Active surveillance has been proposed as an option for patients with low-risk prostate cancer in order to reduce the effects caused by overdiagnosis. Delaying treatment and applying it only if there is evidence of progression requires a careful identification of these patients. Prostate-specific antigen (PSA) serum levels lower than 10 μg/L and Gleason score lower than 7 are the main criteria used to select patients for active surveillance based on experience accumulated in the last two decades. In the selection of patients with active surveillance two points are taken into consideration: (a) Gleason score changes introduced by the Consensus Conference of 2005; (b) differences between assays in the measurement of PSA serum levels, in the selection of patients for active surveillance. Improving the accuracy of patient's selection for active surveillance requires that Gleason score reassignment must be taken into account, as well as the harmonization between PSA assays. The use of incorrect results leads to misclassification of patients, undermining the goals of active surveillance.