Patients were identified from a population-based prospective study of 4990 individuals with symptoms associated with colorectal cancer (CRC). A total of 244 CRC tissue samples were available for immunohistochemical staining of uPAR, semiquantitatively scored at the invasive front, and in the tumor core on cancer cells, macrophages, and myofibroblasts. In addition, the levels of the intact and cleaved uPAR-forms in blood from the same patients are evaluated in this study. In a univariate analysis, the number of uPAR-positive versus uPAR-negative macrophages (HR = 2.26, [95% CI: 1.39-3.66, P = 0.0009]) and cancer cells (HR=1.49, [95% CI: 1.01-2.20, P = 0.047]) located in the tumor core were significantly associated to overall survival. In a multivariate analysis, uPAR-positive versus uPAR-negative macrophages located in the tumor core showed the best separation of patients with positive score associated to poor prognosis (HR = 1.84 [95% CI: 1.12-3.04, P = 0.017]). In a multivariate analysis including clinical covariates and soluble uPAR(I), the latter was significantly associated to overall survival (HR = 2.68 [95% CI: 1.90-3.79, P < 0.0001]) and uPAR-positive macrophages in the tumor core remained significantly associated to overall survival (HR = 1.81 [95% CI: 1.08-3.01, P = 0.023]). Membrane-bound uPAR showed additive effects with the circulating uPAR(I) and stage, giving a hazard ratio of 12 between low and high scores. Thus, combining stage, uPAR(I) in blood and uPAR on macrophages in the tumor core increase the prognostic precision more than tenfold, as compared to stage alone.