Abstract
Background: Chronic myeloid leukemia (CML) is a rare disease in children and accounts for approximately 2% to 3% of all childhood leukemia. Despite of introduction of tyrosine kinase inhibitors, allogeneic hematopoietic stem cell transplantation is still the only proven curative treatment for patients with CML. Few studies have specifically addressed transplantation outcomes in childhood CML.
Patients and Methods: We analyzed the clinical data of 125 children (81 boys and 44 girls) with CML undergoing unrelated bone marrow transplantation (UBMT) through the Japan Marrow Donor Program (JMDP) between 1993 and 2006 (median observation period, 95 months). The recipients’ median age at transplantation was 14 years (range, 1 – 19 years). The disease phases at conditioning were chronic phase (CP) 1 (n=88), CP2 (n=12), CP3 (n=1), accelerated phase (n=11) and blastic crisis (n=13). The median interval from diagnosis to BMT was 14 months (range, 2 – 111 months). 95 (76%) patients received interferon-alpha (IFN-alpha), and 17 (14%) received imatinib mesylate (IM) before BMT. 96 (77%) patients received total body irradiation (TBI) containing a conditioning regimen, and 29 (23%) received a non-TBI regimen. Cyclosporine A (CSA)-based graft-versushost disease (GVHD) prophylaxis was used for 81 (65%) patients, and tacrolimus based prophylaxis was used for 43 (34%) patients. One patient received methotrexate only for GVHD prophylaxis. HLA matching data based on high-resolution DNA typing for HLA-A, -B, -Cw, -DRB1 and -DQB1 alleles were available in 99 (79%) patients; 41 of 99 (41%) patients were “fully matched (10/10)”. The median infused cell number was 314×106/kg (range, 27 – 880×106/kg).
Results: Five-year overall survival (OS) and leukemia-free survival probabilities were 59.3% ± 4.5% and 55.5% ± 4.5%, respectively. Grade II to IV acute GVHD occurred in 50 (40%) patients, and chronic GVHD occurred in 51 (41%) patients. Six (5%) patients did not achieve engraftment. In multivariate models for OS, disease phase (advanced phase/CP1) (p=0.008, RR 2.431 (1.261–4.689)), infused cell number (<314×106/kg/ ≥314×106/kg) (p=0.004, RR 2.917 (1.404–6.060)), HLA-A allele mismatch (p=0.001, RR 4.076 (1.798–9.236)), and age at transplantation (≥15/ <15 years) (p=0.031, RR 2.236 (1.076–4.651)) were independent risk factors. For transplant-related mortality (TRM), disease phase (advanced phase/CP1) (p=0.021, RR 2.341 (1.137–4.820)) and infused cell number (<314×106/kg/ ≥314×106/kg) (p=0.001, RR 3.792 (1.686–8.529)) were significant risk factors. HLA-B, -Cw, -DRB1, and -DQB1 allele mismatch were not significant risk factors for OS and TRM on univariate and multivariate analyses. Subgroup analyses of CP1 patients (n=88) showed that cytogenetic response (with or without major cytogenetic response) at BMT was a strong risk factor for OS (5-year OS: 91.4% ± 6.0% vs 53.4% ± 8.1%, p=0.001). For the 29 patients who achieved major cytogenetic response at BMT, the probability of OS was not statistically different between patients treated with IM (n=15) and those treated with IFN-alpha (n=14) before BMT (p=0.722).
Conclusions: Disease phase, infused cell number, HLA-A allele mismatch, patient age and cytogenetic response at transplantation were important risk factors for transplant outcomes. This is one of the largest cohorts studied to date, and the results provide important information to assess the indications and to improve the outcomes of children with CML undergoing UBMT.
Disclosures: No relevant conflicts of interest to declare.
Author notes
Corresponding author