Clinical Studies of Endometrial Cytology and Cervical Methylation Assays in Endometrial Cancer Screening and Fertility-Preservation Evaluation

Endometrial cancer (EC) represents a significant public health concern for women globally, with nearly 200,000 new cases diagnosed each year. It ranks as the fourth most common gynecological malignancy threatening women's health, particularly prevalent among women in developed countries . In China, EC holds a prominent position in gynecological oncology, with studies indicating that its incidence rate is surpassed only by breast and cervical cancers . In recent years, the incidence of endometrial cancer has been steadily rising, particularly among perimenopausal and postmenopausal women, with a notable trend toward younger age groups .

According to traditional etiological classifications, endometrial cancer can be divided into two types: Type I, which is estrogen-dependent, typically affects younger women, is characterized by lower cellular and tissue atypia, and is generally associated with a better prognosis. In contrast, Type II is independent of estrogen exposure, often exhibits high levels of cellular and tissue atypia, affects older women, and is linked to poorer prognoses.The 2023 FIGO guidelines introduced a molecular classification system for endometrial cancer based on gene expression profiles. This system identifies four molecular subtypes: POLE-mutant, mismatch repair-deficient (dMMR), no specific molecular profile (NSMP), and p53-mutant (p53abn). These classifications have shown improved prognostic accuracy and have been incorporated into the 2023 FIGO staging criteria. As a common gynecological malignancy, recurrent or metastatic endometrial cancer has a poor prognosis, which underscores the critical importance of early detection and diagnosis.

1.2 Early Screening Methods for Endometrial Cancer The search for effective early screening methods for endometrial cancer has garnered increasing attention from researchers and clinicians globally, with a growing body of studies addressing this issue. Currently, the primary method for early diagnosis remains pathological examination through endometrial curettage, which is considered the clinical gold standard. However, curettage has several limitations, including procedural complexity, cost, risk of intrauterine bleeding, potential iatrogenic infection, and patient discomfort.In addition to curettage, imaging techniques such as transvaginal ultrasound (TVU) are widely used for clinical diagnosis. TVU is a non-invasive, safe, and convenient method. However, its diagnostic accuracy is often limited by the apparent thickness of the endometrium, which can be influenced by various factors such as individual uterine anatomy and the patient's menstrual cycle phase. Blood spectrometry can detect endometrial cancer at all stages and may assist in identifying atypical endometrial hyperplasia. Additionally, several potential biomarkers, such as microRNAs, have been discovered. However, blood tests generally have low specificity and sensitivity, making them prone to interference from other pathological factors, which limits their clinical application prospects.

1.3 Endometrial Cytology Examination A review of commonly used screening methods reveals that each has its limitations and cannot be considered an optimal or universally applicable approach. Consequently, there is a pressing need for a more efficient, cost-effective, and low-risk screening method for endometrial cancer.

Liquid-based endometrial cytology is an emerging minimally invasive screening technique. This method involves collecting endometrial cells and tissues from the uterus and its downstream anatomical structures. The samples are preserved and analyzed using either liquid-based cytology techniques or tissue smear methods to reach a pathological diagnosis. Compared to invasive procedures such as fractional curettage, liquid-based cytology offers advantages in the screening of endometrial cancer, including being minimally invasive, convenient, and low-risk. In some clinical guidelines, liquid-based endometrial cytology has been recommended as a preferred screening method for endometrial cancer.

Currently, several instruments are used for endometrial cytology sampling, including the Tao Brush, Li Brush, pipelle catheter, and SAP-1 endometrial sampler. The Tao Brush and Li Brush are primarily used in Western countries, while the SAP-1, a domestically developed device, is more commonly applied in China. Recent studies have demonstrated that the Tao Brush provides a sample adequacy rate ranging from 89.9% to 100%, with a diagnostic concordance rate with histopathological findings of 91% to 96%. For diagnosing atypical endometrial hyperplasia and endometrial cancer, the Tao Brush exhibits sensitivity ranging from 91.7% to 100% and specificity of 96%. Similarly, the Li Brush demonstrates a sensitivity of 92.73% and specificity of 98.15% for screening endometrial cancer and endometrial hyperplasia without atypia. The pipelle aspiration method, known for its simplicity and reliability, shows a sample adequacy rate of 73.9% to 100%, with a sensitivity of 77.4% to 94.1% and a specificity of 98.5% to 100% for diagnosing endometrial cancer. Moreover, the pipelle aspiration method demonstrates both high sensitivity (99.6%) and specificity (98% to 100%) for detecting endometrial cancer and its precursor lesion, atypical endometrial hyperplasia. The SAP-1 endometrial sampler has also proven effective, with sensitivity, specificity, positive predictive value, and negative predictive value of 91.7%, 100.0%, 100.0%, and 99.3%, respectively, for diagnosing endometrial cancer. For endometrial hyperplasia without atypia, the SAP-1 device offers sensitivity, specificity, positive predictive value, and negative predictive value of 82.0%, 100.0%, 100.0%, and 99.4%, respectively. Among these methods, Pipelle suction has been most widely studied in clinical applications, with advantages in specimen quantity and accuracy. However, it is limited by a lack of visual control during sampling, which often restricts the sample range. Brush sampling covers a broader range and is effective for cell collection, but may not adequately capture small tissue samples. The SAP-1 sampler was designed to balance the advantages of both, though larger-scale studies are still needed to further evaluate its effectiveness.

