Multiparametric MR Imaging with High-Resolution Dynamic Contrast-enhanced and Diffusion-weighted Imaging at 7 T Improves the Assessment of Breast Tumors: A Feasibility Study

Abstract

Purpose: To ascertain whether multiparametric magnetic resonance (MR) imaging of the breast in combination with dynamic contrast material-enhanced (DCE) imaging and diffusion-weighted imaging (DWI) at 7 T is feasible and improves diagnostic accuracy.

Materials and methods: From December 2011 to December 2013, 40 patients with suspicious breast lesions were included in this institutional review board-approved prospective study. Before bilateral multiparametric MR imaging of the breast at 7 T, all patients gave written informed consent. Lesions were classified according to Breast Imaging Reporting and Data System (BI-RADS) and assessed for apparent diffusion coefficient (ADC) values by two readers independently. For combined analysis of DCE MR imaging and DWI, the BI-RADS-adapted reading algorithm, which adapted ADC thresholds to the BI-RADS assessment category, was used. Diagnostic values of multiparametric, DCE MR imaging, and DWI were calculated. Receiver operating characteristic curve analysis was performed. Image quality and interreader agreement were assessed. Histopathologic results were used as the highest standard.

Results: There were 29 malignant and 17 benign lesions (range, 6-95 mm; mean, 23.3 mm). Multiparametric MR imaging yielded a sensitivity of 100% (29 of 29 lesions), a specificity of 88.2% (16 of 18 lesions), and an area under the curve of 0.941, which was greater than for DCE MR imaging (P = .003), which had a sensitivity of 100% (29 of 29 lesions), a specificity of 53.2% (nine of 17 lesions), and an area under the curve of 0.765. DWI had a sensitivity of 93.1% (27 of 29 lesions), a specificity of 88.2% (15 of 17 lesions), and an area under the curve of 0.907. Multiparametric MR imaging at 7 T of the breast eliminated all false-negative findings and reduced false-positive findings, from eight false-positive findings with DCE MR imaging to two false-positive findings. Thus, if used clinically, 7-T multiparametric MR imaging may have potentially obviated unnecessary breast biopsies in six of eight lesions (P = .031). Multiparametric MR imaging demonstrated either excellent or good image quality and interreader agreement (κ = 0.89-1.00).

Conclusion: The clinical use of 7-T multiparametric MR imaging is feasible, provides good or excellent image quality, and has the potential to improve diagnostic accuracy.