(385–394):20120222. https:// doi. org/ 10. 1016/ s1470- 2045(11) 70404-1 5. Pectasides D, Pectasides E, Psyrri A et al (2006) Treatment issues in clear cell carcinoma of the ovary: a different entity? Oncologist 11:1089–1094. https:// doi. org/ 10. 1634/ theon colog ist. 11- 10- 1089 6. Machida H, Matsuo K, Yamagami W et al (2019) Trends and characteristics of epithelial ovarian cancer in Japan between 2002 and 2015: A JSGO-JSOG joint study. Gynecol Oncol 153(589– 596):20190321. https:// doi. org/ 10. 1016/j. ygyno. 2019. 03. 243 7. Sugiyama T, Kamura T, Kigawa J et al (2000) Clinical charac- teristics of clear cell carcinoma of the ovary: a distinct histologic type with poor prognosis and resistance to platinum-based chemo- therapy. Cancer 88:2584–2589 8. Suzuki K, Takakura S, Saito M et al (2014) Impact of surgical staging in stage I clear cell adenocarcinoma of the ovary. Int J Gynecol Cancer 24:1181–1189. https:// doi. org/ 10. 1097/ igc. 00000 00000 000178 9. Ogawa S, Kaku T, Amada S et al (2000) Ovarian endometrio - sis associated with ovarian carcinoma: a clinicopathological and immunohistochemical study. Gynecol Oncol 77:298–304. https:// doi. org/ 10. 1006/ gyno. 2000. 5765 10. Park JY, Kim DY, Suh DS et al (2018) Significance of ovarian endometriosis on the prognosis of ovarian clear cell carcinoma. Int J Gynecol Cancer 28:11–18. https:// doi. org/ 10. 1097/ igc. 00000 00000 001136 11. Bai H, Cao D, Yuan F et al (2016) Prognostic value of endome- triosis in patients with stage I ovarian clear cell carcinoma: expe- riences at three academic institutions. Gynecol Oncol 143(526– 531):20161013. https:// doi. org/ 10. 1016/j. ygyno. 2016. 10. 009 12. Orezzoli JP, Russell AH, Oliva E et al (2008) Prognostic implica- tion of endometriosis in clear cell carcinoma of the ovary. Gynecol Oncol 110(336–344):20080718. https:// doi. org/ 10. 1016/j. ygyno. 2008. 05. 025 13. Boyraz G, Selcuk I, Yazıcıoğlu A et al (2013) Ovarian carcinoma associated with endometriosis. Eur J Obstet Gynecol Reprod Biol 170(211–213):20130709. https:// doi. org/ 10. 1016/j. ejogrb. 2013. 06. 001 14. Wang S, Qiu L, Lang JH et al (2013) Clinical analysis of ovarian epithelial carcinoma with coexisting pelvic endometriosis. Am J Obstet Gynecol 208(413):411–415. https:// doi. org/ 10. 1016/j. ajog. 2012. 12. 004 15. Mangili G, Bergamini A, Taccagni G et al (2012) Unraveling the two entities of endometrioid ovarian cancer: a single center clinical experience. Gynecol Oncol 126(403–407):20120515. https:// doi. org/ 10. 1016/j. ygyno. 2012. 05. 007 16. Noli S, Cipriani S, Scarfone G et al (2013) Long term survival of ovarian endometriosis associated clear cell and endometrioid ovarian cancers. Int J Gynecol Cancer 23:244–248. https:// doi. org/ 10. 1097/ IGC. 0b013 e3182 7aa0bb 17. Cuff J, Longacre TA (2012) Endometriosis does not confer improved prognosis in ovarian carcinoma of uniform cell type. Am J Surg Pathol 36:688–695. https:// doi. org/ 10. 1097/ PAS. 0b013 e3182 4b6eed 18. Bounous VE, Ferrero A, Fuso L et al (2016) Endometriosis-asso - ciated ovarian cancer: a distinct clinical entity? Anticancer Res 36:3445–3449 19. Kumar S, Munkarah A, Arabi H et al (2011) Prognostic analysis of ovarian cancer associated with endometriosis. Am J Obstet Gynecol 204(63):e61-67. https:// doi. org/ 10. 1016/j. ajog. 2010. 08. 017 20. Tsuchiya A, Sakamoto M, Yasuda J et al (2003) Expression pro- filing in ovarian clear cell carcinoma: identification of hepato- cyte nuclear factor-1 beta as a molecular marker and a possible molecular target for therapy of ovarian clear cell carcinoma. Am J Pathol 163:2503–2512. https:// doi. org/ 10. 1016/ s0002- 9440(10) 63605-x 21. DeLair D, Oliva E, Köbel M et al (2011) Morphologic spectrum of immunohistochemically characterized clear cell carcinoma of the ovary: a study of 155 cases. Am J Surg Pathol 35:36–44. https:// doi. org/ 10. 1097/ PAS. 0b013 e3181 ff400e 22. Bulun SE, Monsavais D, Pavone ME et al (2012) Role of estro- gen receptor-β in endometriosis. Semin Reprod Med 30(39– 45):20120123. https:// doi. org/ 10. 1055/s- 0031- 12995 96 23 Sampson JA (1925) Endometrial carcinoma of the ovary, arising in endometrial tissue in that organ. Arch Surg 10:1–72. https:// doi. org/ 10. 1001/ archs urg. 1925. 01120 10000 7001 24. Scott RB (1953) Malignant changes in endometriosis. Obstet Gynecol 2:283–289 25. Ju UC, Kang WD, Kim SM (2019) The effect of concurrent endometriosis on the prognosis of women with ovarian clear cell or endometrioid carcinoma. Int J Gynaecol Obstet 146(177– 183):20190529. https:// doi. org/ 10. 