Clinicopathological characteristics and prognostic analysis of synchronous endometrioid endometrial and ovarian carcinomas: a retrospective cohort study.

The clinicopathological features of SEEOCs, including menopause status, histological diagnosis, myometrial invasion, LVSI, lymph node metastasis, presence of carcinomatous ascites or peritoneal washes, and 2023 FIGO staging, are summarized in Table  1 . In most cases, the histological subtypes and grades of the endometrial and ovarian carcinomas were uniform. In the three cases of mixed carcinoma, both histological components were observed in the endometrium and ovary. Among the 47 cases of pure endometrioid carcinoma, four showed inconsistent histological grading between the endometrium and ovary—two cases were grade 1 in the endometrium and grade 2 in the ovary, and two were grade 2 in the endometrium and grade 1 in the ovary. All four cases were ultimately classified as grade 2, as all tumor components were regarded as a single entity, fulfilling the criteria for grade 2. Table 1 Clinicopathological features of synchronous endometrioid endometrial and ovarian carcinomas Parameters Number of cases (percent) Menopause  No 29 (58.0%)  Yes 21 (42.0%) Histological diagnosis  EC, G1 16 (32.0%)  EC, G2 20 (40.0%)  EC, G3 11 (22.0%)  Mixed carcinoma 3 (6.0%) Myometrial invasion  ≤ 50% 31 (62.0%)  > 50% 19 (38.0%) LVSI  None 29 (58.0%)  Focal 8 (16.0%)  Substantial 13 (26.0%) Lymph node metastasis  Positive 9 (18.0%)  Negative 9 (18.0%)  Untested 32 (64.0%) Carcinomatous ascites/peritoneal washes  Yes 10 (20.0%)  Suspicious 4 (8.0%)  No 28 (56.0%)  Untested 8 (16.0%) 2023 FIGO staging  ⅠA3 5 (10.0%)  ⅢA1 21 (42.0%)  ⅢB + ⅢC 15 (30.0%)  Ⅳ 9 (18.0%) EC Endometrioid carcinoma, LVSI Lymphovascular space invasion Clinicopathological features of synchronous endometrioid endometrial and ovarian carcinomas EC Endometrioid carcinoma, LVSI Lymphovascular space invasion Immunohistochemical results were retrieved from either the endometrial or ovarian tumors, including 36 cases tested for MMR proteins, 38 for p53, 27 for PD-L1 (22C3), and 28 for HER2. Deficient MMR (dMMR) was identified in 13 cases (36.1%), with 10 cases showing MLH1/PMS2 loss and three showing MSH2/MSH6 loss. Mutant p53 expression was detected in five cases (13.2%). PD-L1 (22C3) positivity was observed in 17 cases (63.0%), with CPS values ranging from 1 to 30. HER2 expression results were: 0 in seven cases (25.0%), 1 + in 15 cases (53.6%), and 2 + in six cases (21.4%). The positivity rate of HER2 (1 + and 2 +) was 75.0%. Representative images were seen in Fig.  1 . Fig. 1 Representative histological images from a mixed endometrioid and clear cell carcinoma. A Hematoxylin–eosin staining showing mixed carcinoma—endometrioid on the left and clear cell on the right. Both components were present in the endometrium and the ovary. B ER immunostaining—diffuse positivity in the endometrioid component, negative in the clear cell component. C Napsin A immunostaining—negative in the endometrioid carcinoma, scattered cytoplasmic positivity in the clear cell carcinoma. D MLH1 immunostaining—both components negative. E MSH2 immunostaining—positive in both components. F p53 immunostaining—wild-type pattern in both components. All images at × 50 magnification Representative histological images from a mixed endometrioid and clear cell carcinoma. A Hematoxylin–eosin staining showing mixed carcinoma—endometrioid on the left and clear cell on the right. Both components were present in the endometrium and the ovary. B ER immunostaining—diffuse positivity in the endometrioid component, negative in the clear cell component. C Napsin A immunostaining—negative in the endometrioid carcinoma, scattered cytoplasmic positivity in the clear cell carcinoma. D MLH1 immunostaining—both components negative. E MSH2 immunostaining—positive in both components. F p53 immunostaining—wild-type pattern in both components. All images at × 50 magnification Among the 50 SEEOC cases, five were classified as FIGO IA3 stage. Table 2 presents the detailed clinicopathological and immunophenotypic features of these cases. All patients were