Abstract
Endometrial cancer is an increasingly common cancer in affluent Western countries, largely due to the increasing obesity of those populations. Type I endometrial cancer is the more common phenotypic picture, being that of an obese postmenopausal woman, and comprises approximately 80% of endometrial cancers. Type II endometrial cancer describes a woman who is often younger and thinner, with a more aggressive histological type that is non-estrogen dependent, of either serous or clear-cell histology, and has a more aggressive clinical course and resultant poorer prognosis. The standard surgical procedure is an extrafascial total hysterectomy with bilateral salpingo-oophorectomy, with or without pelvic and para-aortic lymphadenectomy. The safety of a laparoscopic approach in the surgical management of uterine cancer has not yet been demonstrated in prospective randomized trials. Surgical staging of endometrial cancer, including a pelvic and para-aortic lymphadenectomy, allows appropriately tailored postoperative radiotherapy. Teletherapy or vaginal brachytherapy may reduce the risk of pelvic recurrent disease in high-risk patients.
Endometrial cancer is the fourth most common cancer in women, behind breast, bowel and lung cancer, and the seventh leading cause of death from malignancy in females. This disease tends to affect affluent, postmenopausal, obese, middle-aged and elderly women of low parity. The incidence of endometrial cancer in affluent Western countries in increasing and this is largely due to the increasing incidence of obesity. The risk of developing endometrial cancer increases with age, with a mean age at diagnosis of 61 years; less than 5% of patients with this disease will be less than 40 years of age. While most women with endometrial cancer present with postmenopausal bleeding, only 10–20% of patients with this underlying symptom will have an endometrial cancer. Menorrhagia is the commonest presenting symptom in premenopausal patients and the fact that this is a common presenting symptom in gynecological patients and the relative rarity of this disease in premenopausal patients often leads to a delay in diagnosis. At presentation, 75% of patients will have surgical Stage I disease and surgery remains the mainstay of therapy, with adjunctive radiotherapy being advocated in selected patients to improve local disease control without improvement in overall survival. Chemotherapy and hormonal therapy is usually reserved for patients with systemic or recurrent disease. Globally, approximately 42,000 deaths were attributed to uterine cancer in 1990, of which 27,500 occurred in developed countries and 14,400 in developing countries. In the USA, death rates from endometrial cancer have doubled between 1988 and 1998, probably due to a combination of increased life expectancy and the increasing obesity rates (with associated comorbidities) [1,2].
Classification
It is now recognized that there are two types of endometrial cancer patient. Type I endometrial cancer is the more common phenotypic picture, being that of an obese postmenopausal woman, and comprises approximately 80% of endometrial cancers. Such a patient is more likely to have a low-risk, estrogen-dependent, endometroid subtype, superficially invasive Grade 1 tumor and a good prognosis. Such tumors are thought to progress through a premalignant phase of endometrial hyperplasia in a large proportion of cases (Figure 1). Type I carcinomas frequently show mutations of DNA-mismatch repair genes (mutL homolog 1 [MLH1], mutS homolog 2 [MSH2], mutS homolog 6 [MSH6]), phosphatase and tensin homolog (PTEN) gene, proto-oncogene K-Ras2A (k-ras) and β-catenin genes (catenin-β1 [CTNNB1]) [3,4].
Bisected uterus showing an endometrial cancer.
Type II endometrial cancer describes a woman who is often younger and thinner with a more aggressive histological type that is non-estrogen dependent, of either serous or clear-cell histology, and has a more aggressive clinical course and resultant poorer prognosis. In contrast to Type I cancer, they are more likely to be associated with an atrophic endometrium or uterine polyps. There is also no clear epidemiological profile of Type II cancers. Type II cancers are more likely to involve activation of the v-erb-b2 erythroblastic leukemia viral oncogene homolog 2 (ERBB2) oncogene and mutation of the tumor protein p53 (TP53) tumor suppressor gene [5].
Screening and surveillance
As most patients with endometrial cancer present with symptoms and are at an early stage of disease, screening is unlikely to be cost effective or reduce the mortality rate. Thus routine screening is not indicated [6]. Patients with a family history of endometrial cancer, those with a history of hormone replacement therapy with less than 12–14 days of progestogens, long-term use of tamoxifen, hereditary nonpolyposis colorectal cancer family syndrome, Cowden's Syndrome and Peutz-Jeghers Syndrome, history of breast cancer and obesity are at risk of developing endometrial cancer and may be candidates for endometrial screening. While ultrasound, endometrial biopsy and serum cancer antigen 125 levels have been utilized, they have limited accuracy and the merit of screening in this high-risk subgroup awaits prospective data [7].
