The most prevalent uterine tumours are leiomyomas, which are benign and have a prevalence of about 50% at menopause. The incidence of endometrial cancer and uterine sarcomas is about 25 per 100,000 and 0.7 per 100,000, respectively. Reported risk factors for endometrial cancer are advanced age, unopposed oestrogen stimulation, late menopause, obesity, diabetes mellitus, nulliparity, feminising ovarian tumours, polycystic ovarian syndrome, tamoxifen and belonging to a hereditary non-polyposis colorectal cancer family. Unopposed oestrogen stimulation and tamoxifen have also been confirmed to induce uterine sarcomas. Cervical cytology, endometrial sampling and ultrasound have been proposed in the early diagnosis of endometrial cancer. No pathognomonic ultrasound, magnetic resonance imaging or computed tomography features are able to differentiate between a leiomyoma and a uterine sarcoma, and reliable serum markers for sarcomas are lacking. To date, mass screening for uterine malignancies is not feasible or effective.
Introduction
According to World Health Organization definitions, ‘mass screening’ should be differentiated from ‘selective screening’. Mass screening is a term used to indicate large-scale screening of whole population groups, whereas selective screening is used for screening of selected high-risk groups in the population. To qualify for mass screening, a disease should have a high incidence, high mortality, mortality rate, or both, the disease should be preceded by a treatable precursor, and a screening test available that is accurate, patient friendly and affordable. The beneficial effects of cervical cancer screening is well established and widely applied. In this chapter, we evaluate whether tumours of the uterine corpus (i.e. endometrial cancer or uterine sarcomas) fulfil the screening criteria.
We first discuss risk factors for uterine malignancies, and subsequently focus on endometrial cancer and uterine sarcomas.
Risk factors for uterine tumours
Knowledge of risk factors of a disease is important for its diagnosis and prevention. Reported risk factors for endometrial cancer are advanced age, postmenopausal status, arterial hypertension, obesity, diabetes, long-term oestrogen exposure (caused by anovulation, ovarian granulosa-cell tumour or after unopposed oestrogen replacement therapy), and tamoxifen therapy.
Unopposed oestrogen stimulation
Oestrogen exposure, without the counterbalancing effects of progesterone, is referred to as unopposed oestrogen. Unopposed oestrogen stimulation increases the risk for endometrial cancer two- to three-fold if used for more than 2 years. Grady et al., in a meta-analysis of 37 observational studies, reported that women receiving progestins in conjunction with oestrogen have no increased risk for endometrial cancer; however, if the oestrogen is given without progestins, the relative risk for developing endometrial cancer compared with non-users is 2.3. The risk persists for several years after the discontinuation of oestrogen therapy. Combined oral contraceptives provide protection against endometrial cancer. This may be especially important among high-risk groups, such as women with ovulatory dysfunction, obesity and late menopause.
Late menopause
Throughout the perimenopausal period, the frequency of anovulatory cycles increases. This results in increased exposure to unopposed oestrogen. Women who experience menopause at a later age are more likely to be exposed to unopposed estrogen for a longer time, and are therefore at higher risk for uterine tumours.
Obesity
Obesity is associated with an increased risk of abnormal uterine bleeding, ovulatory dysfunction and endometrial cancer. In premenopausal women, obesity has little effect on the serum concentration of oestradiol, but causes an increase in the frequency of anovulatory menstrual cycles, leading to longer episodes of unopposed oestrogen. After menopause, obesity causes an increase in the serum concentration of bioavailable oestrogens through aromatisation of androstenedione to oestrone in adipose tissue.
Diabetes mellitus
Diabetes has been associated with an increased risk of endometrial cancer in some epidemiological studies. This increased risk is thought to be mainly due to overweight. Anderson et al. found a moderately strong positive age-adjusted association between diabetes and endometrial cancer. They concluded that diabetes does not increase the risk of endometrial cancer in women with a body mass index (BMI) of 27.4 kg/m 2 or less, and may confer a modest additional risk to women with a BMI of greater than 27.4 kg/m 2 . If diabetes contributes to an increased risk of endometrial cancer, independent of obesity, there are two possible mechanisms. The first is that insulin acts to increase unopposed oestrogen levels. In postmenopausal women, the ovaries no longer produce significant amounts of oestrogen, and most oestrogen is produced through aromatisation of androstenedione to oestrone in adipose tissue. The adrenal glands are the major source of androgens in women, but the ovaries also contribute to the pool of androgens. Insulin stimulates ovarian testosterone release, and thus hyperinsulinaemia associated with diabetes may increase oestrogen levels. A second proposed mechanism is that insulin decreases hepatic insulin-like growth factor binding protein-3 and consequently increases circulating insulin-like growth factor-I (IGF-I). IGF-I receptors exist in the endometrium, and IGF-I stimulates cell proliferation in vitro . Insulin is also a weak analogue of IGF-I in the endometrium. In addition, hepatic insulin-like growth factor binding protein-3 may have a regulatory role in cell-growth control and cancer, apart from its effect on IGF-I.
