Canonical NF-κB Signalling Is a Potential Target in FLT3/ITD AML.

Abstract 2447

Internal tandem duplication (ITD) of the fms-like tyrosine kinase 3 (FLT3) receptor is common in acute myeloid leukemia (AML) and is associated with a dismal outcome. Despite initial response, FLT3/ITD AMLs often relapse early, suggesting a residual population of resistant leukemia stem cells (LSC). Clinically, FLT3 inhibitors asmonotherapy have yet to improve outcome significantly and therefore, targeting additional pro-survival pathways may be necessary for this group of AML patients. Mice genetically egineered to express a hemizygous FLT3/ITD mutation develop a progressive, fatal, myeloproliferative neoplasm. Lin⁻cells isolated from the bone marrow of FLT3/ITD or control mice were subjected to gel shift analysis using a radio-labeled NF-kB binding site. This analysis demonstrated high levels of nuclear activation of NF-kB in the FLT3/ITD-expressing cells, suggesting its activation downstream of mutant FLT3 signaling. MV4–11 is a human AML-derived cell line harboring a homozygous FLT3/ITD mutation. Cells expressing high levels of aldehyde dehydrogenase (ALDH) have been shown to be enriched for LSC in primary AML samples and cell lines. High ALDH expressing MV4–11 cells were isolated using FACS and analyzed for NF-kB activation. Western blot analysis demonstrated preferential phosphorylation of NF-kB p65 by activated IKK on Ser536 in this subpopulation, compared with cells with low ALDH activity. These findings indicate activation of NF-kB in MV4–11 LSCs.

We wanted to next test the requirement for NF-kB signaling in transformation by FLT3/ITD mutations. NF-kB p65 null mice die in utero. We therefore established C57BL/6 p65(flox/flox);Mx1-Cre mice. Intra-peritoneal injection of pIpC every other day for 7 doses efficiently deletes the RelA/p65 gene, resulting in expression of <1% of the corresponding RNA or protein. Despite effective excision of p65, the mice survive. Bone marrow cells harvested from control or p65(del/del) mice were transduced with a FLT3/ITD-expressing lentivirus and seeded in methylcellulose without cytokines. Equal transduction rate was verified by measurement of GFP expression by flow cytometry. Reproducibly, p65(del/del) marrow transduced with FLT3/ITD was ineffective in forming cytokine independent colonies, in contrast to wild-type marrow (5 +/− 0.6 vs. 55 +/− 6 colonies per 1E5 cells, P<0.001), and the few p65(del/del) colonies that resulted were smaller than those from p65 expressing wild-type marrow cells. Cells transduced with a lentiviral vector expressing GFP but not FLT3/ITD did not form colonies without cytokines, and p65(del/del) marrow formed normal numbers of colonies of normal size and distribution in the presence of IL-3, IL-6, and SCF.

Sorafenib inhibits FLT3 signaling and kills MV4–11 cells with an IC₅₀ of approximately 10 nM. Reproducibly, a sub-toxic dose of sorafenib (5 nM) combined with sub-toxic levels of the IKKb inhibitor IMD-0354 (400 nM) resulted in synergistic cell killing as indicated by the calculated combination index of 0.55.

Currently, clinical efforts in FLT3/ITD leukemia concentrate on FLT3 inhibition alone. Our data suggest that canonical NF-kB may be an important pathway in FLT3/ITD AML and that simultaneously targeting FLT3 and NF-kB in this disease may be an effective approach.