Endometrial Carcinoma

INTRODUCTION

Background: Corpus cancer is the most frequently occurring female genital cancer. Approximately 38,300 cases of corpus cancer will be diagnosed in the US this year, making it the fourth most common cancer among women; of these women, approximately 6600 will die from the disease.

In developed countries, adenocarcinoma of the endometrium is the most common gynecological cancer; however, in developing countries, it is much less frequent than carcinoma of the cervix. In the US, cancer of the cervix in the early part of the 20th century killed more women than any other cancer, but in the ensuing decades, the incidence for that malignancy decreased precipitously. The impact of screening with the Pap smear has been credited with decreased incidence. In less-developed countries, screening for cervical cancer is performed very infrequently, and therefore, cancer of the cervix is quite prevalent.

Frequency:
　

• In the US: Surveillance, Epidemiology, and End Results (SEER) data notes a total age-adjusted incidence of 21.2/100,000 people, with white being 22.5 and blacks 15.0.

Mortality/Morbidity: In the US, endometrial carcinoma is expected to be diagnosed in 31,000 women in 2000; approximately 5900 of these women will die as a result of the disease. Mortality is higher in blacks than in whites, with a mortality ratio of 5.8/100,000 in blacks and only 3.1/100,000 in whites.

Sex: Endometrial carcinoma occurs in females only.

CLINICAL

History: Since about 75% of women with endometrial cancer are postmenopausal, the most common symptom is postmenopausal bleeding. Investigate all bleeding during menopause unless the patient is on cyclic replacement therapy with normally anticipated withdrawal bleeding. The duration or amount (staining vs gross) of bleeding does not make any difference. Perform investigation as noted below.

The fact that only about 20% of postmenopausal bleeding is due to cancer is appreciated, but obviously, that diagnosis must be eliminated in these patients. Since 25% of endometrial cancers are in patients who are perimenopausal or premenopausal, symptoms suggestive of cancer may be subtler. The idea that any type of bleeding during the perimenopausal period is probably due to menopause is a common misconception. This irregular bleeding often is ignored by the patient and even healthcare providers. Remember that the normal bleeding pattern during this time should become lighter and lighter and further and further apart. Heavy frequent menstrual periods or intermenstrual bleeding needs to be evaluated.

Physical: Bleeding leads to evaluation of the endometrium. In the vast majority of cases, no gross evidence of disease is noted. The uterus may be a normal size on pelvic exam. The cancer can be present on cervical evaluation and, less frequently, in the upper vagina or periurethrally.

Causes: There are multiple epidemiological risk factors that have been identified in patients who have adenocarcinoma of the endometrium.

• Endogenous factors

• Obesity increases the risk for developing endometrial cancer, and some data suggests that a 10-fold increase in risk occurs if an individual is more than 50 lb heavier than their ideal weight.

• Nulliparity also increases risk 2- to 3-fold, compared with the parous woman.

• An individual who has a late menopause (older than 52 years) also appears to have an increased risk.

• Unopposed estrogen, either as replacement therapy or endogenously produced (eg, granulosa cell tumor, polycystic ovarian disease), increases the risk of endometrial cancer several times. Obesity is known to increase endogenous estrogen, as the presence of fat appears to be responsible for conversion of androstenedione to estrogen compounds at a much higher rate than if fat is not present. Anovulation, which may be secondary to unopposed estrogen, also appears to contribute to this situation.

• Tamoxifen, the most widely used anticancer drug, is suggested by some studies to cause an increased incidence of adenocarcinoma of the endometrium.

• This data was derived from retrospective analysis in which adenocarcinoma of the endometrium was not an endpoint in multiple prospective randomized studies evaluating the role of tamoxifen in patients with breast cancer.

• A recent case control study using the SEER database indicates that when confounding factors have been corrected, an increased risk of endometrial cancer in patients on tamoxifen does not appear to exist. This study is very reassuring because the potential for an increased number of women taking tamoxifen is becoming apparent, particularly as the prophylactic role of tamoxifen is being explored.

• In contrast, increasing data exist noting that the use of combination oral contraceptives (OCs) decreases the risk of developing endometrial cancer.

  　

  • Several studies have noted that women who use OCs at some time have a 0.5 RR of developing endometrial cancer compared to women who had never used OCs.

    　

  • This protection occurred in women who used OCs for at least 12 months, and the protection continued for at least 10 years after OC use. Protection was most notable for nulliparous women.

