Safety and efficacy study of CRISPR/Cas9 treatment of sickle cell disease in clinically relevant conditions highlights disease-specific response

Reactivation of fetal hemoglobin (HbF) expression through clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9-mediated disruption of regulatory elements involved in {gamma}-globin gene repression is a promising gene therapy strategy for the treatment of sickle cell disease (SCD). However, preclinical studies aimed at optimizing the genome editing process and evaluating the safety of the editing strategy are necessary to translate this approach to the clinics. This is particularly relevant in the context of SCD, a disease characterized by inflammation, which can affect hematopoietic stem and progenitor cells (HSPCs), the target cell population in gene therapy approaches for hematopoietic disorders. Here, we describe a genome editing strategy leading to therapeutically relevant reactivation of HbF expression by targeting the binding sites (BSs) for the leukemia/lymphoma related factor (LRF) transcriptional repressor in the HBG1 and HBG2 {gamma}-globin promoters. Electroporation of Cas9 ribonucleoprotein and single guide RNA (sgRNA) targeting the HBG promoters in healthy donor (HD) and patient-derived HSPCs resulted in a high frequency of LRF BS disruption and potent HbF synthesis in their erythroid progeny differentiated in vitro and ex vivo after transplantation into immunodeficient mice. LRF BS disruption did not impair SCD and HD HSPC engraftment and differentiation, but was more efficient in SCD than in HD cells. However, SCD HSPCs showed a reduced engraftment and a myeloid bias compared to HD cells. Importantly, in HSPCs, we detected off-target activity and the intra- and inter-chromosomal rearrangements between on- and off-target sites, which were more pronounced in SCD samples (likely because of the higher overall editing efficiency), but did not impact the target gene expression. Off-target activity was observed in vitro and in vivo, thus indicating that it does not impair engraftment and differentiation of SCD and HD HSPCs. Finally, transcriptomic analyses showed that the genome editing procedure results in the upregulation of genes involved in DNA damage and inflammatory responses in both HD and SCD samples, although gene dysregulation was more evident in SCD HSPCs. Overall, this study provides evidences of feasibility, efficacy and safety for a genome editing strategy based on HbF reactivation and highlights the need of performing safety studies, when possible, in clinically relevant conditions, i.e., in patient-derived HSPCs.