Biparametric Quantitative MRI for Prostate Cancer Detection

Area of Science:

• Radiology
• Oncology
• Medical Imaging

Background:

• Prostate cancer (PCa) detection in biopsy-naïve men often relies on imaging interpretation.
• Current methods like Prostate Imaging Reporting & Data System (PI-RADS) have limitations in accuracy.

Purpose of the Study:

• To prospectively evaluate a novel quantitative biparametric prostate magnetic resonance imaging (MRI) protocol for PCa detection.
• To compare the accuracy of fractional order calculus (FROC) diffusion and quantitative T2 imaging against PI-RADS.

Main Methods:

• A prospective pilot study enrolled 50 biopsy-naïve men undergoing 3T MRI with T2 and diffusion-weighted imaging (DWI).
• Fractional order calculus (FROC) diffusion modeling was applied to DWI data (b-values: 50-4,000 s/mm2).
• Men with PI-RADS lesions ≥3 underwent targeted and systematic biopsy; analysis focused on detecting Gleason grade group ≥2 (GG≥2) PCa.

Main Results:

• The study included 48 men, with a mean age of 61.5 years; 52% were African American.
• The area under the curve (AUC) for detecting GG≥2 PCa was 0.63 for PI-RADS, 0.82 for ADC, and 0.87 for the FROC model.
• Prostate cancer (GG≥2) was detected in 7% of PI-RADS 3, 28% of PI-RADS 4, and 36% of PI-RADS 5 lesions.

Conclusions:

• A novel quantitative biparametric MRI protocol is feasible for detecting prostate cancer in biopsy-naïve men.
• The FROC diffusion model demonstrated superior accuracy compared to PI-RADS for PCa detection.