premenopausal. One case exhibited focal LVSI in the endometrial carcinoma, while the remaining four had no LVSI. The ovarian tumor diameters ranged from 1 to 16 cm. Two cases (cases 1 and 5) were dMMR, with MLH1/PMS2 and MSH2/MSH6 loss, respectively. Notably, all five cases exhibited wild-type p53 expression and lacked MELF pattern invasion. Four among five patients accepted postoperative therapy (detail in Table  2 ). No recurrence was observed during the follow-up period. Table 2 Clinicopathological characteristics and outcome of synchronous endometrioid endometrial and ovarian carcinomas in IA3 stage Case number 1 2 3 4 5 Age 46 years 48 years 51 years 50 years 44 years Menopause No No No No No Histological grade G2 G1 G1 G1 G1 LVSI Focal None None None None Diameter of ovarian tumor 1 cm 8 cm 10 cm 16 cm 8 cm MMR protein dMMR* pMMR pMMR pMMR dMMR** p53 Wild type Wild type Wild type Wild type Wild type PD-L1 (22C3) CPS = 8 CPS < 1 CPS < 1 CPS = 1 CPS < 1 HER2 2 +  2 +  1 +  0 1 +  MELF invasion Absent Absent Absent Absent Absent Postoperative therapy Absent CRT CRT CRT CRT Follow-up time 9.2 months 9.0 months 21.0 months 46.0 months 61.6 months Outcome NED NED NED NED NED LVSI Lymphovascular space invasion, MMR Mismatch repair, dMMR deficient MMR * Loss of MLH1 and PMS2 proteins ** Loss of MSH2 and MSH6 proteins, pMMR proficient MMR, CPS Combined positive score, MELF Microcystic, elongated, and fragmented, CRT Chemoradiation therapy, NED No evidence of disease Clinicopathological characteristics and outcome of synchronous endometrioid endometrial and ovarian carcinomas in IA3 stage LVSI Lymphovascular space invasion, MMR Mismatch repair, dMMR deficient MMR * Loss of MLH1 and PMS2 proteins ** Loss of MSH2 and MSH6 proteins, pMMR proficient MMR, CPS Combined positive score, MELF Microcystic, elongated, and fragmented, CRT Chemoradiation therapy, NED No evidence of disease Clinicopathological parameters including menopausal status, histological grade, myometrial invasion, LVSI, lymph node metastasis, carcinomatous ascites/peritoneal washes, and 2023 FIGO staging were analyzed as potential risk factors for PFS. The analysis revealed that postmenopausal status exhibited a trend toward association with poorer PFS, though this association did not reach statistical significance (Fig.  2 A, P  = 0.060). Among all evaluated parameters, only carcinomatous ascites/peritoneal washes failed to demonstrate statistical significance ( P  = 0.279) in the Kaplan–Meier survival analysis, while all other variables emerged as significant predictors (Fig.  2 B-F). Kaplan–Meier pairwise comparisons showed no significant difference between the focal and the substantial LVSI groups ( P  = 0.733), but a significant difference between the none and the focal LVSI groups ( P  < 0.001), which indicated that the presence of LVSI, even focal, was an unfavorable parameter in predicting PFS. The overall 5-year PFS rate in this cohort was 66.3%. As detailed in Table  3 , the 5-year PFS rates were further stratified according to various clinicopathological parameters. Fig. 2 Progression-free survival using Kaplan–Meier method and log-rank test. A Stratified by menopausal status. B Stratified by histological grade. G1, G2, G3 refer to endometrioid carcinoma grades; “Mixed” refers to mixed carcinoma. C Stratified by depth of myometrial invasion. D Stratified by lymphovascular space invasion (LVSI). E Stratified by lymph node metastasis. F Stratified by 2023 FIGO staging Table 3 5-year progress-free survival (PFS) based on different clinicopathological parameters Parameters 5-year PFS rate in % 95% confidence interval Menopause  No 77.0 59.4–94.6  Yes 52.5 28.4–76.6 Histological grade  G1 + G2 74.7 58.4–91.0  G3 + Mixed carcinoma 43.3 12.6–74.1 Myometrial invasion  ≤ 50% 79.0 62.3–95.7  > 50% 48.3 23.0–73.6 LVSI  None 95.7 87.3–100  Focal 16.7 0–96.8  Substantial 33.2 3.6–62.8 Lymph node metastasis  Negative 90.4 77.7–100  Untested 50.8 17.5–84.1  Positive 22.9 0–50.5 2023 FIGO staging  ⅠA3* / /  ⅢA1 94.1 82.9–100  ⅢB + ⅢC 39.2 10.0–68.4  Ⅳ 29.2 0–63.1  Total 66.3 43.1–78.6 * All IA3 patients had follow-up periods under 5 years, with no case of disease progression observed Progression-free survival using Kaplan–Meier method and log-rank test. A Stratified by menopausal status. B Stratified by histological grade. G1, G2, G3 refer to endometrioid carcinoma grades; “Mixed” refers to mixed carcinoma. C Stratified by depth of myometrial invasion. D Stratified by lymphovascular space invasion (LVSI). E Stratified by lymph node metastasis. F Stratified by 2023 FIGO staging 5-year progress-free survival (PFS) based on different clinicopathological parameters * All IA3 patients had follow-up periods under 5 years, with no case of disease progression observed

