Predictive factors of stable deep molecular response in chronic myeloid leukemia patients treated with imatinib standard dose: a study from the Gruppo Triveneto LMC.

Background. A sustained deep molecular response (MR4 or better) is a validated criterion for discontinuation of Tyrosine Kinase Inhibitors (TKI) in patients with Chronic Myeloid Leukemia (CML), but only a minority of patients attain this response. Predictive factors of stable MR4 have been reported in patients treated with different dosages of imatinib (i.e from 400 to 800 mg daily) but data about long-term treatment with standard dose imatinib are lacking. Moreover, various definitions of sustained MR4 have been used in these studies. Aims and methods. To assess the probability and the predictors of a stable MR4 we restrospectively analyzed our cohort of chronic phase CML patients treated with imatinib 400 mg daily, as first-line therapy or after interferon (IFN) failure. IFN-treated patients already in complete cytogenetic response at the time of imatinib start were excluded from the analysis. Major molecular response (MMR) was defined as BCR-ABLIS ratio <0.1%. Deep molecular response (MR4) was defined as BCR-ABLIS ratio ≤0.01% or undetectable disease with ≥10,000 ABL copies. Patients with MR4 lasting ≥ 2 years and at least a Q-PCR test every 6 months were defined as stable MR4. Patients with any sample >0.01% BCR-ABLIS after the achievement of MR4 were defined as unstable MR4. Baseline factors (age, sex, Sokal, Hasford and EUTOS risk score, type of BCR-ABL transcript, pre-treatment with IFN) and response to imatinib at 3, 6, and 12 months according to the European LeukemiaNet (ELN) 2013 recommendations have been examined for the association with stable MR4. Frequencies were compared by Fisher’s exact test. Univariate and multivariate regression analysis were performed using the competing risk model of Fine and Gray, where the achievement of the response was the event of interest, and cessation of imatinib 400 mg daily for any reason (including dose increase for resistance, and death) were the competing risks. The significance of individual parameters comprising more than 2 variables was determined by the Wald test. Results. A total of 320 patients (260 treated with imatinib front-line and 60 after IFN) was evaluated. Median age at diagnosis was 57 years (range 20-88). Sokal distribution was 42%, 40% and 15% for low, intermediate and high risk, respectively (3% were not evaluable). After a median follow-up from imatinib start of 74 months, 146 patients (46%) never reached MR4, 84 patients (26%) obtained an unstable MR4 and 90 patients (28%) achieved a stable MR4. The cumulative incidence of stable MR4 was 26.8% (95% CI: 20.8-32.3%) at 5 years and 39.3% (95% CI: 32.1-45.7%) at 10 years. Median time to first MR4 for patients subsequently obtaining a stable MR4 was 25.3 months and all but 5 stable responders achieved the MR4 within 5.5 years of imatinib. As compared to patients with unstable MR4, those with stable MR4 tended to have a higher frequency of e14a2 transcript (63% vs 53%; p=0.07) and had a marginally significant lower frequency of IFN pre-treament (16.7% vs 29.8%; p=0.048) while no differences were observed concerning the other baseline factors, including sex. Predictors of stable MR4 were: type of transcript (e14a2 vs e13a2 HR 2.07; p=0.003), pre-treatment with IFN (no IFN vs IFN HR 2.45; p=0.002), BCR-ABL level at 3- (≤10%IS vs >10%IS HR 3.48; p=0.004), 6- (<1%IS vs 1-10%IS HR 9.95; p=0.001) and 12-months (MMR vs no MMR HR 12.5; p<0.001). Considering only patients with optimal response to imatinib, rates of stable MR4 were significantly higher for e14a2 than e13a2 or e13a2/e14a2 patients both at 3 (47.8% vs 23.2% and 21%; p=0.003), 6 (48.6% vs 28.8% and 33.3%; p=0.027) and 12 months (53.7% vs 25.9% and 40%; p=0.037). Conclusions. Biological variables such as the type of BCR-ABL transcript could influence the achievement and the stability of MR4, also in patients with optimal kinetics of response to imatinib. Moreover, our results suggest that a rapid attainment of MR4 during the first years of imatinib treatment is a predictor of a subsequent sustained response, a prerequisite for TKI discontinuation.