OBJECTIVE The purpose of this study was to evaluate sonographic findings of breast papillary lesions and the effectiveness of the American College Radiology Breast Imaging Reporting and Data System sonographic assessment system for differentiation of benign and malignant papillary lesions. METHODS We retrospectively reviewed breast sonographic findings of 46 surgically proven benign papillomas and 22 papillary carcinomas. All sonographic images of patients were interpreted by 2 radiologists. Sonographic findings were analyzed according to the Breast Imaging Reporting and Data System classification. RESULTS The shape of the lesion was round or oval in 33 benign lesions (71.7%) and 13 papillary carcinomas (61.9%). As for the margin, a circumscribed margin was found in 31 benign papillomas (67.4%) and 12 malignant lesions (57.1%). Differences in the predominant shape and margin between the 2 groups were not statistically significant (P > .05). Fourteen benign papillomas (30.5%) and 12 papillary carcinomas (57.1%) showed a complex echo pattern. It was more frequently observed in malignant lesions; however, it was not statistically significant (P = .09). A nonparallel orientation, an echogenic halo, posterior acoustic enhancement, and associated microcalcification were more frequently found in malignant than in benign lesions (P < .05). When the presence of any suspicious sonographic feature (nonparallel orientation, echogenic halo, posterior enhancement, or calcification) was considered to indicate malignancy, interpretation of the sonographic features gave sensitivity of 85.7%, specificity of 64.9%, a positive predictive value of 47.4%, and a negative predictive value of 92.5% for detection of malignant papillary lesions. The only differential finding between noninvasive and invasive papillary cancers was a circumscribed margin (P < .05). CONCLUSIONS Sonographic features more specific to malignancy include a nonparallel orientation, an echogenic halo, posterior acoustic enhancement, and associated microcalcification.