AAV-FVIII gene therapy dose-response study relates vector design (potency) and dose to initial FVIII exposure kinetics, steady-state levels, and time to loss of FVIII activity. (A) This design schematic depicts the 4 vectors administered to male exon 16–disrupted hemophilia A mice. (B) AAV-FVIII infusion was followed by longitudinal plasma collections and an ET3i protein challenge course for a subset of the AAV-ET3–treated animals. (C) Vectors were administered at threefold dose intervals within the range indicated in the table and ranked based on relative predicted potency. The median FVIII activity over time is shown graphically for the 6E13 vg/kg dose of each vector (ET3 transgene, red lines; HSQ transgene, blue lines; E06 promoter, solid lines; HCB promoter, dashed lines). (D) Terminal VCN was quantified by qPCR on liver DNA (E06.TTR-ET3, red circles; HCB-ET3, red squares; E06.TTR-HSQ, blue upwards triangle; HCB-HSQ, blue downwards triangle). Data points represent the median VCN, and error bars indicate the interquartile range (IQR). Values below the red dashed line indicate undetectable VCN. The dose factor (P = .0004) and vector design factor (P = .0198) are significant, with only the E06.TTR-ET3 vs HCB-HSQ vector design comparison showing significance (P = .0158; main effects 2-way ANOVA with Tukey’s multiple comparisons). There is no difference between the vectors when comparing median VCN values within each individual vg/kg dose group. (E-H) These graphs profile the median FVIII activity (IU/mL) over time for each dose administered of AAV-E06.TTR-ET3 (E) and  AAV-HCB-ET3 (F), AAV-E06.TTR-HSQ (G), or AAV-HCB-HSQ (H), with resulting FVIII levels color coded according to vector dose. FVIII activity (IU/mL) was measured by chromogenic plate assay for all samples assayed in this study. (C,E-H) IQR error bars were removed for visual clarity but are shown for all vectors in supplemental Figures 1 and 2.