Guideline••

Chinese expert consensus on clinical application of molecular classification of endometrial cancer (2024)

•,,,,,,Expert Committee of Obstetrics and Gynecology of the Chinese Research Hospital Association.

...

Endometrial cancer (EC) is one of the most common gynaecologic malignancies, with a rising incidence in China. In 2022, an estimated 84 520 new cases were diagnosed, and 17 543 deaths were reported in China.1 The traditional Bokhman classification, proposed in 1983, categorised EC into type I (oestrogen-dependent) and type II (non-oestrogen-dependent) based on clinicopathological features.2 However, with rapid advances in oncology, the diagnosis and treatment of EC have evolved beyond conventional clinicopathological classifications. Integrating molecular features into precision medicine has emerged as a key challenge in the field. While the Cancer Genome Atlas (TCGA) molecular classification of EC3 was introduced over a decade ago and has garnered increasing attention from Chinese experts and scholars,4 5 its clinical implementation remains in the exploratory phase. In response to this situation, a panel of Chinese gynaecologic oncologists and pathologists has developed this consensus statement, based on current literature and clinical experience, to guide the application of molecular classification in clinical practice.

To ensure the representativeness and authority of the expert panel, experts were selected based on their academic achievements in the field of endometrial cancer, clinical experience and the clinical volume of their respective institutions. The list of experts was recommended by relevant academic societies and confirmed through peer review to ensure broad expertise in both clinical practice and research. The consensus process employed the Delphi method, spanning 13 months and consisting of three rounds of discussions. After each round, panel members re-evaluated the recommendations based on updated data and feedback. These issues were resolved through multiple rounds of discussion and revisions, ultimately leading to the final core recommendations of this guideline.

The recommendation levels and their meanings in this consensus are shown in table 1.

Table 1

•

Recommendation levels and implications in this consensus

Main molecular characteristics of EC patients in China

Methods of molecular classification for EC

The TCGA molecular classification of EC integrates genomics, transcriptomics, proteomics, gene copy number and methylation data, dividing EC into four subtypes: POLE ultramutated (POLE mut), microsatellite instability-high (MSI-H), copy-number low (CNL) and copy-number high (CNH).

As this molecular classification strategy requires the integration of multi-platform and multi-omics data, its clinical application is complex and limited in promotion. Subsequently, many scholars have simplified the methodology and optimised the process of EC molecular classification, proposing the Trans-Post Operative Radiation Therapy for EC (TransPORTEC)6 and Proactive Molecular Risk Classifier for Endometrial Cancer (ProMisE)7 molecular classification strategies. These approaches use immunohistochemistry (IHC), Sanger sequencing and next-generation sequencing (NGS) technologies for detection, classifying EC into POLE mutant, mismatch repair deficiency (MMRd), no specific molecular profile (NSMP) or p53 wild-type (p53wt) and p53 abnormal (p53abn) subtypes. These methods are simpler to operate while ensuring consistency with the TCGA molecular classification.

In 2020, EC molecular classification was integrated into the National Comprehensive Cancer Network (NCCN) guidelines8 and the WHO Classification of female genital tract tumours (fifth edition).9 The NCCN guidelines adopted the TCGA classification nomenclature with some modifications, while the WHO guidelines drew from the TransPORTEC and ProMisE classification methods. In 2023, the International Federation of Gynaecology and Obstetrics (FIGO) new staging for EC also incorporated the TCGA molecular classification into EC management. It is recommended that all EC patients undergo molecular classification and the subscript ‘m’ be added to the FIGO stage. For FIGO stages I and II, patients with POLE mut molecular subtype are adjusted to stage IAm_POLE_mut; those with p53abn and myometrial invasion are adjusted to stage IICm_p53abn; NSMP and MMRd subtypes do not affect staging, but the subscript can be added to the FIGO stage for data collection purposes. For FIGO stages III and IV, the stage is not affected by molecular classification, but the subscript can still be added to the original FIGO stage for data collection. The FIGO 2023 staging system integrates histology and molecular classification to more accurately guide prognosis and treatment.10 In this consensus, we have retained the original nomenclature used in the referenced literature, as shown in table 2, while in the narrative and recommendation sections, we uniformly use the WHO molecular classification nomenclature. This study adopts the10 EC staging system to ease reading and understanding.