• Cigarette smoking apparently decreases the risk of developing endometrial cancer.

• The effects of smoking are related to body weight. Heavier women who smoke have the greatest reduction in risk.

• Women who smoke are known to undergo menopause 1-2 years earlier than women who do not smoke.

• Although smoking apparently reduces the risk of developing early stages of endometrial cancer, this advantage is strongly outweighed by the increased risk of lung cancer and other major health problems associated with smoking.

• There are associated medical conditions in which an increased incidence of endometrial cancer may be found.

• Breast, colon, and ovarian cancers frequently are seen in women with endometrial cancer.

• Data suggest that women who have had breast cancer have a 2- to 3-fold increased risk of subsequently developing endometrial cancer.

• Individuals with a family history of endometrial cancer also appear to be at an increased risk.

• At one time, a classic phenotypic characteristic for a woman who would develop endometrial cancer was thought to exist. This phenotype was the patient who was obese, nulliparous, and anovulatory in many instances. More recently, the existence of 2 pathogenic types of endometrial cancer was appreciated.

• The first type occurs in women who fall into the classic category. These women are obese; have hyperlipidemia; have signs of hyperestrogenism, such as anovulatory uterine bleeding, infertility, and late onset of menopause; and may have hyperplasia of the ovarian and endometrial stromas. These patients tend to be white, obese, nulliparous, and have a well-differentiated, superficially invasive cancer that is highly sensitive to progestin. These patients have a very favorable prognosis with removal of the uterus. Extrauterine disease is unusual in this group of patients.

• In contrast, the second pathogenic type occurs in women who have none of the disease states present in the classic presentation. These individuals tend to have a poorly differentiated tumor, deep myometrial invasion, a high degree of metastasis in the lymph nodes, decreased sensitivity to progestin, and a very poor prognosis. These latter patients tend to be thin and multiparous, and they are more frequently African-American. Most patients with endometrial cancer do fall into this latter category.

DIFFERENTIALS

Other Problems to be Considered:

Bleeding from the lower genital tract can occur from the cervix, vulva, or vagina. If the bleeding is due to neoplasms, gross inspection usually can identify these lesions. If cervical cytology is abnormal and no gross lesions are identified, further evaluation needs to be performed. Atrophic changes in the vagina may lead to bleeding, particularly postcoital. Bleeding from the uterus may be due to many benign lesions, such as polyps or endometritis, or hormone replacement therapy.

WORKUP

Imaging Studies:
　

• Vaginal ultrasound: During the recent past, the increased use of the vaginal ultrasound to evaluate the endometrial stripe has been reported.

• Some investigators feel that this should be the first diagnostic procedure because vaginal ultrasound is less invasive than endometrial biopsy.

• One of the difficulties with using the endometrial stripe as criteria for further diagnostic tests (eg, endometrial biopsy) is that several conditions that give a false reading on the endometrial stripe may be present. This is particularly true in a patient who might have an endometrial polyp or who has been taking tamoxifen.

• Hydroultrasound: If a thickened endometrium is present, perform a hydroultrasound to make sure a false-positive result is not present. This is accomplished by placing a small volume of saline into the endometrial cavity and then repeating the vaginal ultrasound.

• In many instances in which the original vaginal ultrasound shows significant endometrial thickness, a sonoultrasound can differentiate other pathology from true endometrial thickness.

• Another problem that arises is that the thickness of the endometrium can vary considerably, depending upon different factors. The thickness of the endometrium depends upon whether or not the patient is perimenopausal or postmenopausal (and for how long), whether she is on hormone replacement therapy, and whether the therapy includes estrogen alone or estrogen plus progesterone. Currently, no generally accepted guidelines for each of these different clinical scenarios exist.

Procedures:
　

• Endometrial biopsy: Although fractional dilatation and curettage (D&C) historically was the definitive diagnostic procedure to rule out endometrial cancer, today the endometrial biopsy as an office procedure is quick, well tolerated, and quite sensitive for making the diagnosis.

• If endometrial pathology is not present on biopsy and the patient has no further bleeding, no additional diagnostic tests need to be performed.

• If the patient continues to be symptomatic, then further evaluate the endometrial cavity.

• Hysteroscopic directed biopsy: Another diagnostic procedure that has been advocated by some as an even more accurate way of determining the status of the endometrium is hysteroscopic directed biopsy; however, studies have shown that when results are compared to the histopathology, both false-positive results and false-negatives results may be noted using this technique.

