Granulocyte Colony-Stimulating Factor Prior to Low-Dose Total Body Irradiation as Conditioning Regimen in Hematopoietic Stem Cell Transplantation for Gene Therapy.

Genetic blood diseases can potentially be treated by transplantation of autologous hematopoietic stem cells (HSC) transduced with the functional gene. Nonmyeloablative conditioning regimens such as low-dose total body irradiation (LDTBI) can enhance competitive engraftment in the absence of a selective advantage for HSC. Granulocyte colony-stimulating factor (G-CSF) treatment of recipients prior to LDTBI has been shown to enhance engraftment of fresh bone marrow (BM) cells, but the underlying mechanisms and whether this combined approach can enhance long-term engraftment of genetically transduced donor cells have not been defined. We examined whether pre-treatment of recipient mice with G-CSF prior to LDTBI improves long-term donor engraftment of retroviral vector-transduced HSC in murine X-linked chronic granulomatous disease (X-CGD). In initial studies, we verified that pre-treatment of wild type mice with 4 ug G-CSF SC BID for 5 days prior to irradiation with either 160 cGy or 300 cGy significantly increased long-term engraftment of congenic fresh BM cells (see Table).

We next investigated 2 potential mechanisms by which G-CSF prior to LDTBI might enhance long-term engraftment of HSC. First, analysis of host BM 20 hours after transplantation showed a significantly lower frequency of homed donor HSC in G-CSF+160cGy treated mice compared to 160 cGy-conditioned mice (1 ± 0.13% vs 3.5 ± 0.7% in one experiment, and 3.4 ± 0.8% vs 9.5 ± 2.3% in another experiment, n= 20 mice). Second, in a competitive repopulation assay, we observed a substantial decrease in marrow long-term repopulating ability (LTRA) in mice treated with G-CSF + 160 cGy compared to 160 cGy alone (10.5% vs 48.6%, n= 5 mice in each group; p= 0.0002). These results suggest that administration of G-CSF prior to LDTBI does not increase initial homing of donor cells but further impairs LTRA of recipient BM compared to LDTBI alone. In the context of gene therapy, we transduced male X-CGD BM with MSCV-gp91Neo vector and infused these transduced cells into female X-CGD mice treated ± G-CSF + 300 cGy. Donor cell engraftment, measured by FISH for Y chromosome, at 3 months post-transplant was significantly increased in recipients pretreated with G-CSF (see Table). NADPH oxidase activity (by 123-dihydrorhodamine assay) was reconstituted in 9.75 ± 5.5% of neutrophils compared to 6.6 ± 2.5% without G-CSF pretreatment. The calculated fraction of donor cells with oxidase activity was similar in both groups (18.8 ± 6.6% vs 21.4 ± 11%, respectively) (p= 0.67). These results suggest that the use of G-CSF prior LDTBI provides a competitive advantage for long-term engraftment by decreasing LTRA in the host and may be a useful strategy to increase engraftment of fresh and retroviral transduced donor BMC.