Prophylactic cranial irradiation dose effects on late cognitive function in children treated for acute lymphoblastic leukemia.

Prophylactic central nervous system treatment has dramatically improved the disease-free survival of children with acute lymphoblastic leukemia (ALL). Long-term neuropsychological sequelae are documented in children who received 2400 cGy prophylactic cranial irradiation. The dose was reduced to 1800 cGy. Available reports on developmental consequences, with short follow-up, have yielded inconsistent results. This study assesses radiation dose effects on cognitive function in children with leukemia who received central nervous system prophylaxis with 2400 cGy versus 1800 cGy whole brain radiotherapy. All leukemic children also received intrathecal methotrexate. A control group of children (treated for Wilms' tumor) received no central nervous system therapy. Nineteen children were treated with 2400 cGy, 16 children with 1800 cGy. The 12 control children received no irradiation. All patients were off therapy for at least 70 months. The 1800 cGy and 2400 cGy patient groups were off therapy for equivalent periods of time (range 70-123 mo) at follow-up testing. Mean age at diagnosis was 49 months, at testing: 142 months. The male to female ratio was 1/1. Standardized psychological tests were administered. Full-Scale, Verbal, and Performance IQ were measured with the Wechsler Intelligence Scale for Children-Revised. Wide Range Achievement Testing evaluated reading, spelling, and arithmetic abilities. Children treated with 1800 cGy performed significantly better than those who received 2400 cGy, and at the same level as controls. There were statistically significant differences between the 1800 cGy and 2400 cGy subjects in all measures. 2400 cGy patients had deficiencies in IQ and academic performance. 1800 cGy patients scored approximately 12 points higher than 2400 cGy children. Eleven children, two in the control group, two in the 1800 cGy, and seven in the 2400 cGy group had IQ scores of less than 90. Eight of the nine irradiated children with deficits had radiotherapy before age 5. These results indicate a mild, but diffuse information processing deficit in children who received 2400 cGy, but not in children who received 1800 cGy. These findings with a minimum of 6 years of follow-up provide new information on late effects of CNS prophylaxis in ALL. Reducing the cranial RT dose from 2400 cGy to 1800 cGy reduced neurotoxicity to acceptable levels.