Breast Density Masking Calls for Precision Screening: Danielle B. Holt, MD, MSS

Nearly half of women undergoing routine mammography have dense breasts, a characteristic that raises breast cancer risk and reduces mammography's sensitivity, according to Danielle B. Holt, MD, MSS, of the Uniformed Services University of the Health Sciences. In this interview with the American Journal of Managed Care^® (AJMC^®), Holt discusses that most screening risk models focus on genetic and inherited risk while overlooking density-driven masking, which systematically disadvantages women whose main vulnerability isn't family history, but the simple fact that dense tissue hides tumors on a mammogram. 

The conversation centers on a topic Holt addressed directly in a recent JAMA viewpoint. Dense breast tissue both elevates breast cancer risk and reduces mammography's sensitivity, because dense tissue and tumors appear similarly radiopaque on imaging. But because nearly half of screening-eligible women have dense breasts, using density as a standalone criterion for supplemental screening would lead to substantial overuse of additional imaging and the harms that come with it, including false positives and unnecessary biopsies.

Holt notes that roughly half of her patients' cancers are found through self-detection, while the remainder are identified through routine screening imaging. When cancers are not caught early, staging shifts, often determining whether a patient proceeds to surgery or neoadjuvant chemotherapy first. More advanced disease, particularly stage 2B or higher, is generally associated with worse long-term outcomes.

Instead, Holt advocates for what she terms "precision screening," an approach that layers multiple risk factors rather than relying on any single variable. She points to the WISDOM trial (NCT02620852), which last year demonstrated that replacing annual, age-based screening defaults with individualized risk stratification was noninferior for detecting advanced disease. Holt discusses the relative strengths of existing models, noting that the Tyrer-Cuzick model incorporates a broader set of risk factors than the more commonly used Gail model, though she emphasizes that genetic and familial risk models alone still miss key contributors to breast cancer risk and detectability, including lifestyle and imaging-modality factors.

Looking ahead, Holt highlights the growing role of artificial intelligence (AI) in this space. AI models analyzing radiographic features, including breast density and tissue texture, are beginning to help clinicians estimate not just the likelihood that a cancer is present, but the likelihood that it would go undetected given the imaging modality used. Combining these AI-driven insights with established risk-stratification tools, she suggests, may eventually allow clinicians to identify the specific subset of women who are both at elevated risk and most likely to be missed by current screening protocols.

Reference

1. Holt DB. Breast density masking and the need for precision screening. JAMA. Published online March 30, 2026. doi:10.1001/jama.2026.2443