Uterine corpus cancer is the most prevalent gynecologic malignancy in American women with over 60,000 new cases expected during the next year and accounting for nearly 11,000 deaths. Endometrial carcinomas account for the greatest number of these cases, as fewer than 10% of uterine corpus cancers are sarcomas. Endometrioid carcinomas compose more than 83% of uterine corpus cancers. More virulent serous and papillary serous carcinomas make up some 4% to 6% of endometrial carcinomas, and 1% to 2% are clear cell carcinomas. It is essential to differentiate type 1 endometrioid from type 2 serous endometrial carcinomas and other highly aggressive non-endometrioid carcinoma histotypes to understand, manage and possibly prevent these diseases.
Present consensus holds that pathogenesis of most endometrial endometrioid carcinomas begins with uninterrupted endometrial proliferation, hormonally stimulated by endogenous or exogenous estrogen unopposed by progesterone or progestins, progressing through states of simple to complex forms of endometrial hyperplasia (EH). Arising in this milieu, histologically recognizable atypical premalignant lesions, defined as endometrial intraepithelial neoplasia (EIN), may transform to endometrioid carcinoma, characterized by stromal and/or myometrial invasion, PTEN mutations, and often KRAS2 mutations, microsatellite instability caused by mismatch repair (MMR) defects, and near-diploid karyotype. Given the hormonal etiology, EIN and endometrial endometrioid carcinomas usually express estrogen and progesterone receptors (ER and PR). Other suspect etiologic factors, including insulin resistance and hyperandrogenemia, are being investigated, but these endometrial carcinogenic mechanisms have not be worked out.
Most endometrial endometrioid carcinomas are lower grade cancers.
International Federation of Gynecology and Obstetrics (FIGO) 3-Grade Assessment of the Glandular Component
Grade 1: Less than or equal to 5% solid non-squamous growth pattern
Grade 2: Six percent to 50% solid non-squamous growth pattern
Grade 3: Greater than 50% solid non-squamous growth pattern, upgrading by 1 grade when there is notable nuclear atypia
Recent Binary System
This system classifies endometrioid carcinomas with less than 50% solid component and absence of marked nuclear atypia as low-grade, and those with greater than 50% solid component and/or marked nuclear atypia are graded as high-grade carcinomas with either endometrioid, serous, clear cell or undifferentiated histology.
Type 1, low-grade endometrial endometrioid carcinomas with less than 50% myometrium invasion have favorable prognosis approaching 100% 5-year survival rates. Low-grade endometrioid carcinomas with more than 50% myometrium invasion and high-grade endometrial cancer have much poorer prognoses with higher proportions of lymph node metastases and post-treatment recurrences. Using binary grading, endometrioid carcinomas classified as high grade on the basis of 2 or more architectural features, which included greater than 50% solid growth pattern without distinction between squamous or non-squamous differentiation, diffuse infiltrative growth and/or tumor necrosis, were associated with only 46% 5-year survival rates compared to 93% 5-year survival rates in patients with low grade endometrioid carcinomas. Aggressively malignant high-grade endometrioid carcinomas with such poor prognosis are more akin to type 2, high-grade non-endometrioid histotypes than they are to low-grade endometrioid carcinomas.
The most prevalent non-endometrioid, type 2 endometrial cancers are high-grade, serous, undifferentiated, and clear cell carcinomas. High-grade serous and clear cell carcinomas overall have been associated with significantly poorer 5-year survival rates than high-grade endometrioid carcinomas. Whether confined or spread beyond the uterine corpus when diagnosed, clear cell carcinomas have been associated with higher 5-year survival rates than serous carcinomas; however, together, type 2 endometrial serous and clear cell carcinomas are associated with significantly poorer survival rates than high-grade endometrioid carcinoma. While the aggressive nature of these major high-grade endometrial cancer histotypes is similar, the prevailing evidence indicates that at least early carcinogenesis differs. Mutations in TP53 with loss of heterozygosity, identified by DNA sequencing and/or aberrant immunohistochemical (IHC) expression of p53, are present in some 90% or more of endometrial serous carcinomas (ESC).
In contrast to low-grade endometrioid carcinomas, ESC typically arises in atrophic, resting, or weakly proliferative endometrium. In postmenopausal women on exogenous hormone replacement therapy, ESC may be diagnosed in a background of the proliferative endometrium and rarely even in the hyperplastic endometrium. Identical TP53 mutations have been found in foci of benign-appearing endometrium (“p53 signatures”), endometrial glandular dysplasia (EmGD) and associated serous endometrial intraepithelial carcinoma (EIC) and invasive ESC, giving credence to a concept of endometrial serous carcinogenesis with accumulation of TP53 mutations recognized first as p53 signatures, progressing to EmGD, to EIC, and ultimately to ESC. Although the initiating event in carcinogenesis from precursor lesions to EIC and invasive ESC has not been identified, an inordinately high proportion of ESC relative to endometrial endometrioid carcinomas has been found in hereditary carriers of BRCA1 mutations, and inordinately high proportions of BRCA1 mutation carriers have been reported in series of women diagnosed with ESC.
Even small ESC tumors are prone to lymphatic infiltration and transtubal dissemination. The propensity of both ESC and non-invasive EIC to metastasize may result from extracellular adhesion alternations in the primary neoplasms due to reduced expression of E-cadherin, CD44 and/or other cell adhesion molecules. Amplification of other genes, such as HER/neu and p16, can be seen in advanced cases of ESC, and p16 protein overexpression (92% to 100% IHC positive) in ESC and mixed ESC-endometrioid carcinomas and high-grade endometrioid carcinoma (25% IHC positive) may help differentiate these cancers, especially when they are associated with low-grade endometrioid carcinoma (7% IHC positive).
The final mutational profile of high-grade endometrial clear cell carcinoma determined by IHC staining for p53 as a surrogate for TP53 mutations is somewhat different than either high-grade endometrioid carcinoma or ESC. Diffuse moderate to strong p53 nuclear staining in greater than or equal to 70% of the tumor cells correctly identifies the presence of TP53 missense mutations, and complete absence of p53 staining correlates with frameshift mutations in TP53. While p53 staining is consistently aberrant in most or all ESC (93% to 100%), p53 is normal in the majority of high-grade endometrioid and clear cell endometrial carcinomas. At the same time, ER staining was found to be positive in most ESC (54% to 93%) and many high-grade endometrial endometrioid carcinomas (50% to 80%) but in only a few clear cell carcinomas (7% to 9%). On the other hand, Heptocyte nuclear factor-1beta (HNF-1beta) stained positively in 100% of clear cell carcinomas and 40% of ESC but in none of the high-grade endometrioid carcinomas; while none of the clear cell carcinomas contained PTEN mutations, which characterize endometrioid carcinomas. When cases of endometrial clear cell carcinomas were segregated for outcome analysis of those with aberrant p53 IHC staining, indicating TP53 mutations, compared to those with wild-type p53 staining, the group of patients who had clear cell carcinomas with aberrant p53 staining trended to worse overall survival (p = 0.07) and had significantly shorter progression-free survival (p = 0.01) than the group with wild-type staining on univariate analysis, but not on multivariate analysis when tumor stage, mitotic index, architectural pattern, necrosis, and patients’ age were considered. Other researchers have confirmed that worsened patient outcomes and advanced stage were associated with p53 overexpression in endometrial clear cell carcinomas.
The molecular etiologies of high-grade endometrioid and clear cell endometrial carcinomas lack the definition so far achieved by research on the pathogenesis of endometrial low-grade endometrioid carcinomas and ESC. Alhough the molecular pathways leading from initial mutational events in early neoplastic endometrial lesions to high-grade endometrioid carcinomas or clear cell carcinomas may be quite different from each other and different from ESC, cascading mutations during carcinogenesis of some eventually high grade endometrioid and clear cell carcinomas may result in the progressive accumulation of TP53 mutations and p53 overexpression in more advanced aggressive cancers demonstrating malignant behaviors similar to ESC.