Preoperative work up & staging
While most patients with endometrial cancer present with abnormal or postmenopausal bleeding, 90% of patients with such symptoms do not have cancer. These patients are elderly and, due to a combination of obesity and associated comorbid conditions, are at high risk of anesthetic complications. A transvaginal ultrasound and an endometrial biopsy are able to provide a diagnosis in most patients without a formal curettage under general anesthesia. Transvaginal ultrasound using a 5 mm cut-off is associated with a detection rate of cancer of 96% and a detection rate of any endometrial pathology of 92% [8]. A recent meta-analysis of the value of the Pipelle® outpatient endometrial biopsy calculated a sensitivity of 81–99% and specificity of approximately 98%. Accuracy is enhanced in symptomatic patients and postmenopausal women [9–11]. However, in patients with suboptimal outpatient endometrial assessment or those at high risk of underlying endometrial cancer, formal dilatation and curettage is necessary.
Surgery is the mainstay of treatment and surgical staging is used to assess the extent of disease and the need for adjuvant therapy. Patients with endometrial cancer should be referred to a gynecological oncology unit to facilitate optimal management. A preoperative anesthetic assessment and investigation and treatment of comorbid conditions are essential. Preoperative imaging including a chest x-ray and a computer-assisted tomography scan is useful to detect extra-uterine disease [12].
Corpus cancer is staged surgically according to the International Federation of Gynecology and Obstetrics criteria (
International Federation of Gynecology and Obstetrics staging of cancer of the uterine corpus.
Surgical approach
A generally recommended surgical protocol would be that the abdomen should be opened with a vertical midline abdominal incision and peritoneal washings taken immediately from the pelvis and abdomen, followed by careful exploration of the intra-abdominal contents. The omentum, liver, peritoneal cul-de-sac and adnexal surfaces should be examined and palpated for any possible metastases, followed by careful palpation for suspicious or enlarged nodes in the aortic and pelvic nodal areas. The standard surgical procedure should be an extrafascial total hysterectomy with bilateral salpingo-oophorectomy. Adnexal removal is also recommended, even if the adnexa appear normal, as they may contain micrometastases. Vaginal cuff removal is not necessary nor is there any benefit from excising parametrial tissue in patients with clinical Stage I disease. If cervical stromal involvement is suspected then a radical-type hysterectomy should be performed with an adequate cuff of normal parametrial tissue resected around the cervix. A pelvic and para-aortic lymphadenectomy is an essential part of the staging procedure. A pelvic lymphadenectomy consists of a representative sample of nodal tissue taken from the distal common iliac, the external iliac and obturator node groups. The para-aortic sampling should include a sample from the upper common iliac group and nodes anterior to the inferior vena cava and aorta, caudal to the origin of the inferior mesenteric artery. No attempt should be made to carry out a formal lymphadenectomy. Although required for accurate staging, whether lymphadenectomy has any therapeutic benefit is as yet unproven. While the extent of lymph-node dissection has been shown to be prognostic in some series, this may be due the greater likelihood of resecting subclinical nodal disease [19,20]. Lutman and colleagues noted an improved progression-free and overall survival in patients with high-risk histological subtype endometrial cancer who had 12 lymph nodes or more resected at surgery. In the UK, the ongoing Medical Research Council Study in the Treatment of Endometrial Cancer trial is randomizing all women undergoing surgery for presumed Stage I endometrial cancer to pelvic lymphadenectomy or no lymphadenectomy [19,21].
A more intense surgical approach may be warranted for Type II endometrial cancer, where extra-uterine spread is more common, with lymphatic and transperitoneal routes common. In all patients with extra-uterine disease, maximum surgical cytoreduction is recommended and the extent of residual disease at the completion of surgery impacts on overall survival [22–25].
Laparoscopic surgery
Improvements in laparoscopic instrumentation and integration of laparoscopy into training programs have led to its widespread use in general gynecology. The advantages of minimally invasive surgery include a reduction in surgical scarring, decreased blood loss, decreased postoperative pain, early discharge from hospital and resultant deceased length of stay in hospital, with a potential for cost savings and earlier return to work [28]. While retrospective studies have clearly demonstrated the safety of laparoscopic lymphadenectomy, there is to date no prospective clinical trial assessing the complications and patient outcome when treated by laparoscopic surgery.
