Reduced Intensity Delayed Intensification in Standard-Risk Patients Defined By Minimal Residual Disease in Childhood Acute Lymphoblastic Leukemia: Results of an International Randomized Trial in 1164 Patients (Trial AIEOP-BFM ALL 2000)

From 07/2000 to 06/2006, 4741 eligible pts with ALL (age range 1-17 years) were enrolled in the trial AIEOP-BFM-ALL 2000 (NCT 00430118 (BFM) and NCT 00613457 (AIEOP)). 1164 patients from Germany, Italy, Austria, and Switzerland entered the randomized comparison of protocol III (P-III) versus standard protocol II (P-II) for delayed reintensification in order to reduce treatment burden in the group of patients with best treatment response and lowest risk of relapse. P-III is shorter than P-II (duration 29 vs 49 days), the dose of dexamethasone in P-III is 30% less and the dose of vincristine, doxorubicine, and cyclophospamide 50% less than in P-II. The intention was to prove non-inferiority of the reduced intensity treatment as compared to standard treatment. Only patients who were defined as MRD standard risk (MRD-SR) being negative at both days 33 and 78 after start of induction therapy, using at least 2 molecular markers with sensitivity of at least 10^-4, were eligible for the randomization. The main analysis was planned as a per protocol analysis of the 4-year disease free survival (4y-DFS). Stratification, induction and consolidation therapy and the results of a randomization in induction, dexamethasone at 10 mg/m²/d (DEXA) vs prednisone at 60 mg/m²/d (PRED), have been described earlier (A. Möricke et al, BLOOD 2016, M. Schrappe et al, BLOOD 2011, and V. Conter et al, BLOOD 2010).

With a median follow-up of 8.6 years, 4y-DFS was 91.8% (SE 1.1%) and 95.8% (SE 0.8%) for patients who received P-III (N=584) or P-II (N=579), respectively (log-rank p=0.04). The lower limit of the one-sided 95% confidence interval for the DFS-rates was -6.4%, far below the non-inferiority range of -4%, (p-value on difference of the 4y-DFS estimates: 0.005). The 4-year cumulative incidence of relapse (CIR) was 6.3% (SE 1.0%) and 3.2% (SE 0.7%), for P-III and P-II, respectively (Gray p=0.09). The results at 8 years were 89.2% (SE 1.3%) and 92.3% (SE 1.2%) for DFS, and 8.7% (SE 1.2) and 6.4% (SE 1.1%) for CIR (P-III vs P-II). The 8-year overall survival (OS) was 96.1% (SE 0.8%) and 98.0% (SE 0.6%, p=0.06) and the 8-year cumulative incidence of secondary malignancies was 1.3% (SE 0.5%) and 0.6% (SE 0.4%) for patients who received P-III and P-II, respectively. An intent-to-treat analysis revealed almost identical results. There were no major differences of the treatment effect in clinical and biological subgroups and in the subgroups of the two induction regimens (DEXA vs PRED) with the exception of age at diagnosis. For patients below 10 years of age the 4-DFS was 90.7% (SE 1.0%) vs 92.5 (SE 1.2, p=0.26) and for patients 10 years or older 81.6% (SE 4.0%) vs 90.3 (SE 4.1%, p=0.04), for P-III vs P-II, respectively. The pattern of relapse was similar after P-III and P-II. The incidence of death in remission was comparable being 0.9% (SE 0.4%) and 0.7% (SE 0.3%) for P-III and P-II. The rate of grade III/IV infections used as an indicator for relevant toxicity was essentially the same in P-III and P-II.

In conclusion, in childhood ALL patients with most favorable prognosis, as predicted by complete MRD response already by day 33, an attempt to reduce the burden of chemotherapy by lower intensity delayed intensification achieved by replacing the traditional BFM protocol II with protocol III, was not successful due to evidence of more relapses observed in patients treated less intensively. Interestingly, almost identical rates of 5-year DFS and CIR (91% and 8% in AIEOP-BFM ALL 2000 vs 93% and 6% in study DCOG 11) were observed. The DCOG authors considered the result of their non-randomized study sufficient to prove that therapy reduction is safely feasible in MRD negative patients defined identically (R. Pieters et al, JCO 2016). Our study underlines the importance of well-designed, sufficiently powered studies for prospective evaluation of treatment reduction in childhood ALL, a major challenge for developing future therapy strategies.