A brief review of adnexal high-grade serous carcinoma (HGSC) with near ubiquitous TP53 mutations can provide insight into another possible etiology of endometrial high-grade endometrioid carcinomas. Arising in the fallopian tube, the evolution of adnexal HGSC from serous tubal in situ carcinoma (STIC) associated with p53 signatures in tubal endothelium has been established by evidence through molecular, transitional, and epidemiological research. From this, the developmental similarities between adnexal HGSC and ESC are evident for their mutual associations with p53 signatures in morphologically normal epithelium through the apparently in situ phases of STIC transitioning to HGSC and, likewise serous EIN transitioning to ESC with the potential intra-abdominal dissemination of very small, early cancerous lesions. High-grade adnexal carcinomas with aberrant p53 but non-papillary serous, endometrioid and transitional cell (SET) morphology and aggressive behavior are now classified as HGSC-SET carcinomas. In contrast to IHC demonstrating aberrant p53 staining in greater than or equal to 50% of high-grade, non-endometrioid adnexal tumors with associated poor prognosis, a study found that none of the adnexal tumors ultimately classified as endometrioid carcinomas showed aberrant p53 staining, and these patients had no recurrences and better survival. On the basis of translational and pathological research, endometrial high-grade endometrioid carcinomas with aberrant p53 should be classed as type 2 endometrial cancers with expected guarded prognosis.
It is controversial and far from settled whether or not hereditary germline mutations in BRCA1 and/or BRCA2 convey increased risks for ESC such as the known susceptibility of BRCA1 and BRCA2 mutation carriers to adnexal HGSC with nearly ubiquitous TP53 mutations. High-grade adnexal carcinomas with SET morphology were more frequently linked to germline BRCA1 or BRCA2 mutations than to classical HGSC, which were associated with inactivation by BRCA1 promoter methylation or somatic BRCA1 and BRCA2 mutations. And high-grade, adnexal carcinomas with germline BRCA1 and BRCA2 mutations tended to more frequently have SET morphology than HGSC. HGSC were not associated with any BRCA1 or BRCA2 dysfunction. Presently, from what is known of adnexal HGSC and ESC carcinogenesis and embryology of the mullerian system, all epithelial derivatives, including not only the endosalpinx but also the endometrium and endocervix, may be at increased risk for high-grade, type 2 carcinomas associated with BRCA dysfunction.
The prevalence of endometrial cancer in the United States is 25.7/100,000 women per year. The lifetime risk for developing this disease is approximately 2.8% in American women. The peak ages of diagnosis are between ages 55 and 64 years (median 62 years). Though the risk for endometrial cancer is slightly lower in American black women than white women (24.8 versus 26.3 new cases/100,000 per year), the lethality of endometrial cancer in black women significantly exceeds the lethality of this disease in white women (8.1 versus 4.2 deaths/100,000 per year). The discrepancy might be explained by a significantly higher incidence of advanced uterine corpus cancers and aggressive endometrial cancer histotypes, such as high-grade endometrioid and clear cell carcinomas, ESC, carcinosarcomas, endometrial stromal sarcomas in black women. Interestingly, the reported incidences of new endometrial cancer cases and deaths from the disease are lower in Asian/Pacific Islanders, American Indians, and Hispanics than in either American blacks or whites.
As briefly reviewed in the section on Etiology, the development of the most common endometrial cancers and their intraepithelial precursors - type 1, low-grade endometrioid carcinoma, EmGD, and EIC - depends largely on a hormonal milieu of persistent estrus uninterrupted by progesterone or progestins. Thus, estrogen-secreting, ovarian tumors and polycystic ovaries (PCO), which interfere with regular ovulation and menstruation, result in elevated levels of free-estrogens leading to endometrial hyperplasia. Anovulation, of course, conveys infertility, but also nulliparity per se has been associated with elevated risk for endometrial cancer, even after adjusting for infertility. Early menarche and late menopause increase the lifetime exposure to endogenous estrogens and are associated with increased risks for endometrial cancer.
High body mass indices (BMI), as well as diabetes type II and insulin resistance, anovulation, menstrual disruption, amenorrhea, and infertility, are consistently associated with increased risks for low-grade endometrioid carcinoma and in recent studies high-grade endometrioid carcinoma, as well. Endometrial cancer is twice as common in overweight women and more than 3 times as common in obese women.
Obesity and hyperinsulinemia, a common feature of insulin-resistant type II diabetes, lower the levels of sex hormone binding globulin (SHBG) to result in higher circulating levels of free estrogen. In a reported large case-control study, diagnosis of insulin-dependent diabetes type I was found to have an even higher odds ratio for endometrial cancer than was diabetes type II.
Androgen precursors, produced by PCO, ovarian hilus cell hypertrophy, or the adrenal cortex, are converted to estrogen in adipose tissues. Large amounts of subcutaneous adipose tissue and more efficient conversion of androgens to estrogen may result in states of persistent estrus. In younger women, the effect of negative feedback on the hypothalamus from constant levels of circulating estrogens interferes with gonadotrophin-releasing hormone (GnRH), mitigating the normal mid-cycle surge of follicle-stimulating hormone (FSH) and luteinizing hormone (LH) to stimulate ovulation. The ovarian follicle, therefore, continues producing estrogens and precursor androgens, thereby exacerbating endometrial growth without progesterone stimulated maturation and cyclic menstrual shedding. Hyperestrogenic states occur in postmenopausal women when androgens produced by the ovaries and adrenal glands are converted in adipose tissues to estrogens which continue endometrial stimulation, growth, and hyperplasia.
Risk factors for type 2, high-grade endometrial cancers also include obesity, though the effect of increasing risk per each element of weight gain over matched control women was less for type 2 endometrial cancers than for type 1, low-grade endometrioid carcinomas. A large prospective study in Norway found no difference in the mean age of diagnosis between type 2 and type 1 endometrial carcinomas. Considering the similarities between ESC and adnexal HGSC and HGSC-SET carcinomas and the observations of associated BRCA mutations and dysfunction, discussed in the Etiology section, it is interesting to find that the association of breast cancer diagnoses with endometrial cancers is dramatically greater for type 2 ESC than type 1 endometrioid carcinoma.
Members of Hereditary Breast Ovarian Cancer families associated with BRCA1 or BRCA2 mutations often are treated with tamoxifen for breast cancer or as prophylaxis against contralateral breast cancer. Although exogenous estrogen HRT unopposed by progestin is associated with increased risks for endometrial endometrioid carcinomas in general populations, a case-control study specifically focused on BRCA1, and BRCA2 mutation carriers with endometrial cancers demonstrated no significant increased risk (OR = 0.23, 95% CI 0.03 to 1.78) in patients who were taking estrogen-only HRT. In contrast, significantly increased risks (OR = 6.91, 95% CI 0.99 to 98) for endometrial cancer were found in mutation carriers taking progesterone-only HRT (p = 0.003) or those who had histories of tamoxifen use. Contrary to its anti-estrogenic effect on breast tissue, tamoxifen has a distinctly stimulating effect on the endometrium, and use of the drug is related to increased risk for endometrial cancer in postmenopausal women. It has been suggested that women treated with tamoxifen, especially at higher dosages, may be susceptible to more aggressive uterine sarcomas and high-grade carcinomas, but as yet the evidence is insufficient to support these observations.
Exogenous estrogen, iatrogenic, and in some cases, possibly environmental, unopposed by progesterone or progestins are associated with increased risks for endometrial endometrioid carcinomas (see above: Etiology, and below: Management, Prevention, Differential Diagnosis).
Family history of endometrial, colorectal, ovarian, and other cancers that characterize Lynch syndrome and related MMR deficient hereditary diseases increase risks for endometrial cancer. While classical Lynch syndrome, linked to germline mutations in the MMR genes MLH1, MSH2, MSH6, PMS2, and inherited EPCAM interference with MMR, accounts for fewer than 5% of all endometrial carcinomas, the “lifetime risk” through age 70 years of mutation carriers for endometrial cancer is approximately 12% to 60%, depending on the mutated gene. Lynch syndrome mutation carriers also may have “lifetime risks” of 5% to 10% for ovarian cancer. Inheritance of aberrant PTEN mutations may convey increased risks for endometrial cancers as well as a myriad of other tumors.
Several epidemiologic factors are associated with decreasing the risks for endometrial cancers. Estrogen-progestin combined oral contraceptives (COC) use reduced risk for endometrial cancer with each increasing five years of use by as much as 33% compared with women who had never used these formulations, and the risk reduction persisted for more than 30 years after COC were discontinued.
The risk for endometrial cancers, both type 1 and type 2 carcinomas, was found to decrease with advancing age of first childbirth, overall by 11% in a large epidemiologic study, and the risk for women who first gave birth after 40 years age was 44% lower than women who gave birth before age 25 years. The finding of endometrial cancer risk reduction with older age at the time of first birth confirmed a previous study, which paradoxically showed that endometrial cancer risk was reduced by advancing parity, short intervals between births, and between the last birth and menopause. Breastfeeding reduced the risk for endometrial cancer by 11% in another large epidemiologic study.
