Combined Use of Ziftomenib and Selinexor Is Effective in NPM1 Mutant Acute Myeloid Leukemia

Introduction: Menin scaffolds the oncogenic histone-lysine-N-methyltransferase MLL1 (KMT2A)-fusion protein (FP) complex in MLL-rearranged (MLL-r) AML. In NPM1-mutant AML (NPM1c), MLL1 is an oncogenic regulator of HOXA9 and MEIS1, promoting self-renewal of myeloid progenitor cells. Treatment with the orally bioavailable menin inhibitor (MI) ziftomenib disrupts binding its interaction with MLL1-FP and MLL1/2, leading to differentiation and apoptosis of AML cells expressing MLL-FP or NPM1c. Early results from a Phase 1 study of ziftomenib monotherapy in R/R NPM1-m AML demonstrated promising activity, supporting the investigation of combination strategies in the R/R setting.

In NPM1-mutant AML, the aberrant mislocalization of NPM1 in the cytoplasm contributes to leukemogenesis, partly by dysregulating PU.1, a master transcription factor. Selective inhibition of nuclear export (SINE) using an XPO1 inhibitor (selinexor) can retain NPM1 within the nucleus along with PU.1, restoring its interaction with CEBPA/RUNX1 and toggling their switch back to activators of granulocyte and monocyte terminal differentiation genes. We previously demonstrated synergy in KMT2Ar models using the combination of ziftomenib and selinexor. To build on these independent mechanistic reasons for activity, we investigated the effects of this combination in NPM1-mutant AML models both in vitro and in vivo.

Methods: An ATP-based cell proliferation assay was performed to assess cell viability and growth inhibition. CalcuSyn Version 2.0 synergy software was used to evaluate the synergy score. Stem-like progenitor cells were isolated using the StemSpan CD34+ expansion kit (StemCell Tech). Colony formation efficiency was determined using a Methocult assay (StemCell Tech). Gene and protein expression and cell death were detected using quantitative real-time PCR, western blotting, and annexin-V/PI-stained flow cytometric analysis. We used ziftomenib and selinexor in PDX models.

Results: Ziftomenib in combination with selinexor synergistically inhibited the growth of NPM1^MT AML cell lines (IMS-M2 and OCI-AML3) (CI<1). The combination significantly suppressed colony formation of CD34+ progenitor cells and was not toxic to normal CD34+ human hematopoietic progenitor stem cells. Western blot analysis showed ziftomenib to downregulate the expression of menin, HOXA5, HOXA10, MEIS1, PBX3, and other cofactors. The nuclear export inhibitor selinexor also reduced HOXA5, HOXA10, PBX3, and MEIS1 protein levels. The combined use of MI and XPO1 inhibitors resulted in further downregulation of menin, HOXA5, HOXA10, MEIS1, and PBX3, providing a molecular basis for the synergy between the two compounds. RT-PCR showed significant downregulation of HOXA9, HOXA10, MEIS1, STAT5B, and PBX3 for the combination. Both agents together also significantly increased apoptotic cell death in NPM1-mutated AML cells compared to the effect of each agent alone. Nuclear cytoplasmic protein fractionation and western blot study showed enrichment of NPM1c in the nucleus with XPO1 inhibitor and the combination. In NPM1-mutant PDX experiments, we found that co-treatment with ziftomenib and selinexor resulted in a more significant reduction in AML burden than treatment with either agent alone or vehicle control. Extramedullary sites, including the liver and spleen, showed a significant reduction of CD34+ leukemic cells in the combination arm in post-necropsy samples from PDX mice. Ongoing experiments with the NPM1-mutant CDX model OCI-AML3, the impact of the drugs on transcriptomics in NPM1-mutant models, and immunofluorescence studies will be reported.

Conclusion: This preclinical study supports the robust synergy of ziftomenib and XPO1 inhibition beyond KMT2A-r and could be a potential strategy for treating NPM1^-mutant AML. The synergy we demonstrated in KMT2A-r and NPM1-mutant models sets a premise for translational potential.