AML-MRC is uniquely sensitive to Lip-C6. Human patient samples of AML-MRC (n = 15) (A) or DN-AML (n = 15) (B) were exposed for 48 hours to 20 µM Lip-C6 or control nanoliposomes without ceramide (Lip-Ghost). Apoptosis within the leukemia stem cell (CD34⁺CD38^–) and bulk leukemia fractions were quantified by flow cytometry (1-way analysis of variance [ANOVA] with Tukey’s post hoc comparison). *P ≤ .0042 compared with controls, **P < .0001 compared with controls. (C) Colony-forming assays evaluating dose-response of Lip-C6 were performed using bone marrow samples harvested from wild-type, transgenic AML-MRC, or transgenic DN-AML mice. *P ≤ .0071, **P ≤ .0001, and ***P ≤ .0468 for mouse AML-MRC compared with both wild-type and DN-AML (2-way ANOVA with Tukey’s post hoc comparison). Error bars represent the standard error of the mean (SEM). Mouse wild-type, n = 3; mouse DN-AML, n = 4; and mouse AML-MRC, n = 4. (D) Therapeutic efficacy of Lip-C6 as a standalone treatment was evaluated using transgenic mouse models of AML-MRC (Nup98-HoxD13) and DN-AML (Flt3-ITD). Mice were treated for 10 days with daily injections of Lip-C6 (11.6 mg/kg) or Lip-Ghost (volume-matched). Mice were then euthanized, and bone marrow was isolated and prepared for flow cytometry in which Gr-1⁺CD11b⁺ cells were evaluated and quantified as representative of leukemia burden. *P = .045 (n = 4 per group; unpaired Student t test with Welch’s correction) comparing the leukemia burden in the bone marrow of AML-MRC (Nup98-HoxD13) transgenic mice treated with Lip-C6 with those treated with control nanoliposomes without ceramide (Ghost). See supplemental Figures 4 and 5 for representative blood smears and bone marrow flow cytometry histograms. (E) Colony-forming assays evaluating a dose-response of Lip-C6 were performed using human mononuclear cells obtained from patients with AML-MRC or DN-AML. ^#P = .047 and ^##P ≤ .003 for human AML-MRC compared with DN-AML (2-way ANOVA with Sidek’s post hoc comparison). Error bars represent the SEM; human DN-AML, n = 6; human AML-MRC, n = 5. (F-G) Patient samples were exposed for 24 hours in vitro to 10 µM Lip-C6, and the metabolism of C6-ceramide was evaluated by lipidomic analysis. (F) Comparing the ratios of C6-ceramide prodeath to neutral prosurvival metabolites in AML-MRC and DN-AML (unpaired Student t test with Welch’s correction; *P = .0081). Error bars represent the SEM; n = 4. (G) Patient AML-MRC preferentially converts C6-ceramide to the prodeath metabolite sphingosine as well as endogenous/physiological ceramides rather than neutral or prosurvival metabolites (S1P, C6-sphingomyelin [SM], C6-cerebrosides [includes both C6-glucosylceramide and C6-galatosylceramide]) (1-way ANOVA with Tukey’s post hoc comparison; *P < .0001 compared with all other metabolites; **P ≤ .0294 compared with endogenous ceramides, C6-SM, and C6-cerebrosides; ^#P < .0001 compared with C6-SM; ^##P < .0001 compared with all other metabolites; ^$P = .028 compared with C6-cerebrosides; ^$$P ≤ .0219 compared with S1P, C6-SM, and C6-cerebrosides; ^&P ≤ .0014 compared with S1P and C6-cerebrosides; ^0026amp;&P ≤ .0057 compared with C6-SM and C6-cerebrosides). Error bars represent the SEM; n ≤ 4. (H) Metabolism of C6-ceramide to proapoptotic or neutral or prosurvival metabolites. AC, acid ceramidase; CERS, ceramide synthases; GCS, glucosylceramide synthase; SMS, sphingomyelin synthases; SPHK, sphingosine kinase.