A Randomized Trial To Evaluate the Impact of Cytokines (G-CSF or GM + G-CSF) on Dendritic Cell and T-Cell Content and Function When Mobilizing Normal Donors for Allogeneic Progenitor Cell Transplant.

Introduction: Optimal cellular immunity following allogeneic HPC transplant is a tenuous balance between effective anti-tumor immunity and the avoidance of life threatening GvHD. The events necessary for potentially curative GvL are related to tumor burden, T-cell content, NK cell content, and dendritic cell content. Our group previously demonstrated that DC content following allogeneic BMT can impact EFS (Waller et al, Blood 2001), and that cytokines used during mobilization impact DC content of the autologous graft (Lonial et al, BBMT 2004). The current trial was designed with the hypothesis that the use of G + GM-CSF will result in fewer tolerogenic DC2 cells within the graft, and increased Th1 phenotype among the mobilized T-cells.

Methods: 40 normal donors were randomized to receive G-CSF alone (7.5 mcg/kg BID) or G-CSF (7.5mcg/kg qd) and GM-CSF (7.5 mcg/kg qd) for 5 days or longer if additional apheresis were necessary. Side effects between the 2 cytokine regimens were documented using a questionnaire that was administered within 2 weeks of successful collection. Graft content was evaluated using multi-color flow cytometry, and pre-mobilization blood was collected to test for T-cell and DC content as well as T-cell function. T-cell proliferation was assessed using thymidine incorporation assays following mitogen exposure, and T-cell activation profile was assessed using ELISA assays of secreted cytokines following mitogen stimulation. DC1 (myeloid DC) were defined as Lin-/HLA-DR+/CD11C+/CD123−, while DC2 (lymphoid DC) were defined as Lin-/HLA-DR+/CD11C−/CD123+.

Results: 18 patients received G+GM-CSF and 22 patients received G-CSF alone. All 40 donors were successfully mobilized, though more patients in the G-CSF arm required multiple days of collection (mean number of collections 1.5 G-CSF vs 1.0 G+GM-CSF, p=0.01). Donors mobilized with G-CSF alone had a higher overall cell count (765e8/ml compared with 615e8/ml for G+GM-CSF, p=0.04), though the CD34 content of the grafts were comparable. Again, there was a significant reduction in delivered DC2 content of graft mobilized with G+GM-CSF when compared with G-CSF alone (5.1e6/kg compared with 2.85e6/kg, p=0.001), with a trend towards fewer DC1 cells as well (2.6e6/kg vs 1.7e6/kg, p=0.05). Overall, the DC1:DC2 ratio favors DC1 among donors mobilized with G+GM-CSF. Additionally, grafts collected with G-CSF alone had nearly twice the total T-cell content of grafts mobilized with G+GM-CSF (420e6/kg vs 220e6/kg, p=0.0001). The reduction in total T-cells was evenly distributed between CD4 and CD8 cells with similar magnitude of reduction in both subsets amongst those receiving G+GM-CSF. Interestingly, grafts collected from the recipients of G+GM-CSF produced more IL-2 at rest, and secreted significantly more IL-12 in response to mitogen. Toxicity between the 2 arms were similar with 1 patient in the G+GM-CSF arm having to stop growth factors due to bone pain and fever.

Conclusion: The combination of G+GM-CSF mobilizes fewer DC2 cells as well as 40–50% fewer T-cells, and is more likely to result in a successful single PBSC collection than the use of G-CSF alone. Preliminary data suggests that the combination of G+GM-CSF shifts T-cells towards a Th1 phenotype, despite fewer overall T-cells. Further laboratory evaluation of the graft, survival, and GvHD data following transplant will also be presented.