1002/ ijgo. 12861 26. Qiu L, Wang S, Lang JH et al (2013) The occurrence of endome - triosis with ovarian carcinomas is not purely coincidental. Eur J Obstet Gynecol Reprod Biol 170(225–228):20130715. https:// doi. org/ 10. 1016/j. ejogrb. 2013. 06. 015 27. Chiaffarino F, Cipriani S, Ricci E et al (2024) Histologic subtypes in endometriosis-associated ovarian cancer and ovarian cancer arising in endometriosis: a systematic review and meta-analysis. Reprod Sci 31(1642–1650):20240304. https:// doi. org/ 10. 1007/ s43032- 024- 01489-9 28. Wepy C, Nucci MR, Parra-Herran C (2024) Atypical endometrio- sis: comprehensive characterization of clinicopathologic, immu- nohistochemical, and molecular features. Int J Gynecol Pathol 43(70–77):20230313. https:// doi. org/ 10. 1097/ pgp. 00000 00000 000952 29 Leone Roberti Maggiore U, Bogani G, Paolini B et al (2024) Endometriosis-associated ovarian cancer: a different clinical entity. Int J Gynecol Cancer 34:863–870. https:// doi. org/ 10. 1136/ ijgc- 2023- 005139 30. Bristow RE, Tomacruz RS, Armstrong DK et al (2002) Survival effect of maximal cytoreductive surgery for advanced ovarian carcinoma during the platinum era: a meta-analysis. J Clin Oncol 20:1248–1259. https:// doi. org/ 10. 1200/ jco. 2002. 20.5. 1248 31. Fukunaga M, Nomura K, Ishikawa E et al (1997) Ovarian atypi- cal endometriosis: its close association with malignant epithelial tumours. Histopathology 30:249–255. https:// doi. org/ 10. 1046/j. 1365- 2559. 1997. d01- 592.x 286 Archives of Gynecology and Obstetrics (2025) 312:273–286 32. So KA, Hong SR, Kim NR et al (2021) Association between atypical endometriosis and ovarian malignancies in the real world. J Ovarian Res 14(110):20210828. https:// doi. org/ 10. 1186/ s13048- 021- 00865-2 33. Guo DH (2008) Pang SJ and Shen Y [Atypical endometriosis: a clinicopathologic study of 163 cases]. Zhonghua Fu Chan Ke Za Zhi 43:831–834 34. Varma R, Rollason T, Gupta JK et al (2004) Endometriosis and the neoplastic process. Reproduction 127:293–304. https:// doi. org/ 10. 1530/ rep.1. 00020 35. Board Whocwhocte 2020 Female Genital Tumours. International Agency for Research on Cancer 36. Lenz J, Chvatal R, Fiala L et al (2021) Comparative immuno- histochemical study of deep infiltrating endometriosis, lymph- node endometriosis and atypical ovarian endometriosis including description of a perineural invasion. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub 165(69–79):20200309. https:// doi. org/ 10. 5507/ bp. 2020. 006 37. Wiegand KC, Shah SP, Al-Agha OM et al (2010) ARID1A muta- tions in endometriosis-associated ovarian carcinomas. N Engl J Med 363(1532–1543):20100908. https:// doi. org/ 10. 1056/ NEJMo a1008 433 38. Zhang M, Xu T, Tong D et al (2023) Research advances in endo- metriosis-related signaling pathways: a review. Biomed Pharma- cother 164(114909):20230519. https:// doi. org/ 10. 1016/j. biopha. 2023. 114909 39. Mandai M, Yamaguchi K, Matsumura N et al (2009) Ovarian can- cer in endometriosis: molecular biology, pathology, and clinical management. Int J Clin Oncol 14(383–391):20091025. https:// doi. org/ 10. 1007/ s10147- 009- 0935-y 40. AlAshqar A, Reschke L, Kirschen GW et al (2021) Role of inflam- mation in benign gynecologic disorders: from pathogenesis to novel therapies†. Biol Reprod 105:7–31. https:// doi. org/ 10. 1093/ biolre/ ioab0 54 41. Ye S, Zhou S, Zhong S et al (2022) The frequency and clini- cal implication of mismatch repair protein deficiency in Chi- nese patients with ovarian clear cell carcinoma. BMC Cancer 22(449):20220423. https:// doi. org/ 10. 1186/ s12885- 022- 09588-z 42. Gadducci A, Multinu F, Cosio S et al (2021) Clear cell carcinoma of the ovary: epidemiology, pathological and biological features, treatment options and clinical outcomes. Gynecol Oncol 162(741– 750):20210708. https:// doi. org/ 10. 1016/j. ygyno. 2021. 06. 033 43. Yamamoto S, Tsuda H, Takano M et al (2012) Loss of ARID1A protein expression occurs as an early event in ovarian clear-cell carcinoma development and frequently coexists with PIK3CA mutations. Mod Pathol 25(615–624):20111209. https:// doi. org/ 10. 1038/ modpa thol. 2011. 189 44. Worley MJ Jr, Liu S, Hua Y et al (2015) Molecular changes in endometriosis-associated ovarian clear cell carcinoma. Eur J Can- cer 51(1831–1842):20150606. https:// doi. org/ 10. 1016/j. ejca. 2015. 05. 011 Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.