Conventional dynamic contrast-enhanced breast (DCE) MRI sequences require a tradeoff between spatial resolution and temporal resolution. A novel DCE-MRI technique named GRASP (Golden-angle Radial Sparse Parallel) MRI uses compressed sensing and parallel imaging to acquire simultaneous high spatial and temporal resolution¹. Continuous golden-angle radial acquisition allows high-resolution data acquired during a single contrast injection to be reconstructed at any desired time point during the acquisition time based on a set of consecutive radial spokes grouped around the given time-point. This technique has been successfully demonstrated for breast imaging at our institution with high-quality, low-artifact images as assessed by breast radiologists^1,2. However, GRASP DCE-MRI has not yet been evaluated in clinical practice for the assessment of lesion conspicuity compared to conventional DCE-MRI with Cartesian sampling. The purpose of this study was to evaluate lesion conspicuity between conventional and GRASP DCE-MRI methods for benign and malignant lesions and by morphology type.This HIPAA-compliant retrospective study was conducted with 32 women with 48 biopsy-proven benign and malignant breast lesions imaged consecutively between March and August 2015. This data set consists of an enriched patient group with indeterminate breast lesions that were subsequently biopsied. All patients underwent a conventional DCE-MRI exam on a 3T whole-body MRI system (Siemens; Tim Trio) with a 7-channel breast coil (Invivo) and a sagittal T1-weighted fat-suppressed 3D volumetric interpolated breath-hold examination (VIBE) sequence (TR/TE, 4.01/1.52; flip angle, 12°; matrix, 384 x 384; field of view, 270 mm; section thickness, 1mm) performed before and after contrast injection. Subsequently, all women underwent MRI-guided breast biopsy within one month using GRASP DCE-MRI with a fat-suppressed radial “stack-of-stars” 3D FLASH sequence with golden-angle ordering (sagittal slab orientation, field of view  = 280 × 280 × 144 mm3, flip angle  = 12°, TR/TE = 3.6/1.47 ms, and bandwidth = 710 Hz/pixel). A total of 2280 spokes were acquired for each of the 35 partitions during free breathing to cover one breast planned for biopsy. Two-fold readout oversampling (512 sample points/spoke) was used to avoid spurious aliasing along each spoke. The reconstructed image matrix size per frame is 256 × 256 × 72 with zero padding along the slice direction. GRASP imaging was reconstructed with 55 spokes (8.3 s/frame) as previously described.² Three breast imagers evaluated each lesion at the second post-contrast time point (approximately 3 minutes) conventional VIBE subtracted images and the corresponding frame of the GRASP subtracted images (Figure 1) for conspicuity (1 to 5 point scale, 1=poor, 5=excellent), lesion morphology, enhancement and BIRADS assessment. Pathologic correlation and statistical analysis was performed.Clinical indications for diagnostic breast MRI were: 30 for staging of known breast cancer and two high-risk screening. 24/48 (50%) biopsied lesions were nonmass enhancement (NME); the other 24/48 (50%) were masses. 23/48 (48%) were malignant (IDC, 7/23 (30%); ILC, 3/23 (13%); DCIS, 13/23 (57%)) (Figure 2). Wilcoxon signed-rank tests were used to evaluate GRASP and VIBE lesion conspicuity; there was no significant difference between the two sequences for all readers (R1 p=0.21, R2 p=0.19, R3 p=0.46) (Figure 3), masses only (R1 p=1.0, R2 p=0.48, R3 p=0.7) or NME (R1 p=0.18, R2 p=0.08, R3 p=0.64). Interclass correlation coefficients demonstrated strong reader agreement in evaluating conspicuity (ICC=0.735; 95% CI: 0.627-0.815). For all three readers, conspicuity scores were higher for masses than NME as assessed by Mann-Whitney U tests (R1=0.047, R2=0.023, R3=0.047). Two of three readers rated benign lesions as less conspicuous than malignant lesions as assessed by Mann-Whitney U tests (R1 p=0.048, R2 p=0.04); the third reader had no significant difference (R3 p=0.26).GRASP DCE-MRI has demonstrated similar spatial image quality to routine DCE-MRI when previously evaluated in body, breast and head/neck imaging^1-4. However, image conspicuity and morphology have not formally been assessed on a per-lesion basis compared to diagnostic DCE-MRI. Breast MRI biopsies utilize compression, which has been shown to reduce conspicuity of both masses and NME⁵. However, our results demonstrate biopsy GRASP images have no significant statistical difference in conspicuity compared to conventional diagnostic DCE-MRI, with excellent inter-reader agreement. Specifically, there was no significant difference between GRASP and conventional VIBE DCE-MRI on a per lesion basis for masses, NME, benign or malignant lesions within this enriched cohort.GRASP DCE-MRI is comparable to conventional DCE-MRI imaging for both masses and NME and provides diagnostic-quality high spatial resolution with flexibility of temporal resolution.