Outcome of Relapse After Allogeneic Transplantation for Childhood Acute Lymphoblastic Leukemia In Complete Remission. A Study of the Pediatric Diseases and Acute Leukemia Working Parties of the EBMT Group

Childhood acute lymphoblastic leukemia (ALL) relapse occurring after hematopoietic cell transplantation (HCT) has a very dismal prognosis. Its treatment is still controversial and ranges from palliative treatment or chemotherapy to donor lymphocyte infusions, second transplant or experimental approaches.

The aim of this study is to assess the actual outcome in a pediatric population. The primary endpoint of this study is the 2-year probability of survival of children with ALL relapsing after allogeneic HCT; the secondary endpoint is the relationship between outcome and time of relapse after transplant, for which the following categories were considered: <3, 3–6, 6–12, > 12 months.

Patients younger than 18 years of age undergoing first HCT from any allogeneic donor for ALL in first (CR1) or second (CR2) remission between January 1^st 1998 and December 31^st 2007 reported to the EBMT were eligible for the study.

Out of 3628 transplanted children with ALL reported to the EBMT, 836 (median age 9 years, male 66%) relapsed at a median of 6 months (range 1–67; 25^th, 75^th 4, 12 months) after HCT. The HCT was performed in CR1 (60%) or CR2 (40%) for a B-lineage (60%) or T- (13%) or unknown (27%) immunophenotype ALL, from an HLA-matched related (44%), unrelated (59%) or mismatched related (7%) donor, with marrow (61%), peripheral (28%) or umbilical (11%) stem cells.

Out of 836, 81% died at a median of 2 months (25^th,75^th centiles:1,7) and 19% were reported as alive at last follow-up at a median of 22 months after relapse (range: 1–130). The 3-year probability of overall survival (3y-OS) was 14% (SE 1). As to immunophenotype, disease phase and donor type, 3y-OS was 15% (SE 2) in B-lineage and 8% (SE 3) in T-ALL, 18% (SE 2) in patients transplanted in CR1 and 11% (SE 6) in CR2 and 17% (SE 2) in patients transplanted from an HLA-identical sibling and 12% (SE 2) from any other donor. According to time of relapse after transplant, 3-year OS was 6% (SE 2), 10% (SE 2), 15% (SE 2) and 27% (SE 4) in those who relapsed in the first quarter, second quarter, second semester or after the first year, respectively.

Donor lymphocyte infusions were reported for 7% and a second HCT for 16% of the 836 relapsed children. The probability of undergoing a second HCT within 1 year after relapse was 17% (SE 1); this probability was 6% for relapses occurring <6 months and 25% for later relapses. 3y-OS of those who underwent a second HCT was 32% (SE 5).

The multivariate analyses confirmed the prognostic role of disease phase and immunophenotype, but not of the type of donor, assessed the strong prognostic impact of the time elapsed in CR after HCT before relapse, being earlier relapses at worse outcome compared with later relapses, possibly due to the chance of undergoing a second HCT, which role per se was not statistically significant.

No relevant conflicts of interest to declare.