AALL0434
Children's Oncology Group AALL0434: a phase III randomized clinical trial testing nelarabine in newly diagnosed T-cell acute lymphoblastic leukemia
Phase 3N = 1,562NCT00408005
- Population
- Children and young adults aged 1-31 years with newly diagnosed T-ALL (1,562 evaluable).
- Design
- Augmented Berlin-Frankfurt-Munster (ABFM) regimen with a 2 x 2 pseudo-factorial randomisation of methotrexate (escalating-dose methotrexate without leucovorin rescue plus pegaspargase, C-MTX, versus high-dose methotrexate with leucovorin rescue, HDMTX). Intermediate- and high-risk patients were also randomly assigned after induction to six 5-day courses of nelarabine or none. Induction failures were non-randomly assigned to HDMTX plus nelarabine; patients with CNS3 disease received HDMTX and were randomised to nelarabine or none. All except low-risk patients received cranial irradiation. Enrolled 2007 to 2014.
- Primary endpoint
- DFS
- Arms
- ABFM plus nelarabine (experimental, n = 323): Augmented BFM with six 5-day nelarabine courses (randomised, intermediate- and high-risk patients)
- ABFM without nelarabine (control, n = 336): Augmented BFM without nelarabine (randomised, intermediate- and high-risk patients)
- People
- Dunsmore KP, senior author Hunger SP, Children's Oncology Group
Δ 60-month DFS+6.1 pointsΔ 60-month OS+2.4 pointsΔ 60-month CNS relapse−5.6 points
Show the numbers
| Endpoint | Analysis | ABFM plus nelarabine | ABFM without nelarabine | Effect |
|---|---|---|---|---|
| DFSPatients with T-ALL randomised to nelarabine (n=323) or no nelarabine (n=336)Disease-free survival at 5 years, nelarabine randomisation | finalDunsmore 2020 | 88.2% at 5 y | 82.1% at 5 y | P=0.029Reported as 88.2% ± 2.4% and 82.1% ± 2.7% (± are standard errors, Peto method). Four-arm comparison P=0.01, with no interaction between the methotrexate and nelarabine randomisations (P=0.41); the best-performing arm (C-MTX plus nelarabine) had 5-year DFS of 91% (n=147). The power calculation for the nelarabine comparison used a one-sided log-rank test; other tests were two-sided unless stated. |
| OSPatients with T-ALL randomised to nelarabine (n=323) or no nelarabine (n=336)Overall survival at 5 years, nelarabine randomisation | finalDunsmore 2020 | 90.3% at 5 y | 87.9% at 5 y | P=0.168Reported as 90.3% ± 2.2% and 87.9% ± 2.3% (± are standard errors). Not statistically significant; not stated in the abstract. |
| CNS relapsePatients randomised to nelarabine (n=323) or no nelarabine (n=336)Five-year cumulative incidence of isolated and combined CNS relapse, nelarabine versus no nelarabine (randomised patients) | finalDunsmore 2020 | 1.3% at 5 y | 6.9% at 5 y | P=0.0001Reported as 1.3% ± 0.63% and 6.9% ± 1.4% (± are standard errors). The abstract does not state the time point; the full text gives 5-year cumulative incidence in randomised patients. CNS-involved relapses: 4 with nelarabine (1 isolated, 2 with marrow, 1 with marrow and another site) and 23 without (14 isolated, 8 with marrow, 1 with marrow and another site). |
| Endpoint | Analysis | ABFM plus nelarabine | ABFM without nelarabine | Effect |
|---|---|---|---|---|
| DFSPatients with T-ALL randomised to nelarabine (n=323) or no nelarabine (n=336)Disease-free survival at 5 years, nelarabine randomisation | finalDunsmore 2020 | 88.2% at 5 y | 82.1% at 5 y | P=0.029Reported as 88.2% ± 2.4% and 82.1% ± 2.7% (± are standard errors, Peto method). Four-arm comparison P=0.01, with no interaction between the methotrexate and nelarabine randomisations (P=0.41); the best-performing arm (C-MTX plus nelarabine) had 5-year DFS of 91% (n=147). The power calculation for the nelarabine comparison used a one-sided log-rank test; other tests were two-sided unless stated. |
| OSPatients with T-ALL randomised to nelarabine (n=323) or no nelarabine (n=336)Overall survival at 5 years, nelarabine randomisation | finalDunsmore 2020 | 90.3% at 5 y | 87.9% at 5 y | P=0.168Reported as 90.3% ± 2.2% and 87.9% ± 2.3% (± are standard errors). Not statistically significant; not stated in the abstract. |
| CNS relapsePatients randomised to nelarabine (n=323) or no nelarabine (n=336)Five-year cumulative incidence of isolated and combined CNS relapse, nelarabine versus no nelarabine (randomised patients) | finalDunsmore 2020 | 1.3% at 5 y | 6.9% at 5 y | P=0.0001Reported as 1.3% ± 0.63% and 6.9% ± 1.4% (± are standard errors). The abstract does not state the time point; the full text gives 5-year cumulative incidence in randomised patients. CNS-involved relapses: 4 with nelarabine (1 isolated, 2 with marrow, 1 with marrow and another site) and 23 without (14 isolated, 8 with marrow, 1 with marrow and another site). |