Background and Importance Tyrosine-kinase inhibitors (TKIs) have shown to be effective in chronic myeloid leukaemia (CML) treatment. Recent clinical trials show selected patients with deep molecular response (DMR) can safely discontinue treatment.

Aim and Objectives Describing clinical experience of discontinuing treatment with TKIs in CML patients.

Material and Methods A retrospective observational study analysed TKIs discontinuation and maintenance of major molecular response (MMR) after discontinuation in all CML patients treated at our centre from the moment they started TKIs until September 2023.

Discontinuation protocol stipulates patients must have been treated for five first generation TKIs) or three (second generation TKIs) years and must have achieved 2 years of DMR (molecular response (MR) =4 or greater). After discontinuation they have monthly monitoring visits for 6 months (period when most patients lose MMR), afterwards controls are spaced out over time. If patients lose MMR (MR=3) treatment should restart.

Variables age, gender, TKI, start date, response, DMR achieving date, TKI switch before discontinuation and cause, discontinuation and date, withdrawal syndrome (WS), WS treatment, restart date and TKI, last consultation date.

Results Sixty-two CML patients were treated with TKIs and 48.4%(30) discontinued. Median age of patients who discontinued was 57.8 years [interquartile range (IQR): 50.1–67.1], 63.3% were female.

We found 73.3% discontinued with 1st-line TKIs, 26.6% received various TKIs before discontinuation due to: toxicity (60%) and suboptimal response(40%).

For those who discontinued median TKI treatment until discontinuation was 6.2 years [IQR: 4.9–12.1], and median time with DMR was 4.9 years [IQR: 3.3–8.1]. When they discontinued, they were treated with: imatinib (63.3%), nilotinib (23.3%), dasatinib (6.7%), bosutinib (6.7%).

Five patients developed WS: osteomuscular pain (4), panniculitis (1). One patient received corticosteroids and two received analgesics.

63.3% maintained discontinuation, follow-up median of 3.4 years [IQR: 0.9–4.5].

36.7% patients lost MMR, follow-up median until restart was 5.3 months [IQR: 4.2–6.9]. Seven patients restarted with previous TKI, four changed to second generation TKIs. One had a late relapse at 19.4 months. All patients regained MMR after restarting treatment.

Conclusion and Relevance Our results are in line with current literature showing controlled discontinuation is a viable and potentially long-term option. Discontinuation is already part of the standard of care in selected patients since it’s cost-effective, representing savings for Healthcare System and improving patient’s life quality.

Conflict of Interest No conflict of interest.