62 Sean Kehoe1,2 1 University of Birmingham, Birmingham, UK 2 St Peters College, Oxford, UK In the UK there were over 9000 cases of endometrial cancer in 2015, which means that it is now the commonest gynaecological malignancy (Fig. 62.1). Some plausible explanations include the overall increased life expectancy, obesity and also the reduction in death rates from other related malignancies, in particular breast cancer. Combined with this is the 50% reduction in the rates of hysterectomy since the early 1990s [1]. Whilst the disease affects mainly postmenopausal women, approximately 20% of cases occur in premenopausal women. The staging of endometrial cancer (using the FIGO classification) has become well embedded in practice since its introduction in 2009 [2]. Recent and ongoing randomized clinical trials continue to try to identify the optimum therapeutic approaches to care, which though primarily surgery still require an evidence base as emphasized by the international variations in care [3]. There has also been progress in understanding the genetics of endometrial cancer, and in recent years an acceptance that some tumour types, in particular serous histologies, should be considered – in surgical management terms at least – somewhat similar to serous ovarian cancers. This chapter provides some updates on the management of this disease. Fig. 62.1 Rates of endometrial cancer in the UK. Source: Cancer Research UK. The recognized risk factors for endometrial cancer are shown in Table 62.1. Besides these, there are other environmental factors that influence the disease, as suggested by the variation in the disease in different countries (Fig. 62.2). The main risk is excessive exposure of the endometrium to oestrogen, which has a direct association with both obesity and diabetes. Other factors such as older age and hypertension are inter‐related with the factors above. Age remains the main risk factor, with the highest incidence recorded in women over the age of 50 years. Table 62.1 Risk factors for endometrial cancer. Fig. 62.2 Age‐standardized (European) incidence and mortality rates, body of uterus cancer, EU‐27, 2008 estimates. Only one condition has been specifically identified as having a hereditary relationship with the development of endometrial carcinoma. Known as hereditary non‐polyposis colon carcinoma or Lynch type II syndrome, this is an autosomal dominant condition; the main associated primary malignancy is colonic carcinoma, normally developing in women under 40 years of age. The Amsterdam criteria are used to define those families who conform to the definition of this condition, with the final diagnosis genetically determined and including a variety of mutations, of which MSH2 and MLH1 are the commonest [4]. In such families the lifetime risk of developing endometrial carcinoma is about 40%, and also relevant is the fact that the lifetime risk of developing ovarian cancer is 12%, nearly 10 times the population risk. At present there are no strategies except prophylactic surgery to prevent these women developing endometrial cancer. Reports on a small series of women with this condition undergoing annual scans and endometrial sampling concluded that such an approach seemed to detect earlier cancers, suggesting that screening may be beneficial. However, the paucity of cases makes this very much an assumption, though an acceptable approach in managing those women desiring to retain reproductive function. A possible alternative is the potential of the Mirena® intrauterine system in preventing the disease in such women [5,6]. A clinical trial to evaluate this failed to recruit and was abandoned. Although this is intuitively a reasonable approach, women agreeing to this therapy need to be informed of the limited evidence. When the woman’s family is completed, recommending a hysterectomy is a reasonable course of action, removing any need for ongoing unproven screening systems and likely alleviating the psychological aspects associated with the potential to develop this disease. The increased risk of ovarian cancer may also justify removal of the ovaries, though each case must be managed in accordance with the woman’s needs and wishes. Nulliparity is associated with a risk of endometrial cancer. However, this may only be one of many factors involved, with some relationship to the individual’s hormonal profile during life. Nulliparous women have significantly increased episodes of endometrial shedding events during their menstrual lives in comparison with parous women. In ovarian cancer, there is a direct association between the lifetime number of ovulatory cycles and the risk of ovarian cancer (the higher the number of ovulatory cycles, the greater the risk), and this could be also applied to the risk of endometrial cancer. The follicular phase of the cycle exhibits increasing proliferation (which increases the probability of abnormal cellular