Recent reports indicate that alcohol consumption may increase, and coffee and tea consumption may reduce, the risk of endometrial cancer. Though hardly recommended, tobacco cigarette smoking is associated with very significant reductions of risk for endometrial cancers in both prospective and case-control studies, which show the greatest decrease of risk in heavy smokers over increasing years of use.
Uninterrupted by intervention, the natural history of endometrial carcinomas begin as preinvasive intraepithelial lesions, which progress to full-blown invasive cancers (see above: Etiology and Epidemiology) involving endometrial stroma, then penetrating ever more deeply into the myometrium to engage lymphatic capillaries that carry the malignancy to regional lymph nodes, whence metastases may occur through vascular channels. Tumorous involvement of the uterine cervix and stroma probably is mostly through lymphatic channels, particularly by dedifferentiated carcinomas; although surface spread can occur from endometrial cancers in the lower uterine segment (LUS). Lymphatic capillaries also may carry endometrial carcinoma cells to the adnexa, in other words, fallopian tubes and ovaries. Anatomically, lymphatic channels usually follow the corresponding veins; thus a regional extension of cancers confined to the uterine corpus or extended only to adnexal organs is considered to be the para-aorta and para-cava nodes; whereas, a regional extension of invasive endometrial cancers involving the uterine cervix and LUS is to para-ureteral and pelvic lymph nodes. Locally advancing endometrial cancers may penetrate fully through the myometrium and uterine serosa to involve surrounding peritoneum, supporting tissues and other pelvic organs. Low-grade, type 1 endometrioid carcinomas tend to remain confined to the uterus and are characterized by rather a favorable prognosis; whereas, high-grade, type 2 endometrioid and non-endometrioid carcinomas with TP53 mutations often metastasize via the lymphatic system or transit through the fallopian lumens to disseminate throughout the pelvis and abdomen, manifesting at advanced stage and portending grave prognoses.
Over 90% of endometrial cancers are epithelial malignancies, in other words, carcinomas. Endometrioid carcinoma histotypes constitute the majority of endometrial carcinomas, and low-grade (FIGO Grades 1 and 2; see above: Etiology) endometrioid carcinomas account for 80% to 90% of all endometrioid carcinomas, serving as the prototype for the WHO (2014) classification of type 1 endometrial cancers. High-grade serous carcinomas (HGSC) are the prototype for the WHO (2014) classification of type 2 endometrial cancers. About 75% of type 2 endometrial cancers are morphologically high-grade serous (greater than 50%), clear cell (12% to 14%), or undifferentiated (5%) carcinomas. They can also be composed of mixed elements of serous, clear cell, or endometrioid carcinoma (7% to 8%). Carcinosarcomas constitute most of the remaining type 2 endometrial epithelial malignancies.
Type 1, low-grade endometrioid carcinomas are generally confined to the uterus when they are diagnosed and have relatively favorable prognoses compared with HGSC and other type 2 endometrial carcinomas, which have significantly poorer prognoses and often are disseminated when diagnosed. The epidemiologic and clinical factors associated with type 1 endometrial carcinomas are different and distinct from type 2 endometrial carcinomas. A majority of endometrial low-grade endometrioid carcinomas show MSI and carry PTEN mutations in contrast to endometrial HGSC in which nuclear pleomorphism is common, and TP53 mutations are nearly ubiquitous. The clinical behavior and some epidemiologic aspects of endometrial high-grade (FIGO Grade 3) endometrioid carcinomas are similar to type 2 endometrial HGSC, and aberrant p53 IHC staining may be found in some poorly differentiated endometrioid carcinomas (see: Etiology, Epidemiology, above, and Management, below). Endometrial carcinomas with high grade solid, endometrioid and/or transitional cell-like (SET) morphologies behave clinically like other type 2 carcinomas and, like morphologically pure HGSC, many HGSC-SET carcinomas demonstrate evidence of TP53 mutations. Carcinosarcomas make up most of the remaining type 2 endometrial epithelial cancers.
Endometrial stromal sarcomas comprise less than 1% of all uterine malignancies and less than 10% of endometrial cancers. The WHO (2014) classifies endometrial sarcomas into:
Typically, ESN and LG-ESS usually show strongly positive diffuse staining for ER and PR and CD10 in IHC but are negative for Cyclin D1; while HG-ESS with YWHAE-FAM22 translocations usually shows diffuse positive nuclear staining for Cyclin D1, but ER, PR, and CD10 staining are absent or only weakly and focally positive. UUS arising in the endometrium may be negative or show weakly positive patchy staining for ER, PR, and CD10, but endometrial UUS can have diffusely present Cyclin D1 staining sometimes coexisting with CD10. ESN is often confined to the endometrium or have limited invasion with good prognosis. LG-ESS usually are intrauterine with greater than 90% 5-year survival, but lymphatic and/or distant metastases may accompany deep invasion or adnexal involvement, thereby worsening the prognosis. HG-ESS with YWHAE-FAM22 translocation may present with localized or metastatic disease and generally carry prognoses between LG-ESS and the less favorable prognoses associated with more aggressive UUS. The most common clinical presentations of endometrial stromal sarcomas are abnormal vaginal bleeding and pelvic-abdominal mass and/or pain.
Thorough history and physical examination, including all systems, habits, medications, allergies, present acute and chronic illnesses and previous illnesses, significant injuries and surgeries, and a meticulous family medical-surgical history with multigeneration family cancer history, engaging cancer genetics counseling when hereditary disease is suspect, are mandatory when undertaking the medical care of women in whom genital cancer or precursors are known or suspected from chief complaints, symptoms, presenting signs, known or revealed risk factors and/or family cancer pedigree (see above: Etiology and Epidemiology and below: Evaluation and Management).
Abnormal uterine bleeding (AUB) is the most frequent symptom of endometrial cancer (see below: Differential Diagnosis). Irregular uterine bleeding associated with complex atypical endometrial hyperplasia and cancer can affect women of all ages, becoming more prevalent during the fifth decade of life, especially in women over age 45 years. About 10% of postmenopausal bleeding (PMB) is related to endometrial cancer. In patients with PMB, reliance on the results of endometrial sampling, unless diagnostic for cancer, is insufficient with the failure rate over 40% and the presence of precancerous atypical hyperplastic lesions in 7% of patients. Atypical endometrial hyperplasia may progress to cancer in 5% to 25% of patients, and although uterine curettage has been considered the standard for histologic diagnosis in the evaluation of PMB, endometrial cancer has been demonstrated in the hysterectomy specimens of 43% of patients previously diagnosed with atypical hyperplasia. Overall, about a third of women undergoing uterine curettage will be found to have uterine polyps. When endometrial polyps are diagnosed in the presence of PMB the risk for malignancy is especially high, particularly in women with family histories of endometrial cancer and those who are 59 years of age and older, when the risk is approximately 12%. The measurement of endometrial thickness by transvaginal ultrasound (TVU) scanning has been advocated to reduce interventions with endometrial sampling and/or curettage. Meta-analysis of published reports concludes that limiting this measurement to less than or equal to 3 mm yields a 98% sensitivity and 35% specificity for diagnosing endometrial cancer in women with PMB. A thin, endometrial stripe, however, does not reliably exclude type 2 endometrial cancers.
While the prevalence of endometrial cancer in premenopausal women younger than 50 years with AUB is 1% or less, the disease may be diagnosed in a few women during their third decades of life and rarely even in teenagers. Therefore, it is essential to appropriately pursue accurate diagnoses and management in all cases of AUB, defeminization, abdominal-pelvic pain or bloating and unusual weight gain or loss, including women in the reproductive age group. Diagnosis and successful management of endocrinopathies (including obesity, diabetes, thyroid dysfunction, hypersecreting hormonal tumors, cysts and hyperplasias of the ovaries, adrenal or pituitary glands); bleeding diatheses; benign uterine tumors; adenomyosis; exogenous hormones; tamoxifen; and anti-epileptic and antipsychotic medications can save such patients from consequent morbidity and premature mortality. Resumption of regular ovulatory menstrual bleeding or progestin intervention in premenopausal women with abnormal uterine bleeding may prevent endometrial atypical hyperplasia and cancer.
