A phase 2 trial of decreasing tyrosine kinase inhibitor dose in chronic myeloid leukaemia patients with stable major molecular response: data from the British DESTINY study

BACKGROUND: Discontinuation of tyrosine kinase inhibitor (TKI) therapy is feasible for some patients with chronic myeloid leukaemia (CML) with deep molecular responses, defined as stable MR4 (BCR-ABL1/ABL1 ratio <0.01%). However, patients in stable major molecular response (MMR; MR3; BCR-ABL1/ABL1 ratio consistently < 0.1%) but not MR4 have not hitherto been studied. In addition, the effect of treatment de-escalation rather than outright stopping has not been investigated so far. PATIENTS and METHODS: This study recruited 174 British adult CML patients in first chronic phase who had received TKI for ≥3 years and were either in stable MR4 (the ‘MR4 cohort’ n=125) or in stable MMR but not MR4 (the ‘MMR cohort’; n=49) for ≥12 months. Participants received half their standard TKI dose for 12 months. Molecular recurrence was defined as loss of MMR (>0.1%) on two consecutive samples. The study endpoint is the proportion of patients who lose their MMR on de-escalation and regain MMR on TKI resumption. The trial was registered at https://clinicaltrials.gov/ as NCT 01804985. FINDINGS: During the 12 months of half-dose therapy, 12 patients had molecular recurrence, all of whom regained MMR within 4 months of full dose TKI resumption. Recurrence was lower in the MR4 cohort (3 of 121 evaluable patients; 2.5%, 90% CI: 0.2-4.8%) than in the MMR cohort (9 of 48 evaluable patients; 18.8%, 90% CI: 9.5-28%) (p = 0.0007), but was unrelated to prior TKI or TKI therapy duration. Many adverse events improved during the first 3 months of de-escalation, though not thereafter. Overall, de-escalation saved 46.7% from an expected TKI budget (without de-escalation) of £4,156,969. INTERPRETATION: TKI de-escalation is safe for the vast majority of patients with excellent responses to TKI therapy, and is associated with improvement in symptoms and significant financial savings. The data imply that lower TKI doses may maintain responses in these patients.