With long-term relapse-free survival in children and young adults with acute lymphoblastic leukaemia (ALL), immediate and long-term treatment-related toxicity remains a major concern. In the UKALL 2003 trial, relapse rates dropped significantly, but treatment-related deaths (5%), and serious non-fatal toxicities (10%) persisted, negatively affecting quality of life (QoL) and long-term health.1,2
Study design
The UKALL 2011 trial is a multicentre, investigator-led umbrella protocol comprising three randomised open-label phase III questions. It evaluated strategies to reduce treatment toxicity and CNS relapse in children (> 1 year) and young adults (under 25 years) with Philadelphia-negative ALL or lymphoblastic lymphoma. Randomisation 1 (R1) compared dexamethasone for 28 days (6 mg/m2; standard) with fourteen days (10 mg/m2; short) to test whether shortening dexamethasone treatment could reduce toxicity. Randomisation 2 (R2) was a factorial randomisation with two aims: to lower risk of CNS relapse by replacing standard interim maintenance (SIM) block of therapy with a high-dose methotrexate (HDM) block, and to improve QoL during maintenance by omitting dexamethasone and vincristine pulses. The primary endpoints were reduction in steroid-related toxicity, CNS relapse rate (CNSR), and bone marrow relapse rate (BMR). EFS was an additional primary endpoint for both randomisations.
Results
In total, 2,750 patients were randomly assigned to R1 (N= 1902) and R2 (N= 1570) with a median follow-up of 99.1 and 87.0 months, respectively. Overall, the 5-year EFS was 83.9% and OS 92.2%. For R1, there was no difference in steroid-related toxicity between short and standard dexamethasone (23.8% vs 25.5%, p= 0.41), which was consistent within subgroups. The EFS was also similar between groups (HR[95%CI]:1.21[0.98–1.49], p= 0.069), indicating no clear advantage of the modified dexamethasone.
As for R2, there was no significant difference in CNSR between SIM and HDM arms, with similar 5-year cumulative incidences of 5.3% and 5.5%, respectively. Furthermore, BMR in the no-pulses arm was non-inferior, showing a 1.7% higher 5-year incidence compared to pulses (HR[95%CI]:1.19[0.87–1.62], p= 0.27). While EFS was significantly worse without pulses (HR[95%CI]:1.34[1.05–1.73], p= 0.021), the difference in relapse risk alone was not statistically significant (HR[95%CI]:1.24[0.96–1.62], p= 0.10).3
Conclusion
A shorter duration of induction dexamethasone does not reduce steroid-related toxicity. Furthermore, HDM does not improve CNSR within a UKALL treatment backbone. Omission of maintenance pulses is non-inferior for BMR.3
References
Top
