UK Uterus (Womb) Cancer incidence statistics

This page presents uterine cancer incidence statistics by age, geographic and ethnic variation, and trends over time.

The ICD code for uterine cancer is ICD9 179+182, and ICD 10 C54+55. A small percentage (6%) of uterine tumours cannot be assigned to a specific part of the uterus and in this section these ‘parts unspecified’ (ICD10 55) uterine cancers have been included with those of the corpus uteri unless stated otherwise.

Uterine cancer incidence by age

Cancer of the corpus uteri (uterus or uterine cancer) is the fourth most common cancer in women in the UK with 7,045 cases diagnosed in 2006 1-4, accounting for around 5% of all female cancers ( Table 1.1).

Incidence of uterine cancer is rising in postmenopausal women but five-year survival rates have improved to more than 75% 5.

Download this table (33.5KB)

Uterine cancer is the most common cancer of the female genital tract.

Figure 1.1 shows the body of uterus and cervix, and outlining endometrium and muscle walls.

Download this chart (6.73MB)

The vast majority (93%) of uterine cancer cases are diagnosed in women aged over 50 years with very few women diagnosed under the age of 35. Incidence rises rapidly to a peak of 78 per 100,000 females in their early seventies. Uterine cancer incidence rates decline after the age of 75 (Figure 1.2)1-4.

Download this chart (36KB)

Geographical and ethnic variation in uterine cancer incidence

Corpus uteri cancer is primarily a cancer of the developed world with incidence rates more than four times higher than in developing countries 6. By contrast, developing countries have much higher rates of cervical cancer and worldwide, there are more than twice as many cases of cervical cancer as corpus uteri cancer diagnosed each year.

In North America, Australasia and many European countries, the incidence of cervical cancer has fallen with the introduction of screening and cancer of the corpus uteri is now the most commonly diagnosed gynaecological cancer 6. Uterine cancer incidence rates are highest in North America; up to ten times higher than in parts of Africa and almost double those in Northern Europe ( Figure 1.36).

Download this chart (28KB)

Uterine cancer incidence rates in the UK are among the lowest in the European Union (EU), while the highest rates are found in Malta, Czech Republic and Luxembourg ( Figure 1.4). Romania, Greece and Portugal have the lowest uterine cancer incidence rates 6.

Download this chart (20KB)

Evidence from the US suggests important ethnic differences; uterus cancer incidence rates are significantly higher in white compared with black women (23.5 vs. 18.0 per 100,000 females) 7. The rate of uterine cancer of unspecified parts is significantly higher in black than white women (1.3 vs. 0.7) 7, which may contribute to poorer survival rates in black American women as these tumours are usually large and diagnosed at a late stage.

Trends in uterine cancer incidence

The age-standardised incidence of corpus uteri cancer in Great Britain remained stable between 1975 and 1993, and increased by 34.0% between 1993 and 2006.

Considered separately, tumours of the corpus followed the overall trend, while rates for uterine part unspecified tumours fell slightly over the period 1974-2006 ( Figure 1.51-3).

Download this chart (40KB)

There are important variations in uterine cancer incidence by age. Incidence rates in the over 55 age group have been rising since the late 1970's. Between 1986-2005 the most significant incident rate increase has been in the 60-79 age group, where rates have increased by over 60%. For those aged under 55 incidence rates have remained largely stable.( Figure 1.61-3).

Download this chart (37.5KB)

The uterine cancer incidence trend for the UK is shown in Figure 1.7.

Download this chart (34.5KB)

Uterine cancer incidence trends in most other European countries are similar 8, but are quite different in the US where age-standardised rates have recently been stable in white women but have increased in black American women by almost 2% a year between 1992-2002. 7

UK Uterine (Womb) Cancer mortality statistics

This page includes uterine cancer mortality statistics by age, geographic variation and trends over time.

Uterine cancer mortality by age

Corpus uteri cancer (uterine or womb cancer) includes tumours of the body of the uterus and uterine tumours, parts unspecified. There were around 1,660 deaths from corpus uteri cancer in the UK in 2007, accounting for only 2.2% of all cancer deaths in UK females 1,2,3. Uterine cancers, parts unspecified made up around 30% of these deaths ( Table 2.11,2,3). It is likely that these tumours are advanced at diagnosis, which may explain why the precise site of their origin is unspecified.

Download this table (41.0KB)

Mortality rates increase lineally with age peaking at around 35 per 100,000 among women aged 85 and over ( Figure 2.14).There are very few deaths under the age of 50.