development) and strategies which reduce this may protect against abnormal mitosis. While this remains to be proven in endometrial cancer, it is a reasonable hypothesis. Of course, the use of the combined oral contraceptive, even with menstruation, affords some long‐term protection (risk reduction of 50%) giving some support to the theory. Obesity is thought to be related to the development of about 30% of cancers in humans. In obese women, excess oestrogen is produced by the conversion of androgens in the fat, and thus the endometrial tissue has increased exposure to estrone. This is particularly pertinent considering the increased incidence of obesity in the Western world, and the concomitant increased rates of endometrial cancer. Hyperplasia is defined as excessive proliferation of normal cells, and atypical hyperplasia is associated with a high risk of endometrial cancer. In a number of series, the risk of underlying malignancy has been shown to be higher than expected. A Gynecologic Oncology Group study in the USA of 348 women diagnosed with atypical hyperplasia on sampling were eventually found to have a frank malignancy in 46.8% of cases [7]. Similar data from the UK showed a risk of 45.9% [8]. Thus it may be preferable to consider these patients as having endometrial cancer and expedite surgery. Postmenopausal bleeding is the commonest presentation in endometrial cancer, and up to 10% of these women will have a diagnosis confirmed. As such, their access to diagnostics must be deemed urgent. The final diagnosis is confirmed by histology. Based on the UK National Institute for Heath and Care Excellence (NICE) guidelines, all women with postmenopausal bleeding should have a transvaginal ultrasound and those with an endometrial thickness greater than 5 mm should undergo sampling of the endometrium. However, many use a cut‐off of 4 mm, based on cost‐efficacy and detection rates. Sampling can be undertaken in three ways: Pipelle, outpatient hysteroscopy or hysteroscopy and curettage under a general anaesthetic. All methods of sampling will miss some cancers but the rate of failure is not significantly different between outpatient sampling and hysteroscopy. Naturally, avoiding a general anaesthetic is preferable where possible and only patients in whom an endometrial sample cannot be obtained in outpatients should hysteroscopy under a general anaesthestic be considered, accepting the inevitable exceptions. Abnormal bleeding in the premenopausal women, particularly intermenstrual bleeding in women over the age of 45, should prompt investigations including ultrasound and endometrial sampling. Whilst the overall incidence of malignancy in this group is small, nevertheless about 20–25% of all endometrial cancers occur in the premenopausal/perimenopausal age group. Endometrial cancer is categorized into type I and type II disease. The defining factors associated with these are shown in Table 62.2. Table 62.2 Endometrial cancer types. The main histological subtypes are shown in Table 62.3 and some of these may well be related to certain agents. In particular, the use of adjuvant tamoxifen in breast cancer has been proposed to be associated with some of the rarer tumours such as mixed Müllerian tumours, now called carcinosarcomas. There are no specific indicators regarding the other subtypes. Leiomyosarcomas are usually an unexpected finding following removal of a fibroid uterus. The mitotic activity of these tumours relates to their metastatic potential, with mitotic counts of less than 5 per high power field associated with a very good outcome, whereas a count above 10 per high power field is associated with a worse prognosis. Other subtypes recognized as more aggressive include papillary serous and clear cell tumours, accounting for 10–15% of all tumour types. In most cases adjuvant therapies (after surgery) would be considered. Table 62.3 Histological subtypes in endometrial cancer. In some rare cases of sarcomas, preoperative diagnosis may be suspected when the endometrial sampling indicates this diagnosis histologically, or when a preoperative chest X‐ray shows evidence of metastatic disease, normally called ‘cannon‐ball’ metastases. The differentiation of the disease is also important, as the grade in association with other factors will influence the recommendation for adjuvant therapy. The FIGO staging for endometrial cancer is shown in Table 62.4. This was redefined in 2009, with positive cytology, previously allocated to stage III disease, now abandoned as part of the staging process. As part of staging, there are agreed preoperative investigations which can be performed. A chest X‐ray is recommended, although a CT scan may replace this. MRI/CT can be of value in some situations to ascertain if any extrauterine disease exists, which may influence the role of surgery. The value of MRI in defining the depth of tumour invasion has been examined with a