We retrospectively reviewed 1160 consecutive patients diagnosed with endometrioid endometrial carcinoma who underwent total hysterectomy and bilateral salpingo-oophorectomy at Peking University Cancer Hospital from 2009 to 2024. Synchronous endometroid carcinoma found in both the endometrium and the ovary was defined as SEEOC in this study, regardless of whether cancer involved other sites. All SEEOC cases adopted the Sectioning and Extensively Examining the Fimbriated End (SEE-FIM) sampling protocol. Fifty patients met the criteria for SEEOC, including 47 with pure endometrioid carcinomas and three with mixed histologies (two with clear cell components and one with small cell neuroendocrine carcinoma). For mixed histologies, two components were observed in both the uterus and ovary. These cases were included in our study because the mixed components were considered to share the same origin as the endometrioid component and were grouped with G3 histology for analysis. Among the 50 cases, three were previously diagnosed as primary ovarian cancer, 28 as primary endometrial cancer, and 19 as independent primaries. None had a history of prior malignancy. The mean age at surgery for SEEOC patients was 50.3 ± 7.3 years, and the mean follow-up duration was 50.5 ± 45.3 months (median: 42.0 months). FIGO 2023 staging was assigned based on clinicopathological criteria in the absence of molecular classification. All patients, with the exception of one in the IA3 group, received standard postoperative treatment according to the primary diagnoses and the prevailing NCCN guidelines at that time, which consisted of six cycles of paclitaxel plus carboplatin chemotherapy with or without concurrent radiotherapy. During the follow-up period, 16 patients experienced disease progression, and three died of cancer (two with FIGO stage IIIC2 and one with stage IVB). Histological sections were reviewed by senior pathologists (BY, LQ, and YQ) to assess tumor differentiation, myometrial invasion, LVSI, lymph node metastasis, and immunohistochemical expression of MLH1 (IHC409, Endele Biotech, Beijing), MSH2 (IHC510, Endele Biotech, Beijing), MSH6 (IHC026, Endele Biotech, Beijing), PMS2 (EP51, DAKO), p53 (DO-7, Gene Tech, Shanghai), PD-L1 (22C3 pharmDx, Agilent Technologies, CA), and HER2 (4B5, ROCHE). The evaluation of LVSI was based on the examination of all carcinoma-containing sections and was not confined to the sections from endometrial carcinoma. LVSI was strictly classified as none, focal (< 5 tumor emboli), or substantial (≥ 5 tumor emboli), as this distinction has reported prognostic significance according to the 2020 WHO Classification for Female Genital Tumors. Four MMR proteins were evaluated based on nuclear staining and recorded as either positive or negative. p53 expression was interpreted as wild-type or mutant according to the established criteria for ovarian carcinoma [ 13 , 14 ]. Mutant p53 expression was defined by one of the following patterns: strong diffuse nuclear staining in at least 80% of tumor cells, complete absence of expression with intact internal control, or diffuse cytoplasmic staining with weak nuclear staining. PD-L1 (22C3) expression was assessed using the Combined Positive Score (CPS), with CPS ≥ 1 considered positive, as previously reported [ 15 ]. HER2 expression was evaluated according to the 2016 ASCO/CAP guidelines for gastroesophageal adenocarcinoma and categorized as 0, 1 +, 2 +, or 3 + [ 16 ]. The presence of a microcystic, elongated, and fragmented (MELF) invasion pattern in the invasive front of endometrial carcinoma [ 17 ] was also assessed for IA3-stage cases. Time-independent and categorical variables were analyzed using the chi-square test or Fisher’s exact test, as appropriate. Progression-free survival (PFS) was defined as the time from surgery to disease progression or the last follow-up. Kaplan–Meier survival curves and the log-rank test were used for PFS analysis. A P value < 0.05 was considered statistically significant. Statistical analysis was conducted using SPSS 20.0 software.

