<b><i>Background:</i></b> Endostatin is a 20-kDa C-terminal fragment of collagen XVIII, known for its ability to inhibit the proliferation of capillary endothelial cells. Previous studies suggested that circulating endostatin independently predicts incident chronic kidney disease. However, the impact of endostatin on graft loss level in kidney transplant recipients (KTRs) remains unknown. <b><i>Methods:</i></b> We conducted a prospective observational cohort study in 574 maintenance KTRs. Patients were followed for kidney graft loss and all-cause mortality during a median follow-up of 48 months. Serum-, and urine-samples and clinical data were collected at baseline. Serum Endostatin concentration was analyzed by an ELISA. <b><i>Results:</i></b> Among 574 patients, 37 patients had graft loss and 62 patients died. For graft loss, the optimal cut-off value based on receiver operating characteristics analysis (area under the curve 0.79, 95% CI 0.71–0.86, <i>p</i> &#x3c; 0.001) of endostatin was 147.3 pmol/L. Kaplan-Meier curves revealed that higher serum endostatin concentrations positively correlated with graft loss (<i>p</i> &#x3c; 0.001). Multivariable Cox regression analyses showed that baseline endostatin concentrations were significantly associated with graft loss after adjusting for graft loss risk factors (adjusted hazard ratio [HR] 8.34; 95% CI 2.19–31.72; <i>p</i> = 0.002). The adjusted HRs for classical graft loss risk factors such as baseline estimated glomerular filtration rate and urinary protein excretion were lower (1.91 and 5.44, respectively). In contrast to graft loss, baseline endostatin concentrations were not associated with all-cause mortality. <b><i>Conclusion:</i></b> Increased serum endostatin at baseline is independently associated with the risk of graft loss in KTRs.