Because the incidence of endometrial cancer is so low (0.1/100,000) before age 20 years, once systemic disease has been ruled-out and regular ovulatory menstrual cycles have been reestablished, endometrial assessment generally is not recommended in teenagers with AUB, unless the young woman is obese and has experienced 2 to 3 years of anovulatory bleeding. As the risk for endometrial cancer increases with age, this rule may be followed for adult women younger than age 35 years. However, after excluding pregnancy and evaluating for endocrinopathies and systemic and iatrogenic causes, all women with AUB older than 35 years should have the endometrial assessment to rule-out hyperplasia or malignancy.
Synchronous ovarian cancers occur with some 10% of endometrial cancer cases and may be more prevalent in premenopausal women 50 years of age and younger. Although a coincidence of ovarian malignancy with endometrial cancer as high as 25% has been reported in young women aged 24 to 45 years, synchronous ovarian cancers are disproportionately non-serous carcinomas and tend to be the same histotypes as the primary endometrial cancers. Most synchronous ovarian cancers, therefore, are low-grade endometrioid carcinomas. Lynch syndrome mutation carriers with endometrial cancer have a 22% rate of synchronous ovarian cancers of the same histotypes, 89% of which are endometrioid carcinomas. Young patients with synchronous ovarian cancers and endometrioid carcinomas with microsatellite instability (MIS) should be advised to have further Lynch syndrome screening. Because of the coincidence of endometrial and ovarian cancer, presentation with a pelvic mass and/or abdominal signs and symptoms may herald the presence of synchronous primary carcinomas.
While AUB, especially PMB, is by far the most common presenting symptoms and signs of endometrial cancer, occasionally abnormal vaginal discharge or cervicovaginal cytology may lead to the diagnosis. Examination of Pap smear taken within 6 months prior to the diagnosis of endometrial cancers reported that 38% were abnormal, 21% were consistent with adenocarcinoma, and 13% showed atypical glandular cells.
Definitive diagnosis of endometrial cancer depends on histology. A complete personal and family medical history and thorough physical examination are fundamental. Abnormal findings will determine the next appropriate steps to take toward establishing differential diagnoses. The possibility of normal intrauterine or abnormal pregnancy should always be considered in women of reproductive age with AUB or amenorrhea unless postmenopausal status has been confirmed. When symptoms, signs, and/or family history suggest the possible presence of gynecologic pathology, basic laboratory evaluation, cervical-vaginal Pap smear, and transvaginal TVU scanning generally are considered the next steps in diagnosis. Abnormal Pap smears at any age demand evaluation. Suspicious and abnormal physical findings should lead to TVU, and trans-abdominal ultrasound scanning also may be judged appropriate in some cases. If non-gynecologic etiologies or benign gynecologic conditions are diagnosed as the cause of AUB in women 35 years or younger with low-risk for endometrial pathology, the workup may forego TVU, unless the patient does not respond to medical therapy. Recurrent AUB in younger women and the evaluation of anovulatory bleeding in women older than 35 years require evaluation with TVU and either outpatient office endometrial sampling or examination under anesthesia (EUA) with dilation and curettage (D and C). Abnormal findings by TVU, unsuspected abnormalities on EUA and/or suspicious, atypical or malignant endometrial sampling require further scanning and tissue diagnosis. Because management of endometrial cancer requires documentation of the extent of disease, many will favor separate endocervical cytologic samples or endocervical curettage (ECC) prior to endometrial sampling to assess for cervical involvement. Any suspicious vulva, vaginal, or cervical lesions should be biopsied for tissue diagnosis. Fractional D and C first of the endocervix before sounding the uterus and dilation of the internal os for systematic endometrial curettage offers an opportunity to collect tissues from both uterine segments, which becomes more important as the incidence of endometrial atypical hyperplasia and cancer and more extensive disease becomes greater in women during their fifth decades and beyond. Although the results from ECC are subordinate to the current surgical-pathological staging of endometrial cancer, these may be valuable when planning individual patient management. EUA also affords an opportunity for hysteroscopic examination of the intrauterine cavity. Although this technique has been advocated to obtain endometrial tissue for histologic diagnosis, the sensitivity for the histologic diagnosis of endometrial cancer and hyperplasia may not be as good as D and C, although hysteroscopy is highly effective for identifying submucous myomas and endometrial polyps, that may be missed on TVU. Hysteroscopy may be a useful adjunct to D and C to enhance the diagnosis and removal of polyps, which may be associated with endometrial cancer, especially in older postmenopausal patients. Although higher rates of positive peritoneal cytology follow diagnostic hysteroscopy, there is no significant difference in the prognoses of endometrial cancer after hysteroscopy compared with no hysteroscopy or D and C.
TVU can focus the need for tissue diagnosis in women with PMB. Although a consensus panel set a limit of less than or equal to 5 mm endometrial thickness measured by TVU before endometrial biopsy is considered necessary to evaluate PMB cost-effectively, a more recent systematic review and meta-analysis has recommended a cutoff level of less than or equal to 3 mm endometrial thickness for the exclusion of endometrial cancer in women with PMB. Moreover, TVU does not reliably exclude type 2 endometrial carcinomas in which cases the endometrial stripe may be less than 4 mm or indistinct in postmenopausal women. With strict criteria of endometrial thickness less than or equal to 3 mm and no other endometrial, uterine, or adnexal abnormalities on TVU, outpatient office endometrial biopsy may be used as the next step for evaluating PMB so long as sufficient tissue is obtained for histologic examination in otherwise low-risk patients. If the endometrial biopsy is inconclusive or PMB recurs, EUA and fractional D and C should be done, and the diagnosis may be enhanced by hysteroscopy for detection of focal lesions.
Although rarely positive in patients with lower grade endometrioid carcinomas, a preoperative chest radiograph should be scrutinized for evident metastases, which would advance the disease and call for consideration of adjuvant antineoplastic chemotherapy. If parametrial invasion or adnexal involvement and/or synchronous ovarian tumors are suspected from physical examination, preliminary TVU may be used, reserving magnetic resonance imaging (MRI) for more difficult cases. Routine preoperative intravenous pyelograms (IVP) in patients with previous history of endometriosis, pelvic inflammatory disease, intra-abdominal surgery or pelvic relaxation are not cost-effective, but IVP may be individually selected when the probability of abnormally is deemed to be high. Unfortunately, imaging techniques, such as MRI, computed tomography (CT), and positron emission tomography (PET)/CT are not efficacious for the diagnosis of lymph node metastases, which still relies on surgical dissection and histological examination. Because of high risks for colorectal, ovarian, and urothelial cancers in Lynch syndrome mutation carriers, preoperative sonographic imaging, colonoscopy, and cystourethroscopy are indicated and should be selected in other suspect cases.
Of primary significance is the individual endometrial cancer patient’s status, specifically, their age, reproductive status and plans, habitus, health and other chronic and/or acute diseases, understanding and psychology, socio-economic factors, risks, among others, well beyond the scope of this article. Next, histotype, grade, tumor size, uterine, local/regional invasion, nodal status, and presence of intraperitoneal, and/or extra-abdominal metastases, must be determined. With this information, a management strategy can be developed.
Because of the association of endometrial cancer with overweight status and obesity, maintaining wholesome body form through a healthful diet and exercise is the most important measure most women can implement to reduce their risks for endometrial cancer. Even significant weight loss can reduce endometrial cancer risk because the risk increases incrementally with weight gained between the beginning of the third decade and middle age. Calculated body mass index (BMI) is the easiest and least expensive parameter to measure obesity. Bariatric surgery should be considered for women who have made serious efforts at weight reduction but still have a BMI greater than or equal to 40 kg/m efforts at weight reduction to reduce the risk of endometrial as well as other cancers and the morbidity/mortality related to obesity,
Once regular cyclic ovulatory menstruation has established in the teens, women who develop oligomenorrhea or amenorrhea and signs of defeminization, such as excessive facial, presternal and pubo-perineal hair growth and/or usual cephalic hair loss, particularly in the temporal regions, even before obvious hirsutism and balding, should be evaluated for PCO and other androgen or estrogen-secreting tumors considered in the workup. Diagnosed pathology, such as PCO, thyroidopathy, pituitary and hypothalamic disease, and hormone-secreting tumors or hyperplasia, should be appropriately treated medically and/or surgically as indicated to achieve regular ovulation and fertility, if possible. If regular ovulation and menstruation cannot otherwise be achieved and the uterus remains intact, cyclic use of COC with regular withdrawal bleeding may protect the endometrium and reduce the risk of endometrial cancer. Levonorgestrel-releasing intra-uterine devices are also effective at reducing the risk of endometrial cancer, because of the protective effects of continuous progestins. The addition of continuous progestins to estrogen HRT reduces the risk for endometrial cancer but might elevate the risk for breast cancer.
