BACKGROUND AND OBJECTIVE: Since 1988 the AIEOP has used BFM-based chemotherapy for childhood ALL. Current organization and results and role of cranial irradiation in the AIEOP-ALL 91 study are reported. DESIGN AND METHODS: From 1991 to 1995, 1194 children (< 15 years) with non-B ALL, were enrolled and assigned to the standard risk [SR: age > 1 year, non-T-ALL, BFM risk factor (RF) < 0.8], intermediate risk (IR: RF > or = 0.8 but < 1.7, or with RF < 0.8 and age < 1 year, or T-ALL), or high risk [HR: RF > or = 1.7, or t(9;22), or t(4;11) or prednisone poor response or late response or CNS involvement] groups. All patients received initially protocol Ia. Thereafter SR patients received HD-MTX 2 g/m2, a modified protocol II, and continuation therapy with triple intrathecal chemotherapy (TIT); IR patients received protocol Ib, HD-MTX 5 g/m2, protocol II and continuation therapy with TIT; HR patients received 9 polychemotherapy blocks, cranial irradiation and continuation therapy. Duration of treatment was 24 months. A randomized study was conducted to evaluate the impact of high-dose asparaginase in non high risk patients: the results of this study cannot be disclosed yet. RESULTS: One thousand one hundred and fifty-two (96.5%) patients achieved CR. Overall EFS (SE) at 5-years was 71.0% (1.4), with a survival of 80.3% (1.3). Relapse occurred in 262 children (21.9%), either in the marrow (n = 192 isolated and 32 with other sites, 18.7%), in the CNS (n = 18, 1.5%), or elsewhere (n = 20, 1.7%). 5-year EFS (SE) was 83.3% (2.4) in SR, 74.7% (1.8) in IR, and 39.7% (3.5) in HR groups, respectively. INTERPRETATION AND CONCLUSIONS: Overall cure rate was higher than in the previous AIEOP-ALL 88 study. Treatment intensification with polychemotherapy blocks did not improve results in HR. Cranial irradiation can be safely omitted in over 80% of children treated with BFM based chemotherapy.