Carcinoma of the uterine endometrium is the most common pelvic malignancy in women. The USAand Canada have the highest incidence rates in the world, whereas developing countries and Japan have inci- dence rates four to five times lower. Epidemiologic data indicate that there are two forms of endometrial cancer. One is directly related to estrogen exposure and is most common in the USA. The other is unre- lated to estrogen and occurs throughout the world. Estrogen-related type I tumors occur among younger perimenopausal women and carry a good prognosis. In fact, type I lesions are potentially preventable through recognition of patient risk, diagnosis of the precursor lesion (atypical endometrial hyperplasia) and proper treatment. Non-estrogenrelated type II tumors occur in older postmenopausal women without a history of estrogen exposure and have a poorer prognosis. The molecular genetic alterations present in type I and II endometrial carcinomas are distinct and may help to explain their clinical characteristics.
Cells of the Müllerian tract can differentiate into a wide range of tissue types. This is demonstrated by the variety of histologic subtypes seen among the endometrial cancers. The vast majority are endometrioid adenocarcinomas. Prognosis for patients with endometrioid adenocarcinoma is determined largely by its degree of differentiation or histologic grade (well, moderately or poorly differentiated). In fact, histologic grade is a prognostic factor independent of stage at diagnosis. Less common histologic subtypes include mucinous adenocarcinoma, serous adenocarcinoma, clear cell adenocarcinoma, squamous cell carcinoma and a variety of rare mixed and undifferentiated tumors. For all subtypes other than endometrioid adenocarcinoma, prognosis depends more on histologic subtype than on histologic grade.
Endometrioid adenocarcinoma first invades the stroma of the underlying uterine tissue by destroying the glandular basement membrane. It then invades the myometrium and cervix. Endometrioid adenocarcinoma typically spreads via the pelvic and periaortic lymphatic channels rather than hematogenously. Vascular invasion is usually seen only with high-grade, non-estrogen-dependent lesions.
Treatment of endometrial cancer usually involves surgical removal of the uterus, fallopian tubes and ovaries. Patients with deep myometrial invasion or disease outside of the uterus may be treated postoperatively with radiation, chemotherapy or progestin-based hormonal therapies. Pretreatment analysis of endometrioid adenocarcinoma specimens for estrogen and progesterone receptor status may help to direct postsurgical therapy. There is a good correlation between tumor differentiation and receptor content. Well-differentiated tumors usually have greater numbers of estrogen and progesterone receptors. Because receptor content predicts response to progestin therapy, patients with well- differentiated tumors may be good candidates for progestin therapy.
The survival rate for endometrial cancer is relatively good. Overall, survival approaches 70% at both 5 and 10 years. Patients with stage 1 disease, in which the tumor has not invaded through more than half the myometrial thickness, have a 5-year survival rate of over 90%. Because of its high prevalence, endometrial cancer can be considered a neoplasia of high morbidity and relatively low mortality in developed countries.
Epidemiology Of Endometrial Cancer
Endometrial cancer is largely a disease of the postmenopausal woman. About 80% of cases diagnosed are in women aged 50–75 years of age, with peak incidence in those aged 55–70. A woman entering meno- pause has double the chance of developing endometrial cancer com- pared with her chance for developing carcinoma of the cervix or the ovary. The incidence of endometrial cancer varies dramatically from country to country. This geographic pattern follows that of breast and ovarian cancer, with the highest rates in industrialized countries. It is exactly the opposite of patterns observed for cervical cancer.
An association between estrogen exposure and endometrial cancer has been apparent for over 50 years. Many of the risk factors listed in Table 43.1 are thought to increase the risk because of their close association with high estrogen levels, typically unopposed by progesterone. The single most important and best defined risk factor for adenocarcinoma of the uterus is obesity. Adipose tissue has active aromatase enzymes. Adrenal androgens are rapidly converted to estrogens within the adipose tissue of obese individuals. These newly synthesized estrogens also have excellent bioavailability because the metabolic changes associated with obesity inhibit the production of sex hormone-binding globulins by the liver. Obese individuals may have dramatic elevations in their circulating bioavailable estrogens and this exposure can cause hyperplastic growth of the endometrium. Close links exist between the risk of endometrial cancer, a high-fat diet and gross national product, which suggests that level of industrial development may affect incidence of endometrial carcinoma by influencing food consumption. A high-fat diet is also associated with obesity and type 2 diabetes mellitus. Amount and type of dietary fat influences estrogen metabolism. For example, diets rich in beef or in fats increase estrogen reabsorption from the bowel.
White women are three times more likely to be diagnosed with endometrial cancer than black women. Again, this is exactly the oppo- site of what is seen for cervical cancer.