Download this chart (35.0KB)

Geographic variation in uterine cancer incidence

There were about 50,300 deaths from corpus uteri cancer worldwide in 2002, showing far less variation than with incidence. This reflects the lower incidence rates, but poorer survival, in less developed countries. The highest mortality rates were in Eastern Europe, while the lowest rates are found in Eastern Asia, Central and Western Africa ( Figure 2.2).

Download this chart (43.5KB)

Within the European Union the highest death rates are in Cyprus and the Eastern European countries, while the lowest rates are in Greece and Ireland ( Figure 2.3).

Download this chart (41.5KB)

Trends in uterine cancer incidence

The mortality rate from corpus uteri cancer for UK women fell by 27% between 1971-2007. There was a 33% decline in the rate of deaths coded as body of uterus while the death rate for cancer of the uterus, unspecified parts remained fairly stable over the same period ( Figure 2.41,2,3).

Download this chart (42.5KB)

Declines in the age-standardised mortality rate masks a rising trend in women aged over 85, among whom the mortality rate rose by 60% 1971-1986, then declined until the late 1990s. Despite the decline since 1986, mortality rates in the over 85's remain elevated with an increase of 25% from 1971-2007. The fastest declines in the mortality rate have been in women aged under 70 ( Figure 2.5).

Download this chart (45.0KB)

Uterus (Womb) Cancer survival statistics

This page contains survival statistics for uterine cancer by trends over time, by age and geographical and socio-economic variations.

Ten-year relative predicted survival rates (75%) among women with corpus uteri cancer are second only to those for malignant melanoma of the most commonly diagnosed cancers for women in England and Wales, 2000-2001 1. Five-year survival is around 85% for stage I tumours, compared with 25% for women diagnosed with stage IV 2. Grade also has a bearing on prognosis, with around 81% five-year survival in patients with grade 1 stage Ic tumours compared with 42% for grade 3 tumours of the same stage, although the effect of grade is less marked with tumours diagnosed at stage Ib or Ia 2.

Trends in uterine cancer survival rates

There have been modest improvements in survival over the last 30 years. Five-year relative survival estimates for women diagnosed with cancer of the body of the uterus in 2000-01 are 77% 3, compared with 61% for women diagnosed in 1971-75 4, an improvement of 16% over the period. There have been similar improvements in shorter and longer term survival rates ( Figure 3.1) 1,4,5.

Download this table (17.5KB)

* PLEASE NOTE: Ten-year estimates for patients diagnosed in 1971-75, 1976-80 and 1981-85 and all estimates for 2000-01 are not age-standardised

Uterine cancer survival rates by age

As with many other types of cancer, survival is significantly lower in the oldest age groups. Women diagnosed with corpus uteri cancer aged 80-99 in England between 1998-2001 had estimated five-year relative survival of 52%, compared with 85% for women diagnosed aged 50-59 ( Figure 3.2) 6.

Download this table (20KB)

Geographic and socio-economic variations in uterine cancer survival rates

For women diagnosed in 1996-1999 there is a significant gap in five-year survival between the most deprived and least deprived women of 4.5% 1. Regional differences in survival rates reflect this deprivation gradient. Wales had the lowest relative one and five-year survival rates for women diagnosed in 1986-90, while the South and West region in England had the highest rates 4. One and five-year survival rates for women in England and Wales diagnosed with cancer of the corpus uteri in 1990-94 were significantly below the European average (90% and 76%) 7. Women in the United States diagnosed with cancer of the corpus uteri in 1995-2001 had five-year survival of 85% 8.

In the United States there are significant differences in survival between African American and white women 9. While this could partly be due to differences in treatment between the two groups, African American women are significantly more likely to get high-grade tumours and to present at a later stage 9.

Uterine (Womb) Cancer risk factors

This section contains information on endogenous hormones and reproduction, excess bodyweight, exogenous hormones, diabetes and insulin, physical activity, smoking and other factors.

Endogenous hormones, reproduction and uterine cancer risk

Most of the established risk factors for uterine cancer are the result of excess exposure to oestrogen unopposed by progestagens , a process that stimulates proliferation of the cells of the womb, increasing cancer risk 1. In addition, insulin and insulin-like growth factors may increase the effect of oestrogen on uterine tissue 2. As all oestrogen exposure is unopposed in postmenopausal women, high circulating levels of oestrogens and androgens in these women is known to increase risk. Being overweight increases oestrogen levels in postmenopausal women 3 and can disrupt ovulation and progesterone production in premenopausal women, resulting in continuous exposure of the womb to unopposed oestrogen 4 (see section on bodyweight below). However, women with the highest levels of sex hormone binding globulin (which governs the bioavailability of oestrogens and androgens) have a significantly reduced risk 2,5.

