Timo Soeterik

71 Impact of MRI on Risk Classification As extraprostatic extension is a strong independent predictor for biochemical recurrence following radical prostatectomy, 21,22 presence of stage ≥T3a on mpMRI can potentially be a valuable prognostic factor indicating the potential need for treatment intensification (eg, non-nerve sparing prostatectomy and additional pelvic lymph node dissection). In this cohort, mpMRI stage ≥T3a was not associated with decreased progression-free survival when compared with mpMRI stage ≤T2. Graphically, most noticeable difference was seen in the intermediate-risk group (Figure S2B). However, this difference was not statistically significant ( p = 0.099). Absence of an association may be explained by either a lack of prognostic relevance of non-organ-confined disease on mpMRI or the fact that patients with stage ≥T3a on mpMRI were accurately assigned to a more intensive treatment regimen (eg, non-nerve sparing during RARP). The second hypothesis is the most attainable explanation, as this is an observational study wherein patients are assigned to certain treatments based on their prognostic risk factors. In addition, present study was limited by a relatively small sample size and short follow-up duration, and significant findings could still occur following study prolongation. Moving forward, further research is needed regarding the harms and benefits of mpMRI as a staging tool. Despite the relevance of our findings, some important limitations need to be acknowledged. First, a limitation of our study is that we did not account for the potential impact of mpMRI target biopsy on stage migration. Only the highest ISUP grade found on either systematic or either mpMRI-targeted biopsy was registered. Thereby, there were differences regarding use of prebiopsy MRI between the hospitals. Fact that prebiopsy MRI was not performed standard in all patients may limit the generalisability of our results to cohorts in which prebiopsy MRI is standardly performed, as upstaging occurs less in hospitals who use prebiopsy mpMRI potentially resulting in lower rates of biopsy sampling error compared to only systematic biopsies. 23 Other limitations of our study are the heterogeneity of mpMRI protocols used, the wide variety of urologists and residents performing the DRE and radiologists performing mpMRI reading. However, the heterogeneity reflects a real-world situation and as we used multi-centre data, our results might provide an accurate overview of the clinical impact of mpMRI local staging in newly diagnosed patients. 4