Assessing Lipid Nanoparticle RNA Delivery Including CRISPR-Cas9 Based Therapy to Bone Marrow Cells with Emphasis on Leukemic Blasts - a Proof-of-Principle Study

Acute myeloid leukemia (AML) with the translocation t(8;21)(q22;q22.1) resulting in the fusion oncogene RUNX1::RUNX1T1 is a well-described subtype of AML. Generally perceived as associated with a favorable prognosis, the main cause of mortality in these patients remain relapse, occurring in an estimated 40% of patients leading to increased mortality. Gene editing technology CRISPR-Cas9, has in previous research been demonstrated to be able to disrupt the RUNX1::RUNX1T1 fusion gene. The disruption leads to inhibited leukemic cell growth and proliferation, suggesting its potential as a future therapeutic in the treatment of AML (Neldeborg et al. PMID: 37464068). One of the main challenges in translating RNA-based technologies, such as CRISPR-Cas9, to clinical testing remains in vivo delivery. Lipid nanoparticles (LNPs) are emerging non-viral vectors capable of delivering RNA-based therapies, such as CRISPR, in vivo. In this study, we investigated LNPs as a potential vector for delivery of RNA, including CRISPR components, to leukemic cell lines as well as to patient peripheral blood (PB) and bone marrow (BM) cells. LNPs were synthesized from SM-102 or D-lin-MC3-DMA, DSPC, Cholesterol and DMG-PEG2000 in molar ratio 50:10:39:1. The LNPs were loaded with: (i) dual guide RNAs (gRNA) targeting introns in RUNX1::RUNX1T1, (ii) Cas9_mRNA, or (iii) GFP_mRNA. Human AML cell line Kasumi-1, RUNX1::RUNX1T1 positive, was used for in vitro experiments. Evaluation of CRISPR cleavage and subsequent non-homologous end-joining, when treated with our dual-gRNA and Cas9-mRNA approach, was evaluated by PCR as well as with digital PCR to assess disruption efficiency. Paired primary patient PB and BM samples from 11 patients and BM from 3 patients evaluated for hematological disease or treatment response were used to assess LNP transfection when treated with LNPs loaded with GFP_mRNA. A multiparameter flow cytometry panel was utilized to quantify GFP expression in different hematopoietic cell subsets. LNP transfection efficiency in Kasumi-1 cells was detected with high GFP expression (>90%). Treatment of Kasumi-1 cells with dual LNP-gRNA and LNP-Cas9_mRNA resulted in consistent disruption of RUNX1::RUNX1T1 albeit at low absolute efficiency as compared to electroporation of the ribonucleoprotein complex. The transfection potential of LNPs was interrogated in an ex vivo analysis of primary patient PB and BM cells. In the BM compartment, high GFP expression was detected in the immature myeloid- and leukemic blast subset (median 50.2% (95% CI 42.4% - 72.2%)), T-cells (median 29.1% (95% CI 19.9% - 37.4%)) and the monocytes (median 58.7% (95% CI 35.2% - 79.0%)). There was no apparent GFP expression detected in the B-cells nor granulocytes. A similar tendency was observed in the PB. Comparing the two LNP constructs, the LNPs synthesized using ionizable lipid SM-102 showed higher efficiency in the immature myeloid- and leukemic blast subset (p=.008) and the T-cells (p=.001) as compared to the LNPs synthesized using D-Lin-MC3-DMA. Our data demonstrates that LNPs can deliver RNA cargo to several types of primary patient cells in PB and BM. Importantly, at high efficiency in the leukemic blast subset. Furthermore, our data indicates that different LNP constructs may change the transfection efficiency. Lastly, we have demonstrated that LNPs can deliver dual-gRNA and Cas9-mRNA to Kasumi-1 cells and disrupt the RUNX1::RUNX1T1 fusion gene. Collectively, our results suggest a future role of LNPs as vectors of RNA-based therapies in hematology, including CRISPR components aimed at disrupting the RUNX1::RUNX1T1 fusion oncogene. Future research efforts will aim to optimize the CRISPR-mediated disruption efficiency of the RUNX1::RUNX1T1 fusion gene when delivered in LNPs in order to elucidate its therapeutic potential.

No relevant conflicts of interest to declare.