screening for breast cancer

An introduction to screening for breast cancer

Although mortality from breast cancer has fallen over the last decade, it is still the leading cause of death from cancer among women in the UK. In 2000,over 40000 cases were diagnosed and there were 1ñ4 The aetiology of the disease nearly 13000 deaths.
is not fully understood, although the risk is known to be associated with reproductive and family
history. Screening for breast cancer has been shown to advance the diagnosis of the disease, which can lead to more successful treatment and therefore reduced mortality. This chapter details the evidence upon which breast screening has been introduced on a national scale, with particular reference to the UK.

The natural history of breast cancer is well The disease is believed to usually documented.
have a pre-invasive stage where the carcinoma cells are confined to the duct system within the breast. This may then become invasive and begin to invade the surrounding tissue, and thereafter possibly spread to the lymph nodes or other secondary sites within the body. Breast tumours may disseminate at different stages in their natural history. In some women, and for some types of tumour, this could take years, while in others metastatic spread may take only weeks, depending on the aggressiveness of the cancer. Ideally, screening should detect tumours
while they are still small and before any metastases In order for screening to be effec-have developed.tive, the disease must have a recognisable early stage. In the case of breast cancer, this is the preclin-ical detectable phase where a tumour can be seen on a mammogram but before it becomes palpable (about 1cm in diameter). Tumours in this phase are more likely to be non-invasive or, if already invasive, less likely to have local regional or distant spread.

For screening to be beneficial, treating breast cancer at this earlier stage must also improve the prognosis compared with more advanced cancers. It is not,however, sufficient to compare the survival of women with screen-detected cancers with the survival of those who present symptomatically without removing the effect of various biases.As screening will advance the date of diagnosis,the survival time will automatically be longer even if there is no effect on the actual date of death. Also,less aggressive, slowly growing cancers will spend
more time in the preclinical detectable phase than will rapidly growing cancers, which are more likely to present symptomatically. Screening will therefore detect proportionally more of the slow growing, or non-invasive, cancers, which in turn have a better prognosis. This is known as length bias. There is also the problem of selection bias in which those who attend for screening are more likely to be health-conscious individuals than those who refuse, and would probably have a better prognosis anyway. These biases can be removed by comparing mortal-
ity in a population that was offered screening witht hat in a population that was not offered screening,in the context of a randomised controlled trial.

The suitability of any screening test depends on its accuracy.

It must be able both to detect the majority of women who have breast cancer (high sensitivity) and therefore give few false-negative results, and to eliminate the majority of women who do not have the disease (high specificity), thereby minimising the number of false-positive results. Sensitivity is defined as the proportion of all those with breast cancer present who test positive; specificity is defined as the proportion of all those without the disease who test negative. Ideally, both sensitivity and specificity would be 100%, but there is an inevitable compro-mise as no test is perfect and the two are inversely related to one another. Different screening modalities
will be discussed in more detail later.

For public health, the acceptability of a screening test by the general population is of paramount
importance. The acceptability will be reflected by the rate of compliance with invitation to screening.Levels of between 80% and 90% have been seen in A level of 70% compliance has been Sweden.shown to be effective in reducing breast cancer Latest figures mortality in the target population. show that an acceptance rate of 75.3% has been although there are achieved in the UK as a whole, regional differences and specific groups where more work is needed. Women from minority ethnic groups and women with learning disabilities have
particular needs. A major research project is now underway to look at the information needs of
women from a diverse range of social and cultural backgrounds. Another important consideration in a screening test is that it should do no harm, either physical or
psychological. The potential physical hazards from screening by mammography are the risk from ionis-ing radiation (X-rays) and unnecessary surgical procedures resulting from overdiagnosis of tumours that may never have become invasive in the lifetime of the patient. The radiation risk from mammogra-phy has been very much reduced in recent years.

Due to technical advances, the maximum dose is now about 2.2mGy (compared with 2ñ3cGy in the There is no evidence that this level of past). radiation induces breast cancer, but from the excess breast cancer incidence seen in women exposed to higher doses, it has been inferred that modern mammography may induce one breast cancer in a population of two million women above the age of 50, after a latent period of 10 years. When compared
with the expected natural breast cancer incidence of 1400 cases per million women per year at age 50 and 2000 cases per million women per year at age 65, the risk is considered to be insignificant compared with the potential benefits.

High-quality screening techniques and highly trained technical and radiological staff should
minimise the risk of overdiagnosis. Ideally, the recall rate should be as low as possible. It is also
important that adequate assessment and treatment facilities exist to ensure that women are seen as quickly as possible. In addition, there is a range of non-invasive investigative techniques stopping short of open biopsy that should be employed to reduce the risk of unnecessary surgery. These include ultra-sound, spot views, micromagnification, fine needle aspiration and needle core biopsy. These techniques have greatly reduced the benign biopsy rate.

Screening for breast cancer was introduced in the UK following the recommendations of the Forrest The NHS Breast Report published in 1986. Screening Programme (NHSBSP) initially offered single mediolateral oblique-view mammography to women aged 50ñ64 with an interval of 3 years. Following further research, the programme has been expanded to invite women up to the age of 69 and to offer two-view mammography at all screens.

Screening must be repeated at regular intervals to ensure its effectiveness, as the risk of developing breast cancer increases with age and the growth rate of the disease is variable. There has to be a compro-mise taking into account the cost and practicality of screening too frequently while aiming to let as few cancers as possible escape detection by screening. In the UK, the interval was initially set to 3 years, which reflected the 33-month interval in the
Swedish Two-County Trial, with a recommendation for more research into the screening interval.

The Nottingham Prognosticannual screening.

Index (NPI) was used to compare predicted mortal-ity in the two groups. No significant improvement in predicted mortality was seen in the annual screening group. The screening interval in the NHSBSP therefore remains unchanged at 3 years.

Related Post: