Treatment with atovaquone as a single agent delays engraftment and reduces disease burden in a PDX model. NSG mice were injected by tail vein with 1.2 × 10⁶ cells from primary patient sample AML3. Treatment with atovaquone (200 mg/kg per day; n = 7 [female, n = 4; male, n = 3]) or VC (n = 9 [female, n = 5; male, n = 4]) by daily oral gavage began on the day of cell injection. (A) Representative dot plots of human CD45 and CD33 at 2 time points (days 72 and 106) are shown for 1 vehicle-treated and 1 atovaquone-treated mouse. Depicted mice were females from the same litter and were cohoused. (B) Composite flow cytometric data for human CD45 and CD33 for all treated mice demonstrated decreased peripheral blood disease burden in atovaquone-treated mice. Bars represent means and standard errors of the mean (P < .0001 by analysis of variance). (C) Survival analysis of the same cohort of mice demonstrated improved survival in atovaquone-treated mice compared with vehicle. *P < .05 by Student t test.