Pregnancy and parity reduce risk of uterine cancer by 30% for a woman’s first birth and by 25% for each successive birth 1, and later maternal age at last birth has also been shown to reduce risk 1. Pregnancy and childbirth are thought to offer reduced risk through the elimination of pre-malignant cells with the “sloughing-off” of cells during delivery, and/or the break in unopposed oestrogen exposure during pregnancy. Research seems to suggest a greater protective effect of parity on post-menopausal than pre-menopausal women 6,7, but results have been inconsistent 8. Early menarche and late menopause significantly increase risk, because they prolong years of oestrogen exposure 1. Other factors associated with increased risk include lack of, infrequent or irregular periods, periods, annovulatory menstrual cycles and more days of menstruation 4,7.

Excess bodyweight and uterine cancer risk

Evidence suggests that risk of uterine cancer is 2-3 times higher in overweight and obese women respectively and it is estimated that around 36% of cases of uterine cancer in the UK are caused by excess bodyweight 9-,13. Prospective studies, mainly among post-menopausal women, report a significant association with BMI at enrolment, but also suggest that women’s BMI in younger adulthood is significant and that the increase in risk begins at a relatively modest raised BMI 8,9. Some evidence suggests that waist circumference and waist-to-hip ratio may be more important than BMI, with significant increases in risk reported in women in the two lowest quartiles of BMI but the highest quartile of waist circumference 10. However, more studies are needed to clarify the effects of central and peripheral obesity and risk of uterine cancer in both premenopausal and postmenopausal women.

Exogenous hormones and uterine cancer risk

Recent results from the UK Million Women study report an increased risk of 50% among current users of oestrogen-only hormone replacement therapy (HRT) and 80% in those using tibolone preparations. A meta-analysis of published studies reported in the same paper found there was a non-significant reduced risk for ever versus never users of continuous combined therapy, and a small but significant increase in risk for cyclic combined 14. The study did not have enough past-users to calculate risk after stopping HRT, but a Swedish case-control study reported an ongoing significant increase in risk in women who had used oestrogen-only HRT but had stopped five or more years previously 15. It is probable that BMI modifies the effect of HRT; the effects of tibolone and oestrogen-only HRT in the Million Women Study were limited to women whose BMI was lower than 25, while there was little if any risk increase in obese women 14. Risk of developing endometrialhyperplasia increases with oestrogen dose; one study reported a 32-fold increase in risk among women taking 1.25mg/d of unopposed oestrogen supplements for two years compared with those taking a 0.3mg/d dose 16.

Users of combined oral contraceptives have a lasting reduction in risk as they have fewer days of unopposed oestrogen exposure each month. The risk reduction is about 10% for each year of use 1 and has been reported in some studies to persist for as long as 20 years after stopping use 17.

Tamoxifen, an oestrogen receptor modulating hormone used to treat and prevent breast cancer, has been shown to treble risk of uterine cancer 18. This relationship doesn’t appear to be dose dependant, although it has been suggested that risk increases with duration of therapy. Excess deaths from uterine cancer among women taking tamoxifen is estimated to be about 2 per 10,000 women each year 18.

Diabetes, insulin and uterine cancer risk

A recent meta-analysis reported that women with diabetes have double the risk of uterine cancer. 32 Although risks associated with diabetes are difficult to separate from those of excess bodyweight, there is evidence of an independent effect 19 and results of a recent serum study suggest that circulating insulin and free insulin-like growth factor 1 may play a role in uterine cancer 20. Other studies show a higher risk of uterine cancer in women with higher blood and dietary glucose 21-,23,33.

Physical activity and uterine cancer risk

A meta-analysis of prospective studies reported that the most active women have a 23% reduction in risk of uterine cancer. 34

Smoking and uterine cancer risk

Smokers have a modest reduction in uterine cancer risk 27. This effect may be linked with the fact that smokers metabolise oestrogens into less active metabolites than non-smokers. Recent results from the US Nurses’ Health study show a significant risk reduction in past as well as continuing smokers 28.

