INTRODUCTION: Up to 50% of patients will have disease reclassification while on active surveillance (AS) for their prostate cancer. Determining which patients will have reclassification that will impact their survival is difficult. We investigated clinicopathologic factors associated with disease reclassification and differences in both overall and metastasis free survival between those treated and those remaining on AS. MATERIALS AND METHODS: We performed a retrospective review of patients who were enrolled in an AS protocol between 1994 and 2000. Inclusion criteria for AS were: < cT2a disease, PSA < 10 ng/mL, < 50% of single core involvement, and Gleason score < 7, as well as sufficient follow up for evaluation (at least 1 subsequent transrectal ultrasound guided biopsy after initial diagnosis). RESULTS: There were 102 patients that met the inclusion criteria with median age of 70 years (IQR 68-73), follow up of 9.25 years (IQR 6.1-12.2) and time to disease reclassification of 4.7 years (IQR 2.8-7.9). Only prostate-specific antigen (PSA) density ≥ 0.15 was a significant predictor of disease reclassification with a hazard ratio of 5.5 (95% confidence interval 2.3-13.4, p < 0.01). There was no significant difference in metastasis free and overall survival between patients who received treatment and those that continued on AS despite reclassification of disease; this remained true even while stratifying patients by age ≥ 70 compared to those < 70 years old. CONCLUSIONS: PSA density is a significant predictor of disease reclassification and AS remains a safe option for patients with low risk prostate cancer with up to 10 years of follow up.