Login to Access Video or Poster Abstract: MP70-03
Sources of Funding: None

Introduction

Multi parametric Magnetic Resonance Imaging (mp-MRI) is the most common imaging modality used for detecting residual cancer following Partial Gland Ablation (PGA) for Prostate Cancer. The aim of this study was to assess the diagnostic accuracy of mp-MRI to detect residual cancer in the ablated area following PGA and to assess whether a biopsy of the ablated area may be avoided based on MRI.

Methods

Prospectively maintained institutional database was queried to identify 65 patients with primary PGA from May 2010 to February 2016 and had a postoperative MRI and a control biopsy within 18 months after the procedure. All the MRIs were reviewed by a single radiologist blinded to the biopsy results and visible abnormalities in the ablated area were scored based on a 5-point Likert scale and PIRADS V2. Univariate generalized estimating equation regression with an exchangeable correlation structure to account for within patient correlation was used to test the association between a high MRI score (Likert or PIRADS score ?3) and any- (Gleason ?6) and high-grade (Gleason ?7) disease detected on biopsy of the ablated area.

Results

After exclusions, we identified 38 post-PGA MRI and biopsy pairs with 6 patients having 2 MRI/biopsy combinations. Any-grade disease was detected on 15 biopsies (39%; 95% CI 24%, 57%) and high-grade on 2 biopsies (5.3%; 95% CI 0.6%, 18%). Patient with MRI lesion of Likert and PIRADS score ?3 in the ablated area were 8 (21%) and 7 (18%) respectively. Both the 2 instances of high-grade disease detected on biopsy occurred among the 8 MRIs with a Likert score ?3 (25%; 95% CI 3.2%, 65%) and 7 MRIs with a PIRADS score ?3. No high-grade disease was detected among participants with a Likert or PIRADS score <3, the upper bound of a 97.5% exact confidence interval was 11.6% and 11.2% based on Likert score (n=30) and PIRADS Score (n=31) respectively. One patient had a PIRADS score of 5 and had a Gleason 6 disease detected on biopsy. We did not find sufficient evidence to suggest an association between a PIRADS score and any-grade disease (p = 0.6) nor between Likert score and any-grade disease (p = 0.9), the association was not estimable as we had no participants with high-grade disease with an MRI score less than 3.

Conclusions

This preliminary study was unable to identify an association between MRI scores and presence of prostate cancer in the treated area following PGA therapy. Men with MRI scores of <3 should not be precluded from biopsy given that it is plausible that the rate of high-grade disease is meaningfully high.

Funding

None