Effects of priming with recombinant human granulocyte colony-stimulating factor on conditioning regimen for high-risk acute myeloid leukemia patients undergoing human leukocyte antigen-haploidentical hematopoietic stem cell transplantation: a multicenter randomized controlled study in southwest China.

HLA-haploidentical hematopoietic stem cell transplantation (haplo-HSCT) is an effective and immediate treatment for high-risk acute myeloid leukemia (HR-AML) patients lacking matched donors. Relapse remains the leading cause of death for HR-AML patients after haplo-HSCT. Accordingly, the prevention of relapse remains a challenge in the treatment of HR-AML. In a multicenter randomized controlled trial in southwestern China, 178 HR-AML patients received haplo-HSCT with conditioning regimens involving recombinant human granulocyte colony-stimulating factor (rhG-CSF) or non-rhG-CSF. The cumulative incidences of relapse and graft-versus-host disease (GVHD), 2-year leukemia-free survival (LFS), and overall survival (OS) were evaluated. HR-AML patients who underwent the priming conditioning regimen with rhG-CSF had a lower relapse rate than those who were treated with non-rhG-CSF (38.2%; 95% confidence interval [CI], 28.1% to 48.3% versus 60.7%, 95% CI, 50.5% to 70.8%; P < .01). The cumulative incidences of acute GVHD, chronic GVHD, transplantation-related toxicity, and infectious complications appeared to be equivalent. In total, 53 patients in the rhG-CSF-priming group and 31 patients in the non-rhG-CSF-priming group were still alive at the median follow-up time of 42 months (range, 24 to 80 months). The 2-year probabilities of LFS and OS in the rhG-CSF-priming and non-rhG-CSF-priming groups were 55.1% (95% CI, 44.7% to 65.4%) versus 32.6% (95% CI, 22.8% to 42.3%) (P < .01) and 59.6% (95% CI, 49.4% to 69.7%) versus 34.8% (95% CI, 24.9% to 44.7%) (P < .01), respectively. Multivariate analyses indicated that the 2-year probability of LFS of patients who achieved complete remission (CR) before transplantation was better than that of patients who did not achieve CR. The 2-year probability of LFS of patients with no M4/M5/M6 subtype was better than that of patients with the M4/M5/M6 subtype in the G-CSF-priming group (67.4%; 95% CI, 53.8% to 80.9% versus 41.9%; 95% CI, 27.1% to 56.6%; P < .05). This study suggests that the rhG-CSF-priming conditioning regimen is an acceptable choice for HR-AML patients, especially for the patients with no M4/M5/M6 subtype who achieved CR before transplantation.