Cancer of the Uterine Corpus
Amelia M. Jernigan
Amanda Nickles Fader
Endometrial cancer is the fourth most common cancer in women and the most common gynecologic malignancy, accounting for 6% of all female cancers.
EPIDEMIOLOGY OF UTERINE CANCER
In the United States and other developed countries, 1 in 38 women will develop uterine cancer, making it the most common gynecologic malignancy in these settings. The American Cancer Society estimates that there will be 49,560 new cases and 8,190 deaths from uterine cancer in 2013. The incidence has increased 0.8% each year since 1998. Most commonly, these are endometrial cancers; only 2% of uterine cancers are sarcomas. Seventy-two percent of cases will be localized at the time of diagnosis because endometrial cancer often presents with postmenopausal or irregular bleeding.
Risk Factors for Uterine Cancer
A woman’s risk of endometrial cancer increases with age. The median age at diagnosis is 61 years, and the peak incidence occurs from ages 55 to 70 years. Women older than 50 years old account for 90% of the diagnoses of endometrial cancer and 5% develop disease before age 40 years.
Other risk factors are based on increased estrogen exposure.
Estrogen replacement without concomitant progesterone carries a relative risk of 4.5 to 8.0 and persists for 10 years after treatment is stopped (Table 47-1).
Chronic anovulation states, such as seen in polycystic ovarian syndrome (PCOS), lead to constant estrogen stimulation of the endometrium and increase the risk of cancer due to the lack of a corpus luteum to produce progesterone.
Obesity increases endogenous estrogen by peripheral conversion of androstenedione to estrogen by aromatase in adipose tissues. Nearly 70% of early-stage endometrial cancer patients are obese. The relative risk of death increases with increasing body mass index (BMI), and a BMI >30 kg/m2 will triple the risk of endometrial cancer.
Nulliparity (related to infertility) and diabetes mellitus are independent risk factors and have a relative risk of two or three for endometrial cancer, whereas the association of hypertension seems related to obesity.
A woman taking tamoxifen has an annual risk of 2 in 1,000 of developing endometrial cancer and 40% of women will develop cancer more than 12 months after stopping therapy.
Women with hereditary nonpolyposis colon cancer (HNPCC) syndrome have a 39% risk of developing endometrial cancer by age 70 years.
Some factors can decrease the risk of endometrial cancer.
Factors that decrease circulating estrogen, such as cigarette smoking and oral contraceptive pill (OCP) use, may be protective.
TABLE 47-1 Risk Factors for Endometrial Cancer
Risk Factor
Relative Risk
Nulliparity
2.0
Estrogen replacement without progesterone
4-8
Obesity
30-49 pounds
3.0
>50 pounds
10.0
Type 2 diabetes mellitus
2.8
Tamoxifen
2.2
From Barakat RR, Markman M, Randall ME, et al. Corpus: epithelial tumors. In Hoskins WJ, Perez CA, Young RC, eds. Principles and Practice of Gynecologic Oncology, 2nd ed. Philadelphia, PA: Lippincott-Raven Publishers, 1997:884.
OCPs decrease endometrial cancer risk by 40%, even up to 15 years after discontinuation, and this protection increases with length of use. Four years of use reduces risk by 56%, 8 years decreases risk by 67%, and 12 years of use decreases risk by 72%.
Hyperplasia appears to be the precursor lesion for most endometrial cancer. A study that followed women for 10 years after a diagnosis of hyperplasia showed that the risk of progression to cancer increased from simple hyperplasia to complex, and the presence of atypia further increased the risk.
A recent study revealed that 43% of hysterectomies performed in community hospitals for complex atypical hyperplasia will have endometrial cancer on final pathology.
PRESENTATION, EVALUATION, AND DIAGNOSIS
Clinical Presentation
Seventy-five percent to 90% of endometrial cancer cases present with postmenopausal bleeding. In one study of women with postmenopausal bleeding, 7% had cancer, 56% had atrophy, and 15% had endometrial hyperplasia.
The likelihood that postmenopausal bleeding is due to cancer significantly increases with a woman’s age. One study showed that 9% of women in their 50s with postmenopausal bleeding had endometrial cancer, whereas the rate was 16% for women in their 60s, 28% for women in their 70s, and 60% for women in their 80s.
Although endometrial cancer is mostly a disease of postmenopausal women, 20% of cases are diagnosed before menopause. Perimenopausal menometrorrhagia, especially in women at high risk for endometrial cancer, should be investigated with endometrial biopsy.
