Residency · Residency · Obstetrics Gynecology

Endometrial Cancer: Evaluation and Staging

Epidemiology

Endometrial cancer is the most common gynecologic malignancy in the United States, with approximately 66,000 new cases and 13,000 deaths annually. Its incidence is rising, closely linked to the obesity epidemic. The median age at diagnosis is 63 years. The overall 5-year survival is 81%, which is relatively high because most cases are diagnosed at an early stage. A significant racial disparity exists: Black women have higher mortality despite similar incidence, due to higher rates of aggressive histologic subtypes, later stage at diagnosis, and disparities in access to care.

Risk Factors

Estrogen-Related (Type I)

Unopposed estrogen exposure is the central driver of type I endometrial cancer. Obesity with a BMI above 30 carries a relative risk of 2 to 4 because adipose tissue aromatizes androgens to estrogen. Chronic anovulation, as seen in polycystic ovary syndrome, increases risk through unopposed estrogen stimulation of the endometrium. Other estrogen-related risk factors include nulliparity, early menarche, late menopause (all prolonging estrogen exposure), estrogen-only hormone replacement therapy without progestogen in women with a uterus, tamoxifen use (which acts as an estrogen agonist on the endometrium), estrogen-producing ovarian tumors such as granulosa cell tumors, and diabetes mellitus and metabolic syndrome.

Genetic / Non-Estrogen (Type II)

Lynch syndrome (hereditary nonpolyposis colorectal cancer) confers a 40 to 60% lifetime risk of endometrial cancer through germline mutations in mismatch repair genes (MLH1, MSH2, MSH6, PMS2). Other non-estrogen risk factors include age above 55 and prior pelvic radiation. Serous and clear cell histologies often arise independently of estrogen stimulation.

Protective Factors

Combined oral contraceptive pills confer a 50% risk reduction with 5 or more years of use, and this protection persists for 15 to 20 years. Progestin-containing IUDs, multiparity, and physical activity are also protective. Cigarette smoking is paradoxically protective due to anti-estrogenic effects, though it is not recommended as a preventive strategy.

Histologic Classification

Type I (Endometrioid, 80%)

Type I endometrial cancer is estrogen-driven, with low-grade tumors (grade 1 and 2) predominating. Atypical endometrial hyperplasia is the recognized precursor lesion. The prognosis is generally favorable. PTEN loss and microsatellite instability are common molecular features.

Type II (Non-Endometrioid, 20%)

Type II cancers are less estrogen-dependent and often arise in atrophic endometrium. Serous carcinoma is aggressive, frequently presents at an advanced stage, and demonstrates p53 mutations in nearly all cases. Clear cell carcinoma behaves similarly to serous. Carcinosarcoma (malignant mixed Mullerian tumor) is treated as a high-grade carcinoma with a poor prognosis. Dedifferentiated and undifferentiated carcinoma also fall into this category. Type II cancers account for a disproportionate share of endometrial cancer deaths.

<image>Comparison table and histologic images of Type I (endometrioid) versus Type II (serous, clear cell) endometrial cancers showing differences in precursor lesions, molecular features, typical grade, stage at diagnosis, and prognosis</image>

Molecular Classification (TCGA / ProMisE)

Four Molecular Subtypes

The POLE ultramutated subtype carries an excellent prognosis regardless of grade or stage and may not require adjuvant therapy. The microsatellite instability-high / mismatch repair deficient subtype has an intermediate prognosis, responds to immune checkpoint inhibitors, and should be screened for Lynch syndrome. The copy number low (p53 wild-type) subtype has an intermediate prognosis and includes most endometrioid cancers. The copy number high (p53 mutant) subtype has a poor prognosis, includes most serous and some high-grade endometrioid cancers, and benefits from aggressive adjuvant therapy.

Clinical Impact

Molecular classification is increasingly incorporated into treatment algorithms. The PORTEC-4a trial is evaluating molecular classification to guide adjuvant therapy decisions. Immunohistochemistry for p53 and mismatch repair proteins (MLH1, MSH2, MSH6, PMS2) plus POLE sequencing can classify the majority of tumors.

Evaluation of Suspected Endometrial Cancer

When to Evaluate

Any postmenopausal bleeding, defined as any bleeding after 12 months of amenorrhea, warrants evaluation. Approximately 10% of postmenopausal bleeding is caused by endometrial cancer. In premenopausal women, evaluation is indicated for persistent intermenstrual bleeding, menorrhagia, or irregular bleeding in the presence of risk factors, in all women aged 45 and older with abnormal uterine bleeding, and in women under 45 with abnormal uterine bleeding and risk factors such as obesity, PCOS, chronic anovulation, or tamoxifen use. Incidental endometrial thickening found on imaging also warrants investigation.

