Diffusion-weighted magnetic resonance imaging detects significant prostate cancer with high probability

Lauren J Bains; Urs E Studer; Johannes M Froehlich; Gianluca Giannarini; Maria Triantafyllou; Achim Fleischmann; Harriet C Thoeny

doi:10.1016/j.juro.2014.03.039

Diffusion-weighted magnetic resonance imaging detects significant prostate cancer with high probability

J Urol. 2014 Sep;192(3):737-42. doi: 10.1016/j.juro.2014.03.039. Epub 2014 Mar 15.

Authors

Lauren J Bains¹, Urs E Studer², Johannes M Froehlich¹, Gianluca Giannarini², Maria Triantafyllou¹, Achim Fleischmann³, Harriet C Thoeny⁴

Affiliations

¹ Department of Radiology, Neuroradiology and Nuclear Medicine, Institute of Diagnostic, Interventional and Pediatric Radiology, University of Bern, Bern, Switzerland.
² Department of Urology, University of Bern, Bern, Switzerland.
³ Institute of Pathology, University of Bern, Bern, Switzerland.
⁴ Department of Radiology, Neuroradiology and Nuclear Medicine, Institute of Diagnostic, Interventional and Pediatric Radiology, University of Bern, Bern, Switzerland. Electronic address: harriet.thoeny@insel.ch.

PMID: 24641913
DOI: 10.1016/j.juro.2014.03.039

Abstract

Purpose: We prospectively assessed the diagnostic accuracy of diffusion-weighted magnetic resonance imaging for detecting significant prostate cancer.

Materials and methods: We performed a prospective study of 111 consecutive men with prostate and/or bladder cancer who underwent 3 Tesla diffusion-weighted magnetic resonance imaging of the pelvis without an endorectal coil before radical prostatectomy (78) or cystoprostatectomy (33). Three independent readers blinded to clinical and pathological data assigned a prostate cancer suspicion grade based on qualitative imaging analysis. Final pathology results of prostates with and without cancer served as the reference standard. Primary outcomes were the sensitivity and specificity of diffusion-weighted magnetic resonance imaging for detecting significant prostate cancer with significance defined as a largest diameter of the index lesion of 1 cm or greater, extraprostatic extension, or Gleason score 7 or greater on final pathology assessment. Secondary outcomes were interreader agreement assessed by the Fleiss κ coefficient and image reading time.

Results: Of the 111 patients 93 had prostate cancer, which was significant in 80 and insignificant in 13, and 18 had no prostate cancer on final pathology results. The sensitivity and specificity of diffusion-weighted magnetic resonance imaging for detecting significant PCa was 89% to 91% and 77% to 81%, respectively, for the 3 readers. Interreader agreement was good (Fleiss κ 0.65 to 0.74). Median reading time was between 13 and 18 minutes.

Conclusions: Diffusion-weighted magnetic resonance imaging (3 Tesla) is a noninvasive technique that allows for the detection of significant prostate cancer with high probability without contrast medium or an endorectal coil, and with good interreader agreement and a short reading time. This technique should be further evaluated as a tool to stratify patients with prostate cancer for individualized treatment options.

Keywords: diagnosis; differential; magnetic resonance imaging; observer variation; prostate; prostatic neoplasms.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Adult
Aged
Aged, 80 and over
Diffusion Magnetic Resonance Imaging / statistics & numerical data*
Humans
Male
Middle Aged
Observer Variation
Prospective Studies
Prostatic Neoplasms / diagnosis*
Reproducibility of Results