Abnormal cervical cytology can prompt the workup for and diagnosis of endometrial cancer. However, routine Pap smear screening will only detect half of endometrial cancer cases and is not a screening test. Workup should be considered with the following Pap smear results:
Endometrial cells (remote from menstrual bleeding) in a woman older than 40 years
Atypical glandular cells of undetermined significance (the risk of endometrial cancer in women older than 35 years with this Pap result is 23%)
Adenocarcinoma (consider endometrial and cervical sampling)
Some cases of endometrial cancer are discovered incidentally at the time of hysterectomy. If this is the case, a surgeon skilled on endometrial cancer staging procedures should be involved if available. To avoid this, it is recommended that all women with abnormal uterine bleeding have endometrial sampling prior to their hysterectomy.
With a few exceptions, there are no guidelines or recommendations to screen for endometrial cancer.
Tamoxifen
Women who are on tamoxifen have an increased risk of developing endometrial cancer.
Routine screening with ultrasound or endometrial biopsies is not recommended in this setting.
Use of ultrasound for screening is of limited use because tamoxifen causes subepithelial stromal hypertrophy and therefore increases the thickness of the endometrial stripe and may result in unnecessary surgical procedures.
Women on tamoxifen should be counseled on warning signs and followed with yearly pelvic exams. Any episode of vaginal bleeding should trigger an evaluation.
Hereditary Nonpolyposis Colon Cancer
HNPCC, or Lynch syndrome, is inherited in an autosomal dominant fashion, resulting from a germline mutation in one of the mismatch repair genes (MMR genes MLH1, MSH2, MSH6) and comprises the majority of inherited cases of endometrial cancer. It also increases risk of cancer of the colorectum, small intestine, ureter, renal pelvis, and ovary.
Genetic assessment for HNPCC is strongly recommended in women with a 20% to 25% risk of HNPCC.
This includes women with a family pedigree meeting Amsterdam criteria, patients with metachronous or synchronous colorectal and endometrial or ovarian cancers before age 50 years, or those with a first- or second-degree relative with a known germline mutation in an MMR gene.
Women with HNPCC have a high risk of endometrial cancer (as high as 50% lifetime). A prophylactic hysterectomy should be considered.
There is limited evidence regarding endometrial cancer screening in this population, but current recommendations advise women to undergo endometrial biopsies annually starting at the age of 30 to 35 years or 10 years before the age that first case appeared in the family and to undergo hysterectomy and bilateral salpingooophorectomy once childbearing is completed.
Evaluation and Diagnosis of Postmenopausal Bleeding
The appropriate evaluation for postmenopausal bleeding is widely debated. Ultrasound and endometrial biopsy are the two main tools available.
Ultrasound
Pelvic ultrasound measurement of the endometrial stripe can be used with a minimum cutoff of 5 mm in thickness based on the Postmenopausal Estrogen/Progestins
Intervention (PEPI) trial. This trial showed that with a 5-mm cutoff, postmenopausal ultrasound has a positive predictive value of 9%, a negative predictive value of 99%, a sensitivity of 90%, and a specificity of 48% for endometrial cancer.
Meta-analysis shows that the posttest probability of cancer following a pelvic ultrasound with a stripe <5 mm is 2.5%. Conversely, a stripe >5 mm conveys a 32% posttest probability of cancer.
Even with a thin stripe, if a woman persistently bleeds, sampling should be performed.
Biopsy
Endometrial biopsy provides a cancer detection rate of 99.6% in premenopausal women and 91% in postmenopausal women. Specificity is 98%, and sensitivity is 99%. The false-negative rate is between 5% and 15%.
The posttest probability of endometrial cancer is 82% if the biopsy is positive and 0.9% if it is negative.
However, a biopsy read as “insufficient sample” should trigger further evaluation because on further investigation, 20% of these women will have pathology and 3% will have cancer.
Further Workup
No matter which method is used for initial evaluation, if bleeding persists or clinical suspicion is high, further evaluation with a dilation and curettage (D&C) should be pursued. The false-negative rate of a D&C is 2% to 6%.
Hysteroscopy with D&C has a positive predictive value of 96%, a negative predictive value of 98%, a sensitivity of 98%, and a specificity of 95%.
A recent Gynecology Oncology Group (GOG) prospective study demonstrated the difficulty in diagnosing complex atypical hyperplasia (CAH). One third of cases of CAH were deemed to be “less than” CAH by study pathologists, one third were deemed to be “greater than” CAH (i.e., endometrial cancer), and another third of the diagnoses were consistent with the original diagnosis of CAH.