Endometrial Thickness

Transvaginal ultrasound with endometrial stripe measurement is the initial imaging study. In postmenopausal women with bleeding, an endometrial thickness of 4 mm or less has a 99% negative predictive value for cancer. If the endometrial stripe exceeds 4 mm or cannot be adequately measured, tissue sampling is indicated. These thickness thresholds are not applicable to premenopausal women or women taking tamoxifen.

Tissue Sampling

Office endometrial biopsy with a Pipelle device is the first-line sampling method, with a sensitivity of 90 to 98% for cancer detection. Hysteroscopy with directed biopsy is performed if the office biopsy is insufficient, non-diagnostic, or discordant with clinical suspicion. Dilation and curettage is used when cervical stenosis prevents office biopsy and can be combined with hysteroscopy. Saline infusion sonohysterography is useful for distinguishing focal from diffuse lesions before biopsy.

Preoperative Workup (Once Cancer Diagnosed)

Pelvic MRI assesses myometrial invasion depth, cervical involvement, and lymph node status. CT of the chest, abdomen, and pelvis is obtained for high-grade histologies or suspected advanced disease. Laboratory studies include CBC, metabolic panel, and CA-125 (elevated in advanced disease and useful for monitoring). Medical comorbidities are assessed, as many patients are obese with diabetes and cardiovascular disease. Referral to gynecologic oncology is appropriate.

Staging (FIGO 2023, Surgical)

Stage I: Confined to the Uterine Corpus

Stage IA involves tumor limited to the endometrium or invading less than 50% of the myometrium. Stage IB involves invasion of 50% or more of the myometrium. Stage IC, new in the 2023 revision, applies to aggressive histologies (serous, clear cell, undifferentiated, carcinosarcoma, mixed) confined to a polyp or endometrium with no myometrial invasion.

Stage II: Cervical Stromal Invasion

Stage IIA involves endocervical glandular involvement only. Stage IIB involves cervical stromal invasion.

Stage III: Local and/or Regional Spread

Stage IIIA involves tumor invading the serosa and/or adnexa. Stage IIIB involves vaginal and/or parametrial involvement. Stage IIIC1 involves pelvic lymph node metastasis. Stage IIIC2 involves para-aortic lymph node metastasis with or without pelvic nodes.

Stage IV: Distant Metastasis

Stage IVA involves invasion of bladder and/or bowel mucosa. Stage IVB involves distant metastasis, including intra-abdominal and inguinal lymph nodes. The 2023 FIGO revision notably incorporates molecular classification into staging for the first time.

<image>FIGO staging diagram for endometrial cancer showing cross-sectional views of the uterus illustrating Stage IA (less than 50% myometrial invasion), Stage IB (50% or more invasion), Stage II (cervical stromal invasion), Stage IIIA (serosal/adnexal spread), Stage IIIC (lymph node involvement), and Stage IV (distant metastasis)</image>

Surgical Management

Standard Surgical Staging

The standard procedure is total hysterectomy with bilateral salpingo-oophorectomy. Sentinel lymph node mapping is increasingly replacing comprehensive lymphadenectomy. Indocyanine green or methylene blue is injected into the cervix, and sentinel nodes are identified in bilateral hemipelves with any suspicious nodes removed. Ultra-staging of sentinel nodes increases detection of micrometastases. The FIRES trial and SHRINK study support sentinel lymph node mapping as adequate for staging. Comprehensive lymphadenectomy with pelvic and/or para-aortic dissection is still performed in some centers and is indicated when sentinel node mapping fails or for high-risk histologies. Omentectomy is performed for serous, clear cell, and carcinosarcoma histologies for peritoneal staging. Peritoneal cytology is no longer part of FIGO staging but is still often collected.

Surgical Approach

Minimally invasive surgery is preferred when feasible, using laparoscopic or robotic-assisted total hysterectomy and bilateral salpingo-oophorectomy with sentinel node mapping. The LAP2 trial demonstrated fewer complications, shorter hospital stays, and similar recurrence rates with minimally invasive surgery compared to open surgery. Open laparotomy is reserved for large uteri, advanced disease, or situations where minimally invasive surgery is not feasible.