Steroid Hormones And Endometrial Cancer
As noted above, the epidemiologic data on endometrial cancer reveal a striking association between estrogen exposure and cancer development. Interestingly, a direct causal link can only be inferred at this time. The basis for considering estrogen as an etiologic factor comes from three sources: (i) the biologic activity of estrogen and progester- one on the endometrium; (ii) animal and human data on the effects of diethylstilbestrol (DES) on carcinogenesis; and (iii) the association of endometrial cancer with endometrial hyperplasia in conjunction with the association of hyperplasia with prolonged and unopposed estrogen exposure.
The strongest attestation to the high sensitivity of the endometrium to ovarian steroid hormones is the dramatic changes that occur in this tissue during each menstrual cycle (Chapters 10 and 14). In a normally cycling woman, the endometrium changes its morphology on a day- to-day basis. In the follicular phase of the cycle, estrogens stimulate proliferation of the epithelium covering the endometrial glands and of the underlying stroma. Estrogen induces production of its own recep- tor and of the progesterone receptor during this time. Progesterone secreted after ovulation promptly arrests the proliferative activity in the glands and converts the epithelium to a secretory state. The stroma responds to progesterone with angiogenesis and functional maturation. If pregnancy should occur, these changes will prepare the endometrium for implantation. It is believed that the potent mitogenic effect of estrogen on the epithelium of the endometrial glands accelerates the spontaneous mutation rate of predisposing oncogenes and/or tumor suppressor genes. This leads to neoplastic transformation.
Animal and human data gathered after developmental exposure to DES add biologic evidence for the carcinogenic potential of estrogens in the reproductive tract. DES is a nonsteroidal estrogen agonist that was among the first synthetic estrogens to be developed. It was admin- istered to over 2 million women between 1940 and 1970 as treatment for threatened miscarriage. In mice, neonatal exposure to DES produces endometrial cancer in 95% of animals by 18 months of age. In women, prenatal DES exposure leads to structural abnormalities of the reproductive tract (Chapter 27) and to clear cell adenocarcinoma of the vagina and cervix. The carcinogenic action of the DES appears to be mediated in part through activation of the estrogen receptor. Whether prenatal DES exposure will cause endometrial cancer in humans will be determined as this cohort of women continues to be followed through menopause. The molecular genetic mechanism by which DES lead to clear cell carcinoma and naturally occurring estrogens to type I endometrial cancer may be similar. Genetic instability of microsatellite sequences has been demonstrated in both of these tumors.
Molecular biology of endometrial cancer K-ras oncogene mutations and microsatellite instability are most common in type I estrogen-related tumors. Mutations of the PT53 tumor suppressor gene and overexpression of the ERBB2 oncogene are more frequently observed in type II non-estrogen-related tumors.
Endometrial hyperplasia describes a spectrum of changes in the endometrium. These can range from slightly disordered patterns that merely exaggerate the changes seen in the late proliferative stage of the menstrual cycle to irregular, hyperchromatic lesions that are difficult to distinguish from endometrioid adenocarcinoma. Nonetheless, noninvasive endometrial hyperplasia can be divided into two basic types: hyperplasia and atypical hyperplasia. Atypia is characterized by nuclear enlargement, hyperchromasia or irregularities in nuclear shape. Hyperplastic lesions can be further subdivided. Simple hyperplasia describes hyperplastic changes with regular glandular architecture while complex hyperplasia has irregular glandular architecture (Fig. 43.1a). Of the four types of endometrial hyperplasias – simple, complex, atypical simple and atypical complex – only atypical complex hyperplasia poses significant risk for progression to invasive carcinoma. The progression from hyperplasia is slow and may take 5 years or more. About 20% of women with complex atypical hyperpla- sia will develop endometrial adenocarcinoma (Fig. 43.1b). Only 1–2% of those with the other hyperplastic lesions will progress.
Endometrial hyperplasia has the same epidemiologic risk factors as endometrial cancer. Among patients with atypical endometrial hyperplasia, postmenopausal status is associated with the highest risk of progression to adenocarcinoma (33% over 10 years). Endometrial cancer is rare during the child-bearing years. When it occurs, it is usually associated with clinical disorders that cause chronic, unopposed estrogen exposure, including the polycystic ovary syndrome and chronic anovulation (Chapter 31). Estrogen-producing ovarian tumors, such as the granulosa–theca cell tumors (Chapter 42), are also associated with the development of endometrial hyperplasia and adenocarcinoma in premenopausal women.
Progesterone-based therapies are used to halt endometrial proliferation and to convert the endometrium to a secretory state in women with endometrial hyperplasia with low malignant potential. Treatment can be given cyclically or continuously. Atypical endometrial hyperplasia is treated surgically (hysterectomy) unless there is a contraindication to the procedure.