FLT3 and KIT Mutated Pediatric Acute Myeloid Leukemia (AML) Samples Are More Sensitive In Vitro to the Tyrosine Kinase Inhibitor SU11657.

Novel treatment strategies to improve the outcome of pediatric AML are required. Around 30% of pediatric AML patients harbor a mutation in the tyrosine kinases FLT3 (±20%) or KIT (±10%). Patients with FLT3 and KIT mutations have a poor prognosis. It was reported that FLT3/ITD length and allelic ratio (AR) influence prognosis. Tyrosine kinase inhibitors (TKI) are novel drugs specifically targeting activated tyrosine kinases. SU11657 is a selective TKI of FLT3 and KIT. SU11657 is comparable to SU11248 (sunitinib, Sutent®), which is FDA approved for the treatment op gastro-intestinal stromal tumors (GIST) and renal cell carcinoma. In a phase I trial of sunitinib in AML, all 4 patients with FLT3 mutations had complete (n=1) or partial morphologic responses (n=3) compared with only 20% (2/10) of patients with WT FLT3. All responses were of short duration. In this study we investigated whether pediatric AML samples were sensitive to SU11657 in vitro, and whether sensitivity to SU11657 was related to mutations in FLT3 and KIT.

We studied 77 pediatric AML samples for FLT3/ITD, FLT3 D835 and KIT exon 8 and 17 mutations. In case of a FLT3/ITD mutation the ITD length and AR were determined. All 77 samples were also tested for in vitro sensitivity to SU11657 using the 4 day MTT assay (concentration range 0.0098 – 10μM). Two measures of sensitivity were calculated: 1. The LC50 value (the concentration at which 50% of the cells is killed); 2. The percentage of cells surviving (CS) at 0.625 μM SU11657. The study population consisted of 49 boys and 28 girls. Fifty-five samples were taken at initial diagnosis and 22 at relapse. Median age at diagnosis was 9.0 years and median white blood cell count 78.8×10⁹/L. FAB types were M0 5×, M1 9×, M2 11×, M3 3×, M4 21×, M5 14×, M7 2× and unknown 12×. A FLT3/ITD mutation was detected in 22/77 (29%), a FLT3 D835 mutation in 6/71 (8%) and a KIT exon 17 mutation in 4/55 samples (7%). No KIT exon 8 mutations were detected.

There was an approximately 1000 fold difference in LC50 values between the most sensitive and most resistant sample to SU11657. WT FLT3 and KIT samples were relatively resistant to SU11657 (median CS at 0.625 μM SU11657=91%). However, FLT3/ITD positive samples were significantly more sensitive to SU11657 [median CS at 0.625 μM=66% (p<0.0001)], as well as the FLT3 D835 mutated samples [median CS at 0.625 μM=64% (p=0.004)]. There was no relation between the AR or ITD length and sensitivity to SU11657 (Spearmans ρ=−0.11 (p=0.7) and −0.04 (p=1.0), respectively). The 4 KIT mutated samples also were significantly more sensitive to SU11657 than WT FLT3 and KIT samples [median CS at 0.625 μM=70% (p=0.049)].

In conclusion, there was large interpatient variation in in vitro sensitivity to SU11657. FLT3 and KIT mutated pediatric AML samples were more sensitive to SU11657 than samples with WT FLT3 and KIT. There was no relation between FLT3/ITD AR or ITD length and sensitivity to SU11657. Further clinical evaluation of SU11657 or sunitinib combined with chemotherapy, would be of interest since a third of pediatric AML patients potentially would be sensitive.