view to deciding the need to excise lymph nodes but it has proved too inaccurate to be of clinical value. Table 62.4 Carcinoma of the endometrium, FIGO staging 2010. Besides these investigations, cystoscopy, sigmoidoscopy and an examination under anaesthetic are all part of the staging procedure. Notably, endometrial cancer can be staged both clinically and surgically, with surgical staging being most commonly employed. Surgery remains the primary intervention for endometrial cancer. Whilst radiotherapy is an alternative, from retrospective case–control studies it would seem that surgery affords a better survival outcome. It is unlikely that there will ever be a randomized controlled trial comparing primary radiotherapy to surgery. The removal of the uterus and (normally) the ovaries is the recommended basic surgical procedure. This may be performed by open laparotomy or by a laparoscopic approach. Studies have shown the advantages regarding short‐ and long‐term recovery when the laparoscopic approach is taken, which may entail either a laparoscopic‐assisted vaginal hysterectomy or a laparoscopic total hysterectomy [9,10]. Robotic surgery has also become part of the surgical approach, but the consensus is that where possible the minimally invasive approach is preferable. Obviously, it is recommended not to insert any instruments into the uterine cavity during such surgery. Also, clamping/ligating the fallopian tubes at commencement of surgery would seem a reasonable action to prevent any disease dissemination when moving or handling the uterus. In some circumstances, the procedure could be performed vaginally, and indeed this is acceptable as long as the ovaries can be removed. However, a vagina‐only approach will not permit access to the pelvic lymphatics, thus limiting this type of surgery to selected patients. The risk of lymphatic spread in endometrial cancer is influenced by the tumour grade, type and depth of invasion into the uterine wall (Table 62.5). Knowledge of lymphatic disease forms part of the staging process and can influence adjuvant therapy. However, debate still surrounds the value of routine pelvic and para‐aortic lymphadenectomy, and randomized trials are required to resolve this issue. Table 62.5 Lymph node metastases in endometrial cancer. There has only been one prospective randomized trial reported on lymphadenectomy in endometrial cancer. This study, called ASTEC (A Surgical Trial in Endometrial Cancer), randomized over 1400 women with clinically early‐stage disease [11]. The study included two parts: (i) patients randomized to pelvic lymhadenectomy or not, and (ii) patients randomized to adjuvant pelvic radiotherapy or not in high‐risk cases. The use of brachytherapy (outome reported in Blake et al. [12]) was permitted and the decision made locally as to whether this was used. The patient cohort receiving radiotherapy were not necessarily those recruited to the surgical aspects of the study. The conclusions were interesting in that the use of lymphadenectomy did not alter survival rates, and indeed it was suggested that it may have a negative impact on outcome, for reasons yet to be explained. Also, the number of lymph nodes retrieved did not influence outcome. Although the study was confined to patients with disease localized to the uterus (based on clinical examination and imaging in some cases), it does indicate that lymphadenectomy should not be undertaken in this group of patients. There are many non‐randomized reports on the role of lymphadenectomy of pelvic and para‐aortic nodes in endometrial cancer. All these studies naturally suffer from the fact that there are no randomized controlled trials. Some groups suggest that, in particular, para‐aortic nodal excision has a therapeutic effect, based on the fact that those who undergo this therapy have longer survival and a notable reduction in relapse disease affecting the para‐aortic region [13]. Also, such resection either permits the avoidance of adjuvant radiotherapy in some cases, or identifies those where the radiotherapy field can be extended to incorporate the para‐aortic region. The main problem with these debates is the lack of appropriate high‐level evidence that lymphadenectomy has a real therapeutic effect on survival and indeed that extension of the field of radiotherapy influences survival. Without doubt both these procedures increase morbidity. There is an urgent need to address this issue and ensure patients are been managed in a manner whereby the morbidity of the intervention can be justified by the improved patient outcome. It can be anticipated that such trials will be undertaken in the near future. When the disease has obvious macroscopic spread beyond the uterus or indeed the pelvis, then multimodal therapy will be required, be this surgery to alleviate symptoms followed by radiotherapy with or without chemotherapy. The combination of chemoradiotherapy