AbstractEmicizumab is approved for prophylaxis of patients with hemophilia A (HA). Despite its efficacy in reducing bleeding, some patients on emicizumab still experience hemarthrosis, but no tool is yet available to identify those at a higher risk of spontaneous joint bleeding. This study aimed to evaluate whether laboratory measurements (global coagulation assays and emicizumab concentration) and/or arthropathy scores can distinguish patients at higher risk of spontaneous joint bleeding while on emicizumab prophylaxis. A thrombin generation assay was performed upon the addition of tissue factor and synthetic phospholipids. Nonactivated thromboelastography was performed on citrated whole blood. Emicizumab concentrations were measured using a modified 1-stage factor VIII assay. The degree of hemophilic arthropathy was assessed using the Hemophilia Joint Health Score and Hemophilia Early Arthropathy Detection with Ultrasound (HEAD-US) score. A Cox proportional hazards model was used to evaluate the association between variables and bleeding. The predictive power of these variables was investigated using receiver operating characteristic (ROC) analysis. Forty patients with severe HA, with or without inhibitors, on emicizumab prophylaxis were enrolled in an observational cohort study. Ten of 40 developed spontaneous joint bleeding. None of the laboratory parameters were able to distinguish patients with a higher risk of spontaneous joint bleeding. ROC analysis showed that during emicizumab prophylaxis, only the presence of synovitis and a higher HEAD-US score were associated with spontaneous joint bleeding (area under the curve, 0.84). A greater degree of arthropathy and the presence of synovitis could help predict the risk of spontaneous joint bleeding in patients with HA on emicizumab prophylaxis.