Other uterine cancer risk factors

Ultrasound scanning suggests one-quarter of women of reproductive age have polycystic ovaries 29 but only 10% of women have additional symptoms associated with polycystic ovarian syndrome (PCOS) (also called Stein-Leventhal syndrome). These can include prolonged anovulatory menstrual cycles, lack of, infrequent or irregular periods, infertility, obesity, diabetes and hypertension. The hormonal factors governing these symptoms are complicated but are likely to include hyperandrogenism, an increase in production of luteinising hormone, disruption to progesterone production, and insulin resistance. Most of these symptoms are known to increase risk of uterine cancer. 30 Metformin, the combined contraceptive pill or cyclic progestagens are often prescribed to women with PCOS to help prevent endometrial hyperplasia 30.

Some types of ovarian cancers (granulosa-cell and theca-cell tumours) can secrete hormones such as oestrogens or androgens. Women with such tumours have a 10-fold higher risk of uterine cancer 31.

Uterine (Womb) Cancer molecular biology and genetics

A number of molecular changes are observed in endometrial carcinogenesis but the frequency of these events varies with cancer type. Mutations in the tumour supressor genePTEN are the most commonly reported in type I tumours1,2.

Alterations in this gene are early events in cancer progression and have also been detected in precancerous lesions3. Changes in the oncogeneK-ras have also been observed. Type II carcinomas are characterised by inactivation of the tumour suppressor gene p53 and overexpression of the ErbB oncogene. Aneuploidy is also commonly observed in these tumours.

Microsatellite instability is a feature in around 25% of endometrial cancers, predominantly type I tumours 4. This is due to underlying faults in DNA mismatch repair ( MMR) genes. The exact mechanism by which faults in these genes lead to cancer is unclear, but they may permit an increase in the mutation rate leading to an accumulation of faults in genes important in the cancer pathway.

Endometrial cancer commonly occurs as part of the hereditary non-polyposis colorectal cancer (HNPCC) syndrome. Individuals with HNPCC syndrome have faults in a MMR gene (most commonly MSH2, MLH1 or MSH6) and female family members have a 60% lifetime risk of uterine cancer 5. Aside from HNPCC families there are few reported cases of clustering of the disease. A recent study identified a number of cases without faults in MMR genes 6 and a large Swedish study identified that daughters of women with endometrial cancer but without HNPCC syndrome had almost double the risk of endometrial cancer 7, although a major dominant gene for this disease is unlikely to exist 8. Instead, one or more low penetrance polymorphisms may account for the small number of familial cases observed.

Microarrays are being used to construct gene expression profiles for sporadic cases of both type I and type II cancers. In the future this will help to provide better prognostic information and more tailored treatments for this disease 9.

Back to top ^

Uterine (Womb) Cancer symptoms and treatment

This section contains information on symptoms, staging and treatment, risk of recurrence/prognosis and screening.

Symptoms of uterine cancer

Around three quarters of women developing endometrial cancer are postmenopausal and 90% of these present with bleeding 1. As it is an obvious symptom many women seek advice after only one or two episodes and would be referred for investigation under the rapid referral rules for suspected cancer 2. Consequently approximately 75% of all patients with endometrial cancer present with early stage disease confined to the body of the uterus. Other symptoms can include low pelvic pain, vaginal discharge or, in advanced cases, urinary or rectal symptoms.

Distant metastatic disease is an unusual at diagnosis although local metastasis to the lower vagina can lead to a patient presenting with vulvo-vaginal soreness or bleeding. Women with suspected endometrial cancer undergo investigation in a Gynaecological Cancer Unit where clinical examination, ultrasound and hysteroscopy/ biopsy can be carried out.

Staging and treatment of uterine cancer

Surgery is the mainstay of both tumour staging and treatment. A woman diagnosed with endometrial cancer usually undergoes total abdominal hysterectomy (TAH) and bilateral salpingo oophorectomy (BSO), although less invasive laparoscopic techniques for both staging and treatment continue to be developed 3. Imaging of the tumour is carried out using MRI or CT scan to demonstrate the depth of myometrial invasion, extra-uterine spread and nodal involvement.

The NHS advice is that, patients should be operated on by a cancer specialist in either a Cancer Unit or Cancer Centre 4. The ASTEC trial is comparing survival in stage 1 patients treated with conventional surgery and lymphadenectomy or with conventional surgery alone, with the aim of providing clearer evidence on the benefits of this procedure 5, findings are expected in the first half of 2006.