Documented family histories are an essential part of good medical care and should be periodically updated. Patients with personal and/or family histories suggestive for a hereditary cancer syndrome should have focused genetic counseling for cancer pedigree analysis, risk assessment, education, possible molecular testing for cancer-associated mutations, and management planning. Excessive numbers of endometrial, colon, ovarian, and other characteristic Lynch syndrome cancers and multiple cancers occurring at younger ages over several generations with apparent Mendelian dominant transmission are strongly associated with MMR gene and PTEN mutations. Because of the high prevalence of microsatellite instability (MSI) in Lynch syndrome colorectal and endometrial cancers, there has been a movement to examine these tumors with IHC to detect loss of MMR protein expression to identify possible mutation carriers for germline DNA testing. Lynch syndrome mutation carriers should be offered risk-reducing hysterectomy with bilateral salpingo-oophorectomy (RRH-BSO) because of their increased risk for ovarian cancer as well as endometrial cancer. During mean 12 years of follow-up, no gynecologic or peritoneal cancers were diagnosed in LS mutation carriers who had undergone prophylactic surgery compared to the diagnosis of endometrial cancer in 33% and ovarian cancer in 26% during mean 7.4 years follow-up of mutation carriers who did not. Ideally, risk-reducing surgery should be undertaken soon after childbearing has been completed and before the sixth decade of life, when the risk of these cancers begin to increase substantially. Often colectomy is performed in LS mutation carriers for primary colorectal cancer before the diagnosis of gynecologic cancer. In these cases evaluation for gynecologic cancers always should be done before colectomy is undertaken, and consideration given to RRH-BSO when childbearing has been completed. In these cases, estrogen HRT can be prescribed to alleviate menopausal symptoms. Evidence, so far, is inconclusive whether or not COC, aspirin, and nonsteroidal anti-inflammatory drugs are effective for chemoprevention of endometrial cancer in LS mutation carriers.
Although transvaginal ultrasound scanning and endometrial sampling can be useful techniques for evaluating abnormal gynecologic findings and abnormal uterine bleeding, these methods have not been proven effective as screening tests for endometrial cancer in asymptomatic low-risk women. However, there may be a place for regular periodic endometrial biopsies for surveillance of asymptomatic women with increased hereditary risk for endometrial cancer during intervals prior to risk-reducing surgery.
Generally, type 1, grades 1 and 2 endometrioid carcinomas, the most common endometrial cancers, are confined to the uterine corpus and are associated with an excellent prognosis. Conversely, type 2, serous, clear cell and grade 3 endometrioid carcinomas are aggressive cancers, which often present with lymphatic invasion and metastases, even when the primary lesion is confined to the uterine corpus. These are also are associated with advanced disease and poor survival. As such, approaches to management will be quite different in these dissimilar cases.
Endometrial Cancers Confined to the Uterine Body
Unless otherwise determined by the history and physical examination, patients diagnosed with EIN or type 1, low-grade endometrioid carcinomas grossly confined to the uterine corpus, once fitness to undergo surgery is determined, usually will be successfully treated with total hysterectomy and bilateral salpingo-oophorectomy by open or minimally invasive approaches. Minimally invasive laparoscopic or robotic-assisted techniques are presently the favored methods. Patients expected to have EIN or low-grade endometrioid carcinoma confined to the uterine corpus should have an intraoperative examination of the extirpated uterus and adnexa. The accuracy of gross inspection of the uterus for location and depth of invasion may be enhanced and relatively favorable pathology reliably confirmed by intraoperative frozen section examination. When a large tumor, myometrial invasion greater than 50%, cervical invasion, the involvement of capillary-like spaces, unexpected aggressive histology, adnexal mass, intra-peritoneal metastases, or enlarged lymph nodes are found, retroperitoneal pelvic and para-aortic lymphadenectomy should be done for correct staging and assignment for possible adjuvant treatment. Vaginal hysterectomy without laparoscopy is not preferred for surgical management of endometrial cancer, because this approach does not permit proceeding to lymphadenectomy when risk factors are found intraoperatively; although this procedure may have a place for treating endometrial cancers confined to the uterus in women deemed to be at high risk for surgical morbidity.
An early Gynecologic Oncology Group study of endometrial cancers clinically confined to the uterus by preoperative evaluation found no cancer positive nodes in grade 1 carcinomas involving only the endometrium. Pelvic node metastases were reported in 6.4% of 468 grade 1 and grade 2 endometrial carcinomas, and 3.6% had para-aortic node involvement. Grade 3 endometrial carcinomas were associated with 18.3% pelvic and 11.1% para-aortic node involvement. Largely on the basis of this classic report, which noted that most microscopically positive nodes were not grossly enlarged, pelvic and para-aortic lymphadenectomy was recommended generally for initial surgical management of endometrial cancer. Several subsequent studies questioned the efficacy of lymphadenectomy for treating low grade endometrioid carcinomas without additional risk factors, and data on the coincidence and relationship of para-aortic to pelvic node status are inconsistent.
Lymph node metastases are present in approximately 10% to 20% of unselected endometrial carcinomas and over half of the cases with positive pelvic nodes also have metastases to para-aortic nodes, but the presence of positive para-aortic nodes is uncommon in patients with negative pelvic nodes (2% to 6%). Two large multicenter prospective studies of stage I endometrial cancer, including both low and high risk cases, found no significant difference in overall and disease-free survival between patients randomized to pelvic lymphadenectomy in one study and pelvic plus para-aortic lymphadenectomy in the other study versus no lymphadenectomy in the first study and selective lymphadenectomy in the second study.
Metastases to para-aortic nodes, particularly isolated para-aortic nodes, are usually associated with poorly differentiated endometrioid and type 2 carcinoma histotypes, deep myometrial invasion, large tumors and other high-risk factors. Improved overall and progression-free survival benefit has been reported after thorough pelvic and para-aortic lymphadenectomy in high-risk patients, such as these. No survival advantage was demonstrated with pelvic and para-aortic lymphadenectomy in low-risk patients with grades 1 and 2 endometrial carcinomas confined to the uterine corpus. Overall, patients treated with hysterectomy for stage I endometrial cancer and randomized to pelvic lymphadenectomy had significantly longer operating times and hospital stays and suffered more postoperative complications, principally pelvic lymphocytes and lymphedema in 13% compared to none in patients who did not have lymphadenectomies.
Due to the time and expense of lymphadenectomy, the increase in adverse effects, and the absence of prospective evidence that lymphadenectomy provides benefit to patients with endometrial cancer clinically confined to the uterine corpus, there is a growing consensus that patients with grade 1 or 2 endometrioid carcinomas less than 2 cm diameter and confined to the uterine corpus with less than 50% myometrium invasion do not require surgical lymphadenectomy. Sentinel node mapping (SNM) has been advocated to select cases that are most likely to benefit from lymphadenectomy and avoid these extensive dissections in patients with low-risk stage I endometrial cancers and negative nodes. While SNM using several techniques has been successful in identifying positive pelvic nodes with cervical injection, mapping para-aortic nodes has been disappointing. Considering the uncommon presence of isolated metastases to para-aortic nodes in the absence of positive pelvic nodes and/or high-risk endometrial cancers, pelvic and para-aortic lymphadenectomy in patients with small, low-grade endometrioid carcinomas and minimal invasion might be reserved for those with positive pelvic sentinel nodes.
Presently, omentectomy is not standard in the primary surgical management of endometrial cancer. Meta-analysis of 13 relevant studies concluded that the prevalence of metastases to the omentum in clinical stage I endometrioid carcinoma is only 4.4%, but the prevalence of omental metastases associated with stage I, non-endometrioid endometrial cancer rose to 9.8% of which 26.7% were microscopic. Only adnexal metastases was a risk factor for omental metastases in the meta-analysis. A European consensus conference has recommended omentectomy in the surgical management of stage I endometrial serous carcinoma, because of the frequent association of peritoneal dissemination and omental metastases, and a subsequently reported series of 218 non-endometrioid endometrial cancers reported finding 15% omental metastases of which 44% were occult. The high rate of disseminated carcinomatosis with early non-endometrioid carcinoma, even when apparently confined to the uterine corpus, may be a case for omentectomy along with lymphadenectomy for adequate staging and intent to maximize survival.
