Multiparametric MRI (mpMRI) has repeatedly been shown to have diagnostic utility in primary prostate cancer diagnosis. Its role in active surveillance (AS) is debated and currently there is a lack of prospective studies in this area. The current study aims to assess the utility of serial mpMRI to predict progression to clinically significant prostate cancer (csPCa) in men on AS. Moreover, the oncologic safety of baseline MRI and saturation diagnostic biopsy in replacing early confirmatory biopsy during AS was evaluated.
A total of 172 men were enrolled in this single arm prospective trial (HREC/13/SVH/248). Eligibility included men with biopsy proven ≤cT2 prostate cancer (ISUP grade group (GG) ≤2 with ≤10% Gleason pattern 4 in any core, and maximum ISUP GG 2 in 1 core) who chose AS as their management strategy. Men underwent baseline mpMRI and saturation biopsy at baseline, followed by serial annual mpMRI until a 3-year end point after which they underwent repeat saturation biopsy. The standardized 1-year confirmatory biopsy was omitted and biopsies during the protocol were triggered based on new abnormalities on mpMRI (new PI-RADS 3 lesion or persistent PI-RADS 4/5 lesion) and PSA density >0.2 µg/L.
Baseline median age was 63 (IQR 57-68) years and PSA was 4.6 ng/ml (IQR 3.4-6.6). 92% had ISUP GG 1 prostate cancer and mpMRI was negative (PI-RADS 1-3) in 85% of men. Within 3 years 23% experienced pathological progression. The sensitivity, specificity, positive predictive value and negative predictive value of surveillance mpMRI to detect csPCa was 57% (95% CI 39-74), 82% (95% CI 74-89), 50% and 86%, respectively. Both mpMRI and PSA density were significant predictors for progression during surveillance (MRI OR 6.20, 95% CI 2.72-14.16, p <0.001); PSAD OR 6.19, 95% CI 2.14-17.92, p=0.001). Multivariate analysis illustrated the stronger predictive value of MRI in surveillance for csPCa. Based on MRI results, biopsies could be prevented in 73% of cases at the cost of delaying the diagnosis in 15 patients. Except for one, none of the missed prostate cancer cases by mpMRI had locally advanced pathology (≥pT3b) or metastastic disease after a median of 71 (IQR 61-79) months of follow-up.
The final analysis of this single arm prospective trial confirms our interim result that the 1-year confirmatory biopsy in many active surveillance protocols can be safely omitted. However, mpMRI still misses significant prostate cancer and therefore a standardized 3-year systematic biopsy should be performed.