Targeting leukaemia at its root: a lipid nanoparticle platform for fusion gene-specific siRNA delivery
The chromosomal translocation t(8;21) is found in 15% of all pediatric acute myeloid leukaemia (AML) cases and generates the RUNX1/ETO fusion protein. It initiates and maintains leukaemogenic transformation by blocking differentiation and promoting self-renewal. Currently used chemotherapy regimens are accompanied by serious side effects, both short-term and long-term. Although direct therapeutic targeting of RUNX1/ETO has the potential of providing an effective and substantially less toxic treatment approach, developing small molecular weight compounds for such fusion transcription factors has been unsuccessful. Inhibiting fusion gene expression by RNA interference has been an important tool to functionally investigate mechanisms of such fusion genes including RUNX1/ETO. Importantly, its knockdown caused durable changes in its transcriptional networks and impaired leukaemic propagation both in vitro and in vivo by inducing G1 cell cycle arrest, senescence and differentiation.
However, siRNA has very poor pharmacokinetic properties including poor stability, fast renal clearance and inefficient cell uptake. To address these issues we are exploring lipid nanoparticle (LNP) approaches for a safe and efficacious delivery of siRNAs in vivo.
LNPs were prepared using microfluidics mixing techniques, containing siRUNX1/ETO (siRE) or a mismatch control (siMM). The efficacy was examined in cell lines, patient-derived xenografts (PDX) and primary material. Toxicity and on target effect was tested in vivo in two mouse models (Kasumi-1 and RL048 PDX). The knockdown of RUNX1/ETO as well as its impact on downstream gene expression was validated on mRNA and protein level both in vitro and in vivo. To assess leukaemic propagation we performed colony formations assays, FACS analyses and orthotopic re-transplantation assays using immunodeficient mice.
Application of LNPs reduced RUNX1/ETO mRNA and protein levels more than threefold in Kasumi-1 cells. This was associated with reduced proliferation, loss off clonogenicity and G1 cell cycle arrest and senescence. Application of LNPs reduced RUNX1/ETO mRNA and protein levels more than twofold in both primary AML and patient-derived xenografts (PDX).
Using an orthotopic xenotransplantation model, we found that intravenously injected LNPs accumulated in leukaemia-relevant tissues including liver, spleen, and bone marrow. Colocalisation studies demonstrated that the LNPs reached the leukaemic cells. Pharmacodynamics analyses proved RUNX1/ETO knockdown in all animals examined with accordingly changed target gene expression levels. Sequential treatment of transplanted animals with LNPs significantly prolonged survival (p<0.001) compared to animals treated with LNPs/siMM. Interestingly, leukaemic cells obtained from treated animals showed severely reduced proliferation and clonogenicity, and substantially impaired engraftment capacity and leukaemia propagation in secondary recipients suggesting long-term impaired self-renewal capacity upon transient RUNX1/ETO depletion.
Direct and specific inhibition of RUNX1/ETO interferes with leukaemic self-renewal and propagation. LNP-mediated siRNA delivery is a promising new approach for specific targeting of fusion gene depended cancers.