Bone Marrow Stromal Cells Protect Both FLT3-ITD and FLT3-WT Primary AML Cells from the Anti-Leukemic Activity of the FLT3 Inhibitor AC220

FTL3 mutations are found in about 30% of AML patients, conferring a leukemic blast growth advantage, drug therapy resistance in the bone marrow (BM) and poor outcome. Mesenchymal stem/stromal cells (MSCs) are essential components of the bone marrow microenvironment, and growing evidence suggest that MSCs play a critical role in AML chemo-resistance, although the molecular mechanisms involved are poorly understood. The purpose of the study was to (1) establish an novel in vitro co-culture system between primary AML blasts and healthy donor BM-MSCs (HD-MSCs) or AML patient-derived MSCs (AML-MSCs), (2) evaluate the impact of culture with BM-MSCs on the sensitivity of AML cells to AC220 using patients samples with FLT3-ITD (n=4) or FLT3-WT (n=3). We first cultured HD-MSCs (n=5) and AML-MSC (n=3) and observed no phenotypical differences (CD14^- CD34^- CD45^- CD73⁺ CD90⁺ CD105⁺), although HD-MSCs grew faster. We evaluated the effect of co-culturing AML samples (n=6) with HD-MSCs or AML-MSCs for 5 and 12 days on leukemic cell growth and found that both types of MSCs significantly and equally enhanced AML cell proliferation while maintaining blast phenotype. Using clonogenic assays on 4 AML specimens cultured alone or with either HD- or AML-MSCs for 5 and 12 days, we found that co-culture with either source of BM-MSCs drastically increased colony-forming cells number at day 5 and day 12 while CFC number decreased in the absence on BM-MSCs (no colonies at day 12 for the 4 samples), indicating that AML co-culture with HD/AML-MSCs supports the survival and/or proliferation of AML stem/progenitor cells. We next assessed the effect of increasing doses of AC220 (1, 10, 50, 100 and 500nM) on the apoptosis of FLT3-ITD (n=3) and FLT3-WT (n=4) AML cells cultured alone or with HD-MSCs. Exposure to AC220 for 72 hours significantly, and in a dose-dependent manner, increased the apoptosis of AML FLT3-ITD cells in monoculture (n=3, 21±1% of Annexin V positive cells for control, AC220 1nM 29±3.7%, 10nM 31±2.5%, 50nM 32±1.5%, 100nM 34±1.7% and 500nM 38±3.6%). In contrast, AML FLT3-ITD cells co-cultured with HD-MSCs were resistant to the drug (n=3, 21±2.6% of Annexin V positive cells for control, AC220 1nM 23±3%, 10nM 22±3%, 50nM 25±5.7%, 100nM 30±8.3% and 500nM 33±9.5%). Interestingly, we found that AML FLT3-WT are much less sensitive to increasing doses of AC220 compared to ITD samples (n=4, 27±3.9% of Annexin V positive cells for control, AC220 1nM 30±6.5%, 10nM 35±14%, 50nM 37±11%, 100nM 39±13% and 500nM 43±11%), and co-culture with BM-MSCs further decreased the sensitivity of AML FLT3-WT cells to AC220-induced apoptosis (n=4, 19±3.2% of Annexin V positive cells for control, AC220 1nM 17±3.9%, 10nM 20±3.4%, 50nM 19±3.7%, 100nM 21±4.5% and 500nM 26±1%). AC220 treatment for 3 days significantly, and in a dose-dependent manner, inhibited CFCs in AML FLT3-ITD (n=4, with 26±8%, 46±6%, 60±9%, 69±10% and 86±3% inhibition with 1, 10, 50, 100 and 500nM of AC220 respectively) while AML FLT3-ITD co-culture with HD-MSCs were less sensitive (n=4, with 9±10%, 30±6%, 42±9%, 57±11% and 72±7% inhibition with 1, 10, 50, 100 and 500nM of AC220, respectively). Similarly to the AC220-induced apoptosis, we observed that AML FLT3-WT CFCs are less sensitive to AC220-induced growth inhibition compared to ITD samples, although a 3 days exposure to AC220 significantly, and in a dose-dependent manner, inhibited AML FLT3-WT CFCs (n=3, with 38±16%, 44±14%, 58±12%, 70±21% and 81±19% inhibition with 1, 10, 50, 100 and 500nM of AC220, respectively). Interestingly, we observed that co-culture of AML FLT3-WT with stromal cells were significantly more resistant to increasing doses of AC220 (n=3, with 22±7%, 36±5%, 43±8%, 46±8% and 57±6% inhibition with 1, 10, 50, 100 and 500nM of AC220, respectively). Altogether, these results suggest that AML FLT3-ITD cells in monoculture are more sensitive to AC220 treatment compared to AML FLT3-WT primary cells, but more importantly, upon interaction with primary HD-MSCs, both WT and FLT3-ITD primary samples are protected from apoptosis and growth inhibition induced by AC220, indicating a critical role for the BM microenvironment in AC220 resistance. We are currently testing the impact of BM-MSCs co-culture on leukemic stem cell sensitivity to AC220 using transplantation in NSG mice. We will also evaluate if this co-culture model can be predictive of the response to in vivo treatment with AC220 in a patient-derived xenograft model.