The potential role for laparoscopy in the management of endometrial cancer would be to carry out primary treatment with a laparoscopic hysterectomy and lymphadenectomy, evaluate recurrent disease, and determine the feasibility of exenterative surgery. There is to date no prospective large series comparing conventional open surgery with laparoscopic surgery in the management of endometrial cancer. There are case-controlled and cohort studies attesting to the feasibility, safety and adequacy of laparoscopic surgery for endometrial cancer. Potential drawbacks to a minimally invasive approach to uterine cancer surgery include: prolonged operative time, greater expense, a shallow learning curve and increased complications. There are also the real risks of port site recurrence, intraperitoneal dissemination, early tumor recurrence and resultant decreased survival.
The Gynecological Oncology Group (GOG) has recently completed a prospective study (Lap-2), in which laparoscopic-assisted vaginal hysterectomy (LAVH) and laparoscopic lymphadenectomy were compared with abdominal hysterectomy and lymphadenectomy. Despite the fact that the results from the Lap-2 study are currently not available, it would appear the many units around the world have realized the potential impact of minimally invasive surgery, particularly in low-risk patients, and offer this surgical approach as first-line treatment. The newer technique of total laparoscopic hysterectomy is reported to offer greater advantages compared with LAVH, especially in obese patients. [26–37].
Radiation therapy
Radiation therapy can be delivered externally to the pelvis as teletherapy, locally to the vagina as brachytherapy or as a combination of both. In addition, standard pelvic fields may be extended to incorporate the para-aortic areas and in some cases the entire abdomen (whole-abdominal radiation [WAR]).
In early stage disease, the role of pelvic irradiation is restricted to the postoperative, adjuvant setting. Rarely, in medically inoperable patients, radiation may be utilized as primary therapy. Survival after primary treatment of corpus cancer with irradiation therapy is significantly inferior to surgery alone.
Whole-pelvic radiation
The goal of adjuvant pelvic irradiation therapy is to treat the pelvic lymph node regions that contain microscopic resected disease. In those patients where staging was not performed, a decision is then made to prescribe adjuvant radiation therapy based on the risk of lymph-node spread.
In surgically staged patients
Surgically staged patients with low-risk disease have an excellent prognosis without adjunctive therapy, and radiotherapy in this group should be reserved for patients with recurrent pelvic disease.
The GOG recently published a prospective randomized study on adjuvant therapy in intermediate-risk endometrial cancer [38]. Patients with Stage Ib, Ic and II (occult) after total abdominal hysterectomy and bilateral salphingo-oophorectomy and surgical staging were randomized to either no further therapy or pelvic irradiation. While there was a significant reduction in local and regional recurrence in the radiation group (p < 0.01) from 12–3%, there was no difference in survival. As expected the radiation group had increased gastrointestinal complications.
Patients with deeply invasive cancers, poorly differentiated tumors, extensive lymphovascular space involvement or nonendometrioid subtypes are at high risk of pelvic and systemic recurrence. Pelvic radiotherapy is often recommended in this group of patients, although they have a lower survival rate overall. Vaginal vault brachytherapy or careful observation with institution of appropriate salvage radiotherapy at recurrence may also be appropriate.
In unstaged patients
Recently a European consortium published a large trial of postoperative radiation therapy in endometrial cancer. In this contemporary trial patients with intermediate-risk endometrial cancer (grade 1 or 2 and >50% invasion or grade 2 or 3 and <50% invasion) were randomly allocated to no further therapy or pelvic irradiation therapy. No attempt at surgical staging was made in this study. The results confirm, not surprisingly, the findings of the Aalders study. Patients treated with pelvic irradiation had a significant reduction in local recurrence (p < 0.0001) but no survival advantage was obtained from such therapy [39–43].
Extended-field radiation
Potish and colleagues report a 5-year survival of 47 and 43% respectively for patients with surgically confirmed para-aortic spread alone or pelvic and para-aortic spread, with a complication rate of only 2% [44]. The addition of para-aortic irradiation has been reported as improving median survival, ranging from 27–43 months, and is of particular benefit for those patients with microscopic nodal disease [45].
Whole-abdominal radiation
WAR was studied by the GOG for adjuvant treatment of patients with Stage III and IV endometrial cancer. Of the 174 patients evaluable for radiation toxicity symptoms, the frequency of severe or life-threatening adverse effects were as follows: 12.6% had bone marrow depression, 15% had gastrointestinal symptoms and 2.2% had hepatic toxicity. The survival rates were 31% for endometrioid histology and 35% for serous/clear-cell histologies. No patient with gross residual disease survived. The authors conclude that one should consider WAR only in maximally and completely resected advanced endometrial carcinoma. [46]. In comparing WAR (3000 cGy) with chemotherapy (doxorubicin and cisplatin) for advanced Stage III and IV disease, the GOG found a 13% difference in progression-free survival and an 11% difference in survival at 2 years in favor of chemotherapy. Despite this, 55% of patients recurred both within the pelvis and abdomen [47].
Vaginal brachytherapy
Vaginal brachytherapy has a very low morbidity and is used to reduce the risk of vaginal vault recurrence in high-risk, surgically staged, node-negative patients. While cohort studies have suggested that vaginal brachytherapy does not improve overall survival when compared with careful observation and salvage radiotherapy [48,49], it does increase disease-free survival and may improve quality of life.
Future developments in radiation therapy
Advances in radiation therapy delivery include optimized 3D conformal planning and intensity-modulated radiation therapy (IMRT). With optimized 3D conformal planning, computerized tomography planning is performed in three planes. The gross tumor volume is delineated and then a more generous volume – that is, the clinical target volume – is delineated. The advantage of IMRT is that it allows avoidance of normal tissue close to the target volume delineated. Rather than having a single large radiation beam pass through the body, IMRT allows beams to be broken up into thousands of tiny, pencil-thin radiation beams, each with a different intensity, that enter the body from many more angles. The combined effect is to produce a high-dose volume with a sharp dose gradient at its boundaries that can be designed into complex 3D shapes. Intensity-modulated whole-pelvic irradiation therapy provides excellent coverage of the target structures while sparing critical neighboring structures in gynecological patients. Treatment appears to be well tolerated with less acute gastrointestinal toxicity than conventional whole-pelvic irradiation therapy. However, as the procedure is new, more patients and longer follow-up are needed to evaluate the full merits of this approach [50–61].
Chemotherapy
Adjuvant chemotherapy has been studied in only one large, randomized trial. Patients were surgically staged and received postoperative radiotherapy if nodes were positive. Patients were then randomized to adjuvant doxorubicin or no further therapy. Long-term results have not shown a survival advantage in the patients allocated to the doxorubicin arm and thus, at present, there is no role for systemic therapy in patients with Stage I or II disease. When compared with pelvic irradiation therapy in early stage, intermediate-risk endometrial cancer, a prospective Japanese study did not show superiority in terms of survival when combination chemotherapy (cisplatin, cyclophosphamide and doxorubicin) was compared with pelvic irradiation therapy [62].
However, it is interesting to note the results of the recently published prospective study of high-risk, Stage Ic, grade 3 endometrial cancer patients treated with postoperative radiation therapy. Such patients are at a high risk of early distant spread and endometrial cancer-related death and such a high-risk group might benefit from systematic therapy. In this study grade 3 histology was the most important adverse prognostic factor with hazard ratios for relapse and endometrial cancer-related death of 5.4 and 5.5, respectively [42]. The recently published Phase II Radiation Therapy Oncology Group study noted acceptable toxicity in high-risk, node-negative patients treated with concurrent chemo–irradiation [63]. Further prospective trials are needed to assess if this treatment reduces the incidence of distant metastases in this group of patients.
Systemic chemotherapy is typically reserved for women with disseminated primary disease or extrapelvic recurrence. Although the combination of cisplatin plus doxorubicin is commonly used, carboplatin plus paclitaxel represents an efficacious, low-toxicity regimen for managing advanced or recurrent endometrial cancer [64,65].
Recently, the GOG has demonstrated that the addition of paclitaxel to a combination of doxorubicin and cisplatin improves not only the progression-free survival but also the overall survival compared with doxorubicin/cisplatin alone in advanced and recurrent disease [66,67]. In addition, when compared with single-agent doxorubicin, the combination of doxorubicin and cisplatin, while more toxic, has a superior response rate and progression-free survival but similar overall survival [68].
For Type II endometrial cancer, where the spread pattern is not dissimilar to ovarian cancer, adjuvant and neoadjuvant chemotherapy has been recommended [69].
Hormonal therapy
Type I endometrial cancer expresses both estrogen and progestogen receptors, and high levels of the latter correlate inversely with tumor grade, stage and prognosis. In high-risk Stage I patients, the addition of medroxyprogesterone acetate (Provera®) 200 mg twice daily, while reducing relapses, did not affect overall survival in 1012 patients studied by the Clinical Oncology Society of Australia. Interestingly, steroid receptor status had no influence on their outcome [70]. Similarly, in Stage I and II disease a large prospective European study has confirmed these findings [71]. This has been widely prescribed in the past, but a meta-analysis of six randomized trials involving a total of 3339 women has shown no survival benefit for adjuvant progestogen therapy in endometrial cancer. A subsequently published randomized trial of 1012 women also failed to demonstrate any survival benefit [72]. Based on the available evidence, there is no role for adjuvant hormonal treatment in early stage endometrial cancer.