Table 2

•

Molecular classification nomenclature by different organisations

Main molecular characteristics of EC patients in China

Currently, the testing and application of EC molecular classification in China is still in its early stages. In 2024, Peking University People’s Hospital reported molecular classification results for 479 EC cases based on NGS and immunohistochemistry (IHC) methods: POLE mut type, 28 cases (5.85%); MMRd type, 67 cases (13.99%); p53abn type, 60 cases (12.53%); and NSMP type, 324 cases (67.64%).11 This shows that compared with published research results on EC molecular subtypes in Caucasian populations, the proportion of NSMP in Chinese EC patients is significantly higher, indicating a unique situation distinct from European and American populations. Therefore, it is necessary to conduct molecular classification testing at various levels of hospitals using recognised different detection methods and means to explore more precise diagnosis and treatment plans suitable for EC in the Chinese population.

Recommendations: it is recommended that all patients diagnosed with EC undergo molecular classification testing when conditions allow. The testing methods and procedures are recommended to refer to the ‘WHO Classification of female genital tract tumours (fifth edition)’9 and the ‘Chinese Expert Consensus on Molecular Detection of Endometrial Cancer (2021 Edition)’12 (Recommendation level: Category 2A). For MMRd-type patients, with informed consent, it is recommended to perform Lynch syndrome-related germline gene mutation testing and provide genetic risk counselling, assessment and follow-up (Recommendation level: Category 2A).

Molecular classification guiding fertility-preserving treatment

With the increasing incidence of EC in younger populations, the demand for fertility-sparing management strategies in EC patients has become a pressing clinical concern. The ProMisE classification system has been proven to have high consistency between diagnostic endometrial specimens (biopsy or curettage) and whole uterine specimens.13–15 Molecular classification has important reference value for determining whether patients can undergo fertility-preserving treatment, and molecular characteristics help predict the efficacy of endocrine therapy.16–18

POLE mut type

POLE mut-type EC patients tend to be younger at onset, with tumours often being high-grade and having significant tumour-infiltrating lymphocytes (TILs), but with a good prognosis.19 Therefore, POLE mut-type patients may benefit from conservative treatment, and it is worth exploring whether the indications for fertility-preserving treatment can be broadened for this type of patient.20 However, some reports suggest that some POLE mut-type patients experienced disease progression or recurrence after conventional progestogen treatment,16 so further research is needed on whether conventional endocrine therapy should be used for this subset of patients.

Recommendations: POLE mut-type EC patients have a better prognosis. It is recommended to conduct fertility-preserving treatment under strict evaluation and when the lesion can be adequately removed. The treatment plan is recommended to be part of clinical trials (Recommendation level: Category 2B).

MMRd type

Compared with mismatch repair proficient (MMRp) EC patients, MMRd-type patients have the following characteristics: lower body mass index, later clinical stage, higher grade of differentiation, higher proportion of deep myometrial invasion and lymphovascular invasion and shorter overall survival (OS) and progression-free survival (PFS).16 Some reports suggest that EC and atypical endometrial hyperplasia patients with MMRd-type have significantly lower remission rates with progestogen treatment compared with MMRp-type.17 In 2021, Korea published the first study on the impact of molecular classification on the prognosis of fertility-preserving treatment in EC, finding that MMRd-type patients had significantly lower overall response rates and 6 month complete remission rates to progestogen therapy compared with p53wt patients.16 Even after complete remission from treatment, MMRd-type patients had a higher recurrence rate, possibly due to their high tumour mutation burden (TMB-H) (≥10 mutations/Mb), which activates non-progestogen receptor-dependent molecular pathways, thereby affecting the response to progestogen.21 Currently, there are no reports on the efficacy of immunotherapy and targeted therapy in MMRd-type patients undergoing fertility-preserving treatment.

