Long-term outcome in children with relapsed acute lymphoblastic leukemia after time-point and site-of-relapse stratification and intensified short-course multidrug chemotherapy: results of trial ALL-REZ BFM 90.

The multicenter trial ALL-REZ BFM (ie, Acute Lymphoblastic Leukemia Relapse Berlin-Frankfurt-Münster) 90 was designed to improve prognosis for children with relapsed acute lymphoblastic leukemia (ALL) by time-to-relapse- and site-of-relapse-adapted stratification and by introduction of novel chemotherapy elements and to evaluate new prognostic parameters in a large, population-based cohort. Five hundred twenty-five patients stratified into risk groups A (early bone marrow [BM] relapses), B (late BM relapses), and C (isolated extramedullary relapses) received alternating short-course intensive polychemotherapy (in blocks R1, R2, or R3) and cranial/craniospinal irradiation followed by maintenance therapy. Block R3 (high-dose cytarabine and etoposide) was introduced to improve the outcome compared with historical controls. Patients with early BM or T-ALL relapse (poor prognosis group [PPG]) were eligible for experimental regimens. One hundred seventeen patients received stem-cell transplantation (SCT). The probabilities (and standard deviations) of event-free survival (pEFS) and overall survival (pOS) at 10 years were 0.30 +/- .02 and 0.36 +/- .02, respectively. Significant differences existed between strategic groups (pEFS(A) = .17 +/- .03; pEFS(B) = .43 +/- .04; pEFS(C) = .54 +/- .06; pEFS(PPG) = .15 +/- .03; log-rank P < .001). Patients of high-risk groups A plus PPG did better with SCT than with chemotherapy (pEFS = .33 +/- .05 v 0.20 +/- .05; P = .005). The pEFS was similar to trials ALL-REZ BFM 85/87 (.36 +/- .03. v 0.37 +/- .03; P = .419; PPG excluded). Time point, site of relapse, immunophenotype, and SCT were significant predictors of pEFS in multivariate analyses. More than one third of patients in this large, population-based trial were cured. Neither R3 nor adaptation of chemotherapy intensity was capable of improving pEFS or of overcoming prognostic factors. In high-risk patients, remission induction regimens must be improved, and allogeneic SCT should be recommended in patients achieving second complete remission.