University of Toronto, Canada
Title: Prediction of local recurrence in ductal carcinoma in situ: Clinical validation of DCIS Score
Nofech-Mozes obtained her medical degree from Tel Aviv University, Israel. She completed breast and gynecologic pathology fellowships at the University of Toronto, Canada. Dr. Nofech-Mozes is an Associate Professor in the Department of Laboratory Medicine and Pathobiology at the University of Toronto. She is the breast pathology lead in Sunnybrook. She is a member of the institutional Research Ethics Board. Her main research interest is ductal carcinoma in situ. Dr. Nofech-Mozes authored or coauthored more than 55 peer reviewed manuscripts, including provincial guidelines for hormone receptor testing in breast cancer and biomarkers synoptic report for the CAP.
The Ontario DCIS population-based study identified women with pure DCIS from 1994-2003. Clinical validation of DCIS Score (DS) (Rakovitch, SABCS 2014) showed prediction of risk of an ipsilateral local recurrence (LR). Centrally reviewed pathology for: focality, size, grade, subtype, comedo necrosis & clear margins, (CM=no ink on tumor) will be presented. The DS was obtained by quantitative RT-PCR. Cox modeling was used to determine the relationship between independent covariates, DS (hazard ratio (HR)/50 units) & LR. DCIS Score was evaluated in 718 women w/ DCIS tx with BCS alone (571 w/ CM). With a median follow-up of 9.4 years, 100 BCS alone w/ CM cases developed LR (44 DCIS, 57 invasive). In the primary analysis, among 571 pts treated by BCS alone with CM the continuous DS was significantly associated with LR in ER+ pts (HR 2.26; 95% CI 1.41, 3.59; P=0.001) and in all pts (HR 2.15; 95% CI 1.43, 3.22; P=<0.001). The results of univariable and multivariable analyses, hazard ratios for factors associated with in situ and distance local recurrence will be presented. DCIS Scores were widely distributed within each subgroup defined by the clinical and pathology characteristics. For DCIS pts treated with BCS alone the DCIS Score, focality, tumor size and histologic subtype provide independent LR information. Patients with low DCIS score and non-multifocal disease may be considered for BCS alone.