Effect of TRC105 on the ability of AML/ALL blasts to generate leukemia in a xenograft model. (A) Schematic representation of experimental design. Sublethally irradiated NSG mice were IV injected with 5 × 105 human AML or B-ALL blasts isolated from the BM of primary recipient mice that had been injected with primary human leukemic blasts (in vivo expansion). At day 2, mice were randomly divided into groups (n = 7-8 each) and injected with TRC105 or IgG isotype control. (B-D) Effect of TRC105 treatment on AML development. (B) Representative FACS plots show levels of hCD45 in the PB at 4, 8, and 12 weeks postinjection. (C-D) Percentage of hCD45+ cells in PB (C) and BM (D). Bars represent average percentage of hCD45, and error bars indicate SEM for each cohort. Leukemia development in PB was inhibited at week 4 in the TRC105-injected cohort (red), and remained low by week 12 (C). ***P < .001 by ANOVA. Presence of hCD45+ cells in the BM reveals that leukemia development is impaired in the TRC105 cohort at 8 weeks postinjection (n = 3). However, this effect is no longer observed at week 12 (n = 5) (D). ***P < .001 by Student t test. (E-G) Effect of TRC105 treatment on ALL development. (E) Representative flow cytometric plots show levels of hCD45 in the PB at 4 and 8 weeks postinjection. (F-G) Percentage of hCD45+ cells in PB (F) as well as in BM and spleen (G). Bars represent average percentage of hCD45, and error bars indicate SEM for each cohort. TRC105-treated mice exhibited significantly less leukemia in PB by week 4. However, no difference was observed by week 8 (F). **P < .01 by ANOVA. All groups showed massive leukemic cell infiltration in BM by week 8 (G).