Studies have shown that women with Lynch syndrome can show endometrial response to progestogen treatment,22 and there are case reports of Lynch syndrome patients achieving complete remission and fertility after conservative treatment.22–24 The NCCN guidelines do not consider Lynch syndrome as an absolute contraindication for fertility-preserving treatment. However, due to the poor response to progestogen therapy in the corresponding MMRd type, patients need to be fully informed about the possibility of insensitivity to progestogen treatment, and close follow-up is required during treatment, along with screening of other systems such as gastrointestinal endoscopy. The offspring of Lynch syndrome patients have a 50% genetic risk, so preimplantation genetic testing is recommended to prevent inheritance.

Recommendations: MMRd-type EC patients have a poor response to progestogen therapy and an increased risk of recurrence after treatment, so fertility-preserving treatment is not recommended as the first choice (Recommendation level: Category 2A). If fertility-preserving treatment is to be performed, it should be decided cautiously after a comprehensive individualised assessment of the situation and needs and multidisciplinary evaluation. It is recommended to conduct clinical trials on immunotherapy and targeted therapy (Recommendation level: Category 2B). For Lynch syndrome patients, if fertility-preserving treatment is to be performed, they need to be fully informed of the risks of failure, the risk of other tumours and the genetic risk to offspring. During the preservation process, management should be carried out according to the relevant guidelines and consensus for Lynch syndrome patients (Recommendation level: Category 2B).

NSMP type

NSMP-type EC is often low-grade, early-stage, with high expression of ER and PR,13 and responds well to progestogen therapy. Reports have found that among young EC patients under 50 years old, NSMP-type patients are younger than other types, with survival rates lower than the POLE mut group but higher than MMRd and p53abn groups.25 However, not all NSMP-type patients respond well to progestogen. Since most of the Chinese population and fertility-preserving patients are of this type, it is necessary to further refine the subtype characteristics of this group. Immunohistochemical analysis of prognostic markers shows that positive expression of PR and ER is associated with good prognosis in NSMP-type EC, while L1CAM overexpression is associated with poor prognosis.26 Additionally, studies have shown that 1q.32.1 copy-number amplification and DNA damage are associated with worse prognosis in NSMP-type EC patients.27 28 These molecular markers may help further stratify this type of EC.

Recommendations: fertility-preserving treatment is recommended for NSMP-type patients. Pathological and molecular characteristics can be combined for further stratification to assess individual risk and guide individualised treatment (Recommendation level: Category 2A). For NSMP patients, it is recommended to test for oestrogen receptors (ER), as ER-positive cases may benefit from hormone-targeted therapies. NSMP-type patients lack specific molecular characteristics, so further research on re-classification is necessary (Recommendation level: Category 2A).

p53abn type

p53abn EC are predominantly serous or clear-cell carcinomas characterised by high invasiveness and a propensity for lymphovascular space invasion (LVSI), myometrial invasion and distant metastasis.29 These tumours have the worst prognosis among EC subtypes. In principle, fertility-sparing treatment should not be considered for patients with p53abn EC.29 30 Abnormal p53 protein expression detected by IHC is associated with progestogen resistance. Combining p53 status and Ki67 expression levels may independently predict the time to remission for EC patients.18 Due to a 10% inconsistency rate between immunohistochemical protein detection and genetic testing for patients with fertility preservation needs and IHC suggesting p53abn type, genetic testing is recommended to further confirm the classification.

Recommendations: for patients considered p53abn type by immunohistochemistry, genetic testing (next-generation sequencing) is recommended for verification (Recommendation level: Category 2A). p53abn-type patients have a poor prognosis and poor response to progestogen therapy, so fertility-preserving treatment is not recommended (Recommendation level: Category 2A).

