Prostate Cancer Diffusion-Weighted Magnetic Resonance Imaging: Does the Choice of Diffusion-Weighting Level Matter?

Background Diffusion-weighted magnetic resonance imaging plays an important role in multiparametric assessment of prostate lesions. The derived apparent diffusion coefficient (ADC) could be a useful quantitative biomarker for malignant growth, but lacks acceptance because of low reproducibility. Purpose To investigate the impact of the choice of diffusion-weighting levels (b-values) on contrast-to-noise ratio and quantitative measures in prostate diffusion-weighted MRI. Study Type Retrospective and simulation based on published data. Subjects Patient cohort (21 men with Prostate Imaging-Reporting and Data System (PI-RADS) version 2 score >= 3) from a single-center study. Field Strength/Sequence 3 T/diffusion-weighted imaging with single-shot echo-planar imaging. Assessment Both clinical data and simulations based on previously acquired data were used to quantify the influence of b-value choice in normal peripheral zone (PZ) and PZ tumor lesions. For clinical data, ADC was determined for different combinations of b-values. Contrast-to-noise ratio and quantitative diffusion measures were simulated for a wide range of b-values. Statistical Tests Tissue ADC and the lesion-to-normal tissue ADC ratios of different b-value combinations were compared with paired two-tailed Student's t-tests. A P-value Findings about b-value dependence derived from clinical data and from simulations agreed with each other. Provided measurement was limited to two b-values, simulation-derived optimal b-value choices coincided with PI-RADSv2 recommendations. For two-point measurements, ADC decreased by 15% when the maximum b-value increased from 1000 to 1500 seconds/mm(2), but corresponding lesion-to-normal tissue ADC ratio showed no significant change (P = 0.86 for acquired data). Simulations with three or more measurement points produced ADCs that declined by only 8% over this range of maximum b-value. Corresponding ADC ratios declined between 2.6% (three points) and 3.8% (21 points). Simulations also revealed an ADC reduction of about 19% with the shorter echo and diffusion time evaluated. Data Conclusion The comprehensive assessment of b-value dependence permits better formulation of protocol and analysis recommendations for obtaining reproducible results in prostate cancer diffusion-weighted MRI. Level of Evidence 4 Technical Efficacy Stage 2