An initial negative round of targeted biopsies in men with highly suspicious multiparametric magnetic resonance findings does not exclude clinically significant prostate cancer-Preliminary experience

Abstract

Background: Targeted prostate biopsies are changing the landscape of prostate cancer (PCa) diagnosis with the degree of suspicion on multiparametric magnetic resonance imaging (mpMRI) being a strong predictor of targeted biopsy outcome. Data regarding the rate and potential causes of false-negative magnetic resonance imaging-transrectal ultrasound (MRI-TRUS) fusion-targeted biopsy in patients with highly suspicious mpMRI findings are lacking.

Objectives: To determine the rate of clinically significant PCa detection in repeat targeted biopsy or surgery in patients with highly suspicious mpMRI findings and in an initial negative MRI-TRUS fusion-targeted biopsy.

Materials and methods: In this single-center, retrospective study of prospectively generated data, men with highly suspicious lesions (Likert 5 score) on mpMRI and an initial negative MRI-TRUS fusion-targeted biopsy were reviewed. The rate of PCa detection in a subsequent MRI-TRUS fusion-targeted biopsy or radical prostatectomy was determined. Tumors in the intermediate- and high-risk groups according to the National Comprehensive Cancer Network criteria were considered clinically significant.

Results: A total of 32 men with 38 Likert 5 lesions were identified. Repeat targeted biopsy or surgery detected cancer in 42% (16/38) of the Likert 5 lesions with initial negative targeted biopsy. Most of these cancers were intermediate- (69%; 11/16) or high-risk (25%; 4/16) tumors.

Conclusion: A negative round of targeted biopsies does not exclude clinically significant PCa in men with highly suspicious mpMRI findings. Patients with imaging-pathology disagreement should be carefully reviewed and considered for repeat biopsy or for strict surveillance.

Keywords: Biopsy; Diagnosis; Early detection; Magnetic resonance imaging; Prostate cancer.

Copyright © 2017 Elsevier Inc. All rights reserved.

Figures

Figure 1. Eligibility criteria and patient cohort

Figure 2. Clinically significant prostate cancer missed by a first round of targeted MRI-TRUS fusion biopsies

in a 67-year-old man with elevated PSA (8 ng/mL) and 3 previous negative systematic 12-core biopsies. MRI for biopsy planning revealed a highly suspicious lesion prospectively assigned a Likert 5 score in the midline and left paramedian anterior gland with homogeneous low signal intensity on T2-weighted images (dashed line in A), significantly restricted diffusion (B) with low values on the apparent diffusion coefficient map and abnormal contrast kinetics (not shown). Biopsy map generated during the initial targeted biopsy revealed 3 cores (green in C) that seemed to have properly sampled the suspicious area (yellow in C) however no cancer was found. Review of biopsy session data revealed inadequate automated segmentation of the prostate boundaries by the fusion biopsy software (D; yellow line represents the automated segmentation used by the system and green line represents the actual, manually post hoc edited prostate boundaries after the imaging-pathology discrepancy was identified). A repeat targeted biopsy was performed and revealed 3 out of 3 positive cores with Gleason 4+3 disease (E). Whole mount radical prostatectomy specimen confirmed site-concordant Gleason 4+3 tumor (dashed line in F)