Active surveillance (AS) is often restricted to patients with low-risk prostate cancer (PCa) with three or fewer positive cores. We aimed to identify predictors of adverse pathology for low-risk PCa treated with radical prostatectomy (RP) and to determine if a threshold number of positive cores could help the decision process for AS.

A total of 3,359 men with low-risk PCa underwent RP between January 2000 and August 2016. We analyzed the relationship between biopsy core features and adverse pathology at RP, defined as grade group (GrdGrp) ≥3, seminal vesicle invasion (SVI), or lymph node involvement (LNI).

Of the 171 (5.1%) patients with adverse pathology at RP, 144 (4.3%) were upgraded to GrdGrp ≥3, 31 (0.9%) had SVI, and 15 (0.4%) had LNI. Prostate-specific antigen and age were the only predictors of adverse pathology; there was no significant association with number of positive cores, total millimeters of cancer, or maximum percentage of cancer in any core. When we expanded the definition of adverse pathology to include GrdGrp 2 and extraprostatic extension, the association between core features and outcome, while statistically significant, was clinically weak, with no evidence of threshold effects.

There is little basis for excluding patients with otherwise low-risk PCa on biopsy from AS based on criteria such as the number of positive cores or maximum cancer involvement of biopsy cores.