Application of EC molecular classification in surgical approach selection for early-stage EC patients

Two large-scale randomised controlled clinical trials have confirmed that for EC confined to the uterus, there is no significant difference in recurrence and survival between patients receiving minimally invasive and open surgery.31 32 Minimally invasive surgery has the advantages of lower intraoperative blood loss, lower perioperative complication rates and shorter hospital stays compared with open surgery.33 34 Meta-analyses based on multiple randomised controlled trials have shown that minimally invasive and open surgery can bring similar OS and DFS in the treatment of early-stage EC.35 Based on this evidence, current American and European EC diagnosis and treatment guidelines recommend minimally invasive approaches for comprehensive staging surgery in patients with EC confined to the uterus.36 37

However, recent retrospective studies suggest that molecular characteristics may influence the prognosis of EC patients after different surgical approaches. For POLE mut- and MMRd-type patients, there was no significant difference in survival between minimally invasive and open surgery groups. For p53abn-type patients, the PFS of the minimally invasive surgery group was significantly worse than the open surgery group.38 39 For NSMP-type patients, although there is currently no clear research evidence suggesting that the surgical approach has a significant impact on patient prognosis, considering its high proportion in the Chinese population and potential internal heterogeneity, further research is needed to explore new indicators for patient stratification to guide surgical approach selection. Additionally, these conclusions still lack further verification from relevant prospective randomised controlled studies, so their clinical application should still be cautious.

Recommendations: for early-stage EC patients, in regions where conditions allow, the molecular classification results of endometrial biopsy specimens can be referred to for surgical approach selection. For POLE mut-, MMRd- and NSMP-type patients, minimally invasive surgery (including laparoscopic surgery and robot-assisted laparoscopic surgery) is still recommended (Recommendation level: Category 1). For p53abn-type patients, due to the lack of relevant clinical evidence, the pros and cons of open and minimally invasive surgery and the potential survival risks of minimally invasive surgery should be fully explained to the patient. The surgical approach should be decided jointly by the doctor and patient after consultation (Recommendation level: Category 2B).

Application of endometrial cancer molecular classification in adjuvant therapy

The European Society of Gynaecological Oncology (ESGO)/European Society for Radiotherapy and Oncology (ESTRO)/European Society of Pathology (ESP) jointly released EC management guidelines in 2021, incorporating molecular classification into prognostic stratified management,37 making subsequent adjuvant treatment planning clearer and more orderly. Combining existing research reports and relevant guideline recommendations, this consensus summarises different risk factors and corresponding risk stratification for each molecular type, with recommendations for postoperative adjuvant therapy as follows:

POLE mut type

In the 2021 ESGO/ESTRO/ESP EC guidelines, FIGO 2009 stage I-II POLE mut-type EC (regardless of pathological type, tumour grade, LVSI and myometrial invasion status) belongs to the low-risk group10 IAm_POLE_mut stage.10 37 A retrospective analysis of the observation group in the PORTEC-1 trial suggests that the 10 year recurrence-free survival rate of POLE mut-type EC is significantly higher than that of POLE wild-type. In vitro experiments have confirmed that embryonic stem cells with POLE mutations are not sensitive to radiotherapy and chemotherapy.19 The mechanism of good prognosis in POLE mut-type EC is not fully understood. Currently, it is believed that POLE mut-type tumours have an extremely high mutation burden, can produce a large number of neoantigens and are often accompanied by massive lymphocyte infiltration. Therefore, the active tumour immune response induced by tumour cells may be one of the reasons. In addition, TILs in POLE mut-type EC show PD-1 overexpression, making this subtype a potential beneficiary population for immunotherapy.40 41 On the other hand, POLE mut-type tumour cells accumulate a large number of passenger mutations in their genomes, which may disrupt cellular physiological functions or even lead to tumour cell death, which is another possible reason for the good prognosis.42 Clinical evidence for advanced and recurrent POLE mut-type EC is extremely limited, and clinical trials are urgently needed.

