Durable treatment-free remission in CML beyond clinical trials: Lessons from a large single-center real-world experience.

6582 Background: Treatment-free remission (TFR) has become a major therapeutic objective in chronic-phase chronic myeloid leukemia (CML). While clinical trials have established the feasibility of TFR, real-world data evaluating the influence of deep molecular response (DMR) duration, cumulative tyrosine kinase inhibitor (TKI) exposure, and adherence to molecular monitoring remain limited, particularly in resource-constrained settings. Methods: We conducted a retrospective single-centre analysis of chronic-phase CML patients who electively discontinued TKI therapy between May 2019 and October 2025 after achieving sustained DMR (BCR-ABL ≤0.01% IS). Collected variables included demographics, ELTS risk score, TKI type, duration of DMR (<5 vs ≥5 years), cumulative TKI exposure (<8 vs ≥8 years), molecular relapse (loss of major molecular response [MMR]), time to relapse, and time to MMR regain following TKI resumption. Monitoring compliance was defined as completion of at least five PCR assessments in the first year following TKI discontinuation. Relapse-free survival was analyzed using Kaplan–Meier methods. Results: Eighty-eight patients were included (median age 36 years; 53.4% male). Median cumulative TKI exposure prior to discontinuation was 14.5 years, and median DMR duration was 13 years. At a median follow-up of 40 months, 72 patients (81.8%) remained in sustained TFR. TFR rates were 92% at 6 months, 87% at 12 months, and 82% at 24 months. Fourteen patients (15.9%) restarted TKI therapy due to molecular relapse, one due to withdrawal symptoms, and two died from causes unrelated to CML. Median time to relapse was 4.5 months, with the majority occurring within the first six months.No patient progressed to accelerated or blast phase. All evaluable patients who resumed TKI therapy rapidly regained MMR (median 3 months). TFR rates did not differ significantly according to DMR duration (<5 vs ≥5 years; p =0.67), cumulative TKI exposure (<8 vs ≥8 years; p =0.70), or compliance with molecular monitoring (compliant vs non-compliant; p =0.82). Notably, 30.6% of patients were non-compliant with recommended monitoring schedules, yet high TFR rates and excellent salvage outcomes were maintained. Conclusions: In this large single-centre real-world cohort, over 80% of chronic-phase CML patients achieved durable TFR with predictable early relapses, excellent salvage outcomes, and no disease progression. Neither prolonged DMR duration, extended TKI exposure, nor strict adherence to intensive molecular monitoring significantly influenced relapse risk. These findings support broader, pragmatic implementation of TFR strategies and suggest that less stringent molecular monitoring may be safely adopted in carefully selected patients, particularly in resource-limited settings.