Abstract P2-02-12: Utility of Targeted Sonography Using Real-Time Virtual Sonography (RVS) for Breast Lesions That Were Suspicious on MRI

Abstract

Abstract PURPOSE : MRI-detected suspicious breast lesions that are not expected from previous MMG or sonography, are detectable initially on MRI in 48% maximum. Therefore, biopsy is required for definitive diagnosis of MRI-detected lesions. Although MRI-guided biopsy is often necessary, sonographically guided biopsy after targeted sonography is more broadly available and less time-consuming and costly. The aim of this study was to verify the utility of targeted sonography using Real-time Virtual Sonography (RVS) for the identification of MRI-detected suspicious lesions. METHOD AND MATERIALS : From 196 breast MRI examinations, all MRI-detected suspicious lesions and subsequent biopsy were identified between February 2006 and December 2009. All patients were examined using MMG, sonography, MRI and RVS that could synchronize a sonography image and the MRI cutaway images of the same site to be displayed in real time using magnetic navigation system. Following the conventional dynamic enhanced image, MRI was obtained on a 1.5-T imager in the supine position using a flexible body surface coil in order to achieve the same position as in sonography. We searched all cases for MRI-detected suspicious lesions and investigated targeted sonography with or without RVS in identifying the lesions. RESULTS : Of the 196 patients, MRI-detected suspicious lesions were detected in 55 (28%) patients. A total of 67 suspicious lesions in 55 patients comprised this analysis. Of the study lesions, 24 (36%) were malignant and 43 (64%) were benign. Overall mean lesion size was 6.7mm. A total of 46 (69%) mass lesions were identified, compared with 16 (24%) foci and 5 (7%) nonmasslike lesions. 18 (27%) lesions were detected in targeted sonography without RVS and were revealed as cancers in eight, benign lesions in 10. In contrast, 60 (90%) lesions were detected in targeted sonography with RVS and were revealed as cancers in 21, benign lesions in 39. Although seven (10%) lesions were not detected with RVS, all lesions were able to project enhanced MRI information onto a body surface correctly as we checked ultrasound form images without the use of large-scale equipment. CONCLUSION: The present results suggest that targeted sonography with RVS appears to be not time consuming technique which can identify a large part of MRI-detected suspicious lesions. Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P2-02-12.