Nulliparity
Albrektsen et al. showed an effect of parity in women with endometrial cancer: parous women had a significantly better prognosis than nulliparous women. During pregnancy, the hormonal balance shifts towards more progesterone and less oestrogen. It is also possible that pregnancy-related factors influence the biology of endometrial epithelial cells, perhaps in connection with the rebuilding of the endometrium after birth, and that tumours that develop in parous women are less aggressive than those that develop in nulliparous women.
Feminising ovarian tumours
Feminising ovarian tumours may cause endometrial cancer by abnormal, unopposed oestrogen stimulation. Pautier et al. described 45 cases of adult granulosa-cell tumours. The median age was 46.5 years, and 18 women were postmenopausal. Endometrial samples were available for study in 25 women: endometrial carcinoma was diagnosed in two cases (8%) and endometrial hyperplasia was diagnosed in 13 cases (52%).
Polycystic ovary syndrome
Polycystic ovary syndrome is a heterogeneous condition associated with irregular menstrual cycles, anovulation and androgen excess. Most women with Polycystic ovary syndrome, independent of body weight, have insulin resistance and hyperinsulinaemia. As described above, obesity, unopposed oestrogen stimulation due to anovulation, insulin resistance and hyperinsulinaemia are risk factors for endometrial cancer.
Tamoxifen
Tamoxifen is a selective oestrogen-receptor modulator used in the treatment of oestrogen-receptor positive breast cancer. Tamoxifen has an anti-oestrogen activity on breast tissue by competitive inhibition of oestrogen binding to its receptor. In various other tissues, including the endometrium, tamoxifen also has modest agonistic effects. Tamoxifen increases the risk of endometrial carcinoma two- to three-fold in postmenopausal women.
Hereditary non-polyposis colorectal cancer (Lynch syndrome) families
Women belonging to a hereditary non-polyposis colorectal cancer family have an increased risk of developing endometrial cancer. Lécuru et al. showed the high sensitivity and negative predictive value of transvaginal ultrasound in women from hereditary non-polyposis colorectal cancer families. They proposed a maximum endometrial thickness of less than 4 mm (in the absence of polyps or intrauterine abnormalities) in postmenopausal women not receiving hormonal therapy or less than 6 mm in other women. These guidelines, however, await prospective validation. In their systematic review, Aurenen and Joutsiniemi reported that the detection of endometrial cancer improved if routine endometrial sampling was added to transvaginal ultrasound.
Uterine sarcomas
Epidemiological studies suggest that unopposed oestrogen stimulation increases the risk of uterine sarcomas. Exogenous- and endogenous oestrogen stimulation, as well as tamoxifen intake, have been reported to be associated with uterine sarcomas.
It remains to be determined if uterine sarcomas are more common among African women because of increased hormone sensitivity.
Endometrial cancer
Endometrial cancer has an incidence of about 25 per 100,000 women. The overall 5-year survival is 80% and stage related: 85% for stage I disease, 75% for stage II, 45% for stage III and 25% for stage IV. A comparison of overall survival with stage-related survival shows that most cases are diagnosed at early stage. This is because endometrial cancer often presents with abnormal uterine bleeding. The prognosis also depends on the type of tumour. Oestrogen-related endometrioid (type 1) endometrial cancer has a much better prognosis than non-oestrogen-related (type 2) cancer (including serous- and clear cell carcinoma). Type 1 endometrial cancer usually arises on a background of endometrial hyperplasia, and is preceded by atypical hyperplasia, also called endometrial intraepithelial neoplasia. Earlier work by Kurman et al. showed that the risk of developing cancer in women diagnosed with endometrial hyperplasia is 1%, 3%, 8% and 28%, respectively, in cases of simple hyperplasia without atypia, complex hyperplasia without atypia, simple hyperplasia with atypia and complex hyperplasia with atypia, respectively. Type 2 endometrial cancer often occurs against a non-hyperplasic or atrophic endometrial background without identifiable precursor stage, although a serous endometrial intraepithelial carcinoma has been reported as a precursor for serous endometrial cancer. The likelihood of endometrial cancer can be assessed by ultrasound. Endometrial cancer is associated with a thickened endometrium as measured at transvaginal ultrasound. The histological diagnosis of endometrial cancer is usually made by office endometrial sampling. Unfortunately, unlike type 1 disease, an early type 2 endometrial cancer often causes only a subtle and focal thickening of the endometrium on a thin atrophic background, and is more prone to be missed both by ultrasound examination and office sampling.
