Change in mammographic density as a potential predictor of cancer recurrence after breast conservation surgery and adjuvant endocrine therapy: results of the MEDICI study.

[BACKGROUND] Oestrogen-receptor positive breast cancer patients are typically treated with adjuvant endocrine therapy (AET), some develop AET resistance. Previous research suggests mammographic density (MD) may represent an imaging biomarker, with fewer local or distant recurrences occurring with decreasing MD. We investigate whether reduction in MD after 1 and/or 3 years is associated with improved breast cancer specific survival (BCSS), metastasis-free survival (MFS) or disease-free survival (DFS).

[METHODS] This retrospective cohort study was generated from a Mammo-50 trial subset. Participants taking AET (cases) and controls were included. MD was assessed in the AET group using a 0-100% visual analogue scale (VAS). Readers scored mammograms at diagnosis, 1 year and 3 years post-surgery. A decrease in MD was defined as ≥10% reduction from diagnosis. A second reader reviewed paired mammograms and assessed whether there had been a temporal change in MD.

[RESULTS] Data from 1364 cases and 367 controls were included. Median VAS MD was approximately 30% for cases and controls at all time-points; 20% showed decreased MD at 1 year and 21% at 3 years for both cases and controls. Of the AET group, 23 died from breast cancer and 33 developed metastases during follow-up (median 8.7 years post-surgery). The 5-year BCSS rate was 99.6% (95%CI:97.4-99.9) versus 98.3% (95%CI:97.2-98.9) for those with and without a ≥10% reduction in MD at 1 year, p=0.35. The 5-year MFS rate for those with and without a ≥10% reduction in MD at 1 year was 94.2% (95%CI:90.7-96.4) versus 93.6% (95%CI:92.0-95.0) respectively; p=0.47. The 5-year DFS rate for those with a ≥10% reduction in MD at 1year was 92.4% (95% CI:88.5-94.9) versus 92.6% (95%CI:90.8-94.0); p= 0.47. Similar results for BCSS, MFS and DFS were seen for those with and without a ≥10% reduction in MD at 3 years and for those assessed as having a definite reduction in MD compared to those who had not at both 1 and 3 years.

[CONCLUSION] Reduction in MD had no significant association with rates of BCSS, MFS or DFS. Change in MD was not shown to be a useful prognostic indicator in women over 50 years, treated with AET.