In vivo therapeutic efficacy and selectivity of PIK-75 toward human AML in a mouse xenograft model. (A) AML blasts (n = 8 patient samples) or (B) BM cells from normal donors (n = 3) were plated in methyl cellulose in the indicated concentrations of vehicle (V), PIK-75, or PI-103 and CFU-L) or CFU-GM colonies were counted using an inverted microscope. *P < .0001 with error bars representing standard errors. (C) Primary AML blasts or (D) nontransformed BM from normal donors were incubated in vehicle (V), 1 μM PIK-75 (Δ), or DMSO (▪) for 3 hours, washed, and then engrafted into NOD-SCID mice. Engraftment was quantified after 4 to 6 weeks by assessing the percentage of hCD45+ cells in the BM of recipient mice. Each symbol represents the percentage hCD45+ cells observed in a separate mouse. (E) Mice were injected IP with 10 mg/kg of PIK-75 and then sacrificed at the indicated time points, with peripheral blood being collected by cardiac puncture. The peripheral blood was then diluted 1:2 with RPMI/1% FCS and incubated with MV4;11 cells for 24 hours following which apoptosis was quantified by annexin V staining. Each dot represents the observed apoptosis of MV4;11 cells following incubation with a peripheral blood sample collected from a separate mouse, with the horizontal line representing the mean. (F) NOD-SCID mice were engrafted with HL-60 cells and after disease was established (2 weeks), mice were injected IP with either vehicle (V) or PIK-75 for 3 days following which engraftment was quantified by assessing the percentage hCD45+ cells in the BM of recipient mice. Each symbol represents the percentage hCD45+ cells observed in a separate mouse (*P < .05; **P < .01). NOD/SCID/IL2rγ^null mice were engrafted with (G) HL-60 cells or (H) MV4;11 cells and after disease was established (arrow), mice were injected IP 5 times/week with 1 mg/kg PIK-75. P values were calculated using the log-rank Mantel-Cox test.