Background: An overview of trials dating back to the ‘70s and ‘80s has shown that addition of vincristine (VCR)+prednisone/prednisolone pulses to continuation therapy of childhood ALL, i.e. 6-mercaptopurine (6-MP) + methotrexate (MTX), improved disease-free survival (DFS) (Lancet, 1996). However, a recent randomized intergroup trial (IGT) has shown that, when given to intermediate risk (age<1 or >=6 yrs or WBC>20×109/L) patients (pts) treated according to the BFM protocol, VCR+dexamethasone (DEX) pulses in continuation therapy failed to improve DFS and overall survival (Lancet, 2007).Methods: In 12/1998, EORTC CLG started the randomized phase III 58951 trial addressing 3 questions:R1) the value of DEX vs PRED in induction;R2) the value of prolonged (24 injections) vs conventional (12 injections) duration of L-asparaginase (ASPA) courses during consolidation and late intensification;R3) for average risk (AR) pts only: the value of 6 (VCR+corticosteroid) pulses every 10 weeks during continuation therapy.The corticoid of the pulses was that assigned to the patient at R1: DEX (6 mg/sm/d) or prednisolone (PRED) (60 mg/sm/d) during 7 days. AR pts were defined by being neither low risk (hyperploidy > 50 or DNI>1.16, and WBC<10×109/L and no extramedullary involvement) nor very high risk (Leukemia 2000). A total of 81 pts from the EORTC 58951 trial were included in the IGT study. Randomization was done centrally, and the main endpoint was DFS; secondary endpoints were OS and toxicity. Intent-to-treat analysis was used.Results: Between 6.1999 and 11.2004, 411 pts, ALL (N=384) and NHL (N=27), were randomized for the pulse question. In the Pulses group, 101 vs 101 pts were randomized for PRED vs DEX. At a median follow-up of 6.3 years, there were 19 vs 34 DFS events for the Pulses vs No Pulses comparison: BM (10 vs 16), CNS only (1 vs 4), other isolated (2 vs 3), BM+CNS (2 vs 5), BM+other (4 vs 4), death in CR (0 vs 2). The 6-year DFS rate from randomization was 90.6% (SE 2.1%) in the Pulses group and 82.8% (SE 2.8%) in the No Pulses group (hazard ratio (HR) = 0.54, 95% CI 0.32–0.94, 2-sided logrank p=0.027). The impact of pulses on DFS was similar in the PRED group (HR=0.56) and the DEX group (HR=0.59). In girls the treatment difference seemed to be more pronounced (HR=0.25, 99% CI 0.04–1.25; p=0.015) than in boys (HR=0.71, 99% CI 0.30–1.66; p=0.30), and also in those having been randomized to receive conventional duration of ASPA courses (HR=0.46, 99% CI 0.18–1.19; p=0.03) than in prolonged ASPA arm (HR=0.87, 99% CI 0.23–3.29; p=0.78). Grade 3–4 hepatic toxicity was lower in the Pulses group (30% vs 40%); incidence of grade 2–3 osteonecrosis was 4.4% (Pulses) vs 2% (No Pulses) and pancreatitis rate was 4.9% (Pulses) vs 2.9% (No Pulses).Conclusion: VCR+corticosteroid pulses, at long follow-up (median=6.3 yrs), significantly improved the DFS, particularly in pts having received conventional duration of ASPA. Pulses did not increase toxic effects. In future, for AR pts treated according to the BFM protocol, pulses should become a standard component of therapy.