Document Type

Article

Publication Date

10-13-2019

Comments

This article is the author’s final published version in Therapeutic Advances in Urology, Volume 11, October 2019, Pages 1-9.

The published version is available at https://doi.org/10.1177/1756287219882809. Copyright © Leong et al.

Abstract

Background: We examine the practical application of multiparametric MRI (mpMRI) prostate biopsy data using established pre-RP nomograms and its potential implications on RP intraoperative decision-making. We hypothesize that current nomograms are suboptimal in predicting outcomes with mpMRI targeted biopsy (TBx) data.

Materials and methods: Patients who underwent mpMRI-based TBx prior to RP were assessed using the MSKCC and Briganti nomograms with the following iterations: (1) Targeted (T) (targeted only), (2) Targeted and Systematic (TS) and (3) Targeted Augmented (TA) (targeted core data; assumed negative systematic cores for 12 total cores). Nomogram outcomes, lymph node involvement (LNI), extracapsular extension (ECE), organ-confined disease (OCD), seminal vesicle invasion (SVI), were compared across iterations. Clinically significant impact on management was defined as a change in LNI risk above or below 2% (Δ2) or 5% (Δ5).

Results: A total of 217 men met inclusion criteria. Overall, the TA iteration had more conservative nomogram outcomes than the T. Moreover, TA better predicted RP pathology for all four outcomes when compared with the T. In the entire cohort, Δ2 and Δ5 were 16.6–25.8% and 20.3–39.2%, respectively. In the subset of 190 patients with targeted and systematic cores, TA was a better approximation of TS outcomes than T in 71% (MSKCC) and 82% (Briganti) of patients.

Conclusion: In established pre-RP nomograms, mpMRI-based TBx often yield variable and discordant results when compared with systematic biopsies. Future nomograms must better incorporate mpMRI TBx core data. In the interim, augmenting TBx data may serve to bridge the gap.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 License

PubMed ID

31662795

Language

English

Share

COinS