Another controversial issue is whether ovarian conservation should to be considered when hysterectomy is done for endometrial cancer clinically confined to the uterus. Coexisting ovarian malignancies have been described in 26/102 (25%) of young women (aged 24 to 45 years) who underwent a hysterectomy and bilateral salpingo-oophorectomy (86), or hysterectomy with ovarian preservation (16) or subsequent salpingo-oophorectomy (3) for endometrial cancer. Preoperative endometrial biopsies found grade 1 endometrioid carcinoma in 46% of the cases; on final pathology, 69% of the ovarian malignancies were associated with grade 1 endometrial carcinoma, and 58% of all cases involved only the inner myometrium. Four of the patients (15%) appeared to have normal ovaries during the intraoperative assessment. Twenty-three were classified as synchronous ovarian cancer, and 3 were metastatic. To the contrary, a study of 1322 women younger than 50 years who had hysterectomies for stage I endometrial endometrioid carcinomas (80.8% grade 1; 16.2%, grade 2) with ovarian conservation determined only 1.0% and 1.3% 5- and 10-year cumulative recurrence risks, respectively, for ovarian cancer; and there were no deaths from ovarian cancer during 11.6 years follow-up. A large database study of 15,648 women who had hysterectomies for endometrial endometrial carcinomas, compared 14,527 cases treated with bilateral oophorectomy (88.1%, stage IA; 86.6%, grades 1 and 2) to 1121 cases with ovarian conservation (88.9%, stage IA; 88.8%, grades 1 and 2) and found no significant difference in survival through 11 years. Another database study of grade 2 and grade 3 endometrial adenocarcinomas confined to the endometrium in women younger than 45 years found no difference in 5-year survival probabilities of 94.8% for 849 patients who underwent a hysterectomy and bilateral adnexectomy compared to 93.8% of 96 patients who had a hysterectomy with ovarian preservation. One study concluded that women younger than 50 years treated with hysterectomy for small (less than 2 cm) stage I, grade 1 endometrioid carcinomas might enjoy improved overall survival and lower cumulative risk for cardiovascular disease with ovarian conservation than after oophorectomy. Consequently, there may be a place for individualized ovarian preservation in well-counseled patients with small, minimally invasive grade 1 endometrioid carcinomas.
Outcomes utilizing primary radiotherapy for clinically diagnosed stage I endometrial carcinoma of all histologic grades in women with comorbid medically inoperable conditions are comparable to those reported with primary surgery and adjuvant postoperative radiotherapy when indicated. Survival rates reported with low-dose intracavitary brachytherapy in patients with heightened surgical risks, some with high-risk carcinomas, were in the same order as those obtained in other series of stage I endometrial cancers treated with external radiotherapy and later high dose rate brachytherapy. A large database study of patients with inoperable stage I endometrial cancers, grades 1 to 3, found significant survival benefits in the group treated with brachytherapy alone or together with external radiation compared with the group treated with external therapy alone.
While hysterectomy alone will prevent endometrial cancer in women with EIN and this provides 95% probability of relapse-free, 5 year survival in the treatment of stage I, low-risk endometrial cancer, postsurgical adjuvant therapy may have a place in the treatment of endometrial carcinomas with high-risk factors.
Endometrial Cancers Involving the Uterine Body and Cervix
Five-year survival rates for endometrial cancer involving the uterine body, and cervix, that is, stage II, are poorer for all histotypes and grades than survival rates for surgical stage I endometrial cancer. Non-endometrioid histology, large primary tumor, lymphovascular and deep myometrial invasion are common features in stage II endometrial cancers. Considering tumor bulk, parametrial spread, and lymphatic drainage of primary cancers of the uterine cervix, radical hysterectomy often has been favored for the initial surgical treatment of stage II endometrial cancer. Radical hysterectomy for the surgical treatment of stage II endometrial cancer with or without adjuvant therapy has been associated with better overall survival than simple hysterectomy in most studies. While, classically, radical hysterectomy and pelvic-para-aortic lymphadenectomy for the cancerous involvement of the uterine cervix has been by laparotomy, both laparoscopic-assisted and robotic-radical hysterectomies have been demonstrated to be valid alternatives. Non-endometrioid histotypes show high rates of lymphovascular invasion, and for all stages of endometrial cancer, omental metastases have been reported in 22% of serous carcinomas and 19% of clear cell carcinomas. Judging from this, it may be prudent to include omentectomy in primary surgery for non-endometrial, stage II endometrial carcinomas.
Preoperative radiotherapy with or without cytotoxic antineoplastic chemotherapy planned to reduce the size of the stage II tumorous cervix and early parametrial invasion, followed by extrafascial hysterectomy with bilateral salpingo-oophorectomy to extirpate residual endometrial cancer that may be harbored in the myometrium or adnexa has been advanced as a tolerable option that may be considered in elderly patients and those with precarious co-morbidities. To date reported series are too small and follow-ups are too limited to recommend this as the standard approach for managing stage II endometrial cancers. In patients who have not had preoperative radiotherapy, the results from surgical treatment of endometrial cancers found to invade the cervical stroma may be improved with postoperative adjuvant therapy.
For medically inoperable patients with clinical stage II endometrial cancer, radiation therapy alone achieved 5-year survival rates that approached those of operable patients who had standard surgical management. A database study found that combination of external beam and brachytherapy produced the most satisfactory pelvic control of stage II endometrial cancer with relatively low toxicity and the highest survival in medically inoperable patients, many of whom were elderly.
Advanced and Recurrent Endometrial Cancers
The prognosis for stage III, stage IV, and recurrent endometrial cancers may be improved with maximum surgical debulking. The best results with significantly longer survival have been achieved in treating advanced and recurrent endometrial cancers with optimal tumor debulking to less than 1 cm, especially when there was no gross residual disease after surgery. This was true for both type 1 endometrioid and non-endometrioid type 2 endometrial carcinomas. Though experience with neoadjuvant chemotherapy before debulking surgery for advanced endometrial cancer is limited, this approach using platinum-based regimens before definitive surgery in a series of 30 patients with laparoscopically-diagnosed intra-abdominal spread, including 93% non-endometrioid carcinomas, resulted in a favorably high 80% rate of optimal surgical debulking later.
Although recurrent endometrial cancer after initial treatment results in notably diminished long-term survival rates, local recurrences confined to the vagina and pelvis are associated with better overall survival than non-local recurrences. Vaginal and central recurrences are sometimes amenable to surgical resection; and, if patients were not previously treated with radiotherapy, brachytherapy may be used alone or in combination with external beam radiation. Larger vaginal recurrences can be treated with external radiotherapy followed by brachytherapy or interstitial implantation (if the tumor remains greater than 0.5 cm). There may be a place for total or segmental pelvic exenteration of centrally located recurrent endometrial cancers in highly selected patients who have failed previous surgery and radiotherapy. Though these procedures are associated with very high operative morbidity, 20% long-term (greater than 5-year) survival has been reported.
Postsurgical Adjuvant Therapy for Cancers Confined to the Uterus
Studies of postsurgical adjuvant therapy versus observation in patients with high-risk stage I and stage II endometrial cancers have demonstrated that radiotherapy is associated with significant reduction in vaginal and pelvic recurrences. Trials comparing adjuvant vaginal brachytherapy with external beam radiation found nearly equivalent results, reducing vaginal-pelvic recurrences to less than or equal to 5%, but there was significantly more toxicity with external radiation. Though there is insufficient evidence to recommend both external radiation and brachytherapy for standard postoperative treatment of stage I endometrial cancer, some authorities have suggested that pelvic recurrences might be further diminished by this approach in patients found to have poorly differentiated endometrial carcinomas and greater than 50% myometrial invasion or cervical involvement.
A multicenter database study focusing on 250 cases of stage II endometrial cancers found that, following primary hysterectomy, the 5-year, disease-free survival rate but not the overall survival rate was significantly higher for patients treated with postsurgical external pelvic radiotherapy with or without vaginal brachytherapy. In that study, 20% of the endometrial cancers were non-endometrioid, and 20% of the primary surgeries were radical hysterectomies, but there were no significant differences in disease-free or overall survival between groups with non-endometrioid compared to endometrioid carcinomas or according to the type of hysterectomy. Two recent publications reported excellent local control and improved recurrence-free 4- and 5-year survival rates in patients treated for stage II endometrial cancers with adjuvant postoperative vaginal brachytherapy alone; while vaginal brachytherapy with the addition of external radiotherapy resulted in more long-term toxicities.