Recommendations: FIGO 2023 stage IAm_POLE_mut EC patients do not need adjuvant therapy (Recommendation level: Category 2A). There is no data to support whether adjuvant therapy should be performed for stage III–IVm_POLE_mut EC patients without residual lesions, and clinical trials are recommended (Recommendation level: Category 2B). For recurrent, metastatic and refractory POLE mut-type cancer patients, data are lacking. Based on their high mutation burden and PD-1 overexpression, the second-line treatment can consider the adjuvant use of immune checkpoint inhibitors (ICIs) targeting PD-1/PD-L1 and other molecules, and clinical trials are recommended (Recommendation level: Category 2B).

MMRd type

The 2021 ESGO/ESTRO/ESP guidelines combine FIGO 2009 staging and molecular classification to stratify endometrial cancer prognosis risk: (a) low-risk group: Stage I–II POLE mut EC, no residual disease; Stage IA MMRd/NSMP endometrioid carcinoma + low-grade + LVSI negative or focal; (b) intermediate-risk group: Stage IA MMRd/NSMP endometrioid carcinoma +high-grade + LVSI negative or focal; Stage IA p53abn without myometrial invasion; Stage IB MMRd/NSMP endometrioid carcinoma +low-grade + LVSI negative or focal; (c) intermediate-high risk group: Stage IB MMRd/NSMP endometrioid carcinoma high-grade regardless of LVSI status; Stage I MMRd/NSMP endometrioid carcinoma+substantial LVSI regardless of grade and depth of invasion; Stage II MMRd/NSMP endometrioid carcinoma; and (d) high-risk group: Stage III–IVA MMRd/NSMP endometrioid carcinoma with no residual disease; Stage I–IVA p53abn endometrial carcinoma with myometrial invasion, without residual disease. Advanced metastatic: Stage III–IVA with residual disease; Stage IVB of any molecular type.37 Among MMRd type patients, the above low-risk group roughly corresponds to10 stage IAm_MMRd, the intermediate-risk group corresponds to stage IB-ICm_MMRd, the intermediate-high risk group corresponds to stage IIA-IIBm_MMRd; and the high-risk group corresponds to stage IIC-IVAm_MMRd.

A 2019 multicentre retrospective study analysed the relationship between postoperative adjuvant radiotherapy and prognosis in high-grade stage IB-II EC, finding that MMRd-type patients had better DFS after adjuvant radiotherapy than the untreated group, recommending MMR status as an indicator for predicting radiotherapy efficacy.43 In a retrospective analysis of PORTEC-2, there was no significant difference in OS between vaginal brachytherapy and pelvic external beam radiotherapy for MMRd-type patients, suggesting that intermediate-risk MMRd-type EC is more suitable for vaginal brachytherapy with fewer side effects.44 The PROTEC-3 study evaluated the prognostic impact of adjuvant radiotherapy vs adjuvant chemoradiotherapy in EC patients with high-risk factors, showing no statistically significant difference in recurrence rates and survival rates between adjuvant chemoradiotherapy and adjuvant radiotherapy in MMRd-type patients.45

Furthermore, the heterogeneity within MMRd has attracted attention. MMRd can be roughly divided into two genetic mechanisms: promoter methylation-related (mainly MLH1 gene) and MMR gene mutation-related. Multiple studies have shown that MLH1 promoter methylation is an independent risk factor affecting disease-specific survival.46–49 Research has shown that MLH1 promoter methylation-related MMRd-type EC responds worse to immunotherapy than MMR gene mutation-related patients.50 Still, no studies have systematically explored the sensitivity of these two MMRd subtypes to radiotherapy and chemotherapy. Therefore, future re-stratification of MMRd-type EC and exploration of corresponding adjuvant treatment strategies are of great significance.