On the basis of these characteristics (low incidence, good prognosis, absence of a reliable, easy detectable precursor for type 2 endometrial cancer), mass screening for endometrial cancer is not advocated.
Some tests (i.e. pelvic ultrasound and cervical cytology), however, are frequently carried out for indications unrelated to the uterus (e.g. uro-gynaecological evaluation and evaluation of the ovaries). These tests may hint to an unexpected uterine lesion.
Endometrial cancer
Endometrial cancer has an incidence of about 25 per 100,000 women. The overall 5-year survival is 80% and stage related: 85% for stage I disease, 75% for stage II, 45% for stage III and 25% for stage IV. A comparison of overall survival with stage-related survival shows that most cases are diagnosed at early stage. This is because endometrial cancer often presents with abnormal uterine bleeding. The prognosis also depends on the type of tumour. Oestrogen-related endometrioid (type 1) endometrial cancer has a much better prognosis than non-oestrogen-related (type 2) cancer (including serous- and clear cell carcinoma). Type 1 endometrial cancer usually arises on a background of endometrial hyperplasia, and is preceded by atypical hyperplasia, also called endometrial intraepithelial neoplasia. Earlier work by Kurman et al. showed that the risk of developing cancer in women diagnosed with endometrial hyperplasia is 1%, 3%, 8% and 28%, respectively, in cases of simple hyperplasia without atypia, complex hyperplasia without atypia, simple hyperplasia with atypia and complex hyperplasia with atypia, respectively. Type 2 endometrial cancer often occurs against a non-hyperplasic or atrophic endometrial background without identifiable precursor stage, although a serous endometrial intraepithelial carcinoma has been reported as a precursor for serous endometrial cancer. The likelihood of endometrial cancer can be assessed by ultrasound. Endometrial cancer is associated with a thickened endometrium as measured at transvaginal ultrasound. The histological diagnosis of endometrial cancer is usually made by office endometrial sampling. Unfortunately, unlike type 1 disease, an early type 2 endometrial cancer often causes only a subtle and focal thickening of the endometrium on a thin atrophic background, and is more prone to be missed both by ultrasound examination and office sampling.
On the basis of these characteristics (low incidence, good prognosis, absence of a reliable, easy detectable precursor for type 2 endometrial cancer), mass screening for endometrial cancer is not advocated.
Some tests (i.e. pelvic ultrasound and cervical cytology), however, are frequently carried out for indications unrelated to the uterus (e.g. uro-gynaecological evaluation and evaluation of the ovaries). These tests may hint to an unexpected uterine lesion.
Cervical cytology
Cervical cytology to rule out cervical disease is essential in the evaluation of abnormal uterine bleeding. The value of cervical cytology in the diagnosis of endometrial disease, however, is limited. In a study of 128 consecutive women with postmenopausal bleeding, the presence of endometrial cells on cervical smear had a positive predictive value for endometrial cancer of only 17%. The presence of atypical endometrial cells was associated with endometrial malignancy in half of the cases. On the other hand, in women with even advanced disease, cervical cytology smears alone will detect endometrial cancer in only 25–55% of women.
Cervical cytology may hint to possible endometrial disease, but the absence of endometrial cells on cervical cytology does not rule out endometrial cancer.
Endometrial sampling
Office endometrial sampling is a rapid, safe and convenient way of obtaining an endometrial specimen for histological diagnosis of intrauterine pathology, and is well accepted by most women. As in traditional dilatation and curettage, it remains a blind procedure, and concern remains about non-representative samples. In a study of 148 consecutive women with postmenopausal bleeding, Pipelle ® sampling showed a sensitivity of 100% for endometrial cancer, but only 44.6% for both benign and malignant endometrial disease. This is mainly because of the failure to detect focal intracavitary lesions, such as endometrial polyps and intracavitary myomas. In a systematic review of the accuracy of outpatient endometrial biopsy in the diagnosis of endometrial cancer, Clark et al. found that a negative test result of office endometrial sampling was less accurate than a positive test result. For the Pipelle ® device, the positive likelihood ratio was 64.6 (22.3–187.1) and the negative likelihood ratio was 0.1 (0.04–0.28). Endometrial cancers may infrequently be missed by office sampling (e.g. small lesions or a malignant focus in an endometrial polyp, or in case of a massive haematometrium).
