Research and optimisation of breast screening

Evidence for extending the age of eligibility  

The appropriate age bracket to offer breast screening is often debated. There's limited evidence on the benefits and harms of screening before the age of 50 and after the age of 70. Before any changes to the current eligible age range, any evidence would be assessed by the UK National Screening Committee (UK NSC).  

A UK trial has demonstrated that beginning screening at 39-41 years results in a slight reduction in breast cancer mortality, with little increase in overdiagnosis¹

.​ Authors suggest that further analysis of other trials might help clarify the long-term effects of early screening.  

The Cancer Research UK supported AgeX trial will assess the benefit of inviting women aged 47-49 and 71-73 to breast screening.

Risk stratified screening evidence

Risk assessment of factors, such as breast density and family history, could identify people at a higher or lower risk of breast cancer. 

People with a high risk of developing breast cancer could be offered more frequent screening, or be screened using a different method. Additionally, individuals with a low risk of breast cancer could have longer screening intervals.  

Risk stratification has the potential to improve service efficiencies, reduce screening harms and facilitate earlier detection of cancer²

.​ Studies suggest that risk stratification is likely to be cost-effective and acceptable to the public³.​ However, public engagement would be required to address concerns around high risk scores and any reduction in screening frequency for people classified as low risk.

My Personalised Breast Screening (MyPeBS) is an ongoing randomised trial of risk stratified breast screening across six countries, including the UK. The study is due to complete in 2027 and may show whether risk stratified screening would be effective and feasible.

The structure of a risk stratified programme, the best tools for risk assessment and the impact on inequalities need to be understood. 

Proactive identification of women at a high risk of developing breast cancer

People at a higher risk of breast cancer could be identified in primary care by being proactively offered risk assessment. This could particularly benefit women younger than the current eligible age for screening. People identified may be eligible for additional breast screening and surveillance.

Training and resource to support primary care offering breast cancer risk assessment is vital  ⁴

⁵⁶⁷.​​

Evidence is needed to show if risk assessment tools are applicable across diverse populations. Engagement strategies which consider health inequalities and support informed uptake will be required. 

Evidence for supplementary screening based on breast density

Breasts are made up of fatty and ‘dense’ tissue. People with more dense tissue have a higher risk of developing breast cancer. The cancer may also be missed during routine screening as the dense tissue can mask tumours. Younger women and those with a lower Body Mass Index are more likely to have dense breasts. 

Supplementary screening, using an alternative technique to mammography, may support earlier cancer diagnosis in women with dense breasts⁸

⁹.​ The BRAID trial will compare usual care to supplementary screening for women with dense breasts using three different techniques, to identify which is the most effective.  

A reliable and automated method of measuring breast density is needed that's also feasible to implement. We also need to understand if supplementary screening is acceptable for people with dense breasts. 

Artificial intelligence (AI) in breast screening

Several AI-based tools have been developed to optimise mammography, ultrasound, and MRI imaging. Applications include:  

• Improving speed of image acquisition and/or quality of images. 

• Detecting regions of interest for review by a radiologist.  

• Acting as an image reader alongside a radiologist. 

• Identifying exams most likely to reveal cancer for high-priority review.  

• Classification of ambiguous lesions. 

AI tools could bring several clinical benefits. Tools could improve cancer detection rates, sensitivity and specificity, while reducing false negatives and unnecessary biopsies. Implementing AI could also deliver improved efficiency, especially in low-resource areas.  

There are still evidence gaps around performance and further research is required to understand the barriers to implementation in breast screening. This includes assessing programme costs, IT requirements, and acceptance among patients and practitioners¹⁰

¹¹. ​

Read more about breast screening