Functional Evaluation and Clinical Implications of FLT3 J uxtamembrane Domain Short Variants in Pediatric AML

Introduction: Somatic FLT3 activating mutations are common in AML and occur as internal tandem duplications (FLT3 /ITD) within the juxtamembrane domain (JMD) or in the activating loop of the tyrosine kinase domain (FLT3 /ALM). There is considerable variability in the length of the ITD, varying from 6 to >300 bp. We, and others, have previously demonstrated that ITD length is associated with outcome, where those with longer ITDs based on median ITD length (>48bp) had adverse outcome. However, the underlying etiology of this association has not been established. We hypothesized that longer ITDs are more likely to disrupt the auto-inhibitory properties of the JMD, leading to enhanced receptor phosphorylation and autonomous activation. We aimed to define a biologic threshold for ITD-associated relapse risk based on the length of ITD that causes maximal activation. We also evaluated the functional properties of newly identified single nucleotide variants (SNVs) within the FLT3 JMD.

Methods: To evaluate the impact of FLT3 /ITD length on receptor auto-phosphorylation, FLT3 /ITD clones with a variety of ITD lengths (6, 12, 18, 24, 30, 36, 42, 48, 54 bps) were engineered and expressed in HEK293 cells. Semi-quantitative immunoblotting was used to evaluate the level of phosphorylated FLT3 (pFLT3), and comparison of pFLT3 to non-pFLT3 was calculated and presented as fold change (FC) of pFLT3:FLT3 for the different ITD lengths. We also assessed the functionality of newly discovered JMDSNVs in a similar manner, and further evaluated their response to FLT3 inhibitors. For evaluation of the clinical impact of ITD length, of the 1,022 patients treated on AAML0531, FLT3 /ITD data including ITD length as determined by fragment length analysis and allelic ratio was available from 144 patients and was used for clinical correlation with ITD length.

Results: Initial evaluation of the activation potential of the different ITD lengths showed a correlation between ITD length and FLT3 phosphorylation (Fig 1A). A modest autonomous FLT3 phosphorylation was observed at 6 and 12 bp, with incremental increase in 18bp and 24bp. Maximal pFLT3 of FC was observed at 30 bp, with no significant increases detected with longer ITDs. Therefore, a biologic threshold of 30 bp was used to evaluate the clinical outcomes of patients with short ITDs (S-ITD; <30 bps) and long ITDs (L-ITD; ≥30 bps). Among the 144 FLT3 /ITD+ patients, n=37 had S-ITDs and n=107 had L-ITDs. Overall, based on cooperating cytogenetic and molecular features, a higher number of patients with S-ITDs were considered as low risk as compared to L-ITDs (36% vs. 13%; p=0.003), with a higher incidence of t(8;21) in the S-ITD cohort. There was no correlation observed between ITD length and allelic ratio (Pearson correlation r=0.54). Evaluation of clinical outcome based on ITD length as defined above demonstrated that those with L-ITD had a 5-year overall survival of 47±13% compared to 83±16% in those with S-ITD (p=0.004; Fig1B).

We additionally evaluated the functional significance of newly discovered FLT3 SNVs in the JMD. A total of 20 unique mutations affecting 13 distinct codons were identified. We evaluated the effect of each amino acid variant on FLT3 activation as measured by aberrant pFLT3. Six of the 20 mutations resulted in varying degrees of autonomous pFLT3 (E573D, L576R, V592A, F594Y, F594C, Y559C). Notably, the V592A mutation resulted in potent autonomous receptor activation, with a pFLT3 level greater than that observed the D835 FLT3/ ALM mutation. All variants resulting in autonomous pFLT3 were found to be highly sensitive to the FLT3 inhibitor crenolanib, with IC50s for inhibition of pFLT3 ranging between 1.3-5.6 nM

Conclusion: The JMD is a critical region of FLT3, and disruption through ITDs or SNVs can result in profound dysregulation of the receptor and downstream signaling, thus serving as oncogenic events. We show that longer ITDs result in more potent FLT3 receptor activation and are associated with inferior clinical outcome. FLT3 /ITD and ALMs have recently been shown to be excellent targets for therapeutic intervention with FLT3 inhibitors. Further investigation on the use FLT3 inhibitor therapy should focus on those patients who may derive the most therapeutic benefit from these agents, which may include those with highly damaging JMD variants.

View largeDownload slide

View largeDownload slide

Close modal