Abstract
PURPOSE: Risk-stratified screening for breast cancer might improve benefit-to-harm ratio, however, psychological consequences of risk communication and low-risk women's acceptance of de-escalated screening are unclear. PATIENTS AND METHODS: We carried out a randomized clinical trial allocating women aged 50-67 years attending their routine biennial screening mammography 1:1 to a control and intervention group. Women in the control group continued the standard program without risk stratification. Women in the intervention group had their 10-year breast cancer risk assessed using the comprehensive CanRisk model. Around day 90, risk was automatically communicated as low (<2.44%), intermediate (2.44-5.14%), elevated (5.15-<7.99%), and high (≥8.00%) with suggested screening intervals of four years (low), two years (intermediate), one year (elevated) and one year (high risk), and additional investigations for those at elevated and high-risk. RESULTS: Of the 3,949 invited women, 1,801 (46%) were enrolled. Of the intervention group, 49.6%, 38.8%, 8.4% and 3.1% were at low, intermediate, elevated and high risk, respectively. After 292 days, 3.8% (95% confidence interval = 2.3-6.1) of the low-risk women had rejected de-escalated screening. The most significant change of psychological indicators from baseline to day 180 was a 1.6-point reduction of life quality among controls (p < 0.0001), but also reductions of breast cancer worry in the low and intermediate risk groups vs. controls (-0.27, p = 0.007 and -0.33, p = 0.004). CONCLUSION: Risk-stratified breast cancer screening is feasible and de-escalated screening acceptable for low-risk women. Automated risk communication has minimal - if any - impact on short-term psychological well-being. Final conclusions await prespecified 800 days end-of-follow-up.