In patients with advanced or recurrent disease, high-dose progestin therapy is well tolerated and may produce some dramatic responses, and is a logical choice for treatment prior to the initiation of cytotoxic therapy. Response rates of 15–20% have been reported. However, progestin will downregulate its own receptor, perhaps limiting its long-term usefulness. A benefit may be seen by alternating with tamoxifen, which helps to replenish the progestin receptor [73–75]. The ideal dose would appear to be 200 mg daily medroxyprogesterone acetate, which appears to be as effective as higher doses with reduced side effects [75].
Endometrial cancer cells have been shown to express gonadotropin-releasing hormone receptors and small studies have been reported showing antiproliferative activity in endometrial cancer, but at present there are no substantial data to support their routine use [76].
As postmenopausal estrogen is derived from the conversion of adrenal androgens in peripheral tissue by the enzyme aromatase, there is a rationale for the use of aromatase inhibitors in the management of endometrial cancer. In addition, large amounts of aromatase have also been found in the stroma of endometrial cancer. Aromatase inhibitors have had limited testing in endometrial cancer. The GOG noted only two responses (9%) in a trial in 23 patients [77].
An algorithm outlining the management of surgically staged endometrial cancer is presented in Figure 2.
Management of surgically staged endometrial cancer.
Surveillance after treatment
The rationale for routine surveillance after treatment is to detect early treatable and salvageable recurrences. In addition, regular surveillance will help identify and manage side effects from either treatment or the disease, provide emotional support to the patient and family that things are going well and provide information for database research audits. The timing and pattern of recurrence is dependent upon on many factors including the stage at presentation, the treatment received, the histological cell type, the presence of lymph–vascular space invasion, the depth of myometrial invasion and the presence of lymph node metastases (stage). The median time for recurrence is 14 months for vaginal recurrence and 19 months for distant disease, with 34% of recurrences diagnosed in the first 12 months after treatment, 76% within 3 years and 10% not recurring for more than 5 years after treatment. Approximately a third of patients will have disease detected while asymptomatic [78].
While post-treatment follow-up guidelines vary between institutions and countries, the general theme is to follow patients at high risk of recurrence closely every 3–4 months for the first year or two, then extend the surveillance interval to 6-monthly to complete 5 years of follow up.
While routine surveillance has not been shown to be either cost effective, or to significantly prolong the overall survival of all patients, it does provide a significant psychological benefit to patients in confirming an absence of clinical disease. A strong argument can also be made for intense surveillance in previously nonirradiated patients, to detect vaginal recurrence at the earliest opportunity, given the high salvage rate of such recurrences following radiotherapy [79,80].
Future perspective
The incidence of uterine cancer is likely to increase due to the increasing obesity of our population. Surgical management is likely to be increasingly minimally invasive, with laparoscopic lymphadenectomy and total laparoscopic hysterectomy becoming increasingly popular. There are, however, no prospective data to support this management.
There may be an increasing role for adjuvant chemotherapy in high-risk, deeply invasive, high-grade tumors.
Intensity-modulated radiation therapy is likely to be increasingly utilized to reduce unnecessary side effects.
Executive summary
Postmenopausal bleeding is the commonest presenting symptom of endometrial cancer.
A transvaginal ultrasound and hysteroscopy with a dilation and curettage is essential in detecting endometrial cancer.
At diagnosis, 75% of patients with endometrial cancer will have disease confined to the uterus.
Surgery remains the cornerstone of management of patients with endometrial cancer.
Surgical staging includes a hysterectomy, bilateral salphingo-oophorectomy, and pelvic and para-aortic lymphadenectomy.
While primary radiotherapy results in poorer overall survival rates when compared with surgery, this may be appropriate therapy in patients who are medically unfit for surgical treatment.
Postoperative pelvic irradiation therapy will reduce the pelvic recurrence rate but will not have an effect on overall survival or distant recurrence rates.
Adjuvant postoperative external-beam pelvic irradiation therapy should be restricted to those patients with confirmed pelvic nodal spread.
Close surveillance post-treatment, particularly those patients who have not received radiation therapy, may diagnose recurrence early, while it is still likely to be curable.