STAGING AND PROGNOSIS
Pretreatment Evaluation
Complete history, assessing for hereditary cancer syndromes
Complete physical exam including comprehensive pelvic exam assessing the size and mobility of the uterus and assessment for metastasis (i.e., supraclavicular lymphadenopathy)
Consider cancer antigen 125 (CA-125). Elevated CA-125 levels are associated with metastatic disease and can be used to follow the patient if it was elevated at diagnosis.
Imaging: Chest imaging should be ordered. A plain film is reasonable. A computed tomography (CT) or magnetic resonance imaging (MRI) is not necessary if surgical staging is planned. If no surgery is planned, an MRI is the best modality to assess myometrial or cervical and lymph node involvement.
Surgical Staging Procedures
Staging for endometrial cancer is performed surgically, and because many patients have early-stage disease at the time of diagnosis, this is often the only intervention necessary.
Surgical staging most commonly involves a minimally invasive surgical approach for apparent early-stage disease. The procedure includes total extrafascial
hysterectomy, bilateral salpingo-oophorectomy, a pelvic and periaortic lymph node assessment/dissection, as well as cytoreduction of all visible disease.
Peritoneal washings are not part of surgical staging, but if performed, they should be obtained as the first step once abdominal access has been achieved.
Omentectomy should be performed if serous or clear cell histology is suspected.
Current standard of care is minimally invasive surgery when possible. The Gynecologic Oncology Group Study LAP 2 (GOG LAP-2) randomized trial demonstrated that a laparoscopic approach to endometrial cancer staging was feasible and safe with similar intraoperative complications but fewer postoperative adverse events and a shorter hospital stay.
Long-term follow-up data suggest similar recurrence rates and 5-year overall survival in the open and laparoscopic groups.
The lymphatics of the uterine fundus drain to the aortic nodes; the lower uterine segment drains to the internal and external iliac lymph nodes; and the round ligaments can drain to the superficial inguinal lymph nodes.
Pelvic and para-aortic lymph node dissection is required for complete surgical staging of endometrial cancer. Pelvic lymph node dissection involves the removal of nodal tissue from the distal half of each common iliac artery, the anterior and medial proximal of each external iliac artery, and vein and the distal half of the obturator fat pad anterior to the obturator nerve.
Para-aortic lymph node dissection involves the removal of nodal tissue over the distal vena cava from the inferior mesenteric artery to the mid common iliac artery and between the aorta and ureter from the inferior mesenteric artery to the left mid common iliac artery.
Morbid obesity may render a lymph node dissection more challenging, but it is still a required component of the procedure if indicated based on histologic and pathologic risk factors (see the following text).
One pitfall of lymph node dissection is the occurrence of lymphedema (5% to 20%). The incidence increases with the removal of more nodes and administration of adjuvant radiation.
Furthermore, the performance of lymph node dissection is not clearly associated with improved survival. Retrospective data supports improved survival with more extensive lymph node resection, especially in patients with high-risk features. However, Consolidated Standards of Reporting Trials (CONSORT) and Adjuvant External Beam Radiotherapy in the Treatment of Endometrial Cancer (ASTEC) are two prospective randomized trials which demonstrated no difference in survival when lymph node dissection was performed.
Importantly, ASTEC did not require para-aortic lymphadenectomy and is criticized for including many patients who were low risk and would not have benefited from lymph node dissection in the first place as well as a low median number of lymph nodes resected. In the CONSORT trial, more nodes were removed (median = 30) but only 26% of patients in the lymphadenectomy group underwent paraaortic dissection. Therefore, it is difficult to generalize and state that a full lymph node dissection conveys no survival benefit. Furthermore, many would argue that it is the information gained from the dissection that helps guide adjuvant therapy and may convey a survival benefit, which was not adequately tested in these trials.
The decision as to whether or not to proceed to lymph node dissection is often made in the operating room based on frozen section evaluation of the uterus for histologic cell type, tumor differentiation (grade), and depth of myometrial invasion.
TABLE 47-2 Lymph Node Metastasis by Grade and Depth of Invasion of Endometrial Cancer
Pelvic Lymph Nodes (%)
Grade
No Invasion
Inner 1/3
Mid 1/3
Outer 1/3
1
0
3
0
11
2
3
5
9
19
3
0
9
4
34
Para-aortic Lymph Nodes (%)
1
0
1
5
6
2
3
4
0
14
3
0
4
0
23
From Creasman WT, Morrow CP, Bundy BN, et al. Surgical pathologic spread patterns of endometrial cancer. A Gynecologic Oncology Group Study. Cancer 1987;60:2035-2041.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree