Endometrial carcinoma is the most common gynaecological pelvic malignancy in developed nations. Generally it occurs after menopause, manifesting itself as vaginal bleeding. Any postmenopausal bleeding should alert the examiner to the possibility of endometrial cancer. Epidemiological risk factors include obesity, nulliparity, diet, advancing age, and unopposed oestrogen use. The use of oral contraception and, surprisingly, smoking, correlate negatively with endometrial cancer.

Physical examination is not directly helpful in establishing the diagnosis of endometrial cancer because of its occult location, and pelvic examination and Pap smear are usually normal. Diagnosis is generally made by endometrial biopsy, typically indicated to evaluate abnormal bleeding.

The least complicated method of endometrial biopsy is recommended. The results of today's simple, well tolerated, in-office sampling techniques, such as the Pipelle technique, correlate extremely well with those obtained by the technique used formerly, namely the almost obsolete standard dilatation and curettage under anaesthesia. The latter procedure still has a back-up role when the diagnosis remains unclear after more cost-effective methods of biopsy have been tried.

The endometrial biopsy establishes the diagnosis of malignancy. Of equal value is endocervical curettage which can identify occult endocervical extension of the cancer. Histology, grade, and endocervical involvement, all easily determined, affect choice of treatment. Otherwise, few other preoperative tests in clinically early endometrial cancer are fruitful. However chest radiograph and often CT scan are done prior to surgery to look for distant disease. The ovarian cancer serum tumour marker, CA125, is occasionally elevated in endometrial cancer; changes in such elevations can be useful in monitoring treatment response. Without relevant symptoms, cystoscopy, sigmoidoscopy, and pelvic ultrasound are invariably negative and hence can be avoided. MRI, however, shows promise in determining the extent and depth of uterine involvement and pelvic distribution of disease, but need not be done routinely since it is costly and rarely alters the choice of treatment.

Endometrioid adenocarcinoma is by far the most prevalent malignant histology in the uterus, and together with the other adenocarcinomas of the endometrium, adenosquamous, papillary serous, and clear cell cancers, constitutes 90 per cent of uterine cancer. Differentiating between these histological subtypes is important as each has its own biological behaviour and outcome. For example, the 5-year survival for the typical early stage endometrioid cancer can be 90 per cent. This excellent survival rate contrasts sharply with around 40 and 45 per cent survival reported with papillary serous, and clear cell carcinomas respectively. Additionally, papillary serous carcinomas of the endometrium tend to behave biologically more like epithelial ovarian cancer, often spreading and recurring diffusely on peritoneal surfaces.

The biological spread pattern of the typical advancing endometrioid carcinoma is first expansion within the endometrial cavity, then invasion of the myometrium followed by invasion of the vascular/lymphatic space. This allows metastatic spread to the pelvic lymph nodes and, in turn, to the para-aortic lymph nodes. The para-aortic nodes were thought to be the initial/preferential site of nodal metastases for endometrial cancer, but this clinical axiom has been disproven.

Another important site for lymphatic metastasis is the vagina. Although tumour implantation and direct extension to the vagina can also occur, most endometrial metastases to the vagina are derived lymphatically. This observation has major therapeutic implications, since vaginal metastases/recurrence usually represent only the ‘tip of the iceberg’, rather than an isolated recurrence. As a result, the present International Federation of Gynaecology and Obstetrics (FIGO) staging system has allotted vaginal metastases a high stage (IIIb). In at least one-third of cases, vaginal disease is a ‘marker’ for simultaneous, albeit occult, systemic disease. This reality accounts for the high failure of salvage treatment if it is directed only at obvious local or regional endometrial recurrence.

Further progression of the tumour occurs directly within the pelvis and exfoliated malignant cells can implant throughout the peritoneal cavity. Vascular metastases to the lung, liver, bone, and brain do occur, but late in the disease process. Involvement of the endometrium and ovaries with simultaneous ‘endometrioid’ cancers is not uncommon. This progression may represent either synchronous separate primaries or metastatic disease, one to the other, which is an important distinction to make regarding prognosis and treatment.

By convention (FIGO) staging of endometrial cancer is now done by surgery (Table 1) 412. Changing from the former clinical staging system to a surgical approach brings a more comprehensive understanding of the true disease distribution. The extent of disease was frequently misjudged when only preoperative clinical findings were used to determine staging. For example, the Gynecologic Oncology Group found that clinical staging underestimated the true (surgical–pathological) stage in at least 10 per cent of cases. The greater accuracy of surgical staging realistically identifies and classifies patients by existing risk factors and allows a better statistical comparisons of treatments to be made. FIGO staging of endometrial cancer considers not only disease distribution, but further subcategorizes the disease based on histological grade and degree of differentiation of tumour.

Because patients with endometrial cancer develop abnormal bleeding as an early warning sign, most are diagnosed with early/low stage cancers. Indeed, the majority will have no evidence of disease beyond the uterus, 75 per cent being Stage I and 15 per cent Stage II.

Using the information learned from surgical staging and other prognostic variables patients can be rationally divided into prognostic groups. Data from the Gynecologic Oncology Group's landmark prospective endometrial cancer surgical staging study, protocol No. 33, has simplified and validated the list of prognostic factors. Prognosis in Stage I endometrial cancer is predicted by tumour grade, depth of myometrial invasion, vascular/lymphatic space invasion, age, hormonal receptor status, and tumour histology (clear cell and papillary serous carcinomas have a poorer prognosis).

In the Gynecologic Oncology Group study, patients with well differentiated adenocarcinomas had only a 3 per cent risk of pelvic lymph node involvement, compared with 18 per cent for those with poorly differentiated lesions. If no myometrial invasion was present, the risk of lymph node metastases was only 1 per cent. This risk increased to 25 per cent if the carcinoma invaded the myometrium to the outer third. Patients with no invasion of the vascular/lymphatic space had a 7 per cent risk of pelvic lymph node metastases, whereas those that had such invasion had a 27 per cent risk. Moreover, invasion of the vascular/lymphatic space has been associated with increasing the recurrence rate from 10 to 44 per cent.

Information from DNA ploidy and S-Phase fraction analysis may be useful in planning treatment. Their practical clinical importance, however, remains undefined.

The prognostic value of malignant cytology (washings) from the abdomen and pelvis has been controversial. Although uncommon as the sole risk factor present in a case of endometrial cancer, recent data suggest that positive pelvic washings are an important independent predictor of risk. Indeed, in the Gynecologic Oncology Group series, malignant cytology alone was associated with a 5-year survival of 56 per cent. In recognition of the poorer outcome and risk for diffuse recurrence implied by positive washings, FIGO has raised the stage of patients with malignant cytology from I to IIIa (Fig. 1) 1460.

Another observation relevant to the surgical management of endometrial cancer is that, contrary to former belief, the para-aortic lymph nodes (Stage IIIc) are not the prime site of nodal metastases. Rather, the pelvic lymph nodes are those more commonly involved. Overall, 9 per cent of Stage I endometrial cancer patients had pelvic lymph node metastases, compared to 6 per cent with para-aortic disease. The risk of lymph node positivity depends on the presence of other prognostic factors (e.g. grade, vascular/lymphatic invasion). Up to 61 per cent of Stage I patients, had positive pelvic lymph nodes and up to 30 per cent had para-aortic lymph node involvement. However, the para-aortic nodes themselves are rarely the exclusive sight of nodal metastases, occurring in only 2 per cent of cases overall. Thus, in the absence of other major risk factors identified before or during surgery (e.g. high tumour grade plus deep invasion, intraperitoneal disease) examining subclinical para-aortic nodes by biopsy carries an unacceptably high risk of morbidity.

As suggested, the maximal clinical value of prognostic factors comes when they are combined and used, proactively, during all phases of the intervention. For example, the tumour grade and endocervical involvement can be ascertained before laparotomy even though these may be altered by findings at surgery. When ‘high-risk’ factors are identified preoperatively, such as a poorly differentiated cancer or cervical extension, a need for more complex surgery and staging can be anticipated and a referral to a gynaecological oncologist made.

The proactive use of prognostic factors is even more profound at laparotomy, where new, meaningful information is uncovered which shapes the ongoing operation. For example, despite clinical appearances to the contrary, unanticipated gross extrauterine disease can still be encountered at surgery. Extrauterine tumour is a major poor risk factor associated with a high probability of lymph node metastases (pelvic 33 per cent, para-aortic 8 per cent). Finding extrauterine tumour then justifies a more extensive staging operation than usual including lymph node sampling from both the pelvic and para-aortic regions. Other major risk factors can also be determined intraoperatively. Specifically, the depth of myometrial invasion and endocervical involvement can be quickly identified by frozen section. Surgery is then tailored accordingly using the combined knowledge of both the pre- and intraoperative risk factors. After surgery more detailed pathological analysis becomes the basis for the final risk stratification, which determines the need for adjuvant therapy.

Traditionally, endometrial cancer has been managed with radiation followed a few weeks later by hysterectomy. This sequence was credited with a 70 to 90 per cent 5-year survival for patients with early disease. It is understandable, given the good survival record relative to most other cancers, that the time-honored approach (radiation followed by surgery) went unchallenged. However, it became obvious that the same survival duration is achieved when surgery, not radiation, is the primary/initial management step. Moreover, by not giving radiation as the primary treatment, the true pathological findings are not obscured by radiation damage. Equally effective and as well tolerated by patients after surgery as it was preoperatively, radiation is now employed selectively based on the actual disease distribution.

The first step in the management of most endometrial cancers is surgery. Exploratory laparotomy, cytological washings, extrafascial total hysterectomy, bilateral salpingo-oophorectomy, intraoperative assessment of myometrial invasion and disease distribution, and lymph node sampling, if indicated, are usually done. Adjustments are made in the exact intervention used according to patient age, histological cell type, tumour grade, and intraoperative findings. For example, finding gross extrauterine disease in the upper abdomen may diminish the effectiveness of surgery. In that situation, a limited palliative operation should be considered, preventing the otherwise inevitable extensive bleeding that results from untreated endometrial cancer.

For those patients who are deemed inoperable for medical reasons, primary radiation, including external beam and intracavitary implants, is an alternative to surgery. The cure rate, however, even with early disease, is significantly less than that achieved by hysterectomy alone. Fortunately, despite the general first impression to the contrary, most patients, including the very elderly, tolerate surgery well.

At Women & Infants Hospital/Brown University, all patients with gynaecological cancer are presented prospectively to the multidisciplinary Tumor Board.

Patients with a poor prognosis can be accurately identified at the time of surgery. Pelvic radiation is the most common adjuvant treatment given to those at a high risk for local recurrence. Currently at Women & Infants Hospital/Brown University standard treatment protocols recommend, with exceptions, postoperative radiation therapy for patients with ‘Stage I’ disease and deep myometrial invasion, vascular/lymphatic invasion or extension to the endocervix—assuming that in these cases the lymph nodes are disease free. Alternatively, when the lymph nodes are affected, extrauterine disease is identified, or the washings contain malignant cells, the strategic value of radiation is less clear, and patient entry into experimental advanced disease treatment protocols is recommended.

Although radiation has been used for a long time as a major treatment modality for endometrial cancer, there are surprisingly few data suggesting that pelvic radiation improves survival beyond that achieved by surgery alone. Anecdotal and smaller retrospective studies have suggested improved regional control, and survival is sometimes achieved in subsets of patients given adjuvant radiation. However, larger studies have not confirmed the vival advantage and prospective studies have not been carried out.

Local recurrences, particularly in the vagina, can be reduced by radiation given either as an external beam to the whole pelvis or as a vaginal cylinder, albeit at a possible cost of radiation induced morbidity including major and minor gastrointestinal and genitourinary complications and diminished sexual function. A randomized, prospective study is still needed to determine whether radiation therapy improves outcome in endometrial cancer. In the meantime the patients with a good prognosis can be spared the potential problems of radiation by using primary surgery and postoperatively stratifying for risk.

Effective treatment for advanced endometrial cancer is elusive. Pelvic radiation therapy, with and without progesterone supplementation, is often tried with high-risk or high-stage disease, but with generally discouraging results. This problem is understandable since treating a diffuse malignant process using a local/regional therapy is bound to fail. Research studies of protocols to improve management are required to develop appropriate systemic treatment.

Although logical, and once highly recommended, systemic hormonal therapy with high-dose progesterone has proved, in carefully controlled studies, to be a largely ineffective treatment. Occasionally, however, responses may occur, particularly with grade 1 lesions at sites outside the radiated field. Similarly, anecdotal reports of responses to chemotherapy have been made but larger studies have yet to identify the role, if any, for routine use of currently available chemotherapy. At present, in spite of the lack of convincing evidence, standard systemic management for advanced endometrial cancer combines variations of Adriamycin, cisplatinum, cytoxan, ifosfamide, and VP–16 for their short-term benefit.

Recurrent endometrial cancer has a surprisingly poor prognosis. The actual scope of recurrence is often belied by the seemingly limited extent of clinical disease. A common example would be finding an almost innocuous nodule of recurrent endometrial carcinoma in the upper vagina. Given that the mode of metastasis to the vagina is usually lymphatic, although unidentified clinically and radiographically, there is often occult, systemic/diffuse disease simultaneously being disseminated by the same means. Thus, despite their common application in this situation, local/regional treatments directed at the obvious site of recurrence alone have been unsuccessful. There have been, nonetheless, occasional salvage cures reported using radiation (if not previously given) and still rarer successes with salvage surgery (e.g. pelvic exenteration) if the recurrence is truly a local phenomenon. Unfortunately, most patients have widespread disease and thus are candidates for palliative treatment only, often using systemic therapy protocols.

Surgically restaging patients presumed to have early local recurrence can play an important role in the decision on optimum treatment. Further, where possible, complete surgical resection of the recurrent tumour enhances the salvage therapy.

In general for advanced/recurrent endometrial cancer, surgery and radiation are of limited value for local control and, although desperately needed, systemic therapies are merely briefly palliative.

As with all malignancies, a variety of less common uterine neoplasms exist. The oncological/surgeon should be aware of these ‘special cases’ and their need to be managed accordingly.

Papillary serous carcinoma is a relatively uncommon malignancy in the endometrium. However, papillary serous carcinoma as an endometrial cancer is noteworthy because of its different biological behavior. The spread patterns and prognostic factors characteristic of the typical endometrioid adenocarcinoma of the endometrium do not apply to papillary serous carcinoma: on the contrary, papillary serous carcinoma arising in the endometrium behaves biologically more like papillary serous carcinoma occurring as epithelial ovarian cancer. In addition to the same histology as papillary serous carcinoma in the ovary, the endometrial equivalent behaves similarly, spreading and recurring diffusely on peritoneal/serosal surfaces throughout the entire abdomen. Virulent behaviour from endometrial papillary serous carcinoma can be expected, even in the absence of myometrial invasion or of other poor prognostic factors useful in predicting risk in the stereotypic endometrial adenocarcinoma. Moreover, aggressive behaviour of papillary serous carcinoma can occur with only a small, seemingly, insignificant admixture of the histology, often overlooked in a dominant visual background of simultaneous endometrioid carcinoma.

After surgery, endometrial papillary serous carcinoma is difficult to manage, even with combination chemotherapy using cisplatin, based on the regimen used for ovarian cancer. Nonetheless, systemic intervention seems necessary given that papillary serous carcinoma has a tendency to diffuse recurrence.

Apart from endometrial cancers, other malignancies can occur within the uterus. About 5 per cent of uterine tumours are sarcomas that can emanate from the myometrium or the endometrial stroma. Although not generally realized, the mixed muellerian mesodermal tumour (carcinosarcoma) and not leiomyosarcoma is the most common connective tissue malignancy in the uterus. Leiomyosarcomas are very rare in the general population compared with their benign counterparts, the uterine leiomyomas (fibroids).

Mixed muellerian mesodermal tumours of the uterus have been subcategorized by pathologists in many ways, although from a practical clinical point of view there is little difference among the subtypes. In general, the primary treatment for mixed muellerian mesodermal tumours and leiomyosarcomas is surgery, total abdominal hysterectomy or bilateral salpingo-oopherectomy and staging followed by adjuvant treatment. Both lesions usually recur locally/regionally and distantly. Thus, radiation to the pelvis as adjuvant treatment may reduce local recurrence, but has little, if any, impact on overall survival. Experimental protocols using systemic treatments for these sarcomas are being investigated.

Many other neoplasms that can also occur in the uterine connective tissues (e.g. endometrial stromal tumours) are part of a spectrum of neoplasia with benign to malignant variants. A detailed analysis of any connective tissue tumour to define its position in the continuum of connective tissue neoplasia, and hence what biological behaviour is to be anticipated is essential.

The presence of rare uterine tumours of this type is often unsuspected prior to, or even during, laparotomy. Fortunately, surgery alone is often curative, particularly for low grade early stage lesions. If the tumour recurs depending on the histology, another local aggressive resection or even resection of distant disease can be appropriate. More often, however, advanced or recurrent connective neoplasia is treated systemically. Some of these neoplasms respond dramatically to hormonal ablation or to high-dose progesterone therapy. Others occasionally respond to combination chemotherapy. Unfortunately, the majority of high-grade sarcomas that cannot be cured surgically only respond to systemic treatment for a short time. Agents used include cisplatin, Adriamycin, ifosfamide, and VP–16.

Occasionally, a young woman with a very well differentiated and limited adenocarcinoma of the endometrium will want to conceive. Typically, such patients have a medical history and physical findings consistent with chronic anovulation. After appraisal of the potentially severe risks involved, these special patients may elect for an unconventional management.

Barring evidence of advanced disease or an MRI suggesting myometrial invasion, patients can be offered intense primary hormonal management using high-dose progesterone instead of hysterectomy. The anticipated response to medical management is amenorrhoea. If amenorrhoea occurs, treatment is continued for 3 to 4 months. A thorough dilatation and curettage with hysteroscopy must follow the limited trial of medical management (or sooner if the clinical behaviour does not conform to expectations).

If the endometrial cancer is ‘resolved’ at the completion of hormonal therapy assistance with fertility drugs can be administered before the need for hysterectomy develops.

Although the fallopian tubes have a direct physical relation with the uterus, the rare tubal cancer is biologically and behaviorally more analogous to that of epithelial ovarian cancers. Fallopian tube malignancy is, therefore, staged and managed like that of the ovary rather than of the uterus.

The diagnosis of a tubal malignancy is usually fortuitous or comes about retrospectively, after surgery is performed for what is thought to be an ovarian mass. Preoperative symptoms are sometimes present, the classic one being a profuse, intermittent watery vaginal discharge. Pain and an abnormal Pap smear, suggesting denocarcinoma but not otherwise explained by routine testings (colposcopy, endocervical curettage, and dilatation and curettage), can also lead to the diagnosis of fallopian tube cancer.

Aalders J, Abeler V, Kolstad P, Onsrud M. Postoperative External irradiation and prognostic parameters in Stage I endometrial carcinoma: clinical and histopathologic study of 540 patients. Obstet Gynecol, 1980; 56: 419.

ACOG Technical Bulletin. Carcinoma of the Endometrium. American College of Obstetricians and Gynecologists; No. 162.

Ali S, Wells M. Mixed mullerian tumors of uterine corpus: a review. Int J Gynecol Cancer, 1993; 3: 1.

Burk TW, Morris M. ‘Surgery for malignant tumors of the uterine corpus. In Gerschenson DM, DeCherney AH, Curry SL, eds. Operative Gynecology. Philadelphia: WB Saunders Company, 1993.

Burke TW, et al. Prospective treatment of advanced or recurrent endometrial carcinoma with cisplatin, doxorubicin, and cyclophosphamide. Gynecol Oncol, 1991; 40: 264.

Chambers JT, et al. Uterine papillary serous carcinoma. Obstet Gynecol, 1987; 69: 109.

Creasman WT, et al. Surgical pathologic spread patterns of endometrial cancer. Cancer Gynecol Obstet, 1987; 21: 35.

Grimes, DA. Diagnostic dilation and curettage: a reappraisal. Am J Obstet Gynecol, 1982; 142: 1.

Gurney H, Murphy D, Crowther D. The management of primary fallopian tube carcinoma. Br J Obstet Gynaecol, 1990; 97: 822.

Hacker NF. Uterine cancer. In Berek JS, Hacker NF, eds. Practical Gynecological Oncology. Baltimore: Williams & William, 1989: 285–326.

International Federation of Gynecology and Obstetrics. Annual report on the results of treatment in gynecologic cancer. Int J Gynecol Obstet, 1989; 28: 189–90.

Larson B, Silfversward C, Nilsson B, Petterson F. Prognostic factors in uterine leiomyosarcoma. Acta Oncol, 1990; 29: 185.

Levi F, et al. Chemotherapy of advanced ovarian cancer with 4′–0-tetrahydropyranyl doxorubicin and cisplatin: a randomized phase II trial with an evaluation of circadian timing and dose-intensity. J Clin Oncol 1990; 8: 705.

Meerwaldt JH, et al. Endometrial adenocarcinoma, adjuvant radiotherapy tailored to prognostic factors. Int J Radiol Biol Phys, 1989; 18: 299.

Mittal KR, Schwartz PE, Barwick KW. Architectural (FIGO) grading, nuclear grading, and other prognostic indicators in Stage I endometrial adenocarcinoma with identification of high-risk and low-risk groups. Cancer, 1988; 61: 538.

Morrow CP, et al. Relationship between surgical-pathological risk factors and outcome in clinical Stage I and II carcinoma of the endometrium: a GOG study. Gynecol Oncol, 1991; 40: 55.

Onsrud M, Kolstad P, Normann T. Postoperative external pelvic irradiation in carcinoma of the corpus Stage I: a controlled clinical trial. Gynecol Oncol, 1976; 4: 22.

Parazzini F, La Vecchia C, Bocciolone L, Franceschi S. The epidemiology of endometrial cancer. Gynecol Oncol, 1991; 41: 1.

Phillips GL, Prem KA, Adcock LL, Twiggs LB. Vaginal recurrence of adenocarcinoma of the endometrium. Gynecol Oncol, 1982; 13: 323.

Piver MS. Progesterone therapy for malignant peritoneal cytology: surgical Stage I endometrial adenocarcinoma. Semin Oncol, 1988; 15: 50–2.

Piver MS, Recio FO, Baker TR, Hempling RE. A prospective trial of progesterone therapy for malignant peritoneal cytology in patients with endometrial carcinoma. Gynecol Oncol, 1992; 47: 373–6.

Shimm DS, et al. Management of high-grade Stage I adenocarcinoma of the endometrium: hysterectomy following low dose external beam pelvic irradiation. Gynecol Oncol, 1986; 23: 183.

Sivridis E, Buckley CH, Fox H. The prognostic significance of lymphatic vascular space invasion in endometrial adenocarcinoma. Br J Obstet Gynaecol, 1987; 94: 991.

Wolfson AH, et al. The prognostic significance of surgical staging for carcinoma of the endometrium. Gynecol Oncol, 1992; 45: 142.

Young RH, Clement PB. Pathology of the uterine corpus. Hum Pathol, 1991; 776.