• D&C: The role today of the formal D&C is probably very limited because the diagnosis can usually be made in the office.

• An exam under anesthesia may be necessary in a patient who is bleeding and has a cervical os that is very stenotic. Anesthesia may be required to carry out adequate dilatation for endometrial sampling.

• Exploratory laparotomy, total abdominal hysterectomy, bilateral salpingo-oophorectomy, peritoneal cytology, and pelvic and paraaortic lymphadenectomy are performed in the hopes of removing the cancer.

Histologic Findings: Pathological diagnosis is obviously the criterion standard of evaluation of the endometrial cavity. The high index of suspicion as noted above must be maintained if a diagnosis of endometrial cancer is to be made.

Endometrioid adenocarcinoma is the most frequent histopathology subtype. A squamous component, either benign (adenocanthoma) or malignant (adenosquamous), does not affect prognosis, but the grade of the adeno component does affect prognosis. Papillary serous and clear-cell histotypes are poor prognostic lesions but, fortunately, are infrequent compared to adenocarcinoma. Secretory carcinomas are the least frequently occurring cancers and have a good prognosis.

Table 2. Histopathology, Degree of Differentiation: Cases of carcinoma of the corpus should be classified (or graded) according to the degree of histologic differentiation.

MEDICATION

The goal of pharmacotherapy is to eradicate the carcinoma, reduce morbidity, and prevent complications.

Drug Category: Chemotherapeutic agents -- Used in the treatment of endometrial cancer. Inhibit cell growth and proliferation.

FOLLOW-UP

Complications:
　

• Complications that may occur from therapy include complications that normally are expected from the surgical procedure itself. Because a lymphadenectomy is performed, increased bleeding could occur, but unique complications from the procedure really are not present.

• Postoperative complications can be expected, depending upon the preoperative clinical condition of the patient. As noted previously, many of these patients have comorbidities such as hypertension, obesity, diabetes, and increasing age.

• One postoperative complication that may be somewhat increased is thromboembolism because this is increased in patients who have cancer, are obese, and are older. Most physicians today use some type of prophylaxis, either external pneumatic compression or low-dose heparin.

Prognosis:
　

• Multiple prognostic factors of endometrial cancer exist. These prognostic factors are by and large related to surgical pathological findings. As in all cancers, the stage of the disease is the most important prognostic factor. Obviously, the surgical procedure helps in determining the stage. Listed below are prognostic factors that may relate specifically to the stage of the disease and, thereby, affect overall survival.

• Pathology: The majority of endometrial carcinomas are endometrioid adenocarcinomas. Adenoacanthomas (benign squamous components) and adenosquamous carcinoma (malignant squamous components) make up the next largest category. Clear-cell and papillary serous adenocarcinomas represent approximately 10% of all endometrial cancers and are considered to be poor histopathological subtypes. These latter subtypes tend to have deeply invasive myometrial involvement, and they have a propensity for extrauterine spread even though the myometrium may superficially be involved. Previously, a patient with an adenosquamous carcinoma was thought to have a poor prognostic histotype because of the malignant squamous component. Contemporary data suggest that irrespective of whether a squamous component is present (either benign or malignant), prognosis is directly related to the grade of the adeno component and not the fact that a squamous malignancy is present. If a malignant squamous component is present, a greater tendency to
  have a more poorly differentiated adeno component exists.

• Histological differentiation: The degree of histological differentiation of endometrial cancer has long been accepted as a most sensitive indicator of the prognosis. Patients with well-differentiated adenocarcinomas tend to have involvement of the endometrium or superficial myometrium, and extrauterine disease is unusual. On the other hand, if a poorly differentiated lesion is present, these cancers tend to be much more aggressive, involving significant myometrial invasion, and often have extrauterine metastasis, either with positive peritoneal cytology, retroperitoneal spread, or involvement of the pelvic and/or paraaortic lymph nodes. Since papillary and clear-cell carcinomas carry a relatively poor prognosis, these subtypes usually are not graded but are considered in the same category as a poorly differentiated cancer.