increases morbidity, and in endometrial cancer this poses major challenges, as many patients have other comorbidities whereby such a combination may be deemed inappropriate. Debulking surgery, an approach taken in ovarian cancer for some decades, has been reported in some small series in endometrial cancer, with a suggestion that a smaller residual tumour load correlated with a better survival outcome. This is based primarily on retrospective data or small case–control studies and the evidence for this is very poor. There is no agreement that this should be considered accepted therapy for endometrial cancer at present. However, even in advanced disease, removal of the uterus may yield immediate alleviation of symptoms such as persistent vaginal bleeding, which could justify the intervention. As women are increasingly delaying having children, the issue of fertility‐sparing options has become more common, and this is relevant to many other malignancies. There are limited publications relating to such management and the number of cases is small, making firm conclusions difficult. Longer‐term outcomes are equally lacking and therefore before embarking on this therapeutic option, it is imperative to provide the patient with advice based on current information and the ‘unknown’ risks, particularly as this is a deviation from the normal recommended intervention that could potentially exclude the patient from a recommended curative therapy. In those cases reported, the disease was always well differentiated and on clinical and imaging evaluations confined to the uterus (i.e. FIGO stage 1). Women were then exposed to various progestagenic agents, with careful evaluation of response by curettage at 6 weeks, 3 months and 6 months from commencement of therapy [14]. Evidence of non‐response resulted in immediate surgery and in those with response some pregnancies are reported. A hysterectomy was normally performed following a successful pregnancy. Radiotherapy can be used either as a primary or adjuvant therapy. In primary therapy, this is normally where disease spread renders surgery impossible or inappropriate. It is not considered superior to surgical intervention, with an estimated reduction in 5‐year survival of about 5% when compared with surgery in early‐stage disease, though this is based on retrospective series. The use of radiotherapy in an adjuvant setting continues to be modified. Original studies indicated that the use of brachytherapy with external‐beam pelvic radiotherapy may be beneficial in those with high‐grade disease. This original study from the 1980s by Aalders et al. [15] randomized 540 patients with early‐stage disease to brachytherapy versus brachytherapy and external‐beam therapy after undergoing surgery. The relapse rates in the latter group was reduced, although overall 5‐year survival was equivalent in both groups. Further analysis suggested that patients with grade 3 tumours infiltrating greater than 50% of the myometrium might benefit from the addition of pelvic radiotherapy. Two recent trials, the PORTEC and ASTEC trials, have changed the role of adjuvant radiotherapy [12,16]. In the PORTEC study, 715 women with stage I disease were recruited and randomized to pelvic radiotherapy versus no treatment after undergoing a hysterectomy and bilateral salpingo‐oophorectomy. The 5‐year survival rates were 81 and 85%, respectively. The recurrence rates were lower in the radiotherapy groups (4% vs. 14%), but in those relapsing and then receiving radiotherapy the survival was the same. Analysis showed that radiotherapy was not necessary in women with stage I endometrial cancer who were below 60 years of age with grade 1 or 2 tumours and with more than 50% myometrial invasion. The ASTEC study had two parts where patients were randomized, the second looking at external‐beam therapy to the pelvis. All had undergone surgery, consisting of at least a total hysterectomy and bilateral salpingo‐oophorectomy. The conclusions were that the routine use of external beam with brachytherapy reduced the incidence of recurrent disease, and increased the disease‐free survival, but did not have any positive impact on overall survival. The authors also suggested that there may be a possible survival benefit in those with high‐grade disease. When distant metastatic disease is present, systemic treatments are required. For endometrial cancer, chemotherapeutic agents or hormonal therapies are used. Cisplatin and doxorubicin are the commonest cytotoxics used, with medroxyprogesterons the most used hormonal therapy [17–19]. Many trials reporting systemic therapies are small phase II studies, and the overall response rates range from 7 to 69% depending on the study. As previously stated, the comorbidities within this patient cohort often means that hormonal therapy is the best option due to its ease of administration and lack of adverse side effects. Many smaller