Abstract P4-11-18: CT Simulation Alone Appears To Be Appropriate for Pre-Operative Partial Breast Radiation

Abstract

Abstract Introduction: Pre-operative radiotherapy is being evaluated at several centers as a method to potentially increase patient eligibility for partialbreast radiotherapy. Theoretically pre-operative radiotherapy will decrease the volume of normal tissue irradiated which could decrease the morbidity. This study was performed to determine whether CT based imaging could be used for radiotherapy planning. Material and Methods: Between December 2008 and February 2010, 204 breast cancer patients were seen in the breast evaluation and treatment program clinic at the University of Maryland Marlene and Stewart Greenebaum Cancer Center. Patients with a new diagnosis of breast cancer and no prior treatment were included in the study if they underwent MRI of the breast and a CT, including the breast, within 30 days of each other, and images were available for review. Measurements of maximum tumor dimension obtained from CT and MRI were compared. These measurements were also compared to the maximum pathologic tumor dimension, when pathology was available. Measurements were compared using correlative paired t-tests. Associations between these measurements and T stage, tumor diameter, nodal involvement, receptor status and histologic subtype were explored. Specifically, the risk of CT underestimating the size of the tumor in the pre-operative setting was also evaluated, with a 3 mm underestimation of size chosen as significant. Results: CT and MRI images were available for 40 patients. Twelve patients who underwent imaging after receiving induction chemotherapy were excluded, leaving 28 patients for analysis. In 25 patients, CT images were obtained as part of a staging PET/CT. Pathology was available in 19 patients; nine patients received neo-adjuvant chemotherapy after imaging was obtained. The average maximum tumor dimension was 10% smaller on CT than MRI (4.25 cm (range 1.1 — 9.1 cm) versus 4.72 cm (range 1.3 — 9.0 cm)), but this difference was not statistically significant. Tumor measurements obtained by CT and MRI were only statistically different in the presence of lobular carcinoma (p=0.049). The CT underestimated the MRI measurement by greater than 3 mm in 9 out of 28 (32.14%) patients, six of whom had extensive calcifications on mammogram. When the maximum tumor dimension obtained from CT was compared to pathological size, CT underestimated the pathologic tumor size by greater than 3 mm in only 4 out of 19 (21.1%) patients. All 4 patients had pathologic T3 tumors and were node positive. For comparison, the MRI similarly underestimated the tumor size in 3 of these 4 cases. Limiting our analysis to tumors less than 3 cm in diameter, only 1 out of 7 patients had a pathologic tumor size that was 3 mm greater than the maximum tumor dimension obtained from CT imaging. Conclusions: Measurements of maximum tumor dimension on CT are on average 10% smaller than measurements obtained from MRI, although not statistically different. The risk of underestimating pathologic maximum tumor dimension was similar for CT and MRI. The risk appeared to be less for infiltrating ductal carcinoma and tumors < 3 cm. CT based treatment planning for pre-operative partial breast radiotherapy seems appropriate although additional data is needed to confirm these findings. Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P4-11-18.