Abstract
Between 1980 and 2002, 132 children and adolescents (median age 8.5 years, range 1.1 to 17.3) got an allogeneic HLA-identical hematopoietic stem cell transplantation (SCT) from a sibling donor for acute leukemia. The patients suffered from ALL in 1st remission (n=24), ALL in 2nd remission (n=53) and AML in 1st remission (n=55). The mean follow-up time after SCT was 11.5 years (range 1.3 to 23.3 years). The source of the stem cells was bone marrow in all but 3 cases of cord blood-derived stem cells. The number of nucleated marrow cells in the graft varied between 0.5 and 6.8x108/kg (median 2.75x108/kg). Most patients, 118 of 132, were pre-treated with a total body irradiation (TBI) and cyclophosphamide-based regimen. TBI doses were 1x 7 Gy for children of 3 and 4 years, 1x 7.5 Gy for >4 and <10 years, and 1x 8 Gy (before 1990) and 2x 6 Gy on two consecutive days (after 1990) for children older than 10 years. Initially, graft-versus-host disease (GvHD) prophylaxis consisted of long-course methotrexate (MTX) only (n=24), later a short-course MTX-Cyclosporine A prophylaxis was given (n=102). All patients were nursed in strict protective isolation, using positive-pressure laminar-air flow cubicles or rooms, and got total gut decontamination with non-absorbable antimicrobials. Engraftment, transplant-related complications, relapse of the disease and survival were evaluated retrospectively in relation to disease- and transplant-related characteristics. None of the patients failed to engraft. The mean 5-year probability of overall survival was 63%, 53% and 74% for ALL1, ALL2 and AML1 patients, respectively. Transplant-related mortality was 6%. The incidence of acute GvHD in evaluable cases was 17% (21/124), of whom 7 (6%) got grade II to IV. Chronic GvHD occurred in 10 patients out of 125 evaluable patients (8%). The mean 5-year probability of relapse was 35%, 49% and 26% for ALL1, ALL2, and AML1 patients, respectively. Most relapses (70%) occurred within the first year post-SCT. The patient age and dose of TBI were significantly related with post-transplant relapse (p=0.007). Overall survival was significantly higher in children that received a higher dose of TBI (see figure). Overall, acute GvHD was not significantly associated with relapse, but none of the AML patients (n=9) with acute GvHD got a relapse (likelihood ratio test, p=0.02), which was not the case for ALL first or second remission patients. This retrospective single-center study on HLA-identical SCT for acute leukemia in 132 consecutively enrolled children confirms the beneficial role of slight to moderate acute GvHD in the suppression of relapse of AML, and draws attention to the potential antileukemia effect of a high (fractionated) dose of TBI in the conditioning.
Figure
