Marrow transplants from unrelated donors for patients with aplastic anemia: minimum effective dose of total body irradiation

Abstract

Patients with aplastic anemia who do not have suitably HLA-matched, related donors generally receive immunosuppressive treatment as first-line therapy and are considered for transplantation from an unrelated donor only if they fail to respond to immunosuppressive treatment. In this setting, rates of transplantation-related morbidity and mortality have been high. We conducted a prospective study to determine the minimal dose of total body irradiation (TBI) sufficient to achieve sustained engraftment when it is used in combination with 3 cycles of 30 mg/kg of antithymocyte globulin (ATG) and 4 cycles of 50 mg/kg of cyclophosphamide (CY). We also wanted to determine the tolerability and toxicity of the regimen. The starting dosage of TBI was 3 x 200 cGy given over 2 days following CY/ATG. The TBI dose was to be escalated in increments of 200 cGy if graft failure occurred in the absence of prohibitive toxicity, and de-escalated for toxicity in the absence of graft failure. Twenty-one female and 29 male patients aged 1.3 to 46.5 years (median age, 14.4 years) underwent transplantation at 14 medical centers. The time interval from diagnosis to transplantation was 2.8 to 264 months (median, 14.5 months). All patients had been transfused multiple times and all had received 1 to 11 courses (median, 4 courses) of immunosuppressive treatment and other modalities of treatment. In 38 cases, the donors were HLA-A, -B and -DR phenotypically matched with the patients, and, in 12 cases, the donor phenotype differed from that of the recipient by 1 HLA antigen. Recipients of mismatched transplants were considered separately for TBI dose modification, and this study is still ongoing. Seven patients did not tolerate ATG and were prepared with 6 x 200 cGy of TBI plus 120 mg/kg of CY. Of the HLA-matched recipients prepared with CY/ATG/TBI, all 20 who received 3 x 200 or 2 x 200 cGy of TBI achieved engraftment, and 10 are alive. Of the 13 patients who received 1 x 200 cGy of TBI, 1 failed to engraft, and 8 are alive. Each of 10 patients who received an HLA-nonidentical transplant achieved engraftment, and 3 of 6 who were given 3 x 200 cGy of TBI, and 4 of 4 who were given 2 x 200 cGy are alive. Pulmonary toxicity occurred in 8 of 30 patients who were given 3 x 200 or 2 x 200 cGy of TBI concurrently with ATG and CY at 200 mg/kg, and in 2 of 13 patients who received 1 x 200 cGy of TBI, a pattern that suggests a decrease in toxicity with TBI dose de-escalation. Overall, the highest probability of survival (73%) was observed among patients who underwent transplantation within 1 year of diagnosis, compared with patients who underwent transplantation after a longer period of disease. In addition, younger patients (aged < or = 20 years) were more likely to survive than older patients (aged > 20 years). Thus, for patients with an HLA-matched, unrelated donor, a TBI dose of 200 cGy (in combination with CY/ATG) was sufficient to allow for engraftment without inducing prohibitive toxicity. As in previous studies, patient age and pretransplantation disease duration remain important prognostic factors.