Another characteristic of MMRd-type EC is the presence of a large number of tumour neoantigens and lymphocyte infiltration, which is considered to be the largest beneficiary population for immunotherapy among the four molecular types.41 47 51 In 2018, NCCN guidelines recommended considering TMB testing for recurrent, high-risk EC patients to guide immunotherapy. In 2019, pembrolizumab was recommended by NCCN guidelines, becoming an effective systemic treatment option for MSI-H/MMRd recurrent, metastatic and high-risk EC.52 Drugs such as nivolumab, dostarlimab and avelumab have also subsequently proven their effectiveness in the EC treatment. They are recommended in NCCN guidelines for second-line treatment of recurrent and metastatic EC patients.52–54

Recommendations: FIGO 2023 IAm_MMRd patients do not need adjuvant therapy (Recommendation level: Category 2A). FIGO 2023 IB-ICm_MMRd stage patients are recommended for postoperative adjuvant vaginal brachytherapy (Recommendation level: Category 2A). FIGO 2023 IIA-IIBm_MMRd stage patients are recommended for postoperative adjuvant pelvic external beam radiotherapy (Recommendation level: Category 2A). FIGO 2023 IIC-IVAm_MMRd is recommended for concurrent/sequential chemoradiotherapy (Recommendation level: Category 2A). For recurrent/metastatic MMRd-type patients, second-line treatment with immunotherapy targeting PD-1/PD-L1 is recommended (Recommendation level: Category 2B). MLH1 promoter methylation testing is recommended for MMRd-type patients. MMRd-type EC caused by MLH1 promoter methylation has a poor prognosis, and it is recommended to conduct trials to explore the best adjuvant treatment strategy (Recommendation level: Category 2B).

NSMP type

According to the 2021 ESGO/ESTRO/ESP guidelines, early-stage NSMP-type EC has a risk stratification similar to MMRd-type. Therefore, adjuvant radio-chemotherapy for newly diagnosed NSMP patients can generally follow the risk grouping and corresponding treatment of MMRd patients. For advanced/high-risk cases, the GOG-86P clinical trial, which included 349 patients with advanced or recurrent EC, found through comparative analysis that patients with CTNNB1 mutations had longer PFS when receiving bevacizumab treatment along with chemotherapy.55 A large phase III clinical trial, KEYNOTE-775/Study 309, showed that pembrolizumab combined with lenvatinib had significantly better clinical effects than chemotherapy alone.56 The 2022 NCCN guidelines recommend pembrolizumab combined with lenvatinib as second-line treatment for patients with recurrent, progressive non-MMRd EC.57 Recently, a phase II clinical trial exploring the efficacy and safety of sintilimab combined with anlotinib for recurrent or advanced EC in the Chinese population showed that patients receiving this combination had objective response rates and disease control rates of 73.9% and 91.3% respectively, suggesting the effectiveness of ICIs combined with tyrosine kinase inhibitors (TKIs) in the Chinese population.58

It is worth noting that in the TCGA molecular classification, EC patients not classified as POLE mut, MMRd or p53abn are all classified as NSMP-type. Histopathologically, NSMP is a highly heterogeneous group that can manifest as endometrioid carcinoma, dedifferentiated/undifferentiated carcinoma, carcinosarcoma and mesonephric-like carcinoma. Some traditional immunohistochemical markers (such as ER, PR) are prognostic protective factors for the NSMP subtype, while mutations in KRAS, ARID1A, CTNNB1 and L1CAM overexpression are associated with poor prognosis.59 60 Considering these studies, for this type which accounts for 70% of EC cases in China, clinical practice should not treat it entirely as a single subgroup similar to MMRd-type. Instead, traditional clinicopathological indicators and specific molecular testing results should be combined for comprehensive judgement to explore individualised adjuvant treatment strategies.

