Real-world therapeutic response and tyrosine kinase inhibitor discontinuation in chronic phase-chronic myeloid leukemia: data from the French observatory.

Guidelines for tyrosine kinase inhibitor (TKI)-treated chronic phase-chronic myeloid leukemia (CML) management are essentially based on data from clinical research trials; however, real-world data should be valuable for optimizing such recommendations. Here, we analyzed the data collected in the French CML Observatory database, a multicenter real-world cohort (n = 646), using a first-line "intention-to-treat" analysis strategy. This cohort included patients treated with first-line imatinib (n = 484), nilotinib (n = 103), dasatinib (n = 17), imatinib and interferon (n = 9), or second-generation (2G)-TKIs and interferon (n = 29). The cumulative incidence of major molecular response (MMR), MR4, MR4.5 and MR5 confirmed the faster response kinetics with 2G-TKIs. Multivariate analysis identified being a woman and residual disease at month 6 as the main predictive factors of deep molecular response (DMR). Moreover, 30% of patients met the criteria for treatment discontinuation (5 years of treatment and ≥ 2 years of DMR), but only 38% of them stopped treatment. Among the 92 patients who actually discontinued treatment due to optimal response, 31.5% relapsed (48% of them after > 6 months of TKI discontinuation). Multivariate analysis identified age and TKI duration as factors positively correlated with treatment-free remission maintenance. Late (> 6 months) relapses were more frequent in patients with the e14a2 BCR::ABL transcript. Relapse rate was higher in patients who stopped TKI before than after 5 years of treatment (52.6% vs 26%; p = 0.040). These results advocate caution concerning early treatment withdrawal, including in patients receiving 2G-TKIs. This still recruiting database is a valuable source of information for the real-world follow-up of patients with CML.