Background Poor graft function (PGF) is an important complication of allogeneic hematopoietic cell transplantation (allo-HCT), predisposing to infections, bleeding complications and mortality. The reported incidence and outcome of PGF differ between studies, which may partially be explained by differences in patient cohorts and HCT strategies. Pediatric HCT differs from the adults setting in several aspects, such as the underlying disease and use of relatively young donors, including cord blood. To date, however, information on the incidence and outcome of PGF in pediatric HCT recipients is limited. Aim To determine the incidence, risk factors and outcome of poor graft function in pediatric HCT recipients. Methods Longitudinal blood count and transfusion data from 63.219 measurements were retrieved from the Utrecht Patient Oriented Database (ten Berg et al., Clin Chem Lab Med, 2007) to investigate PGF in pediatric recipients of a first HCT. Patients with less than 5 measurements within the first 30 days or 10 measurements within the first 100 days post-HCT were excluded. In line with published definitions, PGF was defined as cytopenia and/or transfusion dependence for 3 consecutive days after day 28 in ≥2 lineages simultaneously, and patients with relapse, grade III-IV GvHD and/or rejection were excluded (Müskens et al., Blood Reviews, 2023). Cytopenias were defined as thrombocytes <20x10 9/L, neutrophils <0.5x10 9/L and hemoglobin <7 g/dL (Kong et al., Biol Blood Marrow Transplant, 2013). Individual and cohort-wide reconstitution patterns of thrombocytes, neutrophils, and hemoglobin after HCT were analyzed and used to identify patients with PGF. Cumulative incidence and survival analyses were conducted using the {survival} R package. Probability of overall survival (OS) and event-free survival (EFS), defined as retransplantation-free, stem-cell-boost free survival, were calculated using the Kaplan-Meier estimate and compared at one year using the log-rank test. Risk factors for PGF and EFS were assessed in univariable and multivariable Cox proportional hazard models. Results Among 348 pediatric recipients who met the inclusion criteria, we identified 53 patients that fulfilled the criteria for PGF, resulting in a cumulative incidence of 0.15 (95% confidence interval (CI): 0.11-0.18). Risk factors for PGF included HCT for hematological disease (p = 0.03) and from an HLA-mismatched donor (p < 0.01). No difference in PGF incidence was observed between recipients of cord blood and bone marrow donors. The one-year OS (1Y OS) of patients in the PGF group was 71% (95% CI: 59-85%), compared to 91% (95% CI: 87-94%) in the good graft function (GGF) group (p < 0.0001, Figure 1A). Infectious disease was the cause of death in 47% in the PGF group, compared to 27% in the GGF group (p = 0.21). The 1Y EFS was 53% (95% CI: 41-68%) in the PGF group, compared to 90% (95% CI: 86-93%) in the GGF group (p < 0.0001, Figure 1B). Conclusion PGF is a common complication in pediatric HCT recipients. No difference in PGF incidence was observed between bone marrow and cord blood grafts. Both patient-specific (underlying disease) and donor-specific (HLA-matching) may contribute to its pathophysiology. PGF patients show a marked reduction in overall and event-free survival, highlighting the need for therapeutic strategies to prevent and/or treat PGF after HCT.
Read full abstract