P2-15-01: Surgical Patterns of Care after Magnetic Resonance Imaging in the Academic Setting in Patients with Operable Breast Cancer Treated with Neoadjuvant Systemic Therapy: A Secondary Analysis of TBCRC 017.

Abstract

Abstract Background: Neoadjuvant chemotherapy is increasingly used to downstage locally advanced tumors, test new regimens, and improve chances of breast conservation. In parallel, use of breast MRI for surgical planning and as a potential response biomarker has also increased along with an observed increase in mastectomy rates. This secondary analysis reports surgical patterns of care across 8 NCI comprehensive cancer centers in women receiving both neoadjuvant therapy and breast MRIs. Methods: 770 women from 8 institutions were retrospectively identified as having received neoadjuvant systemic therapy with MRI done before and after systemic treatment. Univariate and multivariate analyses of covariates associated with surgical management were performed. Within MRI complete response or not, the Jonckheere-Terpstra Test (JTT) was used to test for a trend in mastectomy rate by T stage. Results: Surgical data was available on 763/770 patients. Table 1 lists MRI response by pretreatment imaging size and final breast surgery. Distribution of pretreatment imaging-defined T stage was as follows: T1 66 (8%), T2 436 (57%), T3 230 (30%), and T4 36 (5%). 347/763 (45%) patients received breast conserving surgery and 416/763 (55%) had a mastectomy. 58% vs 43% (p=0.0003) of those with an incomplete vs complete MRI response had a mastectomy, respectively. Univariate logistic regression analysis of covariates significantly associated with use of mastectomy included ER- status (p=0.01), incomplete MRI response (p=0.0003), higher T stage (p<0.0001), chemotherapy regimen (p=0.006), younger age (p<0.0001), and institution (p=0.002); all except chemotherapy regimen and institution remained independent predictors of mastectomy on multivariate analysis. Patients with larger baseline T stage and an incomplete response on MRI (P<0.001) were more likely to undergo mastectomy. Overall, breast conservation was low (262/495, 53%) of all T1-T2 patients, especially in the T1 subset (26/65, 40%). Conclusions: In a contemporary group of women treated with preoperative chemotherapy at NCI designated cancer centers and with available pre/post MRI, we observed a lower frequency of breast conservation as compared to historical data from prospective randomized trials. Larger tumor size at baseline was associated with higher mastectomy rates. However, even among those with a complete MRI response, 43% had a mastectomy. In addition to clinical response and MRI findings, factors not examined here (e.g., BRCA status, education level, and patient/provider preference) may influence the decision-making process and deserve further evaluation. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P2-15-01.