BackgroundInformation on bleeding phenotype in nonsevere hemophilia may be used to determine target factor levels for prophylaxis or gene therapy in severe hemophilia. ObjectivesTo assess the association between endogenous factor level and bleeding phenotype in children with nonsevere (factor [F]VIII/FIX activity 1%-25%) hemophilia A (HA) and B without prophylaxis. MethodsData on annualized bleeding rate (ABR), annualized joint bleeding rate (AJBR), and onset of bleeding were extracted from the international PedNet cohort including children born since 2000. Mean ABR and AJBR were modeled and compared according to FVIII/FIX endogenous activity (1%-2%, 3%-5%, 6%-10%, 11%-15%, 16%-20%, and 21%-25%) using negative binomial regression. Onset of bleeding was analyzed using Kaplan–Meier survival curves. ResultsEight hundred twenty-five children (40% with moderate hemophilia; 87% with HA) with median follow-up of 7.4 years/child were included. The median age at onset of bleeding and median bleeding rates changed with increasing endogenous activity. From endogenous FVIII 1% to 2% to 21% to 25%, the age at onset of bleeding changed from a median of 1.4 to 14.2 years, ABR from 1.6 to 0.1/y, and AJBR from 0.5 to 0.0/y. From endogenous FIX 1% to 2% to 16% to 25%, the onset of bleeding changed from a median of 1.7 to 6.1 years, ABR from 0.5 to 0.1/y, and AJBR from 0.1 to 0.0/y. The negative correlation between AJBR and factor level was most strongly pronounced up to a factor level of 6% in HA and hemophilia B. ConclusionEndogenous factor activity of >5% was identified as a threshold to significantly lower joint bleeding rate, while FVIII levels >15% and FIX levels >10% were sufficient to achieve the goal of 0 bleeds in this pediatric cohort.