With the use of more intensive regimens including prophylactic CNS treatment, the prognosis of children with ALL has dramatically improved over the last three decades. The aim of this cross-sectional, nationwide study was to comprehensively assess long-term toxicity in ALL survivors, with special attention given to neuropsychological morbidity, and to look for possible differences in cognitive outcome between children having received prophylactic cranial irradiation and those not having received it. Between 1994 and 1996, long-term survivors of ALL were assessed in a multi-center setting according to a standardized protocol which included, besides usual clinical and laboratory investigations, a comprehensive endocrine work-up. Additionally, children having received anthracyclines were checked for possible late cardio-toxicity with echocardiography and ECG. Intellectual performance was evaluated with standardized neuropsychological tests (age-adapted versions of the Wechsler test). One-hundred and fifty patients were eligible for the study. The median age at diagnosis was 5 years and at evaluation 16 years, for a median follow-up of 10 years. Thirty-five patients had cranial irradiation as part of the prophylactic CNS treatment. One-hundred and forty (93%) of the 150 eligible patients were completely evaluated in terms of global long-term toxicity: 117 (83%) long-term survivors had no (n = 61) or only minimal (n = 56) late toxicity; 19 (14%) suffered from moderate impairments; 4 (3%) showed severe somatic or neuropsychological sequelae. Intellectual performance could be assessed in 147 (98%) of the 150 eligible patients. The mean global, verbal and non-verbal IQs (103, 105 and 101 respectively) of the ALL survivors as a group were comparable with those found in the general population. The results of the comparison between children having and those not having received prophylactic cranial irradiation showed: 1) significantly higher scores in chemotherapy-only treated patients, both for the global and the verbal performances; 2) significantly poorer results in specific items of the Wechsler test (short-term verbal memory, arithmetics, concentration/speed of processing) in irradiated children. These findings which show the deleterious role of cranial irradiation correlate well with many other reports found in the literature. However, they could have been influenced by the significantly longer time interval observed between therapy and evaluation in our irradiated patients. Prospective studies are needed to further characterize the potential neuropsychological hazards of chemotherapy and their evolution over time.