Abstract P4-01-12: Optical imaging of breast density correlates with magnetic resonance imaging in patients undergoing neoadjuvant chemotherapy

Abstract

Abstract Background: In addition to being a strong risk factor for breast cancer, data from recent studies suggests that breast density changes correlate with response to preventative or adjuvant tamoxifen treatment for breast cancer. Breast density may be a surrogate biomarker for survival and indication of the need for additional treatments such as radiotherapy. However, quantitative methods to assess breast density in the clinic are limited. In this study, we assess how Diffuse Optical Spectroscopic Imaging (DOSI) may be used to evaluate and understand changes in breast density in patients undergoing neoadjuvant chemotherapy. Methods: DOSI was used to measure hemodynamic and metabolic information from the contralateral normal breast of 28 breast cancer patients undergoing neoadjuvant chemotherapy. DOSI uses near-infrared light to determine absolute tissue molar concentrations of oxyhemoglobin (ctO2Hb), deoxyhemoglobin (ctHHb), water, and lipid content without requirement of an exogenous contrast agent. DOSI measurements were compared to 3.0T MRI measured fibroglandular density before and during therapy. Results: Water (r = 0.843; P<0.001), ctHHb (r = 0.785; P = 0.003), and lipid (r = -0.707; P = 0.010) concentration measured with DOSI correlated strongly with MRI-measured density before therapy. During neoadjuvant treatment measured at ∼90 days after treatment commenced, significant reductions were observed in ctO2Hb for pre- (-20.0%; 95% confidence interval (CI), -32.7 to -7.4) and postmenopausal subjects (-20.1%; 95% CI, -31.4 to -8.8), and water concentration for premenopausal subjects (- 1.9%; 95% CI, -17.1 to -6.7) compared to baseline. Lipid content increased slightly in premenopausal subjects (3.8%; 95% CI, 1.1 to 6.5) and water increased slightly in postmenopausal subjects (4.4%; 95% CI, 0.1 to 8.6). Percentage change in water at the end of therapy compared with baseline correlated strongly with percentage change in MRI-measured density (r = 0.864; P = 0.012). Discussion: For this patient cohort, non-invasive optical measurement may provide an assessment of breast density and yield additional information that complements MRI and mammography techniques. We demonstrate a strong baseline correlation between water and ctHHb with breast density, reflecting the increased water content and rate of metabolism in fibroglandular breast tissue. Significant changes in optical markers for vascular density and supply (ctO2Hb and water) are observed during treatment. A significant decrease in ctO2Hb is observed in both menopausal groups. The steady reduction of ctO2Hb without a corresponding decrease in ctHHb suggests that chemotherapy agents act directly on the normal breast tissue, perhaps by causing a reduction of perfusion. The greater reduction in breast-tissue water in premenopausal subjects suggests that chemo-reduced ovarian hormone levels may exert a role in reducing breast-tissue density. Although from a limited patient dataset, these results suggest that DOSI may provide new functional indices of breast density based on hemoglobin and water that could be used at the bedside to assess response to therapy. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P4-01-12.