Discontinuation of tyrosine kinase inhibitors in chronic myeloid leukemia. a randomized national fi-LMC comparative trial of two therapeutic strategies.

Abstract

Abstract Introduction The permanent discontinuation of treatments with tyrosine kinase inhibitors (TKIs) (treatment-free remission, TFR) has become a major goal in chronic myeloid leukemia (CML) to prevent the occurrence of adverse events, improve quality of life, and reduce treatment costs. Various strategies are being evaluated to improve the TFR rate, including the use of second-generation TKIs (2G-TKIs) and the addition of interferon alpha. Interestingly, greater success has been observed by introducing a dose reduction phase (half-dose) prior to TKI cessation (UK DESTINY trial). Here, we present a comparative, phase III, prospective clinical trial designed to demonstrate that a dose de-escalation strategy prior to treatment discontinuation optimizes TFR rates. Beyond the expected benefits of dose de-escalation, we aim to study the recovery of anti-leukemic immune effectors before and after the de-escalation phase. Methods We are conducting a prospective, randomized, open-label, multicenter, French clinical trial in CML (NCT05753384) to compare TFR outcomes at 24 months (primary endpoint) between a group with sudden treatment discontinuation after a 12-month maintenance phase and a group undergoing a 12-month dose de-escalation phase (dosage reduced by 50%) prior to cessation. After these 12 months of therapy, patients in both arms are eligible in the two arms for treatment discontinuation upon confirmation of sustained deep molecular response (DMR). Inclusion criteria include CML patients in chronic phase (CML-CP), aged ≥ 18 years, with a typical BCR::ABL1 transcript, a molecular response ≥ 4 log (MR4) lasting ≥ 1 year, and treatment duration ≥4 years for imatinib, ≥3 years for 2G-TKIs, or ≥4 years for both, with no treatment changes in the 6 months prior to inclusion. Minimum TKI doses required for eligibility are: imatinib (≥ 300 mg/day), dasatinib (≥ 50 mg/day), nilotinib (≥ 300 mg/day), and bosutinib (≥ 200 mg/day). Main exclusion criteria include uncontrolled chronic disease, ECOG performance status ≥ 3, prior TKI resistance, previous TKI discontinuation attempts, and prior allogenic stem cell transplantation. A dynamic randomization process (minimization) is used to allocate patients while balancing known prognostic factors for TFR (treatment duration, DMR duration, type of TKI: imatinib versus 2G-TKIs) between the two groups. To detect a 25% difference between arms, with a two-sided alpha of 5% and a statistical power of 80%, 60 patients per arm achieving TKI cessation are required. Accounting for possible DMR loss during the treatment, 170 patients must be enrolled. Secondary endpoints include the proportion of patients maintaining DMR and major molecular response (MMR) during treatment, the impact of reduced plasma TKI levels on innate lymphocyte reactivity, and their predictive value for successful discontinuation. Longitudinal immunomonitoring is being conducted at treatment initiation, end of the TKI phase and throughout discontinuation, along with assessment of residual plasma TKI concentrations. Results As of August 5, 2025, a total of 163 patients had been enrolled (95.8% to completion of recruitment). If our hypothesis about the superiority of the de-escalation strategy over sudden discontinuation is confirmed, our findings may inform future recommendations. Our study will provide critical data on dose de-escalation strategy (optimal treatment and DMR durations) as well as early treatment discontinuation: 4 years or 3 years of standard-dose treatment required for imatinib and 2G-TKIs, respectively. We also aim to demonstrate that immunologic surveillance of residual leukemic cells is a critical factor predicting successful TFR. A key objective of the planned immunomonitoring is to identify an innate T cell-based immunological signature predictive of successful treatment discontinuation, which may guide personalized management and inform future immunotherapeutic strategies. Finally, this trial offers a unique opportunity to explore the hypothesis of an inverse correlation between the number and function of anti-leukemic effector cells and the plasma TKI levels. Conclusions We aim to demonstrate that de-escalation of TKI therapy improves the proportion of patients who can successfully discontinue treatment, thus maintaining a stable MR4 and achieving prolonged TFR. We also seek to provide evidence that this benefit is supported by the immune system effectors, particularly by innate T cells.