Analysis of Risk Factors Influencing Outcomes After Unrelated Cord Blood Transplantation In Children with Juvenile Myelomonocytic Leukemia. An Eurocord, EBMT, EWOG-MDS, CIBMTR Study

Abstract 533

Juvenile myelomonocytic leukemia (JMML) is a rare and lethal myeloproliferative disease of young childhood. Currently, allogeneic hematopoietic stem cell transplantation (HSCT) is the only curative treatment option. DFS at 5 years after HLA identical and unrelated HSCT was 55% (n=48) and 49% (n=52) in a largest series of patients published so far and mainly transplanted with bone marrow cells. Unrelated Cord Blood Transplantation (UCBT) is considered an alternative option for patients who lack an HLA-matched donor. We retrospectively analyzed 110 children, given a first single unmanipulated UCBT, from 1995 to 2010, and reported to Eurocord-EBMT and CIBMTR. Median age was 1 year (range 0.08–6.4) at diagnosis and 2 years (0.5-7.5) at transplantation, respectively. Median time interval between diagnosis and UCBT was 6 months (1-58); before transplantation, 88 patients were treated with low- or high-dose chemotherapy and splenectomy was performed in 24 children. Among 100 patients with available cytogenetic data, monosomy of chromosome 7 was the most frequent abnormality (24%). All but 8 patients received a myeloablative conditioning, Busulfan-Cyclophosphamide-Melphalan (BuCyMel) was used in 48 patients, total body irradiation (TBI) and Cyclophosphamide in 19 patients and combination of Busulfan-Cyclophosphamide with other drugs in 21 patients. Cyclosporin+steroid was the most common graft-versus-host disease (GvHD) prophylaxis (80%) and ATG was added in 86% of patients. Nineteen percent of units were HLA-identical (antigen level for HLA-A and B, allelic for DRB1), while 43% and 38% had 1 or 2–3 mismatches, respectively. Median TNC infused was 7.1×10e7/kg (1.7-27.6). Median follow-up was 44 months (3-169). At 60 days, cumulative incidence (CI) of neutrophil (PMN) recovery was 80±4%, with a median time to PMN recovery of 25 days. Grades II-IV acute GvHD developed in 45 patients, 100 days-CI of grade II-IV aGvHD was 40±5%. Among 90 patients at risk, 17 developed chronic GvHD and 4 years-CI was 16±4%. At 4 years CI of relapse was 37±5% (n=38); age older than 1 year at diagnosis was the only independent factor associated with increased risk of relapse (HR 2.3, p=0.038). Of note, among 58 patients with available data for level of fetal hemoglobin (HbF), a higher level of HbF (>35%) seemed to be associated with increased relapse incidence (57% versus 31% for remainders; p=0.05). At 4 years, DFS was 43±5%, in multivariate analysis independent factors associated with better DFS were: age younger than 1 year at diagnosis (53% vs 30%, HR 2.4, p=0.001), graft with 0 or 1 HLA mismatched cord blood unit (48% vs 34%, HR=2.1, p=0.006) and cytogenetic without monosomy 7 (48% vs 26%; HR=1.95, p=0.027). At 4 years, CI of transplant related mortality (TRM) was 20±4%; in multivariate analysis, cytogenetic with monosomy 7 (HR=2.7, p=0.036) and transplantation performed before 2003 (HR=3.7, p=0.015) were factors associated with increased TRM. In fact, CI of TRM was 14% after 2003 compared to 30% before 2003. Estimated overall survival (OS) at 4 years was 51±5%, and in multivariate analysis factors associated with decreased OS were: age older than 1 year at diagnosis (42% vs 60%; HR=2.03, p=0.032), and cytogenetic with monosomy 7, (30% vs 57%; HR=2.6, p=0.004). Fifty-one patients died after transplant, 53% for relapse and 47% for transplant related causes. In conclusion, UCBT may cure approximately 50% of patients with JMML who lack a matched related donor. Presence of monosomy 7 is associated with decreased DFS and increased TRM, independent of other factors. Other patient- (age at diagnosis) and transplantation-related factors (HLA and year of transplantation) were also associated with outcomes. Disease recurrence remains the major cause of treatment failure, and strategies to reduce the risk of relapse are warranted.