Benefit of Targeted Intensification for NCI High Risk Childhood Lymphoblastic Leukaemia: Results of the United Kingdom Medical Research Council Trial ALL97 and ALL97/99.

In November1999, the UK childhood ALL trial, ALL 97, adopted CCG risk stratification and treatment regimens in favour of the UKALL approach due to concerns over lower event-free survival (EFS) in the UK compared to the US. A key new feature in the amended trial, ALL97/99, was NCI risk group and early marrow response targeted intensification. Of 1935 patients registered in the trial between January 1997 and June 2002, 997 were treated on ALL97 and 938 on ALL97/99. EFS was better in ALL97/99 compared with ALL 97 (5 year EFS: ALL 97 = 74.1%, 95% CI 71.4%–76.8% vs ALL 97/99 = 78.9%, 76.0%–81.8%, p=0.002). To investigate whether particular sub-groups benefited more or less with the CCG approach, we compared outcomes for different risk groups within the two parts of the trial in regard to EFS, overall survival (OS) and CNS relapse risk. All p values quoted are two-sided. EFS and OS were significantly better in ALL97/99 compared with ALL97 for NCI high risk (HR) patients (5 year EFS: ALL 97 = 61.8%, 95% CI: 56.9–66.7%, ALL 97/99 = 71.8% 95% CI: 66.7–76.9%, p = 0.0006. 5 year OS: ALL97 = 71.3%, 95% CI: 66.8–75.8%, ALL97/99 = 80.3% 95% CI: 76.0–84.6%, p = 0.005), but did not differ significantly for NCI standard risk (SR) patients (5 year EFS: ALL 97 = 81.7% 95% CI: 78.6–84.8%, ALL 97/99 = 83.3% 95% CI: 79.8–86.8%, p = 0.3. 5 year OS: ALL97 = 91.2% 95% CI: 89.0–93.4%, ALL97/99 = 92.6% 95% CI: 90.4–94.8%, p = 0.5). The incidence of isolated CNS relapse was also significantly lower in ALL97/99 for NCI HR (ALL97 = 8% vs ALL97/99 = 3.5%, p = 0.01) but not NCI SR patients (ALL97 = 3.5% vs ALL97/99 = 2.7%, p = 0.6). Despite restricting the use of cranial radiotherapy to patients with overt CNS disease (CNS 3, <5% of all patients), the incidence of isolated CNS relapse in ALL97/99 was reassuringly low, even for sub-groups at high risk of CNS relapse such as those with WCC > 100 × 10 ⁹/l (4.8%) or T cell phenotype (3.8%). Isolated CNS relapse risk in ALL97/99 patients randomised to dexamethasone (which was compared with prednisolone in the trial) was 1.8%, similar to that reported with use of cranial radiotherapy for a higher proportion of patients. A targeted intensification approach improves EFS for NCI HR patients and, especially when combined with systemic dexamethasone, results in a low incidence of isolated CNS relapse for high risk sub-groups without use of cranial radiotherapy.