Diagnosis of relevant prostate cancer using supplementary cores from magnetic resonance imaging-prompted areas following multiple failed biopsies

Abstract

Objectives: To establish the value of MRI in targeting re-biopsy for undiagnosed prostate cancer despite multiple negative biopsies and determine clinical relevance of detected tumors.

Materials and methods: Thirty-eight patients who underwent MRI after 2 or more negative biopsies due to continued clinical suspicion and later underwent TRUS-guided biopsy supplemented by biopsy of suspicious areas depicted by MRI were identified. Diagnostic performance of endorectal 3T MRI in diagnosing missed cancer foci was assessed using biopsy results as the standard of reference. Ratio of positive biopsies using systematic versus MRI-prompted approaches was compared. Gleason scores of detected cancers were used as surrogate for clinical relevance.

Results: Thirty-four percent of patients who underwent MRI before re-biopsy had prostate cancer on subsequent biopsy. The positive biopsy yield with systematic sampling was 23% versus 92% with MRI-prompted biopsies(p<0.0001). Seventy-seven percent of tumors were detected exclusively in the MRI-prompted zones. Sensitivity, specificity, positive predictive value, negative predictive value and accuracy of MRI to provide a positive biopsy were 92%, 60%, 55%, 94% and 71%, respectively. The anterior gland and apical regions contained most tumors; 75% of cancers detected by MRI-prompted biopsy had Gleason score≥7.

Conclusions: Clinically relevant tumors missed by multiple TRUS-guided biopsies can be detected by a MRI-prompted approach.

Figure 1. Example of suspicious MRI findings in a patient with a large anterior tumor missed by systematic sampling but detected on the MRI-prompted biopsy

A, Axial T2-weighted fast spin echo image of the mid-third of the prostate. Tumor is not clearly demonstrated with this sequence. B, Corresponding color-coded DCE 3D T1-weighted gradient-echo image. The large anterior tumor is easily seen as a cluster of bright red pixels (arrows) with an asymmetric distribution compared to the contralateral side. C, Whole-mount histopathology proven Gleason 4+3 cancer (area delineated by the dotted line). Note that whole-mounts were available in selected patients as part of a large clinical trial, but were not used as the gold-standard.

Figure 1. Example of suspicious MRI findings in a patient with a large anterior tumor missed by systematic sampling but detected on the MRI-prompted biopsy

A, Axial T2-weighted fast spin echo image of the mid-third of the prostate. Tumor is not clearly demonstrated with this sequence. B, Corresponding color-coded DCE 3D T1-weighted gradient-echo image. The large anterior tumor is easily seen as a cluster of bright red pixels (arrows) with an asymmetric distribution compared to the contralateral side. C, Whole-mount histopathology proven Gleason 4+3 cancer (area delineated by the dotted line). Note that whole-mounts were available in selected patients as part of a large clinical trial, but were not used as the gold-standard.

Figure 1. Example of suspicious MRI findings in a patient with a large anterior tumor missed by systematic sampling but detected on the MRI-prompted biopsy

A, Axial T2-weighted fast spin echo image of the mid-third of the prostate. Tumor is not clearly demonstrated with this sequence. B, Corresponding color-coded DCE 3D T1-weighted gradient-echo image. The large anterior tumor is easily seen as a cluster of bright red pixels (arrows) with an asymmetric distribution compared to the contralateral side. C, Whole-mount histopathology proven Gleason 4+3 cancer (area delineated by the dotted line). Note that whole-mounts were available in selected patients as part of a large clinical trial, but were not used as the gold-standard.

Figure 2. Location and Gleason score of the detected tumors

The vertical line separates right from left and the horizontal lines separate the anterior from posterior portions of the prostate at the base, mid-gland and apex. Green, tumors detected by MRI only; blue, tumor exclusively detected by systematic biopsy; pink, tumors detected by both approaches. While 4/13 (31%) were located in the posterior peripheral zone, there were 9/13 (69%) in the anterior gland, 9/13 (69%) in the apex, 2/13 (15%) in the mid-third of the gland and 2/13 (15%) in the base. The majority of the cancers detected (10/13, 77%) had a Gleason score ≥7.