Evaluation of diffusion-weighted MR imaging at inclusion in an active surveillance protocol for low-risk prostate

Diederik M Somford1, Caroline M Hoeks, Christina A Hulsbergen-van de Kaa

  • 1Department of Urology, Canisius-Wilhelmina Hospital, Radboud University Nijmegen Medical Center, The Netherlands. r.somford@cwz.nl

Investigative Radiology
|January 19, 2013
PubMed
Abstract

Insights

Diffusion-weighted MRI apparent diffusion coefficient (ADC) can guide prostate cancer biopsies in patients on active surveillance. This method helps identify high-grade cancer not suitable for surveillance, improving patient selection.

Area of Science:

  • Radiology
  • Oncology
  • Urology

Background:

  • Active surveillance (AS) is a management strategy for low-risk prostate cancer (PCa).
  • Accurate identification of high-grade Gleason components is crucial for AS suitability.
  • Multiparametric magnetic resonance imaging (MP-MRI) aids in detecting cancer-suspicious regions (CSRs).

Purpose of the Study:

  • To evaluate if apparent diffusion coefficient (ADC) from diffusion-weighted MRI can guide magnetic resonance-guided biopsy.
  • To determine if ADC can identify patients with high-grade Gleason components unsuitable for AS.

Main Methods:

  • Fifty-four patients with low-risk PCa underwent 3-T MP-MRI for AS.
  • Magnetic resonance-guided biopsy was performed on all identified CSRs.
  • Median ADC (mADC) was calculated for each CSR, and statistical analyses were conducted.

Main Results:

  • Mean mADC was significantly lower in CSRs with PCa (1.04 × 10⁻³ mm²/s) compared to those without (1.26 × 10⁻³ mm²/s; P < 0.001).
  • CSRs with high-grade Gleason components showed a lower mADC (0.84 × 10⁻³ mm²/s) than low-grade CSRs (1.09 × 10⁻³ mm²/s; P < 0.05).
  • The diagnostic accuracy of mADC for predicting PCa presence in CSRs was 0.73 (AUC).

Conclusions:

  • Median ADC effectively predicts the presence of prostate cancer in CSRs identified by MP-MRI.
  • ADC measurements can aid in grading PCa within CSRs, informing AS decisions.