• Myometrial invasion: The degree of myometrial invasion continues to be a consistent indication of tumor virulence. As the depth of myometrial invasion increases, a greater chance of having extrauterine disease exists. As noted above, grade and depth of invasion, as a generalization, are interrelated. As the grade of the tumor increases, an increase in the depth of myometrial invasion usually occurs; however, exceptions exist in that a grade 1 lesion can have deep myometrial invasion and a grade 3 lesion can be limited to the endometrium. When grade and depth of invasion are evaluated separately, the depth of invasion appears to be a more important prognostic factor than the grade of the tumor.

• Peritoneal cytology: Cytological evaluation of the peritoneum appears to be an important prognostic factor. Although a universal agreement about the significance of cytological evaluation does not exist, the vast majority of data in the literature suggest that it is an independent prognostic factor. Cytological evaluation also appears to correlate with other prognostic factors, such as depth of myometrial invasion and lymph node metastasis. The FIGO staging takes the status of peritoneal cytology into consideration in that if malignant cells are present in the peritoneal cytology without any other evidence of extrauterine disease, the patient is classified as having Stage IIIA disease. If ascitic fluid is not present at the time of the exploratory laparotomy, a saline lavage is carried out of the pelvis and lower abdomen and the specimen is submitted for cytological evaluation.

• Lymph node metastasis: A considerable number of patients who were thought to have clinical Stage I endometrial cancer were, in fact, found to have lymph node metastasis when histopathological evaluation was performed on the lymph nodes. Again, a correlation among multiple prognostic factors has been shown to be present. Patients with poorly differentiated cancers, papillary serous and clear-cell carcinomas, deep myometrial invasion, positive peritoneal cytology, or adnexal metastasis tend to have an increased risk of having lymph node metastasis.

  Subsequent therapy after primary surgery depends upon prognostic factors and spread of the disease. If the disease is limited to the uterus, surgery appears to be adequate treatment, with the possible exception of patients who have poorly differentiated deeply invasive myometrium. In these patients, data suggest that possibly postoperative irradiation may be of benefit. In patients who have disease outside of the uterus, radiation therapy may be effective; however, this has not been evaluated in a prospective randomized study. Most investigators irradiate the appropriate area if lymph node metastasis is present.

  In patients with advanced disease (ie, intraperitoneal disease, disease outside of the peritoneal cavity), systemic chemotherapy may be of benefit. Studies suggest that cisplatin and doxorubicin probably are the drugs of choice when systemic chemotherapy is needed. As previously noted, staging is the most important prognostic factor. Using the FIGO surgical staging classification from the recent annual report (worldwide data evaluation), 5-year survival rates of 87%, 76%, 63%, and 37% were noted for stages I, II, III, and IV of the disease, respectively. Since there are substages that take into consideration prognostic factors, actually 4 of the substages within Stage I have better than 90% 5-year survival.

MISCELLANEOUS

Medical/Legal Pitfalls:
　

• Ignored irregular postmenopausal bleeding could lead to a delay in diagnosis and treatment, which may impact survival.

Special Concerns:
　

• Multiple new prognostic factors of endometrial cancer are being evaluated and are brought about by newer technology, which allows for molecular biological evaluation. Since these evaluations are in their infancy, no general agreement about their importance exists.

• Flow cytometry has been used in ploidy analysis (cellular nuclear DNA content) and to measure the proliferative fraction of tumor cells (S phase).

• The prognostic factors of the ECP-1 score (ie, myometrial invasion, DNA ploidy, and mean shortest nuclear axis) have been evaluated, and in at least one study, multivariant analysis was noted to be important prognostically.

• Several other molecular biological characteristics have been noted to be important prognostically, including HER-2/neu and p53 gene overexpression.

• Newer characteristics are being identified almost daily. Obviously, the necessity for standardization is needed before applicability and conclusions can be reached. As these factors gain experience, then they may very well be the new prognostic factors for endometrial cancer.

• Controversies that are present in the pelvic and paraaortic lymphadenectomy management of adenocarcinoma of the endometrium revolve mainly around the role of.

• There is no question that the procedure is diagnostic. The question that has been raised is whether or not it also might be therapeutic. Certainly in other gynecological cancers, it appears to be therapeutic. Retrospective data by Kilgore and colleagues suggest that lymphadenectomy in endometrial cancer also can be therapeutic.

• When proposed, the FIGO surgical staging classification was questioned as being efficacious by many investigators. Data would suggest, however, that the gynecologic oncology community worldwide has accepted the surgical staging classification. In fact, lymphadenectomies are routinely being performed by these investigators.