Devices other than the Pipelle ® sampler have been used to biopsy the endometrium or to sample endometrial cells. The Endopap ® device that collects endometrial cells for cytological analysis was compared with the Pipelle ® sampler in 106 consecutive postmenopausal women with abnormal uterine bleeding. Endopap ® showed a lower specificity for endometrial disease, and one cancer case was missed. These data tend to favour Pipelle ® against Endopap ® as a diagnostic tool for endometrial disease.
Dijkhuizen et al. carried out a meta-analysis of the diagnostic accuracy of different sampling devices. In their analysis, the detection rate for endometrial carcinoma of Pipelle ® was 99.6% and 91% in postmenopausal and premenopausal women, respectively. They concluded that endometrial biopsy with the Pipelle ® is superior to other techniques in the detection of endometrial carcinoma and atypical hyperplasia.
Ultrasound
Although ultrasound is not used for endometrial cancer screening in asymptomatic women, ultrasound is frequently used for indications unrelated to the endometrial status (e.g. in the evaluation of pelvic masses and pelvic pain or in uro-gynaecology). It is, therefore, important to be able to appropriately interpret incidental ultrasound findings, such as a thickened endometrium or the presence of a polyp. Most of the clinical research involved postmenopausal women with abnormal bleeding who did not take hormone replacement therapy. On the basis of these data, endometrial cancer is most unlikely in women with a thin endometrium as measured on vaginal ultrasound; the proposed cut-off values for total endometrial thickness in women after menopause vary from 3 to 5 mm. Smith-Bindman et al. found a risk of cancer of 7.3% if the total endometrial thickness exceeded 5 mm compared with 0.07% if the endometrial thickness was 5 mm or less. More recently, Timmermans et al. proposed a cut-off value of 3 mm. On the basis of a theoretical cohort, Smith-Bindman et al. calculated that, to reach similar cancer risk figures in asymptomatic women than in women with postmenopausal bleeding, a cut-off value for endometrial thickness of 11 mm should be used (cancer risk of 6.7% if endometrial thickness exceeded 11 mm compared with less than 0.002% for an endometrial thickness of 11 mm or less). These results are, however, based on a number of theoretical suppositions, and have never been validated in a prospective clinical trial. The situation might be different for type 2 cancers as it seems that a thin or indistinct endometrial stripe, especially when associated with other ultrasound abnormalities, does not reliably exclude type 2 endometrial cancer.
After menopause, the endometrial thickness is influenced by hormonal therapy. In a study of 238 women, the mean endometrial thickness was 3.5 mm (standard deviation 1.6) in continuous oestrogen–progestogen therapy compared with 4.1 mm (standard deviation 1.9) in women taking tibolone and 5.5 mm (standard deviation 2.5) in the group using a sequential hormone replacement therapy (HRT) scheme. The endometrial lining in women taking sequential HRT was estimated to be 2 mm thicker than in women taking continuous oestrogen–progestogen or tibolone ( P = 0.0001). Use of the same cut-off values as in women not taking HRT is therefore expected to be associated with a lower diagnostic specificity, especially in sequential HRT.
A two- to three-fold increase in the incidence of endometrial carcinoma in tamoxifen users has also been reported. Moreover, tamoxifen induces stromal changes in the endometrium: it thickens with a ‘Swiss-cheese like pattern’ caused by multiple intra-endometrial cysts of varying size. At microscopy, cystic glandular dilatation and stromal oedema is observed. The endometrial thickening, as measured by ultrasound, relates to the duration of tamoxifen therapy and is seen as early as after 6 months of treatment. Once use of tamoxifen ceases, the endometrium tends to return to normal. Subendometrial cyst formation is benign, but has implications for the ultrasound surveillance of the endometrium precluding any reliable measurement of the endometrium. Because of its low specificity, the current consensus is that women taking tamoxifen should not be screened by ultrasound unless they present with uterine bleeding. In case of tamoxifen-induced (sub)endometrial changes, the diagnosis of intracavitary lesions can be made using fluid instillation sonography or office diagnostic hysteroscopy to distinguish between an empty cavity with a cystic appearance of the subepithelial layer of the endometrium and true lesions inside the uterine cavity. Berlière et al. showed that the risk of atypical lesions during tamoxifen therapy was significantly higher in women who had endometrial lesions before starting tamoxifen treatment. They advocate endometrial assessment before starting tamoxifen, and careful follow up in all high-risk women with initial lesions.
Polyps are often seen at ultrasound after menopause: the reported incidence varies between 13 and 17%. Most polyps are benign. Ferrazzi et al. reported that a risk of malignancy or atypical hyperplasia within a polyp is 1% and 2%, respectively, in women presenting with postmenopausal bleeding, and 0.1% and 1.2% in asymptomatic women.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
