A randomized study of 2 risk assessment models for individualized breast cancer risk estimation.

Abstract

Background: Estimating breast cancer risk involves quantifying genetic and non-genetic factors. This supports health interventions and risk communication to ensure adherence to screening recommendations. This study evaluated the change in risk estimation when incorporating breast density and polygenic risk score (PRS) into the baseline cancer risk assessment and compared the efficacy of two risk-assessment delivery models.

Methods: This two-step study included 663 healthy women with a family history of breast cancer in which no pathogenic variants were identified. First, breast density and PRS were added to the baseline risk assessment for all participants. A randomized intervention study compared two delivery models (in-person vs pre-recorded video) for risk assessment in women at moderate or average risk. All tests were two sided.

Results: Breast density and PRS reclassified the risk group into 33% of the participants, with only 5% reclassified as high-risk. After disclosure of their estimated multifactorial risk, 65% of women aligned their risk perception with their estimated risk, compared to 47% at baseline (p-value < 0.05). No statistically significant differences were found in the primary endpoint cancer worry [mean = 10.2(SD = 3.1) vs 10.1(2.7), between delivery models. In-person delivery had slightly better psychological outcomes (excluding the primary outcome) and higher satisfaction, though few participants in the video group sought in-person clarification.

Conclusions: Incorporating breast density and PRS into risk assessments led to substantial reclassification, with 1 in 5 women facing de-escalated surveillance. Personalized assessments improve objective perceptions alignment. A model using a pre-recorded video-based model matches in-person delivery for moderate and average-risk women and is scalable for population-level implementation.