Each year, over 49,000 women are diagnosed with uterine cancer, and approximately 8000 women succumb to the disease (1). The vast majority of uterine tumors are adenocarcinomas. Less than 5% of these are uterine sarcomas.
Endometrial adenocarcinoma is more commonly diagnosed among older women with the peak incidence occurring in the sixth decade of life. It is also more commonly diagnosed in white women compared to non-white women. These cancers can be broadly categorized into two types based on clinical and pathologic factors (2):
• Type I endometrial carcinomas follow an estrogen-dependent pathway. The precursor lesion is atypical hyperplasia. These tumors comprise the majority of endometrial cancers, are limited to the uterus, and have a favorable prognosis.
• Type II endometrial carcinomas appear to develop independent of estrogen exposure. They occur more frequently among black women and arise in a background of atrophic endometrium. Compared to type I endometrial cancers, women tend to be diagnosed at an older age and at a later stage. They also confer a poorer overall prognosis.
Risk factors for the development of type I endometrial cancer include states related to excess estrogen stimulation. This includes nulliparity, unopposed estrogen administration, tamoxifen exposure, polycystic ovarian syndrome, and obesity. Higher parity, smoking, and use of estrogen-progestin hormonal contraception are known to decrease risk. In contrast, specific risk factors for type II endometrial cancer have not been identified.
Genetic factors contribute to only about 10% of endometrial cancers, mostly due to hereditary nonpolyposis colorectal cancer (HNPCC), also known as Lynch syndrome, and, to a lesser degree, Cowden syndrome. Both of these genetic conditions follow autosomal dominant inheritance patterns. While it is not clear if individuals with BRCA 1 and 2 mutations have increased risk, there does seem to be higher risk for patients with breast cancer, perhaps because of shared risk factors.
There is no single molecular event that gives rise to endometrial cancer. However, type I and type II endometrial carcinomas are associated with distinct molecular changes: type I endometrial adenocarcinomas are associated with mutations involving the PTEN pathway or show evidence of microsatellite instability (3). Type II tumors are associated with p53 mutations and mutations involving HER2/neu (3, 4).
The World Health Organization classifies endometrial adenocarcinoma into multiple types including endometrioid, serous, and clear cell, among the more common variants (5). Another variant of endometrial carcinoma is carcinosarcoma, which is considered a high-risk histology.
Endometrioid carcinoma is the most common endometrial cancer, comprising 75%–80% of all cases. These cancers occur with varying degrees of differentiation, characterized histologically by grade. Well (grade 1) to moderately (grade 2) differentiated endometrioid carcinomas are considered type I endometrial carcinomas. Under the microscope they may have varying degrees of complexity, such as back-to-back, cribiform, or confluent growth (5). In contrast, poorly differentiated (grade 3) endometrioid adenocarcinomas are more aggressive tumors and some experts classify them as a type II tumor. Grade 3 endometrioid adenocarcinomas are characterized by at least 50% of a solid growth pattern or prominent nuclear atypia (5).
Serous carcinomas (previously referred to as papillary serous) account for 5%–10% of endometrial cancer. These cancers are typically irregular in histologic appearance with branching papillae and single cells characterized by large prominent nucleoli (5). These tumors are characteristic of type II endometrial carcinoma. They are uniformly aggressive, regardless of stage.
Clear cell carcinomas are characterized by clear cytoplasm and are associated with a high degree of cytological atypia. Typical architectural patterns include solid, papillary, and tubolocystic variants. As with serous carcinomas, clear cell carcinomas are also considered a type II tumor. They are aggressive variants, which may not respond well to chemotherapy (6).
Undifferentiated carcinomas of the endometrium are characterized by specific findings, including: a predominantly noncohesive proliferative pattern, heterogeneously sized monotonous cells without marked nuclear pleomorphism, and lacking glandular differentiation (5). These tumors are also marked by brisk mitotic activity and extensive evidence of necrosis.
Carcinosarcomas are metaplastic carcinomas composed of epithelial (carcinomatous) and mesenchymal (sarcomatous) elements. The sarcomatous element is histologically defined as being either homologous (i.e., sarcoma arising from tissue native to the uterus) or heterologous (i.e., sarcoma element that is not native to the uterus). Metastatic lesions due to carcinosarcoma most commonly comprise of the epithelial element.
