Risk-based breast cancer screening is safe and acceptable, though it does not reduce biopsy rates, according to research published December 12 in JAMA.
Sharing results from the Women Informed to Screen Depending on Measures of Risk (WISDOM) study, a team led by Laura Esserman, MD, from the University of California, San Francisco found that the rate of stage IIB cancers was noninferior in women undergoing risk-based screening compared with women who undergo annual screening mammography.
The results show that “stratifying screening intensity, modality, and age to start screening by individual risk is safe,” the team wrote.
“Additionally, the … approach stratified risk for developing breast cancer, providing a critical foundation for improving and refining risk-based screening approaches,” Esserman and colleagues added.
Proponents of breast cancer screening based on individual risk say this approach can help shift resources from low-risk women to high-risk women. It can guide screening initiation, frequency, use of supplemental imaging, and preventive measures to improve breast cancer screening, they added.
The WISDOM trial is an ongoing trial that is assessing whether personalized breast cancer screening can prevent as many advanced cancers as annual screening. The study included a randomized cohort of 28,372 women with an average age of 54 years. Risk assessment included sequencing of nine susceptibility genes, polygenic risk scoring, and the Breast Cancer Surveillance Consortium version 2 model.
Of the participants in the study, 77% were non-Hispanic white. The researchers placed the women in either a risk-based screening group (n = 14,212) or an annual screening group (n = 14,160).
The risk-based group had a rate of stage IIB cancers (30 per 100,000 person-years) that was noninferior to that of the annual screening group (48 per 100,000 person-years).
Despite fewer mammograms for the risk-based group, biopsy rates did not significantly differ between the two groups. These included 943 biopsies performed in the risk-based group and 1,029 performed in the annual screening group, a rate difference of 98.7 per 100,000 person-years (p = 0.1).
Finally, the team reported that the cumulative incidence of cancer, biopsy, mammography, and MRI increased as the risk category increased.
“In the observational cohort, 89% of participants chose risk-based,” Esserman and colleagues added.
The study authors highlighted that better risk models and risk-reducing recommendations “hold promise for future improvements, as does more effective risk communication to patients and health care professionals to promote informed, shared decision-making.”
They added that work is ongoing for WISDOM 2.0, the next iteration of the study platform. This will use polygenic risk scores for subtype-specific and ancestry-based risk assessment, along with radiographic risk measures.
ACR responds to study
The American College of Radiology (ACR) issued a response to the study, saying its design and results “do not provide adequate information to support changes to present screening policies.”
The ACR pointed out “substantial nonadherence” to screening among women in the study and that the accrual was lower than expected. They added that the trial only recorded outcomes for stage IIB or higher cancers, “which are about an inch in diameter, and likely already present on clinical breast exams or self-breast exams.”
“The trial plan required intense breast health specialist intervention with patients that may be difficult to replicate in many clinical settings yet still produced poor adherence to risk reduction strategies. This trial has not proven the approach to be better than current recommendations,” the ACR stated. “The polygenic risk scores varied in make-up across the trial, which introduces unknowns, and risk assessment models overall have only moderate discriminating ability.”
The ACR said it continually gathers and reviews peer-reviewed studies and data from multiple sources to find out when and how often screening approaches may need to be updated.
Read the full study here.
