Dexamethasone in Induction Can Eliminate One Third of All Relapses in Childhood Acute Lymphoblastic Leukemia (ALL): Results of An International Randomized Trial in 3655 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 (registered at http://clinicaltrials.gov with identifiers NCT 00430118 (BFM) and NCT 00613457 (AIEOP)). 3655 patients from Germany, Italy, Austria, and Switzerland entered the randomized comparison of dexamethasone (DEX) at 10 mg/m²/d vs prednisone (PDN) at 60 mg/m²/d (both given daily from d8 to d29 of induction plus 9 days tapering after a 7d prednisone prophase). Induction therapy also comprised vincristine (1.5 mg/m²) and daunorubicine (30 mg/m²) both given on d8, 15, 22, 29, as well as L-asparaginase given at 5,000 IU/m² x8 q3d from d12. I.T. methotrexate was given thrice (in CNS-2 and CNS-3 patients for 5 times) in induction. Post-induction therapy was derived from trial ALL-BFM 95 (A. Möricke et al, BLOOD 2008) but stratification and subsequent treatment intensity was based on early response as measured by prednisone response (on d8) and MRD quantification on d33 and d78. With a median follow-up of 4.4 years, 6-year event-free survival (6y-EFS) was 84.1% (SE 1%) and 79.1% (1%) for patients randomized to receive DEX or PDN, respectively (log-rank p=0.0083). The 6-year cumulative incidence (CI) of relapse was 11% (1%) and 18% (1%) for patients randomized to receive DEX or PDN, respectively (Gray p<0.001). The difference between the two groups was observed for isolated BM relapses (8% vs 12%), CNS-relapses (2% vs 4%) and other relapses (2% vs 3%). Higher toxicity was observed in patients treated with DEX: CI for death in induction was 2.0% for DEX and 0.9% for PDN (p=0.003). The latter value was similar to that of study ALL-BFM 95 (0.7%). CI for death in remission was similar being 2.0% and 1.6% for DEX and PDN groups. In each arm, only 0.2% did not achieve CR after intensive consolidation. Severe toxicities, mainly infections, were recorded for DEX treated patients more frequently. Three years into the trial, the DSMC and the TSC decided to halt the randomization for patients aged 10 years or older. CI for relapse/death in induction in this age group is now at 15%/4.5% for DEX, and 20%/2.4% for PDN (p=0.09/0.13). In patients less than 10 years of age, the CI for relapse/death in induction was 11%/1.4% if treated with DEX, and 18%/0.5% if treated with PDN (p<0.001/0.01). If specific biological subgroups were analyzed, a significantly lower CI of relapse for patients randomized to receive DEX was observed in T-ALL as well as TEL/AML1 positive and negative pcB-ALL patients. The reduction was most pronounced in T-ALL patients with good prednisone response after the prophase: CI for relapse in DEX treated patients (n=135) was only 6% (SE 2%) as compared to 20% (SE 4%) in PDN treated patients (n=138; p=0.003). Four patients died in the DEX arm, 3 in the PDN arm for toxicity. In TEL/AML1-positive patients with good prednisone response, the CI for relapse was 4% (SE 1%) in the DEX group and 13% (SE 2%) for PDN treated patients (p<0.001). This large difference became most evident only more than 2 years from diagnosis. The percentage of patients found negative for MRD at d33 was very similar between the randomized groups: It was 47.9% for patients treated with DEX, and 45% for patients treated with PDN. Interestingly, patients with pcB-ALL found MRD negative at d33 had significantly less subsequent relapses if treated with DEX in induction (p(Gray)=0.012). In conclusion, the use of DEX at the same dosage as applied in delayed intensification (10 mg/m²/d for 3 weeks) although associated with a greater risk of severe toxicity leads to a marked reduction of the risk of relapse, translating this into a significant benefit in terms of EFS. This was most evident in patients with in vivo sensitivity to the prednisone prephase, while the efficacy of DEX in poor responding patients was not convincing. In the future, more intensive clinical monitoring and early anti-infective interventions could render the advantage of using DEX even more evident. Moreover, other potentially toxic agents such as anthracyclines may be limited in induction for carefully selected subgroups of patients with the aim of limiting early or late toxicities.