Stockholm Mammography Risk Stratified Trial

Aims: The overarching aim of SMART is to compare the effectiveness of current screening practice with individualised breast cancer screening. In SMART women who recently had a normal mammogram will be randomised to either age based or individualised screening. Women randomised to individualised screening will have their 2-year breast cancer risk assessed upfront. This is done via an artificial intelligence (AI) derived analyses of the mammograms. Women scored with the highest 2-year risk will be offered a contrast enhanced mammography and an additional mammography at 12 months. The possible benefits of individualised screening will be contrasted to the possible harms defined as unnecessary recalls, biopsies and induced anxiety and worry.

Primary Objectives: The primary benefit endpoint is a comparison of the proportion of interval cancers in the intervention and comparison group. The primary safety endpoint is the number of screen-initiated recalls and biopsies which do not lead to a cancer diagnosis.

Secondary Objectives: The secondary benefit endpoint is a comparison of the tumor characteristics of diagnosed cancers in the intervention and comparison group. The investigators will also compare differences in cancer worry, anxiety and / or depression between the intervention and control groups.

Explorative Objectives: To describe the possible side effects related to contrast enhanced mammography and compare them to those experienced after an ordinary mammogram. Cost-effectiveness as measured using incremental cost-effectiveness ratio and willingness-to-pay amount.

Primary Endpoint: For women at high 2-year risk in the intervention arm interval cancers are defined as cancers diagnosed in the interval between contrast enhanced mammography (CEM) and 12-month mammogram plus in the interval between 12- and 24-month interval. For the remaining women in the trial interval cancers are defined as cancers diagnosed between baseline mammography and 24-month mammography.

Information on date of diagnosis and mode of detection (screen detected or interval cancer) is derived from the National Quality Registry for Breast Cancer. To certify the correct diagnosis, the investigators will also use the electronic medical records, Take Care, at Södersjukhuset. Information on number of recalled women and number of biopsies will gathered from the Radiation Information System at the hospital.

Secondary Endpoint: Information on date of diagnosis and tumour characteristics (eg. stage, receptor status) is derived from the National Quality Registry for Breast Cancer. The investigators will compare how the screening modalities are perceived using answers from the self-reported Cancer Worry Scale and State Trait Anxiety Inventory.

Explorative Endpoint: The possible side effects related to contrast enhanced mammography and ordinary mammography will be identified using the self-reported Test Morbidity Index.

Study design: SMART is a Phase III, randomised, non-blinded screening trial. Women who consent to participate in SMART are randomised to either individualised screening or screening according to the Swedish National Breast Cancer Screening Program. Women randomised to individualised screening get their 2-year risk of breast cancer assessed and women with the highest short-term risk are offered contrast enhanced mammography and a mammogram at 12 months.

Study population: Women invited for screening according to the Swedish National Breast Cancer Screening Program at Södersjukhuset, that is, age range 40 - 74 years of age, will be invited to participate in SMART.

Number of included participants: Participants will be matched 1:1 with 35,000 women in each arm.

Intervention: Women randomised to the individualised screening arm and judged to be at high risk will be invited to a contrast enhanced mammography and a mammography after 12 months. The intervention is thus a different selection of women compared to what is done today.

Study Duration: There will be two database locks in SMART. The primary lock is done after approximately four years and two months. Women participating in SMART are part of the initial recruitment phase for a total of two years. They perform mammograms at baseline and 24 months. Adding the 2 year follow up for the lastly recruited participant makes a full trial period 4 years in total, with a further two months allowed to capture women whose scan is delayed. After two months the investigators will consider that screening visit to have been missed. The number of high-risk women will be slightly more than 6,000 in each arm (see 10.1 Population / Power). In addition, a second data lock will be performed after 8 years and two months from the start of the trial, that is, after a fourth round of screens (baseline plus 3 routine follow up mammograms). The same objectives and endpoints will be addressed as after the first data base lock with the exception of questionnaire data. Questionnaire data will only be collected between baseline and the next screen in 24 months. SMART will start in 2024, recruit women for 2 years and follow all women for 6 years. The study will thus end in 2032.