We interrogated data from 217 subjects with primary myelofibrosis (PMF) diagnosed before 2001 to identify covariates at diagnosis associated with long-term survival. Forty-one survived more than 25 years. Long-term survivors were more often female, younger than 50 years and less likely to have anemia, splenomegaly, increased blood CD34-positive cells, a JAK2V617F variant, and more likely to have grade-0 bone marrow fibrosis and increased platelets. In multivariable logistic regression analyses, age <50 years, platelet count >450 × 10⁹/L, and grade-0 bone marrow fibrosis independently predicted long-term survival. These co-variate associations were confirmed in Cox regression with landmark analyses. Long-term survivors had a lower cumulative incidence of anemia and thrombocytopenia during follow-up. Identifying long-term survivors at diagnosis provides clinically meaningful prognostic information in PMF and may influence therapy decisions.
Acute myeloid leukaemia (AML) with FLT3 internal tandem duplication (FLT3-ITD) frequently relapses after FLT3 tyrosine kinase inhibitor (TKI) therapy, highlighting the need to define mechanisms of drug tolerance. Here, we identify the mitochondrial complex I core subunit NDUFS2 as a determinant of FLT3 TKI tolerance in FLT3-ITD AML. A metabolism-focused CRISPR-Cas9 sensitiser screen, integrated with transcriptomic profiling of gilteritinib-resistant cells and primary samples, identified NDUFS2 as upregulated in acquired resistance and preferentially expressed in FLT3-ITD AML. NDUFS2 knockdown impaired complex I activity, oxidative phosphorylation, mitochondrial membrane potential and ATP production, thereby promoting apoptosis and sensitising FLT3-ITD AML cells to gilteritinib. Pharmacologic complex I inhibition with IACS-010759 similarly enhanced gilteritinib activity in FLT3-ITD AML cell lines and primary FLT3-ITD samples, activated AMPK-PERK-ATF4-CHOP stress signalling, promoted apoptosis, and reduced leukaemia burden in a patient-derived xenograft model. Genetic NDUFS2 knockdown also prolonged survival in a xenograft model. A preliminary murine safety assessment showed no overt toxicity under the tested short-term dosing schedule. These findings identify NDUFS2-dependent complex I activity as an adaptive metabolic dependency supporting FLT3 inhibitor tolerance and provide mechanistic insight for further preclinical evaluation.
Chronic myeloid leukemia (CML) is driven by the constitutively active BCR::ABL1 kinase, which promotes uncontrolled cell proliferation. Although protein-tyrosine kinase inhibitors (TKIs) have revolutionized CML treatment, resistance remains a significant barrier to curative therapy, particularly due to BCR::ABL1 mutations and the persistence of leukemia stem cells (LSCs). New treatment strategies that directly inhibit BCR::ABL1 protein expression are needed to achieve more effective therapy. Using a high-content antibody microarray and bioinformatics analyses, we investigated differences in the proteome and phosphorylation landscape of Imatinib (IM)-resistant cells and identified highly activated members of the eukaryotic translation initiation complex (eIF4F), particularly eIF4G1, a core component of eIF4F. These eIF4F complex members are highly expressed and activated in CD34+ CML stem/progenitor cells and in IM-resistant cells. Genetic inhibition of eIF4G1 or pharmacological targeting with the small-molecule inhibitor SBI-756 diminished eIF4F complex assembly, reduced overall translation initiation, reduced BCR::ABL1 expression and key downstream targets, impaired the survival of nonresponder stem/progenitor cells, including BCR::ABL1-T315I mutant cells, and sensitized them to TKIs, with strong synergistic effects in vitro and in a PDX model. These findings underscore the potential of eIF4F complex inhibition as a new therapeutic strategy to overcome TKI resistance in CML and other BCR::ABL1+ malignancies.
