Optimization of Targeted Lipid Nanoparticle for the Treatment of RUNX1::ETO Driven AML

Almost 70% of pediatric Acute Myeloid Leukemias (AMLs) arise from chromosomal abnormalities. Among these, the RUNX1::ETO fusion gene is present in more than 15% of cases and is responsible for originating and promoting the proliferation of leukemic cells (Obszański et al, 2022; Issa et al, 2023). Conventional therapy approaches lack the ability to target this genetic driver specifically and are associated with a range of long-term side effects, including cardiac and hepatic problems, increased susceptibility to infections and infertility (Brianna et al, 2023). Therefore, there is a need to develop more specific treatments with fewer side effects.

In this context, siRNA therapeutics stand out for their high specificity towards the targeted mRNA (Iwakawa et al, 2022). Lipid Nanoparticles (LNPs) are currently the most effective delivery system for RNA-based therapeutics. However, challenges such as poor endosomal escape, immunogenicity and insufficient accumulation in the organ of interest hinder their full potential (Dowdy, 2023; Chen et al, 2021; Simonsen, 2024). In this work, we investigate how the LNP composition influences the physicochemical properties (size, polydispersity index, surface charge and shelf life) and its RUNX1::ETO knockdown performance in patient derived xenografts (PDXs) and primary cells in an ex vivo setting.

To address these challenges, we began by modifying the lipid composition of a Standard LNP based on the marketed Onpattro formulation, incorporating a tripeptide, LDV (Leu-Asp-Val), to target the Very Late Antigen-4 (VLA-4) receptor present on all hematopoietic cells (Swart et al, 2023). To reduce undesired immune recognition, typically associated with the presence of polyethylene glycol (PEG) on the surface of the LNPs, we replaced it with a biodegradable polypeptoid named polysarcosine (pSar) (Nogueira et al, 2020). To improve endosomal escape, we substituted the cholesterol present in the formulation with the plant-based analogue β-sitosterol, which has been shown to enhance the endosomal escape of RNAs (Patel et al, 2020). The addition of pSar in the Standard LNP formulation achieved similar knockdown performance across the range of siRNA doses tested (10 - 150 nM) compared to its PEG counterpart. The β-sitosterol inclusion, improved knockdown efficacy already at a 10 nM dose. However, β-sitosterol and the post-insertion of the LDV peptide impaired particle stability, leading to lipid precipitation within five days after LNP production.

We restored LNP stability by incorporating sphingomyelin as the helper lipid and adjusting the lipid molar ratio. Furthermore, the replacement of PEG for pSar in our lead formulation allowed us to double its molar percentage conferring higher stability without compromising the knockdown efficacy. The optimized LNP formulation achieve a knockdown of 70% in PDXs and 50% in primary cells after a single dose, resulting in the complete loss of the RUNX1::ETO protein. Moreover, our prime LNP maintained its physicochemical properties and knockdown activity after one month of storage at 4°C. Currently, in vivo studies are ongoing to examine the pharmacokinetic and pharmacodynamic properties of this novel LNP formulation.

Our results highlight the importance of fine-tuning the LNP composition to achieve an optimal balance in stability and therapeutic effectiveness. The formulation developed in this work demonstrates superiority over a Standard LNP formulation (Onpattro). These advancements are especially significant for treating fusion gene-driven leukemias, offering the potential for more effective RNA-based therapies with enhanced targeting and reduced side effects. This is an encouraging step forward in improving outcomes for leukemia patients.