Upcoming event

Selecting patients with favourable-risk, GG2 prostate cancer for active surveillance: Does MRI have a role?

  • T. Stonier,
  • A.L. Tin,
  • D.D. Sjoberg,
  • G. Jibara,
  • A.J. Vickers,
  • S. Fine,
  • J. Eastham

Publication: Journal of Urology, November 2020


The National Comprehensive Cancer Network (NCCN) recommends that selected men with GG2 (Grade Group 2) prostate cancer be considered for active surveillance (AS). However, selecting which patients with GG2 disease can be safely managed with AS remains controversial.


The aim of this study was to evaluate the association of multiparametric MRI (mpMRI) with adverse pathology in the radical prostatectomy (RP) specimen of men with favourable-risk GG2 prostate cancer, which could help select patients for AS.

Design, Setting, and Participants

We undertook a retrospective cohort study at our institution to identify patients with favourable GG2 disease who underwent RP between 2010 and 2019. Preoperative mpMRI studies were scored as negative (no identifiable lesion), positive (identifiable lesion(s)), and equivocal.

Outcome Measurements and Statistical Analysis

We defined a multivariable logistic regression model with mpMRI score as the predictor and adverse pathology (upstaging to T3a/b disease, upgrading ≥Grade Group 3, or lymph node invasion) as the outcome, adjusting for preoperative PSA, biopsy Gleason grade, clinical stage, and number of negative and positive prostate biopsy cores. Secondary outcomes of biochemical recurrence (BCR), Grade Group upgrading alone, and the added value of incorporating mpMRI data into the nomogram were also investigated.


and Limitations: We identified 1,117 patients with favourable-risk GG2 disease who underwent RP. Positive mpMRI was associated with higher rates of adverse pathology (odds ratio 2.55; 95% CI 1.75–3.40; p <0.0001) and upgrading (odds ratio 3.89; 95% CI 2.00–7.56; p <0.0001). However, as our study included only GG2 patients who underwent RP, our cohort may represent a higher-risk group than GG2 patients as a whole. Adding mpMRI results to a standard prediction model led to higher net benefit on decision curve analysis.


An identifiable lesion on mpMRI is associated with an increased risk of aggressive pathological features in the RP specimen of patients with favourable-risk GG2 prostate cancer who are potential AS candidates. This information could be used to inform biopsy strategy, counsel patients on treatment options, and guide strategies for those on AS.