Liquid-based cytology testing for endometrial cancer involves collecting specimens from the endometrium, cervix, vagina, and urine. A cohort study investigating vaginal specimens demonstrated that endometrial cells obtained via vaginal smears, analyzed using fluorescence in situ hybridization (FISH) targeting three genes associated with endometrial cancer (CTNNB1 on chromosome 3p22.1, FBXW7 on chromosome 4q31.3, and APC on chromosome 5q22.2), produced significant results in screening for endometrial cells. Among 65 confirmed cases of endometrial cancer, only 2 patients (3%) exhibited no abnormalities in the expression levels of these three genetic loci, resulting in a sensitivity of 97%. In a trial with 70 subjects, setting the optimal threshold by defining a positive result as having >33.60% of cells in the vaginal smear show apparent pathological features yielded a sensitivity of 0.91, specificity of 0.83, and an area under the ROC curve (AUC) of 0.94 (95% CI: 0.88-0.99) . Additionally, a multicenter study divided subjects into five cohorts based on their disease background and employed three methods for collecting endometrial cells: cervical smears, vaginal swabs, and self-collection. These samples were then analyzed using the Women's Cancer Risk Identification Quantitative Polymerase Chain Reaction (WID-qEC) test. The WID-qEC technique uses polymerase chain reaction (PCR) to assess CpG island methylation in specific regions of the ZSCAN12 gene and the GYPC gene for the cytological screening of endometrial cancer. At the optimal threshold, the sensitivity for detecting endometrial cancer (EC) in cervical, self-collected, and vaginal swab samples from symptomatic patients was 97.2% (95% CI, 90.2-99.7), 90.1% (83.6-94.6), and 100% (63.1-100), respectively, while the specificity was 75.8% (63.6-85.5), 86.7% (79.3-92.2), and 89.1% (77.8-95.9), respectively. This multicenter study also included a prospective trial where cervical samples were used to predict the occurrence of endometrial cancer over a defined period. WID-qEC successfully identified 90.9% (95% CI, 70.8-98.9) of endometrial cancer cases a year before the subjects were ultimately diagnosed. The results of this multicenter trial showed consistent performance across different demographic groups, including menopausal status, age, cancer stage, grade, and ethnicity.

In another study that utilized cervical smear samples, WID-qEC screening for endometrial cancer achieved sensitivities of 86% and 90% in 64 confirmed endometrial cancer cases and 225 control subjects, respectively, with an area under the curve (AUC) of 0.92 (95% CI: 0.88-0.97). In a prospective study of 150 healthy Swedish women, WID-qEC screening confirmed 54 cases of endometrial cancer after three years. The sensitivity and specificity were 52% and 98%, respectively, with an AUC of 0.82 (95% CI: 0.74-0.89) . In terms of endometrial sampling, the pipelle catheter is the primary method used internationally. A meta-analysis of 12 studies, published between 1986 and 2022, encompassing 1,607 participants, evaluated the use of the pipelle for endometrial cancer screening. The results indicated that the sensitivity, specificity, positive likelihood ratio, and negative likelihood ratio of the pipelle test were 0.774 (95% CI: 0.565-0.900), 0.985 (95% CI: 0.927-0.997), 97.000 (95% CI: 14.000-349.000), and 0.241 (95% CI: 0.101-0.442), respectively. These results were comparable to those of diagnostic curettage. For conventional diagnostic curettage, the sensitivity, specificity, positive likelihood ratio, and negative likelihood ratio were 0.880 (95% CI: 0.281-0.993), 0.984 (95% CI: 0.956-0.995), 59.300 (95% CI: 14.200-153.000), and 0.194 (95% CI: 0.007-0.732), respectively . A similar prospective study from China examined endometrial cytology using the SAP-1 endometrial sampler. This study compared the results with those obtained through hysteroscopic examination and found that the sensitivity was 87.5%, and the specificity was 100%. However, the sample size of this study was relatively small, necessitating further validation with larger studies .

1.4 Cervical Cell Methylation Testing Cervical cell methylation testing is another promising method for methylation-based screening of endometrial cancer. This approach targets specific genes that undergo methylation in the early stages of endometrial cancer. The combined methylation analysis of four genes (CD01, BHLHE22, ACTB, and SEPTIN9) uses menstrual blood or vaginal secretions as samples, offering advantages such as safety, non-invasiveness, and self-collection. This method has shown high sensitivity and specificity for the early detection of endometrial cancer. Ongoing research suggests that the sensitivity and specificity of cervical cell methylation testing can reach 93.2% and 96%, respectively, although this data is yet to be published .