One hundred ninety-eight children and adolescents were entered in the Associazione Italiana di Ematologia ed Oncologia Pediatrica (AIEOP)-ALL95 study for high-risk acute lymphoblastic leukemia (ALL). Inclusion criteria were poor response to initial prednisone/intrathecal methotrexate (prednisone-poor response [PPR]), resistance to induction therapy, translocation t(9;22), infants with the t(4;11), or CD10(-) ALL. The event-free survival (EFS) rate at 4 years was 56.5% (SE, 3.9%) for the entire group. The overall EFS rate in the current study was significantly better (P =.002) than that obtained in a comparable group of patients treated in the early 1990s in the AIEOP-ALL91 study. In particular, patients with PPR had a 4-year EFS of 61.1% (SE, 4.4%) versus 42.8% (SE, 5.4%) in the ALL 91 study (P =.008). Among PPR patients, those who were PPR-only (60.1%)-that is, they achieved CR and were negative for t(9;22) and t(4;11) translocations-had the best outcomes with this intensive treatment, even when additional adverse features (hyperleukocytosis, T phenotype) were present (4-year EFS, 70.1%; SE, 4.7%). We attribute this improvement to the replacement of 6 alternating blocks of non-cross-resistant drugs with an 8-drug reinduction regimen (Berlin-Frankfurt-Muenster [BFM] protocol II), repeated twice, in the context of a standard BFM-type intensive chemotherapy for high-risk ALL. This modified therapy may lead to high cure rates for patients defined as at high risk for intrinsic resistance to corticosteroids only.