Permanent Discontinuation of Tyrosine Kinase Inhibitor Frontline Therapy in Patients with Chronic Phase Chronic Myeloid Leukemia Patients during the First 36 Months of Treatment: A "Campus CML" Study

Abstract

Background Tyrosine kinase inhibitors (TKIs) have changed prognosis and treatment results of chronic myeloid leukemia (CML). However, the first years of therapy are crucial to achieve optimal response, defined as the achievement of deep molecular response: in this early phase, however, different toxicities or suboptimal response/resistance may occur, leading to permanent frontline TKI discontinuation and need for a second-line switch. Methods To evaluate in a large real-life cohort of CML patients the incidence and the different pattern of events leading to a permanent frontline TKI discontinuation during the first 3 years of therapy, 1168 CP-CML patients diagnosed from 1/2012 and 12/2019 at 21 Hematology Centres and treated with frontline imatinib (IM) or second-generation (2G) TKIs dasatinib or nilotinib were retrospectively analysed. Results Frontline TKI was IM in 737 (63.1%) and 2G-TKIs in 431 (36.9%) cases: the main clinical features at diagnosis of the entire cohort and according to frontline treatment are reported in the Table 1. Commonest comorbidities were arterial hypertension (37.4%), previous neoplasm (14.1%), diabetes (10.9%), peripheral vascular diseases (8.4%), ischemic heart disease (6.9%) and COPD (6.7%). IM-treated patients were older (p<0.001) and had more comorbidities (p=0.002 for COPD and p<0.001 for all other diseases). Total number of patients who had discontinued frontline TKI at the 36th month of observation was 391/1168 (33.5%), being higher with IM (294/737, 39.8%) than with 2G-TKIs (96/431, 22.3%) (p<0.001). Cumulative incidence of permanent TKI discontinuation was 19.6% [95% confidence interval (CI) 17.3-21.9] at 12 months, 29.7% (95%CI 27.0-32.4) at 24 months and 34.2% (95%CI 31.5-36.9) at 36 months, respectively: cumulative incidence of discontinuation according to frontline TKI is shown in the Figure (p<0.001 at any time-point). Among the 391 patients who discontinued frontline TKI, the main causes were hematologic toxicity in 27 (6.9%), extra-hematologic toxicity in 106 (27.1%), primary resistance in 166 (42.4%), secondary resistance in 18 (4.6%), evolution in blastic phase in 17 (4.3%), unrelated deaths in 36 (9.2%) and other causes in 21 (5.4%). At univariate analysis, in addition to IM as frontline TKI, baseline factors predictive for an higher rate of TKI discontinuation were age > 65 years (p=0.001), WBC ≥ 100 x 109/l (p<0.001), Hb < 10 g/dl (p<0.001), spleen enlargement ≥ 5 cm below costal margin (p<0.001) and high Sokal score (p<0.001). At multivariate regression logistic analysis, frontline IM (HR 3.22; 95%CI 2.34 - 4.42, p<0.001), WBC ≥ 100 x 109/l (HR 2.18; 95%CI 1.57 - 3.02, p<0.001), spleen enlargement ≥ 5 cm (HR 1.91; 95%CI 1.22 - 3.00, p=0.005) and Sokal high risk (HR 1.78; 95%CI 1.16 - 2.72, p=0.007) retained an independent predictive role on TKI discontinuation. Conclusions This real-world study on over 1100 CML patients reveals that about one third of newly diagnosed CML patients discontinued frontline TKI during the first 3 years of therapy, mostly for primary resistance or toxicity. Discontinuation rates were higher with IM compared to 2G-TKIs at any time-point: however, the impact of older age, higher risks and heavier burden of comorbidities in IM patients should be considered. Other factors seemed to affect the risk of frontline TKI discontinuation but their role needs further refined investigations. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal