The use of biparametric magnetic resonance imaging in active surveillance of prostate cancer

Abstract

Objectives

This study aims to evaluate the performance of biparametric magnetic resonance imaging (bpMRI) compared to multiparametric MRI (mpMRI) for active surveillance (AS) in prostate cancer (PCa), focusing on lesion detection, PI-RADS classification changes and grade reclassification (GR) rates.

Methods

We retrospectively reviewed our institutional AS database for men who underwent mpMRI followed by bpMRI. Lesion counts and PI-RADS classification were compared. Second, a subgroup of men who underwent mpMRI, biopsy, bpMRI and then repeat biopsy (‘bpMRI biopsy group’) was matched to a control group who underwent mpMRI, biopsy, another mpMRI and then repeat biopsy (‘mpMRI biopsy group’) based on age, PSA density, year of diagnosis and number of positive cores. GR rates at systematic and targeted biopsy of all PI-RADS 3–5 lesions were compared, as were PI-RADS changes.

Results

A total of 129 men with a median of 29 months between mpMRI and bpMRI were included. mpMRI identified 77 PI-RADS 3–5 lesions versus 114 on subsequent bpMRI (0.60 vs. 0.88 lesions/patient, p = 0.01); however, lesion distribution was similar, with rates of PI-RADS 3, 4 and 5 lesions of 53%, 40% and 6% on mpMRI and 53%, 37% and 11% on subsequent bpMRI (p = 0.87, 0.29 and 0.23, respectively). In the matched biopsy subgroups, when comparing men who had two surveillance mpMRIs with those who had a surveillance mpMRI followed by a surveillance bpMRI, lesion stability was not significantly different across PI-RADS categories. GR at last biopsy occurred in 29% of the bpMRI biopsy group versus in 25% of the mpMRI biopsy group (p = 0.7).

Conclusions

bpMRI appears safe to incorporate into AS over the short term, as it was noninferior to mpMRI with similar GR rates and lesion stability.