We identified SEEOCs in 4.3% of endometrioid endometrial carcinoma cases (five IA3-stage). Poor tumor differentiation, > 50% myometrial invasion, LVSI, lymph node metastasis, and the 2023 FIGO staging system predicted unfavorable PFS. Survival analysis was limited by sample size and follow-up duration, particularly in IA3 cases. High HER2 and PD-L1 (22C3) expression suggests potential for targeted therapies. These findings provide insights into SEEOC characteristics and treatment, although larger studies are needed to optimize management.

SEEOC has traditionally been diagnosed as either endometrial primary, ovarian primary, or independent primaries based on clinicopathological criteria, which has often led to inconsistencies in diagnosis among pathologists. In the past decade, molecular research has revealed that nearly all sporadic SEEOCs, and a significant proportion of those associated with Lynch syndrome, are clonally related and originate from the endometrium [ 6 – 9 , 18 , 19 ]. These findings support the concept of SEEOC as a metastatic extension of endometrial carcinoma and have prompted further investigation into the clinicopathological features, immunophenotypes, and clinical outcomes of this distinct entity. In our single-institution study, SEEOC accounted for 4.3% of all endometrioid endometrial carcinomas, and IA3-stage tumors represented 0.43% of the total. The incidence of IA3 in our cohort was consistent with previous reports, including a 0.53% rate from the Commission on Cancer’s National Cancer Database and 0.55% from Fudan University Shanghai Cancer Center [ 20 , 21 ]. During our limited follow-up period, none of the IA3 patients experienced recurrence, supporting a favorable prognosis for this stage and consistent with conclusions from Matsuo et al. and Atallah et al. that this population has favorable outcomes and potential for treatment de-escalation [ 11 , 12 ]. The presumed metastatic route for IA3 tumors may involve exfoliation of endometrial tumor cells, transtubal spread, and implantation in the ovary—representing an indolent metastatic mechanism through natural anatomical pathways [ 9 ]. Another hypothesis is that endometrium harboring cancer-associated mutations sheds and implants in the ovary, which may also explain the frequent presence of endometriosis around ovarian tumors in SEEOC cases [ 8 ]. In our experience, IA3-stage diagnosis must strictly adhere to the established criteria and involve joint decision-making between pathologists and gynecologists. A diagnosis of IA3 should not be made if ovarian capsule rupture is found during surgery or in postoperative pathology, nor if cytology of peritoneal washings or ascitic fluid is positive or suspiciously positive. It is worth mentioning that there is still controversy regarding whether cases with focal LVSI could be classified as IA3 stage. Our data, analyzed using Kaplan–Meier survival curves, showed that focal LVSI was associated with a poor prognosis comparable to that of substantial LVSI. This suggests that cases exhibiting focal LVSI may not be appropriately classified as IA3 stage. Among our IA3 stage cases, one patient had focal LVSI. Although this patient did not experience recurrence, the follow-up period was only three months, limiting any conclusions drawn from this single case. In previous studies emphasizing the importance of the three-tier classification of LVSI, Bosse et al. demonstrated that substantial LVSI, compared to focal or no LVSI, was the strongest independent prognostic factor for pelvic regional recurrence, distant metastasis, and overall survival [ 22 ]. Similarly, Qian et