Monitoring of Minimal Residual Disease After Allogeneic Stem-Cell Transplantation in Relapsed Childhood Acute Lymphoblastic Leukemia Allows for the Identification of Impending Relapse: Results of the ALL-BFM-SCT 2003 Trial

PURPOSE: To elucidate the impact of minimal residual disease (MRD) after allogeneic transplantation, the Acute Lymphoblastic Leukemia Berlin-Frankfurt-Münster Stem Cell Transplantation Group (ALL-BFM-SCT) conducted a prospective clinical trial. PATIENTS AND METHODS: In the ALL-BFM-SCT 2003 trial, MRD was assessed in the bone marrow at days +30, +60, +90, +180, and +365 after transplantation in 113 patients with relapsed disease. Standardized quantification of MRD was performed according to the guidelines of the Euro-MRD Group. RESULTS: All patients showed a 3-year probability of event-free survival (pEFS) of 55%. The cumulative incidence rates of relapse and treatment-related mortality were 32% and 12%, respectively. The pEFS was 60% for patients who received their transplantations in second complete remission, 50% for patients in ≥ third complete remission, and 0% for patients not in remission (P = .015). At all time points, the level of MRD was inversely correlated with event-free survival (EFS; P < .004) and positively correlated with the cumulative incidence of relapse (P < .01). A multivariable Cox model was fitted for each time point, which showed that MRD ≥ 10(-4) leukemic cells was consistently correlated with inferior EFS (P < .003). The accuracy of MRD measurements in predicting relapse was investigated with time-dependent receiver operating curves at days +30, +60, +90, and +180. From day +60 onward, the discriminatory power of MRD detection to predict the probability of relapse after 1, 3, 6, and 9 months was more than 96%, more than 87%, more than 71%, and more than 61%, respectively. CONCLUSION: MRD after transplantation was a reliable marker for predicting impending relapses and could thus serve as the basis for pre-emptive therapy.