FLT3-ITD Mutation Expands HSPC Compartment to Initiate AML Development

Acute myeloid leukemia (AML) originates from the expansion of hematopoietic stem and progenitor cells (HSPCs). Sequential mutations accumulate in HSPCs, conferring enhanced self-renewal and abnormal proliferative capabilities. Mutations in signal transduction genes such as FLT3 and NRAS, typically occurred in the late stages in leukemogenesis, provide a proliferative advantage to pre-leukemic cells. However, we unexpectedly discovered that FLT3-ITD mutations show a wide range of variant allele frequencies (VAFs) (0.006-0.989) from analyzing 1807 AML patients in public cohorts (BeatAML, etc.). Those with high VAFs (cutoff=0.333), accounted for about 18% of patients with FLT3-ITD, are associated with poor prognosis. We hypothesize that apart from its conventional role in promoting proliferation, FLT3 mutations may serve as an initiating event in HSPC to drive AML development. To explore this, we selected 16 FLT3-ITD mutant AML patients, including 8 high VAF (H-VAF) and 8 low VAF (L-VAF) patients. We conducted single-cell DNA and protein sequencing using the Tapestri platform (Mission Bio) to analyze the genotype and immunophenotype profiles of these patients.

We first reconstructed the clonal architecture of each patient based on the mutation and copy number variation status of each cell. Subsequently, using unsupervised clustering, we identified 18 cell types based on the expression of 44 surface markers, including CD34+ HSPCs and lineage-committed cells (monocytes etc.). We observed a significant difference in the evolutionary patterns between two groups (P=0.001). The H-VAF exhibited linear evolution, and FLT3-ITD emerges in the dominant clone (8/8). In contrast, most of the L-VAF showed a branched evolution pattern (7/8), characterized by increased clonal diversity as indicated by the greater Shannon index (1.11 vs. 0.67, P=0.008), and subclones carrying multiple FLT3-ITDs or other signal transduction mutations. Notably, in each of the 5 patients from H-VAF, the founding clone harbored FLT3-ITD along with several known early driver mutations (DNMT3A, IDH2, SRSF2), with a cellular composition of 10.6% lymphoid lineages (T/B cells, etc.), significantly higher than that of its descendant clones (10.6% vs 0.5%, P=0.0002). Strikingly, in the remaining 3 H-VAF patients, the founding clone (6.7-9.0% of all cells) harbored only FLT3-ITD without any additional driver mutations. The cellular identities of these cells with single FLT3-ITD were characterized by a mixed population of approximately 29% HSPCs, 21% myeloid lineages (monocytes, DCs, etc.), and 50% lymphoid lineages. These data indicate that FLT3-ITD can be an initiating event in leukemogenesis originating from HSPCs and can maintain multiple differentiation potentials.

To investigate the impact of initiating FLT3-ITD on AML development, we assessed the immunophenotype of each clone during clonal evolution. In 3 patients with single FLT3-ITD mutant clone, initiating FLT3-ITD accumulation led to an expansion of HSPC compartment compared to the wildtype cells (29% vs 5.6%, P=0.003). Subsequent acquired mutations, such as NPM1, further expand the GMP subset characterized by CD34+CD38+CD45RA+CD123+. Conversely, in L-VAF, subclones with FLT3-ITD mutation retained the same cell type composition as their ancestral clone. These data suggest that initiating FLT3-ITD provides a cellular context with significantly expanded HSPCs for AML development. Additionally, the H-VAF exhibited a higher proportion of cells marked by leukemia stem cell markers, such as CD69 (41.2% vs 11.9%, P=0.008), which was further confirmed using FACS analysis in primary AML patient samples. To assess the leukemia initiating capacity of the AML cells in two groups, we selected 4 patients from each group and transplanted their primary cells into NSG mice. Mice engrafted with cells from the H-VAF demonstrated higher engraftment efficiency (100% vs 42.9%, P=0.03) and shorter latency compared to those from the L-VAF (75.4 vs 102.3±11.2 days, P=0.04), suggesting a robust leukemia-initiating capacity and greater disease aggressiveness in the H-VAF group.

In summary, our single-cell data and functional experiments demonstrate that FLT3-ITD can expand the HSPC compartment and initiate AML development, unveiling a previously unappreciated role of FLT3-ITD in HSPCs beyond promoting cell proliferation.

No relevant conflicts of interest to declare.