series have suggested that the combination of radiotherapy with chemotherapy may improve outcome, by reducing local pelvic recurrences and also extra‐pelvic disease relapse. Such effects have been proven in chemoradiotherapy as used in cervical tumours. PORTEC 3 is an ongoing prospective randomized trial that compares standard radiotherapy with combination treatment, and the results should hopefully give guidance to the best option. The main issues with respect to deciding the best therapy for a patient with relapsed endometrial cancer are (i) prior exposure to non‐surgical interventions, (ii) the site of disease relapse, whether localized or multiple, and (iii) the patient’s physical condition. Thus investigations used are similar to those within the staging system, though where available positron emission tomography (PET) can be useful in selected situations. The commonest site of relapse is the vaginal vault and if the disease is localized and the area radiotherapy‐naive, radiation is the first course of intervention. If the disease is localized but has previously undergone radiotherapy, then surgical excision (partial vaginectomy) can be performed. If there are distant metastases, then systemic therapy is necessary and depending on the patient’s physical condition either chemotherapy or hormonal therapies can be used. The response rates are variable but never very high, and the effect is poorer for disease relapsed within a field of radiotherapy. Exenterative surgery [20], when bladder, vagina and rectum are excised, is only undertaken in very carefully selected patients, and may be occasionally justifiable as a palliative procedure. In the main, many patients have such comorbidities that such surgery is generally deemed unsuitable. Endometrial cancer is a disease increasing in incidence though retains a relatively good prognosis. Primary intervention is mainly surgical, with selected patients having adjuvant therapies. Advances in surgical techniques continue to reduce surgically associated morbidity, though in a population with rising obesity, maintaining morbidity rates is challenging. In some early‐stage disease lymphadenectomy is unnecessary, but in higher‐risk populations trials are required to define the role of lymphadenectomy, both pelvic and para‐aortic. Randomized trials are redefining the role of adjuvant therapies; in particular the role of chemotherapy in high‐risk patients is awaited. Prevention is inevitably the ultimate goal, and can be partially achieved through educational health policies in reducing the incidence of obesity. In the future, screening may detect premalignant or early‐stage disease and thus also improve survival rates. However, the latter still requires further research to establish the optimum modalities to employ. Inevitable, the future will also include a greater understanding of the disease and improved individualized therapy, focused more on the actual disease biology rather than based purely on the disease stage and histological subtype.
Endometrial Cancer
Aetiology
High levels of oestrogen/endometrial hyperplasia
Obesity/hypertension/diabetes
Polycystic ovary syndrome [5]
Nulliparity (never having carried a pregnancy)
Tamoxifen use/breast cancer
Post‐menopausal
Hereditary non‐polyposis colorectal cancer
Genetics
Nulliparity
Obesity
Endometrial hyperplasia
Clinical presentation
Types of endometrial cancer
Type I
Premenopausal and perimenopausal women
History of unopposed oestrogen exposure
Endometrial hyperplasia
Minimally invasive, low‐grade endometrioid type
Good prognosis
Type II
Postmenopausal women
Not associated with increased exposure to oestrogen
High‐grade tumours
Poorer prognosis
Endometrioid adenocarcinoma, 50–60%
Adenosquamous, 6–8%
Serous papillary, 18%
Sarcomas/leiomyosarcomas, 3–5%
Carcinosarcomas, 2–3%
Clear cell, 1–6%
Management
Preoperative investigations
Stage
Description
5‐year survival (%)
IA
Tumour confined to the uterus, no or less than half myometrial invasion
80–90 (I)
IB
Tumour confined to the uterus, greater than half myometrial invasion
II
Cervical stromal invasion, but not beyond uterus
60–70 (II)
IIIA
Tumour invades serosa or adnexa
50–60 (III)
IIIB
Vaginal and/or parametrial involvement
IIIC1
Pelvic node involvement
IIIC2
Para‐aortic involvement
IVA
Tumour invasion bladder and/or bowel mucosa
10–20 (IV)
IVB
Distant metastases including abdominal metastases and/or inguinal lymph nodes
Surgical interventions
Lymphadenectomy in endometrial cancer
Variable
Pelvic/para‐aortic nodal disease (%)
<50% uterine invasion and grade1
0–3
<50% uterine invasion and grade2/3
2–6
>50% uterine invasion and grade 1
15–18
>50% uterine invasion and grade2/3
up to 30
Lymphadenectomy and non‐randomized studies
Debulking in advanced disease
Fertility‐sparing surgery
Radiotherapy
Primary
Adjuvant
Chemotherapy
Relapsed endometrial cancer
Conclusion
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
