Use of G-CSF in AML patients with febrile neutropenia: Friend or foe?

Abstract Background Febrile neutropenia (FN) is a common and potentially life-threatening complication in patients with acute myeloid leukemia (AML), particularly during intensive chemotherapy. Granulocyte colony-stimulating factors (G-CSFs), such as filgrastim, are frequently used to accelerate neutrophil recovery and reduce the duration of neutropenia. However, their use in AML has remained controversial due to concerns that stimulating myeloid precursors might promote leukemic progression or increase the risk of secondary hematologic malignancies such as myelodysplastic syndrome (MDS). While G-CSF has demonstrated benefit in reducing infection-related complications in other malignancies, evidence supporting its safety and efficacy in AML, particularly with respect to long-term outcomes, is limited. This study aimed to evaluate the short-term clinical impact of G-CSF use in hospitalized AML patients with febrile neutropenia and to assess its association with the development of secondary MDS over a five-year period.Methods We performed a retrospective cohort study using the TriNetX US Collaborative Network, which includes data from 66 healthcare organizations across the United States. Adult patients diagnosed with AML, neutropenia, and fever were included. Patients were grouped based on whether they received G-CSF (filgrastim or pegfilgrastim) during hospitalization for febrile neutropenia. To reduce confounding, 1:1 propensity score matching was performed based on demographic factors, comorbidities, and chemotherapy regimens, resulting in two balanced cohorts of 4,868 patients each. We assessed short-term outcomes within 30 days of treatment, including mortality, hospitalization, shock, and thrombotic events. Cox proportional hazards models and Kaplan-Meier analyses were used to evaluate time-to-event outcomes. We also examined the incidence of secondary MDS over a five-year follow-up period.Results Among the matched cohort of 9,736 AML patients with febrile neutropenia, G-CSF use was associated with improved short-term clinical outcomes. The 30-day survival rate was higher in the G-CSF group compared to the non-G-CSF group (90.7% vs. 84.5%, hazard ratio [HR] 0.584, 95% confidence interval [CI] 0.520 to 0.657, p < 0.001). Hospitalization rates were also lower in patients who received G-CSF (40.6% vs. 33.1%, HR 0.776, 95% CI 0.738 to 0.817, p < 0.001). G-CSF was associated with fewer cases of shock (23.6% vs. 28.2%, HR 0.799, 95% CI 0.739 to 0.864, p < 0.001) and deep vein thrombosis (7.5% vs. 9.8%, HR 0.746, 95% CI 0.651 to 0.855, p < 0.001). There were no statistically significant differences between groups in rates of stroke, pulmonary embolism, aortitis, or capillary leak syndrome. Over the five-year follow-up, the incidence of secondary MDS was comparable between groups (25.7% in the G-CSF group vs. 26.2% in the non-G-CSF group, HR 0.944, 95% CI 0.873 to 1.021, p = 0.17), suggesting that G-CSF administration does not increase the long-term risk of secondary hematologic malignancy. Conclusion In this large, multicenter, propensity-matched cohort of AML patients with febrile neutropenia, the use of G-CSF was associated with significantly better short-term outcomes, including lower mortality, fewer hospitalizations, and reduced rates of shock and thrombotic events. Importantly, G-CSF use did not increase the risk of secondary MDS at five years, addressing longstanding safety concerns. These findings not only support the clinical benefit and long-term safety of G-CSF in this population, but also challenge previous studies that reported no survival advantage with its use. Further prospective research is warranted to validate these findings and optimize the role of G-CSF in supportive care protocols for patients with AML.