The classic presenting sign of endometrial cancer is abnormal uterine bleeding. However, other symptoms can be seen, including bloating, pelvic pain, or dyspareunia. However, type II endometrial cancers may not present with symptoms until advanced disease is present, at which time, systemic symptoms (nausea, vomiting, change in bowel habits, anorexia) may be present.
The diagnosis of endometrial cancer is usually made based on endometrial biopsy or dilation and curettage. If the sampling is negative but clinical suspicion of a malignancy is high, however, further evaluation is warranted. This can be performed by repeat endometrial sampling (preferably, biopsy), by diagnostic hysteroscopy, or with pelvic imaging (typically, ultrasound or MRI).
Endometrial carcinoma, which includes carcinosarcoma, is surgically staged using the Federation of Gynecology and Obstetrics (FIGO) staging system. The staging technique requires total hysterectomy and bilateral salpingo-oophorectomy. Whether or not lymphadenectomy is required for all cases of endometrial adenocarcinoma remains controversial (7).
The prognosis of newly diagnosed endometrial cancer are illustrated by the delineation of tumors into risk groups, used for both prognostication and for informing treatment decisions:
Low-risk—These include tumors with low-risk features including low to intermediate grade, minimal to no myometrial invasion, and absence of lymphovascular space invasion (LVSI). These patients have an excellent prognosis following surgery and expected survival is over 90%.
Intermediate-risk—These include cancers confined to the uterus with myometrial invasion (stage IA or IB) or occult cervical stromal invasion (stage II). These patients have approximately an 80% chance of overall survival. Additional factors further divide this group into high- and low-intermediate risk disease, including deep myometrial invasion, grade 2 or 3 differentiation, or the presence of lymphovascular invasion (LVSI).
High-intermediate risk criteria used by the Gynecologic Oncology Group for trial purposes include patients of any age with all three pathologic factors (grade 2 or 3, outer 1/3 invasion, LVSI), patients 50–69 years old with two factors, or patients who are 70 years or older with only one factor (8). In the absence of these additional risk factors, patients are considered to have low-intermediate risk.
High-risk—Women with clear cell or serous carcinomas and women with carcinosarcoma constitute the high-risk group, regardless of stage at diagnosis. This also includes women with stage III endometrial cancer that has been optimally resected. These patients are at an increased risk for both recurrence and death.
Low and low intermediate risk—Women with low- or low intermediate-risk endometrial cancer have an excellent prognosis following surgery alone. As such, the risks of adjuvant radiation therapy likely outweigh any benefit of treatment. In addition, adjuvant endocrine therapy is not indicated. A meta-analysis of adjuvant progestin therapy showed no benefit in the risk of mortality at 5 years compared with postoperative surveillance (9).
High intermediate-risk—Women with high intermediate-risk warrant adjuvant therapy due to an elevated risk of a local recurrence. For most patients, vaginal brachytherapy is sufficient. This was demonstrated in the Post Operative Radiation Therapy for Endometrial Cancer (PORTEC 2) trial, which compared pelvic to vaginal brachytherapy and showed that these treatments were equivalent and resulted in similar rates of locoregional or distant recurrence (10, 11).
There is no indication for adjuvant chemotherapy in this select group of women. However, a randomized trial being conducted by the Gynecologic Oncology Group (GOG 249) aims to characterize the benefit of combined modality treatment (vaginal brachytherapy plus carboplatin/paclitaxel chemotherapy) compared to whole pelvic radiation (12).
High-risk—Women with high-risk disease should be treated with adjuvant systemic chemotherapy, which has replaced the use of radiation therapy. This was based on the results of GOG 122, a randomized phase III trial that compared whole abdominal radiotherapy to 8 cycles of cisplatin and doxorubicin chemotherapy (AP) in stage III/IV disease with minimal residual tumor burden following surgery (13). Chemotherapy significantly increased progression-free (hazard ratio 0.71, 95% CI 0.55–0.91) and overall survival (HR 0.68, 95% CI 0.52–0.89). This translated to 5-year progression-free survival rates of 42% versus 38%, and overall survival rates of 53% and 42% for chemotherapy versus radiation, respectively.