CRISPR/Cas9 Genome Editing to Treat Sickle Cell Disease and B-Thalassemia: Re-Creating Genetic Variants to Upregulate Fetal Hemoglobin Appear Well-Tolerated, Effective and Durable.

Persistent expression of fetal hemoglobin (HbF) beyond the neonatal period is a rare, naturally-occurring condition, referred to as the Hereditary Persistence of Fetal Hemoglobin, which substantially ameliorates the pathology of Sickle Cell Disease (SCD) and β-thalassemia (β-thal). Extensive human genetic and epidemiologic studies have demonstrated that this condition is associated with one of several specific point mutations or deletions that lead to expression of γ-globin, resulting in upregulation of HbF. Our strategy is to use the CRISPR/Cas9 genome editing technology in human primary CD34+ hematopoietic stem and progenitor cells (HSPCs) to efficiently re-create specific genetic variants associated with elevated HbF and to evaluate their safety, effectiveness and durability as a potential therapeutic strategy to treat SCD and β-thal.

We have optimized CRISPR/Cas9 editing of human primary HSPCs mobilized from healthy donors to routinely achieve editing efficiencies of greater than 80%. Improved viability with no reduction in editing efficiency was achieved with Cas9 delivered as a recombinant protein as compared to Cas9 delivered as mRNA. We have further optimized the process at clinical scale in our manufacturing facility to achieve similar viability and editing efficiency with a GMP-compatible process. The high rate of editing in human primary HSPCs mobilized from healthy donors was associated with robust HbF expression following erythroid differentiation. In patient samples, clinically relevant increases in γ-globin mRNA to 39% (as a ratio of γ/α) in one β-thal patient sample and to 38±5% (Mean±SD, as a ratio of γ/(γ+β)) in six SCD patient samples were observed. Similarly high rate of editing was observed in the CD34+CD38-CD90+CD45RA- long-term repopulating subset of human primary HSPCs, as in the overall HSPC population (88±6% vs 90±4% Mean±SD, n=4). We have demonstrated the long-term persistence of transplanted cells in a NOD-SCID-Gamma (NSG) xeno-transplant mouse model at 16 weeks (44±12% of cells in controls versus 44±14% and 41±13% for two target modifications, Mean±SD, n=44-47 for each), and that the proportion of the various hematopoietic lineages derived from the transplanted edited human primary HSPCs was not affected by the editing process. Finally, the high editing rate of 85-90% persisted in human bone marrow cells at 16 weeks, demonstrating durability of the edited cells. Because the NSG model does not appropriately replicate erythroid development, we have also demonstrated normal erythroid potential of the engrafted human CD34+ cells isolated from mouse bone marrow in ex vivo differentiation assays, and the editing level is maintained in the erythroid lineage. In-depth analysis confirmed no detectable off-target sequence disruption at over 5,000 candidate off-target sites that were identified bioinformatically through homology to the on-target site or experimentally through a double strand break identification method. In vivo safety and toxicology studies are ongoing.

Taken together, these findings show that a CRISPR/Cas9-based therapy has the potential to upregulate HbF and treat diseases of b-globin. We have optimized editing conditions to achieve high efficiency and viability of editing in human primary HSPCs without detectable off-target sequence disruption, and in turn, demonstrate the durability of this approach in vivo. We believe that the results to date support the initiation of clinical studies for a CRISPR/Cas9 treatment for patients with SCD and β-thal.