Recommendations: early-stage NSMP-type patients’ adjuvant treatment should follow that of MMRd. For recurrent/metastatic NSMP-type patients, CTNNB1 gene testing can be performed. If this gene is mutated, bevacizumab can be considered in combination with chemotherapy (recommendation level: 2B). For recurrent/metastatic MSS/MMRp patients, second-line treatment recommends ICI combined with TKI (recommendation level: 2B).

p53abn type

According to the 2021 ESGO/ESTRO/ESP guidelines, p53abn type is a high-risk factor. The PROTEC-3 study suggests that in p53abn patients, adjuvant radio-chemotherapy can improve 5 year RFS rates and 5 year OS rates, confirming that p53abn type can benefit from supplementary chemotherapy.45 HER2 protein is a transmembrane protein with tyrosine kinase activity, and about 30% of uterine serous carcinoma patients overexpress HER2/Neu, suggesting the potential application of HER2-targeted drugs. HER2 testing should follow College of American Pathologists guidelines for scoring. This is crucial not only for selecting trastuzumab therapy but also for determining eligibility for trastuzumab deruxtecan treatment. A phase II randomised controlled clinical trial reported in 2018 that for HER2-positive advanced or recurrent uterine serous carcinoma patients, the combination of carboplatin + paclitaxel + trastuzumab improved PFS compared with chemotherapy alone.61 However, another phase II clinical trial published in 2010, which included 28 patients with HER2-overexpressing stage III–IV or recurrent uterine serous carcinoma, showed no significant clinical efficacy after adding trastuzumab treatment,62 the reasons for which require further research. Studies evaluating RAD51 protein function found that 46% of p53abn type endometrial cancers have characteristics of homologous recombination deficiency (HRD),63 and the RED trial in RAINBO proposes to compare the efficacy of adjuvant radio-chemotherapy combined with niraparib or placebo in p53abn patients. These studies suggest that p53abn-type endometrial cancer may be a potential beneficiary population for PARP inhibitors.

Recommendations: FIGO stage IIC-IVm_p53abn EC patients are classified as high-risk and are recommended for concurrent or sequential radio-chemotherapy (recommendation level: 2A). It is recommended to use IHC combined with the FISH method for HER2/Neu overexpression detection. If HER2 overexpression occurs, trastuzumab can be attempted in combination with chemotherapy (recommendation level: 2B). HRD-related testing is recommended, and if HRD exists, PARP inhibitors can be attempted in combination with adjuvant radio-chemotherapy (recommendation level: 2B).

For recurrent and progressive advanced tumours, regardless of the molecular subtype, comprehensive judgement based on clinicopathological and molecular characteristics is needed. Various exploratory treatments, including chemotherapy, radiotherapy, hormonal therapy, targeted therapy, immunotherapy and combinations, can be attempted. For example, in advanced or recurrent EC, NTRK gene fusion detection can be performed, and larotrectinib/entrectinib-targeted drugs can be used as the second-line treatment.57 The triple combination of PARP inhibitors, anti-PD-L1 and bevacizumab as a non-chemotherapy combination treatment for recurrent EC has also received preliminary clinical safety and efficacy certification. In addition, clinical trials of various pan-cancer molecular markers (such as PI3K/AKT/mTOR, FBXW7, PTEN gene mutations) and their corresponding targeted drug treatments for EC are constantly being explored. Due to their relatively low level of evidence and lack of molecular subtype-related specificity, they are not summarised or recommended in this consensus.

Conclusion

The 2021 ‘Chinese Expert Consensus on Molecular Detection in Endometrial Cancer’ provided a comprehensive review and recommendations on EC molecular classification detection methods, Lynch syndrome genetic screening and molecular targeted marker detection. Building on the previous foundation, this consensus integrates recent domestic and international literature and guidelines, focusing on guiding the application and practice of molecular classification in clinical management of EC in China.

It should be noted that although some prospective trials have gradually been initiated, results have not yet been published. Some recommendations lack high-quality evidence-based evidence, and some results remain controversial. Concurrently, there are many blank areas and issues in the current clinical application stage that warrant discussion, such as how molecular classification guides the selection of surgical extent and whether immunotherapy or targeted drugs based on molecular classification results can be recommended as first-line treatment options for fertility preservation or clinical adjuvant therapy.

With the continuous development of molecular biology techniques and the implementation of prospective clinical research trials, precision diagnosis and treatment of EC are expected to be further improved. This consensus will be continuously updated to facilitate clinical popularisation and application.