al. concluded that the presence and extent of LVSI, along with cervical stromal invasion, are important predictors of lymph node metastasis in endometrial carcinoma [ 23 ]. However, the former study focused only on Stage Ⅰ endometrioid endometrial carcinomas, while the latter examined endometrial serous carcinoma. Previous studies showed that patients with stage IA3 had a more favorable prognosis compared to those with stage IIIA1 [ 11 , 20 ]. However, these studies did not include patients with focal LVSI in the IA3 category, as information regarding LVSI was not available in the databases [ 11 , 20 ]. The survival impact of the three-tier classification of LVSI on synchronous SEEOCs requires further investigation with a larger number of cases. In recent years, significant progress has been made in the molecular subtyping of endometrial carcinoma, providing a foundation for personalized prognosis assessment and treatment planning. According to the 2020 WHO classification, endometrial carcinoma is divided into four molecular subtypes: POLE-mutant, mismatch repair-deficient, no specific molecular profile, and p53 mutant. Since our specimens span from 2009 to 2024, only a few recent cases underwent POLE testing, and all were wild-type. The rate of dMMR in our SEEOC cohort was 36.1%, slightly higher than that reported for endometrioid endometrial carcinomas (around 30%) [ 24 – 26 ], but still within a comparable range and much higher than that for endometrioid ovarian carcinoma (around 10%) [ 27 – 29 ]. Mutant p53 expression was seen in 13.2% of cases, consistent with previously reported rates for both endometrioid endometrial and ovarian carcinomas [ 24 – 28 , 30 ]. The relatively high dMMR rate further supports the classification of SEEOC as a variant of endometrioid endometrial carcinoma other than ovarian carcinoma. We also examined PD-L1 (22C3) and HER2 expression, which showed positivity in 63.0% and 75.0% of cases, respectively. These high expression rates suggest potential opportunities for immunotherapy and HER2-targeted treatment, especially in selected advanced or recurrent cases. To date, risk factors for SEEOCs remain poorly defined, partly due to diagnostic inconsistencies stemming from the traditional criteria that often resulted in cases being misclassified as ovarian or independent primaries. In 2004, Soliman et al. reported on 84 cases of synchronous primary endometrial and ovarian carcinomas—most of which involved endometrioid or mixed histology—and evaluated prognostic factors [ 29 ]. Their study found that early-stage ovarian carcinomas tended to have better prognoses than advanced-stage tumors, though the difference was not statistically significant. Meanwhile, prognosis of endometrial carcinoma appeared independent of FIGO stage, histologic grade, myometrial invasion, LVSI, or lymph node metastasis. In retrospect, these findings reflect the limitations of the dual-primary tumor concept. In our study, we regarded all SEEOC tumors as originating from the endometrium and examined PFS risk factors accordingly. We found that poor tumor differentiation, > 50% myometrial invasion, LVSI, lymph node metastasis, and advanced 2023 FIGO stage were predictors of unfavorable PFS according to Kaplan–Meier analysis. Our results emphasize the importance of evaluating not only intrauterine but also extrauterine involvement during pathological diagnosis, with particular attention to LVSI. The limitation of this study is that the limited number of SEEOC cases and limited follow-up duration particularly in ⅠA3 cases precluded a reliable Cox multivariate survival analysis. Although we could not explicitly clarify whether 2023 FIGO staging system provides independent prognostic significance among various risk factors, distinguishing ⅠA3 from ⅢA1 cases reflects the improved precision of FIGO 2023. Previous retrospective and molecular studies have achieved breakthroughs in understanding tumor histogenesis and prognosis of SEEOCs. Our study represents a phenotypic investigation built upon prior research. Future studies with larger case series are needed to further enhance the understanding of this rare entity.

