Original Research Characterization of Breast Lesions Using the 3D FIESTA Sequence and Contrast-Enhanced Magnetic Resonance Imaging Catherine S. Klifa, PhD, 1 * Ann Shimakawa, MS, 2 Zaker Siraj, BS, 1 Jessica E. Gibbs, 1 Lisa J. Wilmes, PhD, 1 Savannah C. Partridge, PhD, 3 Evelyn Proctor, RT, 1 and Nola M. Hylton, PhD 1 Purpose: To determine whether combining 3D fast imaging employing steady-state acquisition (FIESTA) and T1- weighted contrast-enhanced (CE) sequences could help characterize lesions in 32 women with benign, in situ, or invasive breast lesions. Since FIESTA provides both T1 and T2 information on the same three-dimensional (3D) matrix as high-resolution T1-weighted dynamic data, we aimed to verify whether invasive lesions could be separated from in situ and/or benign lesions using quantitative FIESTA mea- sures of tissue intensity and homogeneity. Material and Methods: With the use of CE-MRI data, re- gions of interest (ROIs) were manually delineated in en- hancing lesions and on surrounding normal tissue. These ROIs were then applied to 3D FIESTA data. Quantitative measures between lesion and normal tissue were compared among the lesion groups. Results: On FIESTA most invasive cancer lesions were hy- pointense compared to surrounding normal tissue (mean lesion intensity was 89% of normal tissue intensity), whereas most ductal and benign lesions appeared hyper- intense compared to surrounding normal tissue (lesions at 100.9% and 121.9% of normal tissue intensity, respec- tively). Measures obtained from resampled T2-weighted data showed no significant differences between the invasive and benign lesion groups. Conclusion: We detected significant differences between invasive and noninvasive lesions by quantifying intensity differences between the lesions and surrounding normal tissue on FIESTA. Key Words: breast magnetic resonance imaging; FIESTA sequence; breast lesion characterization; lesion signal vari- ability J. Magn. Reson. Imaging 2007;25:82– 88. © 2006 Wiley-Liss, Inc. CONTRAST-ENHANCED (CE) MRI methods have been developed to characterize malignant breast tumors ac- cording to the enhancement dynamics of the tumor (1,2). Although CE-MRI is highly sensitive for detecting breast tumors, it varies in specificity (from 37% to 97%), as reported in several studies (3–5). In breast MRI, T2- weighted sequences alone cannot provide specificity for cancer diagnosis, although they are valuable for dem- onstrating vessels and cysts (6). When used visually in conjunction with CE-MRI, T2-weighted breast imaging sequences have been shown to improve discrimination between benign and malignant lesions (7). However, while T2-weighted images of the breast present significant signal details of both tumors and surrounding breast tissue, their acquisition matrix makes it difficult to make a direct comparison with high-resolution T1-weighted images. T2-weighted im- ages are obtained using a two-dimensional (2D) fast spin-echo (FSE) sequence, which provides high in- plane resolution information with thick slices (5 mm vs. 2 mm for T1-weighted dynamic data). While 3D FSE methods do exist, they use contiguous 3D slabs to in- crease coverage beyond several centimeters, and thus incur long scan times to provide equivalent coverage (eight minutes for only four contiguous 256  256 slices obtained using 3D FSE, vs. five minutes for 10 times more (or 40 256  256 slices) using 2D FSE). Fast imaging employing steady-state acquisition (FI- ESTA) is a 3D imaging pulse sequence that is designed to produce high signal-to-noise ratio (SNR) images at very short repetition times (TRs). The 3D FIESTA pulse sequence is based on the steady-state free precession (SSFP) technique, and is also known as a true fast imaging with steady-state precession (TrueFISP) se- quence. FIESTA uses a short echo time (TE) and short TR approach, and therefore all tissues with a reason- 1 Magnetic Resonance Science Center, University of California–San Francisco, San Francisco, California, USA. 2 General Electric Healthcare–ASL West, Menlo Park, California, USA. 3 Department of Radiology, University of Washington, Seattle Cancer Care Alliance, Seattle, Washington, USA. *Address reprint requests to: C.K., Magnetic Resonance Science Center, University of California–San Francisco, 1 Irving Street, San Francisco, CA 94143-1290. E-mail: klifa@mrsc.ucsf.edu Received August 30, 2005; Accepted August 3, 2006. DOI 10.1002/jmri.20775 Published online 15 December 2006 in Wiley InterScience (www. interscience.wiley.com). JOURNAL OF MAGNETIC RESONANCE IMAGING 25:82– 88 (2007) © 2006 Wiley-Liss, Inc. 82