Protection from anti-capsid NAbs mediated by exo-AAV vectors. (A) Protocol outline. Naive male C57BL/6 mice (n = 5) were passively immunized with intraperitoneal (i.p.) injection of human IVIg at a dose of 0.5 mg, 2 mg, or 8 mg per mouse. Twenty-four hours later, animals received 5 × 10¹⁰ vg of exo- or standard AAV8-hF.IX vectors. (B) Anti-AAV8 NAb titer measured before vector administration (C-E) Plasma hF.IX levels over time assessed by ELISA in mice immunized with 0.5 mg (C), 2 mg (D), or 8 mg (E) of IVIg. hF.IX levels are reported as percentage of hF.IX levels in naive mice following gene transfer with standard AAV8-hF.IX vector. The symbols represent individual animals and the bars represent the mean ± SEM. *P < .05; **P < .01; ***P < .001; 2-way ANOVA with Bonferroni posttest. (F) VGCN per diploid genome in liver collected at euthanization from the animals in panels C-E and in naive controls. Data are shown as mean ± SEM. *P < .05; **P < .01; ***P < .001; 1-way ANOVA with Bonferroni posttest. (G) NAb assay performed at increasing IVIg dilutions with exo-AAV8 or exo-AAV8 vectors after treatment with 0.075% Triton-X-100, and standard AAV8 vectors alone or formulated in an excess of fivefold empty capsids. exo-AAV8 vectors were retitered after Triton treatment together with all other vectors used in the experiment. Shown are mean ± standard deviation of 2 independent experiments. ***P < .001; exo-AAV8 vs AAV8 vectors. πππP < .001; ππP < .01; exo-AAV8 vs exo-AAV8 after treatment with 0.075% Triton-X-100. δδδP < .001; δδP < .01; exo-AAV8 vs standard AAV8 vector formulated in an excess of fivefold empty capsids. Two-way ANOVA with Bonferroni posttest. (H) NAb assay performed with individual sera from healthy donors carry low (1:1), mid (1:3.16), or high (1:31.6) at increasing dilutions. Results are shown as percentage of max expression control (no inhibition). Each line represents a single donor.