Objective: To identify which active surveillance candidates benefit most from confirmatory biopsies to exclude grade underclassification. Materials and Methods: This observational study includes 556 men diagnosed between 2002 and 2015 with Gleason 3 + 3 (GG1) disease on initial diagnostic biopsy, of whom 406 received a confirmatory biopsy within 12 months for active surveillance. Multivariable logistic regression analysis was performed to determine clinicopathologic features associated with Gleason 7 or higher (GG2+) on a confirmatory biopsy. Regression tree analysis was employed to stratify patients into select risk groups. Results: Eighty-five of 406 patients (20.9%) with initially GG1 disease were reclassified to GG2+ on a confirmatory biopsy. On multivariable analysis, increasing age (per year odds ratio 1.07; 95% confidence interval 1.02-1.12; P <.01) and more positive cores at diagnosis (per core, odds ratio 1.37, 95% confidence interval 1.09-1.72; P <.01) were significantly associated with reclassification, independent of prostate volume, clinical stage, initial PSA, or confirmatory biopsy type (including magnetic resonance imaging-targeted approaches or transrectal saturation random sampling). Recursive partitioning demonstrated that age over 73 and 5 or more positive cores were factors associated with the greatest reclassification risk. Conclusion: In our cohort, both advancing age and additional positive cores were associated with increased odds of reclassification to GG2+ on confirmatory biopsy. In men over age 73 or with 5 or more positive cores, a repeat biopsy within 12 months may be particularly beneficial to minimize tumor grade underclassification.