Fertility-Sparing Management

Select patients with grade 1 endometrioid adenocarcinoma confined to the endometrium with no myometrial invasion on MRI may be candidates for fertility-sparing management. Progestin therapy is administered, using medroxyprogesterone acetate 600 mg daily, megestrol acetate 160 mg daily, or a levonorgestrel IUD. Endometrial sampling every 3 to 6 months documents treatment response. Hysterectomy is recommended after completion of childbearing. The complete response rate is 70 to 80%, but the recurrence rate is high at 30 to 40%. This approach requires close surveillance and strong patient commitment.

Adjuvant Therapy

Low-Risk (Stage IA, Grade 1-2 Endometrioid)

Observation alone after surgery is sufficient, with no adjuvant therapy needed.

Intermediate-Risk

Vaginal cuff brachytherapy reduces local recurrence. GOG 249 and PORTEC-2 data support brachytherapy alone as equivalent to external beam radiation for local control, with fewer side effects.

High-Intermediate and High-Risk

External beam radiation therapy with or without vaginal cuff brachytherapy is indicated. Chemotherapy with carboplatin and paclitaxel is used for advanced-stage or high-grade histology. The PORTEC-3 trial demonstrated that combined chemoradiation followed by chemotherapy improves outcomes for high-risk endometrial cancer.

Advanced / Recurrent Disease

Systemic chemotherapy with carboplatin plus paclitaxel is the standard first-line regimen. Immunotherapy with pembrolizumab plus lenvatinib is used for mismatch repair deficient / microsatellite instability-high tumors, as demonstrated in the KEYNOTE-775/Study 309. Dostarlimab is available for mismatch repair deficient recurrent endometrial cancer. Hormonal therapy with medroxyprogesterone, tamoxifen, or aromatase inhibitors is an option for low-grade recurrent disease.

<image>Treatment algorithm for endometrial cancer organized by risk group: low-risk (observation), intermediate-risk (vaginal cuff brachytherapy), high-risk (EBRT plus chemotherapy), and advanced/recurrent (systemic chemotherapy, immunotherapy for dMMR tumors, hormonal therapy for low-grade recurrence)</image>

Lynch Syndrome Screening

All endometrial cancers should undergo immunohistochemistry for mismatch repair proteins (MLH1, MSH2, MSH6, PMS2) or microsatellite instability testing. Loss of MLH1/PMS2 should prompt reflex MLH1 promoter methylation testing to distinguish sporadic from Lynch-associated cancers. If Lynch syndrome is suspected, genetic counseling and germline testing are indicated. Lynch syndrome carriers should consider prophylactic hysterectomy and bilateral salpingo-oophorectomy after childbearing and require increased surveillance for colorectal cancer.

Clinical Pearls

Any postmenopausal bleeding warrants evaluation, as approximately 10% is caused by endometrial cancer.

An endometrial thickness of 4 mm or less on transvaginal ultrasound in a postmenopausal woman with bleeding has a 99% negative predictive value, but tissue sampling is still indicated if clinical suspicion is high.

Molecular classification with POLE, mismatch repair status, and p53 is transforming adjuvant therapy decisions. Immunohistochemistry for p53 and mismatch repair proteins should be performed on all endometrial cancer specimens.

Sentinel lymph node mapping is replacing comprehensive lymphadenectomy for staging in apparent early-stage disease.

All endometrial cancers should be screened for Lynch syndrome using immunohistochemistry or microsatellite instability testing.

Fertility-sparing management with progestins is an option for select young patients with grade 1 endometrioid cancer confined to the endometrium, but it requires close surveillance and eventual hysterectomy.

References

• ACOG Practice Bulletin No. 149: Endometrial Cancer (2015, reaffirmed 2023)
• NCCN Clinical Practice Guidelines: Uterine Neoplasms (Version 1.2024)
• Cancer Genome Atlas Research Network. Integrated genomic characterization of endometrial carcinoma. Nature. 2013;497:67-73
• de Boer SM et al. (PORTEC-3). Adjuvant chemoradiotherapy versus radiotherapy alone for women with high-risk endometrial cancer. Lancet Oncol. 2018;19:1395-1407
• Rossi EC et al. (FIRES trial). A comparison of sentinel lymph node biopsy to lymphadenectomy for endometrial cancer staging. Gynecol Oncol. 2017;146:46-52
• Berek JS et al. FIGO staging of endometrial cancer: 2023. Int J Gynaecol Obstet. 2023;162:383-394

Read this lecture as Markdown