Skip navigation


Readers of Bandolier frequently request that we carry the best available knowledge about screening programmes. Screenwatch will try regularly to do this.

Current thoughts on ovarian cancer for screening, treatment and follow-up have been summarised in the Journal of the American Medical Association reporting the conclusions of an NIH Consensus Conference (JAMA 1995, 273: 491-7). This is a thorough and useful description of where we are now, and a signpost to future developments. Reading it in detail is a must for those interested in ovarian cancer screening.

Dr Mike Bedford of the North Derbyshire Health Authority has sent us a paper entitled "Ovarian cancer screening - The Evidence". It contains much useful information, especially about the performance of currently available tests, and can be obtained from him at the address below. Mike Bedford has also written the following article on screening for ovarian cancer in asymptomatic women

Ovarian Cancer Screening in Asymptomatic Women

Ovarian cancer is a major public health problem. It is the most common, insidious, and lethal gynaecological malignancy. Despite this there is no mass screening programme to detect it. This has arisen because of the nature of the disease itself (the deep abdominal site of the ovaries, and a lack of a clear understanding of the disease), and the absence of a simple, cheap, and suitable, screening test.

Mortality from the disease depends on what stage it is detected. Changes in treatment over the last 30 years have not significantly improved survival. Hence, research work has centred around screening methods which could diagnose the condition earlier, and thus improve prognosis.

The size of the problem

It is the fifth most common malignancy in females. In England and Wales there are approximately 3900 deaths a year from the condition. The incidence is approximately 17 per 100,000 women per year. Although there is a low incidence, ovarian cancer often accounts for more deaths than cancer of the cervix and uterus combined. This is due to poor survival rates from the disease. Overall 5 year survival is around 35%.

Compared with this, the rates for early stage disease (stage 1) are excellent at around 80%. 5 year survival from late stage disease (stage 4) is around 5%. The death rate increases with age, the main number of deaths falling in the 65-75 year age group.

Risk factors

Migration studies have suggested Western lifestyle may be a contributory factor. Lifetime number of ovulations are closely associated with ovarian cancer risk so that nulliparity, few pregnancies, early menarche, and late menopause, are all risk factors. Talc dusting of the perineum has been suggested, although the evidence for this is debatable.

Oral contraceptive use may decrease lifetime risk by as much as 40%. Other protective effects have been noted with high gravidity, hysterectomy, and sterilisation.

The most significant risk is a positive family history. Heredity is thought to account for between 5 to 10% of ovarian malignancies. There are three types of ovarian cancer namely epithelial (which accounts for the vast majority), sex-cord mesenchyme, and germ cell. Work on genetic inheritance has looked at epithelial cancer.

Three clinical types have been identified [1,2]:

Familial ovarian cancer

  • Breast/ovary syndrome - families with an increased risk of both.
  • Lynch type II syndrome - where clusters of ovarian, endometrial, and colorectal cancer are seen (sometimes called family cancer syndrome).
  • Site specific - families with increased risk for ovarian cancer only.
All probably have autosomal dominant pattern with variable penetrance. Hence, healthy males can transmit the disease to their daughters. There have been suggestions that a woman who has one first degree affected relative (mother/sister/daughter), has an increased lifetime risk from 1.4% to approximately 3%.

Bruce Ponder [3], reviewing population based data [4], suggests that if there is only one close affected relative then this corresponds to a 1 in 40 (2.5%) risk of death from ovarian cancer by the age of 70. For a women with two or more affected close relatives the risks may be in the order of 30 to 40%. There is often an earlier onset with familial risk.

To further study family sets, a register of families with ovarian cancer in two or more close relatives has been set up by the United Kingdom Co-ordinating Committee for Cancer Research (UKCCCR) and Cancer Family Study Group called the UKCCCR Familial Ovarian Register [5]. This provides a resource in this area. Women can be referred to the study for investigation of the family tree, and the study will confirm the histological diagnosis in relatives. The study itself does not undertake actual screening but will make appropriate recommendations for screening to the original source of referral.

Screening tests available

There is a general consensus that pelvic examination, tumour markers, and ultrasound examination form the three most likely candidates for a screening programme. Pelvic examination is particularly poor in screening for ovarian cancer and there is little to recommend it.

The most intensively investigated serum tumour marker is CA 125. Problems encountered include the fact that it is not specific for ovarian cancer, and there is a lack of specificity and sensitivity particularly in early disease.

Both trans-abdominal and trans-vaginal ultrasound have been used in screening trials. Because of the low incidence of the disease it is difficult to attain a high positive predictive value from the screening tool. This is particularly important in ovarian screening where a positive test may result in laparotomy. Increasing the incidence in the screened group by focusing on high risk groups (such as those with familial risk) allows the test to more easily attain reasonable predictive values.

What should purchasers provide?

A major drawback is the lack of a randomised controlled trial to show a reduction in mortality from screening. The European Randomised Trial of Ovarian Cancer (ERTOC) is about to start, and should provide evidence in this area. Ovarian cancer is a high profile media issue. This often results in women requesting advice. Thus a decision must be made on what can be identified as best practice based on the available evidence.

There is no evidence to support screening asymptomatic women in the general population for ovarian cancer.

Screening asymptomatic women with one affected relative should be confined to recognised research studies [3].

Women with two or more first (mother/daughter/sister) or second (grandmother/aunt) degree affected relatives with the disease are at a markedly increased risk. There is no evidence as yet to support actively seeking these women out in the community, as due to the lack of RCT evidence on reduction in mortality, benefit cannot be guaranteed. If a woman with this history presents requesting screening, she should be referred for further advice. This might be to a Gynaecologist or to the UKCCCR Familial Ovarian Cancer Study. Any decision to undergo screening must be based on a fully informed choice.

The current criterion for the UKCCCR study is two first or second degree relatives living or dead with (epithelial) ovarian cancer. The protocol suggests yearly screening from the age of 25 or 30 with a method such as trans-vaginal or trans-abdominal ultrasound.
Dr Mike Bedford
North Derbyshire Health Authority
Scarsdale, Newbold Road, Chesterfield S41 7PF
Tel: 01246 231255 Fax: 01246 206672


  1. HT Lynch, P Watson, C Bewtra, et al. Hereditary ovarian cancer - Heterogeneity in age at diagnosis. Cancer 1991; 67: 1460-6.
  2. HT Lynch, W Kimberly, WA Albano, et al, Hereditary nonpolyposis colon cancer (Lynch syndrome I and II). Cancer 1988; 56: 934-8.
  3. Non-surgical aspects of ovarian cancer. Report of a meeting of Physicians and Scientists, Institute of Cancer Research and the Royal Marsden Hospital, London. Lancet 1994. 343: 335-40.
  4. BAJ Ponder, DF Easton, J Peto. Risk of ovarian cancer associated with a family history. In: Sharp F, Mason W D, Leake R, eds. Ovarian cancer. London: Chapman and Hall, 1990. 3-6.
  5. UK Familial Ovarian Cancer Study, Box 238, Addenbrookes' Hospital, Hills Road, Cambridge, CB2 2QQ.

previous or next story in this issue