Response: The role of G-CSF on the risk of graft-versus-host disease after donor lymphocyte infusions

The letter written by Maury et al raises an interesting question regarding the role of G-CSF on the increased incidence of acute graft-versus-host disease (aGVHD) we observed when giving donor lymphocyte infusions (DLIs) after lymphodepleting chemotherapy.¹  To address their concern we reviewed the patient charts and pharmacy records for our cohort. Because of the known immunologic effects of G-CSF, including its potential to inhibit immune function, the protocol did not prohibit but did allow the use of G-CSF if necessary for patient safety. Of the 8 patients who developed grade III to IV aGVHD, G-CSF was either not given (1 patient) or was given 15 to 22 days after the onset of a definitive diagnosis of severe aGVHD (3 patients), excluding G-CSF as a causative agent in DLI-induced severe GVHD. The remaining 4 patients did receive G-CSF with the following timing after DLI in relation to aGVHD onset: day +3/aGVHD day +16, day +10/aGVHD day +17, day +9/aGVHD day +15, and day +19/aGVHD day +21. Based on these results, we conclude that the use of G-CSF does not fully explain the incidence of severe GVHD seen in our study. In murine studies, G-CSF treatment of donors reduced murine GVHD while maintaining a graft-versus-leukemia effect, presumably due to the induction of a type 2 cytokine profile (Th2) and the preservation of cytotoxic T-lymphocyte activity.²  Similarly, in humans, G-CSF–mobilized peripheral blood stem cells had higher levels of dendritic cell type 2 (DC2) cells, which are capable of inducing T cells to a Th2 phenotype.³  No effects on aGVHD or chronic GVHD were noted by Khoury et al, who studied 2719 patients given G-CSF early posttransplantation.⁴  Lymphodepletion may stimulate aGVHD by release of cytokines (eg, IL-15, IL-7) and suppression of host immunity. In support of the latter mechanisms as an explanation for the more severe GVHD in patients receiving chemotherapy prior to DLI, murine studies have indicated that host T cell–mediated resistance at the time of DLI reduces the severity of GVHD.⁵  In our clinical trial, host resistance would be diminished by lymphodepleting chemotherapy given prior to DLI, thereby facilitating the expansion of donor T cells capable of causing GVHD. Based on the known effects of lymphopenia and concurrent cytokine accumulation on supporting T-cell expansion, Th2-skewing properties of G-CSF in rodents and humans, and the fact that in our study G-CSF did not contribute to severe aGVHD in at least 50% of the patients, we conclude that the immune stimulating milieu induced by lymphodepleting chemotherapy is still likely to be the predominant factor resulting in enhanced immune effects of DLI.