Rationale
Breast cancer is the most common cancer in women and has an lifetime incidence of 1/8 women. It progresses in most cases form pre-invasive disease, to a tumour mass within the breast, eventually developing lymph node metastases and finally distant metastases. In the pre-invasive and early stages, breast cancer is highly curable (> 90%) whereas patients with nodal disease have lower cure rates (50-70%) and those with metastatic disease are essentially incurable. Therefore, there is a strong incentive to screen the population for breast cancer due to the frequency of the disease and potential curability with early disease.
Arguments against breast cancer screening focus on the use of ionising radiation for mammography (which likely causes a very small number of cancers), and that screening likely overdiagnoses a number of patients who would otherwise have remained well. Another important argument is that the introduction of endocrine therapy and more effective systemic therapy may have caused a significant reduction in breast cancer mortality, rather than screening.
Evidence
Mammography
Breast cancer screening with mammography was established by multiple randomised trials and a meta-analysis that demonstrated a reduction in breast cancer mortality of about 33% within 7 years of screening introduction. In the modern setting, it is very difficult to run a randomised study of screening versus nothing due to the general acceptance of screening and lack of equipoise. Case control studies suggest that screening still provides at least a 10% benefit in mortality from breast cancer.
MRI
MRI is more sensitive but less specific than mammography. Unlike mammography, it has not been evaluated in randomised controlled trials, but rather case series in combination with mammography for high risk women. It is particularly useful for young women where active breast tissue obscures the image on mammography. The sensitivity of MRI in this setting is considerably greater (80% versus 30-40%). It is recommended for women with an lifetime risk of breast cancer over 15% by American bodies; this would include women with a strong family history (2 first degree relatives) or germline mutations in BRCA1/2 or TP53.
Examination
Examination by clinicians is thought to add only marginal benefit on top of a mammography program and performs less well than mammography. It is of use in developing countries where mammography is not available.
Self examination has not demonstrable benefit in randomised trials, with the only impacted factors being an increased number of breast biopsies.
Special Issues
Young women
Young women have both a lower incidence of cancer and increased speed of cancer growth, coupled with reduced sensitivity of mammography. Analysis of randomised controlled trials in the 40-50 age group demonstrated a 15% reduction in breast cancer mortality. A trial that specifically examined this age group demonstrated a non-significant reduction in mortality. Guidelines vary; in Australia it is recommended that women 50-70 undergo biannual screening; women 40-50 can receive free screening but are not actively recruited.
Older women
The benefit of population based screening in older women becomes problematic due to competing risks and reduced benefit. Trials on women over 70 have been conflicting with outcomes. I would recommend that women over 70 continue screening if their life expectancy despite competing risks is > 10 years.
High Risk women
Guidelines suggest increased screening frequency, coupled with breast MRI, beginning at the age of 25 or 5 years before the first cancer diagnosis in a relative.