Genome editing for treatment of JAK2 V617F-driven myeloproliferative neoplasms Free

Genome editing is a promising new avenue for cancer therapy as it allows for precise inactivation of oncogenic driver mutations. However, clinical translation is hampered by two main hurdles. Firstly, for aggressive tumors, it is necessary to inactivate virtually 100% of mutant alleles to achieve durable therapeutic benefit, which is not achievable with current genome editing technologies. Secondly, most tissues require in vivo delivery of genome editing reagents by viral vectors, which poses significant challenges for both safety and efficacy.

Myeloproliferative neoplasms (MPNs) are hematopoietic stem cell (HSC)-propagated blood cancers that are driven in most cases by the JAK2 V617F mutation, often as the sole genetic event. Reduction in variant allele frequency (VAF) in response to treatment is a major predictor of favorable long-term outcomes, but current JAK2 inhibitors do not fully inhibit aberrant V617F signaling and fail to robustly clear the mutant clone. We reasoned that JAK2 V617F-driven MPNs are uniquely suitable for the development of a genome editing-based therapeutic strategy because i) the V617F mutation confers an only mild selective advantage, meaning that significant reductions in clonal burden are likely to exert therapeutic effect even if residual unedited cells persist and ii) HSCs are readily accessible through ex vivo editing of CD34+ cells in an autologous transplant setting.

We have developed an allele-selective genome editing strategy for the V617F mutation which exploits the fact that the G>T point mutation creates a de novo target site for the Cas12a nuclease. Electroporation of Cas12a ribonucleoproteins (RNPs) into CD34+ cells from MPN patients with single-hit JAK2 V617F mutations achieves 97% inactivation of V617F alleles, with less than 3% targeting of JAK2 wild-type alleles. Furthermore, off-target assays showed high genome wide-specificity of this genome editing approach. Crucially, we find that both wild-type and heterozygous mutant cells retain colony-forming ability after editing, and that only homozygous cells lose viability. Moreover, bone marrow organoid assays showed that the edited CD34+ cells maintain multilineage differentiation potential and that VAF can be reduced by as much as 99%.

To determine the distinct impact of allele-specific JAK2 V617F genome editing on wild-type versus JAK2 V617F heterozygous or homozygous cells, we have performed single cell combined allelic-level genotyping and RNA sequencing of edited CD34+ cells (TARGET-Seq). These experiments showed that inactivation of the V617F allele reverts the expression of both JAK-STAT targets and erythroid differentiation regulators to the levels found in wild-type cells, demonstrating that editing is sufficient to revert aberrant transcriptional phenotypes. In keeping with our colony-forming assays, we find that homozygous cells show severe defects in cell cycle and metabolic expression signatures upon editing, whereas heterozygous cells more closely resemble wild-type cells after deletion of their V617F copy.

To investigate the therapeutic efficacy of our strategy in an in vivo model, we performed xenotransplantation experiments using the JAK2 V617F mutant SET-2 leukemia cell line. Compared to mice engrafted with control edited cells, V617F editing markedly increased overall survival and reverted the splenomegaly and bone marrow fibrosis phenotypes found in control animals. Importantly, we find that SET-2 cells (which harbor a wild-type JAK2 copy) continue to persist in treatment group animals with virtually complete levels of editing on their V617F alleles, showing that V617F inactivation does not fully abrogate engraftment potential.

In conclusion, we have developed a genome editing strategy which eliminates the most common hotspot driver mutation in hematological neoplasms. Our approach achieves rapid and deep reductions in VAF while maintaining productive hematopoiesis from wild-type cells. Moreover, our extensive single-colony and single-cell genotyping experiments have shown that JAK2 V617F does not cause oncogene addiction, and that heterozygous cells can be reverted to a normal phenotype. Our work provides a proof of principle for the application of genome editing as a viable treatment strategy in cancer, which may be applied to other recurrent driver mutations in blood cancers.