Synchronous endometrial and ovarian carcinomas occupy a small proportion of gynecological malignancies, with an incidence that varies among different populations, ranging from 3.3% to 10.0% in patients with endometrial or ovarian carcinoma [ 1 – 4 ]. These tumors may exhibit similar or different histological types across sites, including endometrioid, clear cell, serous, or mixed carcinoma. Synchronous endometrioid endometrial and ovarian carcinoma (SEEOC) is the most common, and presents a clinical dilemma in determining independent primaries from metastatic disease. Historically, the diagnostic criteria proposed by Scully et al. have guided the differentiation between primary and metastatic SEEOCs based on clinicopathological parameters including tumor size, depth of invasion, atypical endometrial hyperplasia, presence of ovarian endometriosis, and localization of vascular invasion [ 5 ]. However, a pivotal 2016 study by Schultheis et al., using whole-exome massively parallel sequencing or high-depth targeted massively parallel sequencing, revealed that 22 of 23 SEEOC cases were clonally related. The exception was a case with Lynch syndrome [ 6 ]. Another study conducted during the same period yielded similar results [ 7 ]. Subsequently, one study focused on SEEOC cases with DNA mismatch repair (MMR)-deficiency syndromes, and identified that even in Lynch syndrome cases, most SEEOCs were shown to originate from the endometrium [ 8 ], highlighting the limitations of traditional diagnostic paradigms. Several articles proposed two models of ovarian metastasis either of an invasive way (direct invasion, lymphovascular, transcoelomic, etc.) or of an indolent way (transtubal, including exfoliation and intraepithelial spread) [ 8 , 9 ]. The two metastatic patterns differ in tumor genetic changes, tumor growth speed and mode, resulting in distinct biological behaviors and prognoses. These mechanisms may explain ovarian involvement in seemingly early-stage endometrial cancers. Following these pivotal molecular discoveries, it is now widely accepted that almost all SEEOCs, either sporadic or inherited, originate from the endometrium. Consequently, the FIGO 2023 staging system incorporated IA3 stage, representing low-grade endometrioid carcinomas limited to the uterus and ovary and meeting all of the following criteria: (1) no more than superficial myometrial invasion; (2) absence of substantial LVSI; (3) no extrauterine metastasis; and (4) a unilateral ovarian tumor confined within an intact capsule (equivalent to pT1a) [ 10 ]. Although patients with IA3 tumors are considered to have favorable outcomes and may not require adjuvant therapy [ 11 , 12 ], the clinical acceptance of this staging and its treatment recommendations remain challenging for many gynecologic oncologists. This study aimed to investigate SEEOC cases, including mixed histologies with endometrioid components, to assess their incidence, clinicopathological and immunophenotypic features, and identify prognostic factors to enhance diagnostic accuracy and inform treatment decisions.