Calibrated Breast Density Measurements.

Area of Science:

• Radiology
• Oncology
• Medical Imaging

Background:

• Mammographic density is a key breast cancer risk factor.
• Ensuring prior mammography research applies to current technologies is crucial for clinical translation.
• Previous calibration methodologies need validation on contemporary imaging systems.

Purpose of the Study:

• To assess if a previously developed calibration methodology yields breast density metrics predictive of breast cancer risk.
• To evaluate the applicability of this calibration technique to modern mammography and digital breast tomosynthesis systems.

Main Methods:

• A matched case-control study involving 319 pairs was conducted.
• Two calibrated breast density measures (PGsd and PGm) were evaluated using mammograms from Hologic FFDM and DBT systems.
• Calibrated measures were compared against the Cumulus method's percentage of breast density (PD) using logistic regression and ROC analysis.

Main Results:

• The calibrated PGsd measure showed statistically significant associations with breast cancer risk, similar to PD (Az 0.61-0.62).
• Odds ratios for PGsd were comparable to PD, indicating similar predictive power.
• The calibrated PGm measure demonstrated weaker, non-significant associations with breast cancer risk.

Conclusions:

• The calibrated PGsd measure effectively predicts breast cancer risk, validating its clinical utility.
• The calibration methodology is robust and replicates under general conditions, applicable to current breast imaging technologies.
• This study supports the use of calibrated breast density metrics for improved breast cancer risk assessment.