Original Research Contrast-Enhanced MRI of Ductal Carcinoma In Situ: Characteristics of a New Intensity-Modulated Parametric Mapping Technique Correlated With Histopathologic Findings Michelle N. Mariano, MD, 1 Maurice A.A.J. van den Bosch, MD, 1,2 * Bruce L. Daniel, MD, 1 Kent W. Nowels, MD, 3 Robyn L. Birdwell, MD, 1 Kathy J. Fong, MSc, 1 Pam S. Desmond, MSc, 4 Sylvia Plevritis, MD, 1 Lara A. Stables, MSc, 4 Marowan Zakhour, MSc, 1 Robert J. Herfkens, MD, 1 and Debra M. Ikeda, MD 1 Purpose: To identify morphologic and dynamic enhance- ment magnetic resonance imaging (MRI) features of pure ductal carcinoma in situ (DCIS) by using a new intensity- modulated parametric mapping technique, and to correlate the MRI features with histopathologic findings. Materials and Methods: Fourteen patients with pure DCIS on pathology underwent conventional mammography and contrast-enhanced (CE) MRI using the intensity-modulated parametric mapping technique. The MR images were re- viewed and the lesions were categorized according to mor- phologic and kinetic criteria from the ACR BI-RADS-MRI Lexicon, with BI-RADS 4 and 5 lesions classified as suspi- cious. Results: With the use of a kinetic curve shape analysis, MRI classified seven of 14 lesions (50%) as suspicious, including four with initial-rapid/late-washout and three with initial-rapid/late-plateau. Using morphologic criteria, MRI classified 10/14 (71%) as suspicious, with the most prominent morphologic feature being a regional enhance- ment pattern. Using the intensity modulated parametric mapping technique, MRI classified 12/14 cases (86%) as suspicious. Parametric mapping identified all intermediate- and high-grade DCIS lesions. Conclusion: The intensity-modulated parametric mapping technique for breast MRI resulted in the highest detection- rate for the DCIS cases. Furthermore, the parametric map- ping technique identified all intermediate- and high-grade DCIS lesions, suggesting that a negative MRI using the parametric mapping technique may exclude intermediate- and high-grade DCIS. This finding has potential clinical implications. Key Words: magnetic resonance imaging; breast; ductal carcinoma in situ; parametric mapping; histopathology J. Magn. Reson. Imaging 2005;22:520 –526. © 2005 Wiley-Liss, Inc. DUCTAL CARCINOMA IN SITU (DCIS) is a proliferation of atypical epithelial cells of the mammary ducts with- out evidence of invasion through the basement mem- brane into the periductal connective tissue (1). The in- cidence of DCIS has increased and it now accounts for 15–25% of all breast cancers. Cases are typically de- tected because of suspicious microcalcifications on screening mammography (2). Although pleomorphic, fine linear, and linear branching microcalcifications are a highly sensitive mammographic sign of DCIS, DCIS may be noncalcified and more extensive than visualized by mammography (3–7). Contrast-enhanced magnetic resonance imaging (CE-MRI) of the breast has become a well established method for detecting invasive breast carcinoma. It is a more sensitive method than conventional mammogra- phy for detecting invasive tumors, with a sensitivity approaching 100% (8 –16). However, there are few sci- entific data regarding the accuracy of MRI as compared to mammography for evaluating DCIS (17). Previous studies showed variable sensitivities of breast CE-MRI in detecting DCIS, ranging from 40% to 100% depend- ing on the criteria used to define malignancy and the MRI features that were assessed (17–25). DCIS encom- passes a histologically heterogeneous group of tumors 1 Department of Radiology, Stanford University Medical Center, Stan- ford, California, USA. 2 Department of Radiology, University Medical Center Utrecht, Utrecht, The Netherlands. 3 Department of Pathology, Stanford University Medical Center, Stan- ford, California, USA. 4 Lucas Center for Magnetic Resonance Spectroscopy and Imaging, Stanford University Medical Center, Stanford, California, USA. *Address reprint requests to: M.A.A.J.v.d.B, Department of Radiology, Room E.01.132, University Medical Center Utrecht, Heidelberglaan 100, 3584 CX Utrecht, The Netherlands. E-mail: m.a.vandenbosch@azu.nl Contract grant sponsor: NIH; Contract grant number: CA66789. Received December 14, 2004; Accepted June 15, 2005. DOI 10.1002/jmri.20405 Published online 2 September 2005 in Wiley InterScience (www. interscience.wiley.com). JOURNAL OF MAGNETIC RESONANCE IMAGING 22:520 –526 (2005) © 2005 Wiley-Liss, Inc. 520