Knockout of TNF leads to improved survival, reduced BM infiltration, and increased myelopoiesis in mice. (A) Kaplan-Meier survival plot of NSG mice injected with ROSA^{KIT D816V} cells with CRISPR/Cas9-mediated TNF knockout (blue) compared with a nontargeting control (NTC; red) with retained TNF secretion. (B-D) Workup of NSG mice that were injected with ROSA^{KIT D816V} cells with or without TNF knockout and sacrificed after 5 weeks. Infiltration of ROSA^{KIT D816V} cells was assessed by flow cytometric analysis of flushed BM cells for human CD45 expression (B). BM formalin-fixed paraffin-embedded sections (femur) were stained with hematoxylin and eosin (HE) and antibodies against NIMP R-14 (Ly-6G/-6C) (staining murine myelopoiesis). Representative sections are shown in panel C and the fraction of NIMP R-14–positive cells in the BM in panel D. (E-F) Murine BM cells were preincubated with etanercept or infliximab (10 μg/mL) or remained untreated for 4 hours. Human recombinant TNF (100 ng/mL) was then added and active caspase-3–positive cells were measured after 72 hours by flow cytometry (E). Complementary viable cells were evaluated after 24 hours of incubation using annexin V and 4′,6-diamidino-2-phenylindole staining (F). (G) Human BM mononuclear cells were seeded at a density of 100 000 cells per well and treated with human recombinant TNF (100 ng/mL) and/or etanercept and infliximab, and the proliferation rate was evaluated by ³H-thymidine incorporation assay. (H) CFU-GM formation of BM mononuclear cells obtained from patients with KIT D816V⁺ SM compared with those from normal controls (n = 3). Results represent the mean and standard error of the mean of 3 independent experiments. ∗P < .05; ∗∗P < .01; ∗∗∗P < .001; ∗∗∗∗P < .0001.