Introduction

Prostate Cancer (PC) overdiagnosis and overtreatment is a major concern for clinicians and policy makers. Multiparametric MRI (mpMRI) and the PCA3 urine test aim to limit this by identifying fewer cases of indolent cancer and more clinically significant cases. We explore whether the utility of the tests can be maximized by combining them for a group of patients with previous prostate biopsies.

Methods

We collected clinicopathologic data from all patients that underwent a urine PCA3 test from 2011 to June 2016 at the University Health Network at The University of Toronto in accordance with ethics committee approval. This included patients on active surveillance (AS) for low-risk PC and those without PC with previous negative biopsies and suspicion of occult, significant disease primarily based on rising PSA. We explored whether age, PSA, PCA3, mpMRI, DRE, family history and prostate size predicted for clinically significant prostate cancer on repeat biopsy as defined by Epstein criteria. We then stratified patients by mpMRI and PCA3 result to detect whether any particular combination of these test has exemplary negative predictive value (NPV) and considered the optimal sequence of tests.

Results

470 patients met inclusion criteria with median (IQR) age and PSA of 62.5 ng/mL (58-68) and 6.3 (4.6-8.8), respectively. PCA3 was abnormal (≥35) in 32.5% of cases. 18.8% of men had a positive family history and 5.6% had suspicious DRE. Epstein criteria or worse PC was identified in 26.3% of cases. In the multivariate model, only age (OR 1.08, 95%CI 1.01-1.16), mpMRI score 4 (OR 16.6, 95%CI 3.9-70.0) or 5 (OR 28.3, 95%CI 5.7-138), and PCA3 (OR 2.9, 95%CI 1.0-8.8) predicted for clinically significant PC on biopsy. No patients with a negative mpMRI and normal PCA3 test were found to have clinically significant PC on biopsy (0 of 26, 100% NPV for double negative test, p<0.0001). Using mpMRI as the initial test diminishes the number of overall tests (11 fewer tests per 100 patients), adds spatial information for targeted biopsy when available, but is more expensive than starting with PCA3 test for all patients.

Conclusions

Both PCA3 and mpMRI are useful tests for predicting clinically significant PC on repeat prostate biopsy. In the 1 of 6 patients in our cohort with double negative tests no clinically significant PC was found on biopsy, which raises the question whether biopsy can be avoided in this group altogether. This study is limited by its retrospective design and selection bias. A prospective trial at our centre is currently ongoing examining this question for patients on AS.

Funding

none