The Effect of Early High Dose Ara-C in Children with High-Risk Acute Lymphoblastic Leukemia (ALL): Tokyo Children’s Cancer Study Group (TCCSG) Study L99-15.

High dose Ara-C is employed in the treatment of acute myeloid leukemia; however, its effect on ALL has not been evaluated systematically. We tested high dose Ara-C incorporated in an early phase with a randomized fashion in children with high-risk ALL. Patients diagnosed as ALL between 1999 and 2003 were consecutively enrolled on TCCSG L99-15 study. Risk stratification was based on the age, initial white blood cell count, immunophenotype, cytogenetics, and the response to prednisolone monotherapy. Totally, 317 out of 770 children (41%) were categorized in a high-risk group. A standard-risk group constituted 35% and a very high-risk group constituted 24% of all the patients. Induction treatment consisted of a standard 4-drug regimen plus cyclophosphamide. Remission was obtained in 310 patients (98%). After remission was obtained, patients were randomized to receive either early intensification regimen A or B: regimen A consisted of high-dose cytarabine (2g/m2 x 8 times with L-asparaginase); regimen B consisted of a combination of cyclophosphamide, cytarabine (75mg/m2 x 15 times), and 6-mercaptopurine. Then all the patients were given an identical treatment including high-dose methotrexate (3g/m2 x 3 times), reinduction therapy, late intensifications, and maintenance therapy. The treatment was completed 24 months after diagnosis. The number of patients randomized in each arm was 155 cases for each arm. Clinical features including immunophenotype, specific karyotype, and the response to prednisolone monotherapy were not different in patients allocated in each arm. The median follow-up for patients who survived at the time of the analysis was 3.2 years. Event-free survival (EFS) (SE) and overall survival (OS) (SE) at 4 years for the 310 patients were 78.4% (2.9%) and 88.9% (2.3%), respectively. EFS (SE) at 4 years for patients in regimen A and B were 77.0% (4.1%) and 79.7% (4.0%) while OS (SE) at 4 years in regimen A and B were 88.3% (3.2%) and 89.7% (3.2%). The site of relapse was not different in two arms. In 23 cases with T-cell phenotype, patients treated with regimen A did not have better EFS than the others. Similarly, in 25 cases with E2A-PBX1 rearrangement, patients treated with regimen A did not have better EFS than the others. Although no patients in either regimen died of infections, more patients in regimen A experienced bacterial sepsis: 13 cases in regimen A vs. 8 cases in regimen B. In conclusion, we did not find any benefit of employing early high-dose Ara-C in children with high-risk ALL.