Abstract P6-09-19: Breast density change at 6 months is associated with change at 12 months as measured by fat-water decomposition MRI in women on tamoxifen

Abstract Objective: Tamoxifen (TAM) lowers breast cancer recurrence by 40-50% with evidence of individual variability in responsiveness. A ≥10% decrease in mammography-determined breast density (BD) after 12–18 months of TAM use has been associated with clinical benefit. Early determination of changes in BD may offer a strategy to tailor hormone therapy in non-responders; for responders, it may encourage adherence. Fat-water decomposition MRI (FWD-MRI) is an accurate and fast (< 5 minutes) method for measuring BD without ionizing radiation or contrast agent. Here, we examined whether change in FWD-MRI-derived BD predicts decrease in BD at earlier time points than observable with a 12-month measure of BD. Methods: The study population included a subset of 44 pre- and post-menopausal women receiving TAM for treatment of early-stage breast cancer or prevention who were enrolled in a randomized, placebo-controlled trial of diindolylmethane. Eligibility for this analysis included participants with FWD-MRI scans at baseline, 6 and 12 months. Median time on TAM at baseline was 13 months (IQR, 5–26 months). All MRI images were acquired on a 1.5T GE Signa NV-CV/i scanner. Automated breast segmentation was performed using MATLAB software and validated against manual ROI drawings. MRI-based BD was calculated as the ratio of breast voxels with <80% apparent fat fraction (Fra80) over the entire breast, a measure previously shown by our group to be highly correlated with mammography-derived BD. For 40 participants, the unaffected, contralateral breast was analyzed. For 4 patients with two unaffected breasts, BD data from the left breast were analyzed. Change in BD was conservatively defined as > 2 times the test-retest variability of Fra80 (0.032). McNemar's test was used to test the association between change from baseline to 6 months and change from baseline to 12 months. Results and Discussion: At 12 months, 15 (34%) participants had a decrease in BD, whereas 29 (66%) remained unchanged or increased. Of these 29, 28 also had no decrease at 6 months (specificity = 97%), and 9 of the 15 women who showed a decrease at 12 months had a decrease at 6 months (sensitivity = 60%; McNemar's test, P = 0.06). Conversely, for those women with a measured decrease in BD from baseline to 6 months, 9 of 10 had a measured decrease at 12 months. A study limitation is inclusion of participants on TAM for varying duration as the greatest change in BD likely would have occurred earlier. Ongoing efforts will focus on FWD-MRI for measures of change in BD in patients initiating TAM. Conclusion: Use of the specified cut point would fail to detect a decrease in BD at 12 months in 40% of women. However, a decrease in BD from baseline to 6 months was highly associated with decrease from baseline to 12 months and in some women may be useful as an early biomarker of effect. Ongoing effort is needed to determine the impact of factors such as baseline BD, menopausal status, and time on TAM in misclassification of BD change using the 6-month measure. Acknowledgement: NIH grants CA149417, CA161534.Objective: Tamoxifen (TAM) lowers breast cancer recurrence by 40-50% with evidence of individual variability in responsiveness. A ≥10% decrease in mammography-determined breast density (BD) after 12–18 months of TAM use has been associated with clinical benefit. Early determination of changes in BD may offer a strategy to tailor hormone therapy in non-responders; for responders, it may encourage adherence. Fat-water decomposition MRI (FWD-MRI) is an accurate and fast (< 5 minutes) method for measuring BD without ionizing radiation or contrast agent. Here, we examined whether change in FWD-MRI-derived BD predicts decrease in BD at earlier time points than observable with a 12-month measure of BD. Methods: The study population included a subset of 44 pre- and post-menopausal women receiving TAM for treatment of early-stage breast cancer or prevention who were enrolled in a randomized, placebo-controlled trial of diindolylmethane. Eligibility for this analysis included participants with FWD-MRI scans at baseline, 6 and 12 months. Median time on TAM at baseline was 13 months (IQR, 5–26 months). All MRI images were acquired on a 1.5T GE Signa NV-CV/i scanner. Automated breast segmentation was performed using MATLAB software and validated against manual ROI drawings. MRI-based BD was calculated as the ratio of breast voxels with <80% apparent fat fraction (Fra80) over the entire breast, a measure previously shown by our group to be highly correlated with mammography-derived BD. For 40 participants, the unaffected, contralateral breast was analyzed. For 4 patients with two unaffected breasts, BD data from the left breast were analyzed. Change in BD was conservatively defined as > 2 times the test-retest variability of Fra80 (0.032). McNemar's test was used to test the association between change from baseline to 6 months and change from baseline to 12 months. Results and Discussion: At 12 months, 15 (34%) participants had a decrease in BD, whereas 29 (66%) remained unchanged or increased. Of these 29, 28 also had no decrease at 6 months (specificity = 97%), and 9 of the 15 women who showed a decrease at 12 months had a decrease at 6 months (sensitivity = 60%; McNemar's test, P = 0.06). Conversely, for those women with a measured decrease in BD from baseline to 6 months, 9 of 10 had a measured decrease at 12 months. A study limitation is inclusion of participants on TAM for varying duration as the greatest change in BD likely would have occurred earlier. Ongoing efforts will focus on FWD-MRI for measures of change in BD in patients initiating TAM. Conclusion: Use of the specified cut point would fail to detect a decrease in BD at 12 months in 40% of women. However, a decrease in BD from baseline to 6 months was highly associated with decrease from baseline to 12 months and in some women may be useful as an early biomarker of effect. Ongoing effort is needed to determine the impact of factors such as baseline BD, menopausal status, and time on TAM in misclassification of BD change using the 6-month measure. Acknowledgement: NIH grants CA149417, CA161534. Citation Format: Ding J, Thompson PA, Wertheim BC, Roe DJ, Marron MT, Altbach MI, Galons J-P, Wang F, Thomson CA, Huang C, Stopeck A. Breast density change at 6 months is associated with change at 12 months as measured by fat-water decomposition MRI in women on tamoxifen [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P6-09-19.