Tremendous success has now been `achieved for the treatment of adenosine deaminase deficiency severe combined immunodeficiency,1 metachromatic leukodystrophy,2 and other severe monogenic diseases through the use of ex vivo hematopoietic stem and progenitor cell gene transfer (HSPC-GT) and subsequent bone marrow transplant (BMT)3 of the gene-corrected cells. In HSPC-GT/BMT, enriched preparations of hematopoietic stem cells (HSPCs) are collected from patients and transduced ex vivo with a lentiviral vector (LV) or retroviral vector encoding the defective gene. The cells are then transplanted back into patients that have been preconditioned with myeloablative agents so as to deplete the recipient's existing hematopoietic cells, thereby making “space” for the transduced stem cells to engraft and repopulate the patient's hematopoietic system with cells expressing a functional copy of the defective gene.
The promise of HSPC-GT/BMT has led to an interest in using this approach for the treatment of hemophilia A and B.4,5,6 One of the most exciting developments in the use of HSPC-GT/BMT for hemophilia was work by Montgomery and colleagues that demonstrated that transduction of HSPCs with an LV expressing factor VIII (FVIII) under the control of the glycoprotein αIIb (GPαIIb) promoter restricted FVIII expression to megakaryocytes following transplant.7,8 This had the important benefit that FVIII was enriched in megakaryocyte-derived platelets, which were able to release the clotting factor locally into wounds, so as to circumvent neutralization by preexisting FVIII antibodies. However, despite the efficacy of HSPC-GT/BMT in preclinical models of hemophilia, there has yet to be a clinical trial of the approach. Because a factor VIII– or factor IX–encoding LV is expected to have a low oncogenic potential, the possibility of insertional mutagenesis may not be a major hurdle for the translation of this approach. A more definite concern is over the need for myeloablative conditioning in hemophiliacs. BMT carries significant risks, including a high susceptibility to infections and even death. Unlike the case with an untreatable disease such as metachromatic leukodystrophy, for patients with hemophilia these risks are cause for trepidation.
In this issue of Molecular Therapy, Wang and colleagues9 describe a clever approach for hemophilia HSPC-GT that circumvents the need for ablative conditioning. Specifically, they show that direct injection of LVs into the bone marrow (intraosseous, IO) of the tibia of mice can stably transfer a FVIII transgene into HSPCs and their progeny without the need for BMT. Initially they evaluated the efficiency of IO delivery using a LV encoding green fluorescent protein from the ubiquitously expressed human elongation factor 1α promoter, and found that more than 15% of HSPCs (defined as Lin−Sca1+c-Kit+ cells) were transduced. They then inserted a FVIII transgene into the LV and delivered it IO into hemophilia A mice. They were able to detect FVIII production in up to 2% of HSPCs and even to measure FVIII activity in circulating blood. However, FVIII expression was lost over time, and antibodies to FVIII developed in the mice. To overcome this problem, Wang et al. adopted a platelet-targeted expression approach.7,8 They substituted the broadly active human elongation factor 1α promoter with the glycoprotein 1bα (Gp1bα) promoter, whose activity is limited to late-stage megakaryocytes, and injected the vector IO. Although this did not affect the vector's pattern of transduction, FVIII was no longer expressed in HSPCs or other differentiated cells, but was detected in ~2% of CD42d+ platelets. These findings are consistent with the pattern of expression of the Gp1bα promoter. Importantly, these mice exhibited reduced blood loss upon tail clipping. Even more impressively, when the Gp1bα-driven LV was injected IO into hemophilia A mice that had preexisting FVIII antibodies, blood clotting times were still reduced. This indicated that the vector and this delivery approach could mediate functional correction of the bleeding phenotype even in the presence of FVIII inhibitors.
Before IO delivery of LVs can be moved to the clinic, several issues will need to be addressed. In particular, it will be important to determine whether IO injection of the vector causes any disruptions to hematopoiesis. LVs can trigger a transient type I interferon response through activation of plasmacytoid dendritic cells, which are relatively abundant in bone marrow.10 This could be dangerous to patients, though it may be manageable with anti-inflammatory agents.10,11 A long-term concern is that transduction of cells in the stem cell niche could somehow affect the biology of the HSPCs, or that this approach could increase the risk of insertional mutagenesis. It must also be determined whether long-term HSPCs are actually being transduced by IO injection. These issues will require more extended follow-up and even serial transplantation studies. Nonetheless, the work by Wang et al. represents a promising new approach for hemophilia A gene therapy that overcomes several hurdles inherent in existing approaches—in particular, the need for myeloablative conditioning. This strategy could one day benefit patients with hemophilia A and other genetic diseases that have thus far not been considered for HSPC gene transfer. Furthermore, this novel in vivo approach could be considered as an alternative therapy for some of the diseases currently treated by LV gene transfer to HSPCs using conventional ex vivo protocols.