Many pelvic recurrences after surgery and adjuvant radiotherapy for stage I and stage II endometrial cancers are associated with distant metastases. These observations raise the possibility of including anti-neoplastic chemotherapy in the treatment regimens. However, a large database study, just published, confirmed better overall survival for patients receiving adjuvant postoperative radiotherapy for stage II endometrioid endometrial cancer but not with adjuvant chemotherapy, and there was no survival advantage to adding chemotherapy to adjuvant radiotherapy. Authoritative, comprehensive reviews of pertinent studies have concluded that evidence for improved relapse-free and overall survival with platinum-based chemotherapy alone or in addition to adjuvant radiotherapy is insufficient to recommend this for the treatment of stage II endometrial endometrioid carcinoma. These patients may be best served by counseling and offering opportunities to be included in prospective treatment trials.
Because of the propensity for surgical stages I and II endometrial serous carcinomas to recur in the vagina and disseminate throughout the peritoneal cavity, several retrospective studies and clinical trials have explored the results of postsurgical radiotherapy and chemotherapy. The most promising regimens, providing reductions in vaginal recurrences and improved progression-free survival for patients with endometrial serous carcinomas, clinically and surgically confined to the uterus, were those that combined vaginal brachytherapy and systemic platinum-taxane chemotherapy.
Postsurgical Adjuvant Therapy for Advanced and Recurrent Endometrial Cancers
Adjuvant therapy for stage III, stage IV, and recurrent endometrial cancers usually follows maximum surgical effort.
Large series reports, database studies, and authoritative reviews generally conclude that postoperative external radiotherapy to the pelvis with the addition of vaginal brachytherapy for stage III endometrial cancers reduces local-regional recurrences and may improve the lengths and rates of progression-free and overall survivals. External radiation confined to the pelvis in cases with surgically proven pelvic and/or para-aortic node metastases (stage IIIC) is associated with fewer recurrences and higher survival rates, which might be further improved by extended fields to encompass para-aortic nodes.
Many multi-institutional and registry-based studies of postsurgical treatment for stage III endometrial cancers indicate that protocols including both adjuvant radiotherapy and chemotherapy were associated with superior cancer control and survival benefits than does either method alone. A recently reported database study of 12,988 stage III endometrial cancer cases concluded that adding vaginal brachytherapy to external beam postsurgical radiotherapy improved survival in cases with cervical involvement but not in other cases. While fewer stage IV cases were reported comparing the effects of adjuvant chemotherapy to radiotherapy or combining these modalities, the improved progression-free and overall survival rates associated with the addition of pelvic radiotherapy to systemic chemotherapy were attributed to reduction in the number of pelvic recurrences, as there were no substantial differences in the rate of extrapelvic relapse with combined treatment compared to chemotherapy alone.
Presently, systemic chemotherapy is the mainstay for postsurgical adjuvant therapy of advanced primary and recurrent endometrioid and serous carcinomas.
Options for adjuvant chemotherapy for advanced and recurrent endometrial cancer include hormonal, cytotoxic and emerging targeted treatments. Although objective responses have been reported with progestogen treatment of metastatic endometrial cancers, particularly in those with high estrogen receptor (ER) and progesterone receptor (PR) expression, these medications should be reserved as first-line systemic therapy only for slow growing, grade 1 or grade 2 endometrioid carcinomas and in controlled prospective trials. Regimens using the combination of tamoxifen and a progestogen show promise and are on trial. While a comprehensive review of cytotoxic and molecularly targeted chemotherapy far exceeds the scope of this presentation, suffice it to note that the currently favored first-line cytotoxic chemotherapy is paclitaxel and carboplatin (TC), associated with 51% response rate, median 13 months progression-free survival and relatively low toxicity. After prolonged treatment-free and progression-free survival, generally at least 6 to 12 months, platinum alone, other platinum-containing regimens. and unrelated cytotoxic single drugs or regimens have been offered for second-line chemotherapy, but the results with no more than 20% to 30% response rates have been disappointing. These patients may be best served by counseling and enrollment on a legitimate, prospective trial when this is available.
So far, options for molecularly targeted treatment of endometrial cancer also are limited to trials. The [poly (ADP-ribose) polymerase] PARP inhibitor olaparib was approved by the Federal Drug Administration (FDA) in 2014 for treatment of progressive ovarian cancer in BRCA1 or BRCA2 germline mutation carriers who have failed at least three previous anti-cancer drug regimens. Prospective trials of olaparib effectiveness against endometrial cancer are ongoing based on published improvements of overall response rates and mean progression-free survival, though not overall survival, in previously treated patients with ovarian cancers associated with germline or somatic BRCA mutations. The greatest benefit of PARP thus far has been prolonging progression-free survivals in ovarian HGSC with BRCA mutations. Olaparib was FDA approved in 2017 for maintenance therapy of patients with ovarian, fallopian tube, or primary peritoneal cancer in complete or partial response to platinum-based chemotherapy. Another PARP inhibitor, rucaparib, was approved by the FDA in 2016 for the treatment of advanced ovarian cancers associated with germline or somatic BRCA mutations after 2 or more previous chemotherapies. Because PARP inhibition has been demonstrated in PTEN deficient endometrial cancer cell lines, it is thought that these agents may be effective in MMR defective endometrial cancers with MSI (microsatellite instability). It has been considered that mTOR inhibitors and antiangiogenic agents also might have molecularly targeted cancericidal and/or tumor suppressive effects against endometrial cancer, but single and combined drug trials are limited and toxicities are significant, and neither has been FDA approved for the treatment of endometrial cancer. Finally, immunotherapy to block the Programmed Death 1 (PD-1) pathway, which if left unchecked represses cytotoxic immune responses, was associated with responses in patients with MMR deficient endometrial cancers that are characterized by marked tumor infiltrating lymphocytes, which are hallmarks of Lynch syndrome-linked cancers. There is speculation that patients harboring recurrent endometrial cancers showing MIS with large numbers of TIF and/or demonstrated MMR deficiency or high levels of PD-1 neoantigen expression may be better candidates for immunotherapy rather than for second-line chemotherapy trials.
The differential diagnoses confronted in the evaluation of the most common presenting symptoms and signs of endometrial cancer, which are abnormal vaginal bleeding and/or pelvic masses, range from benign localized lesions to systemic diseases and malignancies. Endometrial cancers confined to the uterus may present with minimal symptoms of AUB to advanced disease with pelvic-abdominal discomfort and pain, widespread lymphatic and peritoneal dissemination, ascites and rarely even extra-abdominal metastases.
The evaluation of AUB in women during their reproductive years should be alert to the complications of intrauterine and ectopic pregnancies and the gestational trophoblastic diseases. Both endogenous and exogenous hormone stimulation of the endometrium may cause irregular uterine bleeding during premenopausal years and postmenopausal bleeding (PMB). Systemic and endocrinologic causes of irregular and anovulatory bleeding, such as anorexia and obesity, polycystic ovarian disease, which often is associated with diabetes mellitus, hilar cell hypertrophy and hormone-secreting tumors of the ovary, thyroidopathies, adrenal hyperplasia and tumors, and pituitary tumors should be considered. Endocrinopathies associated with excess androgen secretions may be marked by signs of defeminization, hirsutism, and alopecia. Simple endometrial hyperplasia is commonly associated with periods of anovulation, oligomenorrhea and amenorrhea in the premenarcheal and premenopausal years. Exogenous sources of hormones, environmental and iatrogenic, particularly estrogen, progesterone and progestogens, and androgens, and treatment with tamoxifen must be ruled out. The association of endometrial cancer with exogenous hormones after menopause is most common in women on hormone replacement therapy with unopposed estrogen. Submucosal leiomyomata (fibroids), sometimes accompanied by adenomyosis, especially in the late fourth and fifth decades, can result in irregular endometrial shedding with prolonged or irregular menstruation in premenopausal women, but should not cause PMB. Benign endometrial polyps may be the source of intermenstrual and irregular menstrual bleeding during the reproductive years and uterine bleeding after menopause, when these lesions are more frequently associated with endometrial cancer, particularly with late PMB and in women with family histories of this disease. Primary endometrial disorders due to inflammation or linked to intrinsic enzymatic, hormonal or vasculogenic factors have been associated with AUB. Finally, congenital and acquired bleeding diatheses and exogenous anticoagulants must be ruled out.
