In gynecological malignancies, the coexistence of endometrial cancer and ovarian cancer is a relatively rare yet not uncommon phenomenon. Approximately 10% of ovarian cancer patients and 5% of endometrial cancer patients present with this condition, accounting for 0.63% of female reproductive system malignancies. The most common pathological type is endometrioid adenocarcinoma, accounting for about 46% to 85% ^[1]. Clinical data indicate that about 5% of endometrial cancer patients have concurrent ovarian cancer, and 10% of ovarian cancer patients have concurrent endometrial cancer. The prognosis of such patients is highly heterogeneous, with a 5 - year survival rate ranging from 30% to 90%. Currently, the distinction between primary double cancers and metastatic cancers mainly relies on the histopathological criteria established by Scully, but there is still a misdiagnosis rate of 15% to 30%.

In 2013, The Cancer Genome Atlas (TCGA) first established a molecular classification system for endometrial cancer. Since 2015, the promotion of practical schemes such as ProMisE has provided a new perspective for the precise diagnosis and treatment of synchronous endometrial and ovarian endometrioid carcinoma (SEO - EC). Molecular classification, by analyzing four subtypes: POLE gene mutations, mismatch repair deficiency (MMRd), p53 status, and no specific molecular features (NSMP), can not only clarify the clonal origin of the two lesions (independent primary vs. clonal - related metastasis) but also guide treatment stratification and prognosis prediction.

In the past 5 to 10 years, genomic technology has confirmed through mapping and comparing the genomic landscapes of the uterine and ovarian cancer foci in SEO - EC that the vast majority of SEO - EC is the result of endometrioid carcinoma originating in the uterine body metastasizing to the ovary. A review summarizing and analyzing the literature on genomic studies of SEO - EC found that in 160 cases of SEO - EC, the vast majority (96.9%) of the uterine and ovarian cancer foci were clonal, and only 5 cases (3.1%) were confirmed to be independent primaries, and among them, 3 cases were all Lynch syndrome.

A study included 28 cases, and NGS testing was performed on both the uterine and ovarian foci. It was ultimately found that the histological grades of the uterine and ovarian cancer foci were the same in 23 cases (82.1%), and 96.4% (27 cases) of SEO - EC patients had consistent molecular classification in the uterine and ovarian cancer foci, with only 1 case being different. All patients had shared somatic gene variant sites, with an average of 7.4 (1 to 41) shared somatic gene variant sites per patient. The most common shared somatic variant genes were PTEN (64.3%, 18/28), PIK3CA (46.4%, 13/28), ARID1A (28.6%, 8/28), CTNNB1 (25.0%, 7/28), and KRAS (21.4%, 6/28) ^[2]. Another study included 17 patients (11 with metastasis and 6 with double primaries) and found that 94% (16/17) had clonal relatedness ^[3].

01. The Value of Molecular Classification in SEO - EC

The traditional Scully pathological criteria (1998) distinguish between double primary cancers (low - risk) and metastatic cancers (high - risk) based on morphological features such as tumor size, differentiation degree, and vascular invasion, but they have significant limitations. Molecular classification of endometrial cancer can achieve precise differentiation through the following dimensions:

1. Clonal correlation analysis: Multiple studies have shown that SEO - EC has clonal correlation (same molecular subtype + shared gene mutations), but the clinical behavior remains low - risk, suggesting that "clonal origin ≠ metastasis", and prognosis depends on the molecular subtype rather than the clonal relationship.

2. Specificity of the p53abn subtype: The proportion of p53abn in metastatic cancer is 63.6%, while in double primary cancer it is only 9.7%, indicating that p53abn is a strong predictor of metastatic cancer (OR = 11.07, 95% CI 1.45 - 84.50).

3. Heterogeneity of MMR status: Most SEO - EC double lesions have consistent molecular subtypes, but there are still a few inconsistent cases, suggesting that molecular testing must be performed separately for each lesion to avoid misdiagnosis due to sampling from a single lesion.

Literature findings indicate that the proportion of Lynch syndrome patients in seromucinous endometrial carcinoma (SEO - EC) (ranges from 7% to 14%, which is higher than that in endometrial adenocarcinoma (2% - 6%) and endometrioid ovarian cancer (approximately 6%). This implies that for SEO - EC, during the process of molecular classification, a comprehensive battery of tests should be carried out to exclude hereditary cancer syndromes resulting from germline pathogenic gene variations and to identify patients who may benefit from immunotherapy.

02. Summary

Molecular classification of endometrial cancer in ovarian and uterine double primary cancer has moved from research to clinical practice and has become a core tool for solving diagnostic dilemmas, optimizing treatment strategies, and accurately assessing prognosis. The "zero recurrence" feature of the POLEmut subtype supports a de - escalation treatment approach, the high invasiveness of the p53abn subtype requires intensified treatment, the MMRd subtype provides a target for immunotherapy, and the NSMP subtype requires further stratification in combination with more biomarkers - this "molecular - first" diagnosis and treatment model makes precision medicine for SEO - EC possible.

In the future, with the standardization of detection techniques, the implementation of prospective investigations, and breakthroughs in targeted/immunotherapies, molecular classification is expected to further reshape the clinical management of SEO - EC, ultimately attaining the objectives of "personalized treatment" and "maximizing therapeutic efficacy while minimizing toxicity".

References

[1] Chinese Journal of Pathology, 2025, 54(11): 1240-1246.

[2]  Chinese Medical Journal, 2025, 105(08): 584-591.

[3]  Gynecol Oncol. 2025; 178:234-242.

Notice: This article is only for sharing. If there are any issues such as copyright infringement, please contact us as soon as possible and we will correct it immediately. Thank you!