Relationship of breast MRI to recurrence rates in patients undergoing breast-conservation treatment

The effect of pre-operative MRI on the in-breast tumor recurrence rate (IBTR) of patients undergoing breast-conservation treatment (BCT) remains uncertain. We began to routinely perform pre-operative MRI in 2006. Our goal was to determine the effect of pre-operative MRI on IBTR. Retrospective review of a prospective database of all patients undergoing BCT (n = 1396) from 2000 to 2010. IBTR were calculated using Kaplan-Meier estimates. 664 (47.6%) patients underwent pre-operative MRI. The use of MRI increased from 13.9% in 2000-2005 to 80.7% in 2006-2010. Ten percent of patients who underwent MRI were found to have an additional ipsilateral cancer, with a mean diameter of 1.6 cm. The IBTR for patients with and without MRI were 4% vs. 8% at 8 years (p = 0.04). In multivariate analysis, radiation therapy and endocrine therapy were associated with decreased IBTR, but MRI was not (RR 0.77 (0.45-1.28)). For 1030 patients with invasive cancer, the IBTR at 8 years with and without MRI was 4.2% vs. 7.3% (p = 0.28). For 366 DCIS patients with and without MRI, the IBTR was 3.6% vs. 10.9% (p = 0.06). In the subgroup of DCIS patients who did not receive radiation, the IBTR with and without MRI was 0% vs. 18.2% (p = 0.08). Patients with an additional cancer found by MRI had a higher IBTR at 8 years (10.1% vs. 3.3%, p = 0.02). In a study analyzing BCT patients from one time period who rarely had a pre-operative MRI and a subsequent time period where most patients had MRI, the use of MRI was associated with a decrease in the IBTR on univariate, but not multivariate analysis. Patients who had additional cancers detected had a significantly higher IBTR.