Intended as an international system to standardize nomenclature and facilitate patient care and research, FIGO in 2011 published and recommended a classification of AUB during patients’ reproductive years using the acronym PALM-COEIN, which stands for Polyps, Adenomyosis, Leiomyoma, Malignancy, Coagulopathy, Ovulatory dysfunction, primary Endometrial disorders, Iatrogenic, Not classified.
Abnormal per vaginal discharge, especially if blood tinged, may be the only sign of endometrial complex and atypical hyperplasia, and cancer, and it should be noted and the cause pursued even during the reproductive years, especially during the fourth and fifth decades of life, and after menopause. Benign causes of these signs include inflammatory and benign, premalignant and malignant lesions of the vagina and the ectocervix and endocervix. So, all abnormal Pap smears should be explained and treated appropriately. Exophytic and other suspicious lesions should be positively diagnosed and properly treated and biopsied when tissue diagnosis is necessary. The finding of either benign or atypical glandular cells in postmenopausal Pap smears can herald cervical, endometrial and adnexal premalignant lesions and extra-genital cancers in a significant proportion of patients. As well as uterine, also vulvar-perineal bleeding, whether a complaint or elicited on review, may be the only sign some patients may have of local-regional, urological, colorectal, intestinal or metastatic cancer.
Endometrial cancers frequently arise in uteri of normal size unless there are coexisting fibro-muscular tumors (benign leiomyoma and sarcomas). Uterine size and leiomyoma are expected to regress in size after menopause and diminished endogenous estrogen stimulation from senescent ovaries. Uterine sarcoma should be suspect in patients with an enlarging or even non-regressing irregular “fibroid” uterus after menopause, especially if accompanied by PMB. Although dysfunctional cystic, polycystic, and tumorous ovaries may be enlarged on clinical examination, excessive hormonal activity from hyperplastic and tumorous ovaries of normal size may be sufficient to disrupt ovulation, resulting in AUB or endometrial stimulation, hyperplasia, and polyps.
Also included in the differential diagnoses of physically detected pelvic masses are metastatic cancer, hydrosalpinx and fallopian tube tumors; para-ovarian and broad ligament cysts and tumors, which are usually enlarging benign embryological ruminants; and leiomyomas and pseudocysts secondary to inflammatory, endometriotic, or postsurgical pelvic adhesions, retroperitoneal kidneys, dermoid tumors, tumorous nodes, and primary colorectal and other gastrointestinal cancers, urological tumors, and masses.
For all stages of endometrial carcinoma reported by FIGO, 5-year survival rates were significantly superior for patients primarily treated from 1999 through 2001 with surgery alone compared with patients treated with radiotherapy alone. These results likely were biased by younger patients and those with fewer and less severe co-morbidities being selected for primary surgery. In the early stages of endometrial cancer clinically confined to the uterine corpus (Stage I) or with cervical extension (Stage II), the results with primary radiotherapy alone for aged and medically inoperable patients approached the results with primary surgery. Stage by stage, compared to endometrial cancer patients treated with surgery alone, the survival rates were improved in patients who received postoperative adjuvant radiotherapy. Preoperative radiotherapy with or without adjuvant cytotoxic chemotherapy has been used to reduce the size of endometrial carcinomas clinically involving the cervix before proceeding with extrafascial hysterectomy to remove the uterus and adnexa that may harbor residual cancer.
There are fewer vaginal-pelvic recurrences, and progression-free survival may be improved with postoperative radiotherapy in surgically staged endometrial cancer patients found to have deep myometrial penetration, involvement of the uterine cervix and/or adnexa, non-endometrioid histotypes. Adjuvant pelvic radiotherapy after surgical debulking of advanced primary and recurrent endometrial cancer also reduces the rate of local-regional recurrences and improves progression-free survival rates. Extending radiotherapy to encompass the para-aortic nodes has been advocated for surgically proven metastases. Some series show that the improved results with postoperative radiotherapy after surgical debulking may be enhanced when combined with chemotherapy.
So far, chemotherapy has a limited role in the pre-operative treatment of endometrial cancers clinically and surgically confined to the uterus, though platinum-based regimens have been used with radiotherapy to reduce the size of tumors involving the cervix before surgery. Preoperative platinum-based regimens have also been used before debulking surgery for stage IV and recurrent endometrial cancers.
Postoperative Adjuvant Chemotherapy
Although local-regional recurrence rates have been reduced and progression-free survival rates increased with postoperative pelvic radiotherapy in cases of endometrial carcinoma clinically limited to the uterus, but with surgically demonstrated deep myometrial, cervical and/or adnexal involvement, these favorable results were not further enhanced by the addition of cytotoxic chemotherapy. In surgical stage III endometrial cancers, the combination of chemotherapy with pelvic radiotherapy has been associated in some studies with increased rates and lengths of progression-free and overall survival and reduced local-regional recurrence rates.
Platinum-based cytotoxic regimens have been the most often used chemotherapy for both neoadjuvant and postsurgical chemotherapy for endometrial cancer; although targeted hormonal, enzymatic and cell growth, and proliferation inhibitors and immunotherapies are emerging for advanced endometrioid and for serous carcinomas.
Management of endometrial cancer and its precursor lesions depends upon the histotype and differentiation (see above: Etiology), extent of disease and status of the patient. The extent of endometrial carcinoma is surgically-pathologically classified, or "staged," according to International Federation of Gynecology and Obstetrics (FIGO) criteria published in 2009 as:
Stage I:* Tumor confined to the corpus uteri
Stage II:* Tumor invades cervical stroma but does not extend beyond the uterus
Stage III:* Local and/or regional spread of tumor
Stage IV:* Tumor invades bladder and/or bowel mucosa or distant metastases
*Includes FIGO Grades 1, 2, or 3 (see above: Etiology); Endocervical glandular involvement on is Stage I. Positive peritoneal cytology is reported separately without changing stage.
Although positive peritoneal cytology does not change FIGO staging, this finding is associated with other factors that portend a poor prognosis.
It is important to understand that staging is a means of coding and reporting the extent of disease and not a formula for managing endometrial cancer. Current FIGO classification by stage depends upon both surgical and pathological findings. Generally, preinvasive neoplastic lesions are diagnosed clinically and pathologically without resort to surgery. Sometimes, it is necessary to clinically and pathologically "stage" cancers when surgery is incomplete or precluded because of co-morbidity or other forbidding reasons.
The prognosis for patients diagnosed with endometrial cancers depends not only on the extent of disease found by surgical-pathological staging but also on the histotype, size, grade, and of course, co-morbidities. Using the older FIGO 1988 surgical classification for uterine corpus carcinoma, the overall five year survival results for patients treated in the years 1999-2001, regardless of histotype and grade, were: Stage IA 90.8%, Stage IB 91.1%, Stage IC 85.4%, Stage IIA 83.3%, Stage IIB 74.2%, Stage IIIA 66.2%, Stage IIIB 49.9%, Stage IIIC 57%, Stage IVA 25.5%, and Stage IVB 20.1%. Besides the depth of invasion, lymphovascular involvement and carcinoma positive peritoneal cytology are associated with higher recurrence and poorer survival rates in patients with the Stage I (FIGO, 1988) cancers; although these factors are not used for staging in the current FIGO 2009 classification system (see above: Staging). Reported 5-year survival results using strictly clinical staging were significantly worse for Stages I, II and III (FIGO, 1988), likely reflecting the selection of patients deemed inoperable due to medical complications. Overall, patients with non-endometrioid and dedifferentiating cancers did worse, stage by stage, than those with endometrioid grade 1 tumors.
Astute post-treatment follow-up of patients after primary surgery for endometrial cancer and adjuvant therapy, when indicated, should be intent on detecting recurrent cancer when this can be most successfully and cost-effectively treated. Authoritative groups are consistent in their recommendations for intelligible patient counseling and attention to new symptoms, particularly vaginal discharge and/or bleeding, pelvic or abdominal discomfort and pain, loss of appetite, anorexia, weight loss, lethargy, weakness, and shortness of breath. Thorough cancer-focused general and pelvic examinations are recommended every 3 to 6 months for 2 years and at 6 to 12 months intervals after that. Use of interval vaginal cytology, chest x-ray, CT or PET/CT scans, and serum CA125 determinations are not recommended routinely in the absence of suspicious symptoms and/or for surveillance of most patients after treatment for early endometrial cancer. These tests may be selectively used when following patients after treatment of advanced primary endometrial cancer and during management of recurrent disease.