The tumour is graded and staged on the hysterectomy specimen according to FIGO (International Federation of Gynaecology and Obstetrics) rules 6. Grade will be stated as well, moderately or poorly differentiated (G1, 2, 3). Histological type should be specified and will usually be endometrioid adenocarcinoma. There is no major difference in prognosis between pure endometrioid adenocarcinoma and adenosquamous carcinoma, but the rarer types of serous papillary and clear cell adenocarcinomas have a markedly worse prognosis, with a propensity to distant spread. These should always be classed as G3. Stage and substage is assessed histologically, according to FIGO rules 7, with tumours confined to the body of the uterus being classed as stage I, those involving the cervix as stage II, those with spread outside the uterus as stage III and those with bowel, bladder or distant organ involvement as stage IV ( Table 6.1).

Download this table (18.5KB)

Adjuvant treatment of uterine cancer

For patients with G1 and G2, stage 1a and 1b tumours, prognosis will be excellent and no further treatment should be needed beyond surgery 8. For patients with G3 tumours or stage 1c or higher, who have been rendered macroscopically free of disease by surgery, there may be an advantage to adjuvant radiotherapy both for the control of occult nodal disease 9 and vaginal vault recurrence 10. Many centres will offer these patients adjuvant external radiotherapy to the pelvis, with or without brachytherapy to the vaginal vault. External radiotherapy will give rise to considerably more toxicity than vault brachytherapy 11 and that has to be considered in this relatively elderly group of patients, often with co-morbidities, who may only be at low risk of recurrence 12,13

Radiotherapy has consistently been shown to result in a decrease in local recurrence rates but not a significant increase in overall survival 10, 14, 15. This appears to be due to the fact that the small minority of patients who relapse often do so at distant sites with or without local recurrence 16.

Trials are underway to try to elucidate the roles of external and internal radiotherapy in the adjuvant treatment of high-risk patients and to determine the high-risk prognostic indicators 17. The PORTEC2 trial is currently underway and aims to determine the role of external radiotherapy versus brachytherapy. The ASTEC trial 5, 18 has been designed to assess the therapeutic role of external radiotherapy in high-risk patients and should report some early radiotherapy results in 2007. The role of platinum-based chemotherapy is currently being studied in G3 tumours and in the serous papillary and clear-cell variants as these behave similarly to their ovarian counterparts in which platinum-based chemotherapy is standard 19, 20. One trial has already reported results showing women treated with radiotherapy and chemotherapy for high-risk early-stage tumours had almost half the risk of death of women receiving radiotherapy only over the study period 36. There is no role for progestagens as an adjuvant in first-line treatment 21.

Treatment for residual, inoperable or metastatic uterine cancer

Fortunately, most patients are rendered macroscopically disease-free by surgery. However, there will be a small number who are unfit for surgery, who have residual disease at the completion of surgery or who develop metastatic disease. For these patients, chemotherapy, radiotherapy and hormonal treatment with progestagens, alone or in combination, need to be considered. Radical brachytherapy to the uterine body can be used in obese patients unfit for either surgery or external radiotherapy, but there is a risk of intrauterine recurrence 22.

Risk of uterine cancer recurrence/prognosis

Risk of recurrence is higher in poorly differentiated tumours that have invaded into the myometrium. The PORTEC trial has shown that, in addition to tumour type, grade, stage and substage, the patient’s age is also a factor, with a higher risk of recurrence in older women 23,24. It also appears that lymphovascular invasion is a prognostic factor in some early-stage patients 25. Lymph node status is closely associated with tumour stage and grade, with approximately 10% of stage 1 and 20% of stage 2 patients presenting with positive pelvic or para-aortic lymph nodes 9,26,27, compared to almost one quarter of patients with more than two-thirds invasion into the myometrial wall, and about 15% of patients with grade 3 tumours. Other prognostic factors are uterine size, menopausal status and the presence of progestagen receptors 9,28,29.

Uterine cancer screening

At present, there are not screening methods able to detect uterine hyperplasia or cancer with acceptable sensitivity and specificity to make population screening for endometrial cancer a possibility. Cytological sampling is technically viable, but there is currently no evidence from randomised trials to support its use 30. Screening studies suggest that ultrasound imaging of the endometrial wall can be reliably used to rule out the presence of endometrial cancer in postmenopausal women experiencing bleeding, but the test carries a high false positive rate in asymptomatic women 31,32. The Cancer Genetics Studies Consortium recommends gynaecological screening for women with HNPCC syndrome, but there is no clear evidence that ultrasound screening of the womb in this group is of benefit 33,34,35.