Hemoglobin Changes during Long-Lasting Frontline Treatment with Tyrosine-Kinase Inhibitors in Patients with Chronic Myeloid Leukemia

Introduction Chronic Myeloid Leukemia (CML) has become highly curable after introduction of Tyrosine-Kinase Inhibitors (TKIs). However, several side-effects have been reported with the prolonged use of TKIs: in particular, severe cardiovascular and pulmonary toxicities were observed with 2^nd generation TKIs (2G-TKIs), nilotinib and dasatinib. Imatinib is generally considered safer, even if some concerns were recently raised on its renal toxicity and occurrence of late anemia.

Aims To evaluate the impact of imatinib compared to 2G-TKIs on the hemoglobin (Hb) levels in the long-lasting frontline treatment of CML patients.

Methods From 1/2002 to 12/2015, 365 CML patients were diagnosed and treated frontline with TKIs in 2 different Centres in Italy: of them, 123 permanently discontinued the treatment before the 5^th year from the start due to intolerance (31 cases, 25.2%), primary resistance (41, 33.3%), secondary resistance (16, 13.0%), blastic evolution (4, 3.2%), unrelated death (12, 9.8%) or were lost to follow-up (19, 15.5%). At 5 years, the remaining 242 patients were still receiving frontline TKI treatment and were considered for the present analysis. Hb levels were recorded at baseline and thereafter every 12 months up to the 5^th year of treatment.

Results As to frontline treatment, 186 patients (76.8%) received imatinib and 56 (23.2%) 2G-TKIs (nilotinib in 44 cases and dasatinib in 12, respectively). The main clinical features at diagnosis of the entire cohort and according to frontline treatment are reported in the table: median Hb value at baseline was significantly lower in patients treated with 2G-TKIs. Median Hb values at different time-points according to frontline treatment are reported in the Figure. In patients treated with imatinib, median Hb levels at 12^th (12.5 g/dl, IQR 11.6-13.5) and 60^th month (12.4 g/dl, IQR 11.4-13.3) were stable compared to baseline (12.8 g/dl, IQR 11.3-13.8) (p=0.248 and p=0.075, respectively). On the contrary, median Hb levels at 12^th (13.4 g/dl, IQR 12.2-14.3) and 60^th month (13.6 g/dl, IQR 12.0-14.6) showed a significant increase compared to baseline (11.8 g/dl, IQR 10.6-13.7) in patients treated with 2G-TKIs (p<0.001 in both cases). As a consequence, during the treatment median Hb values became significantly higher at any different time-point in patients treated with 2G-TKIs compared to patients treated with imatinib (p=0.005 at the 12^th month, p=0.010 at the 60^th month). At baseline, the rate of patients with mild to moderate anemia (Hb < 11 g/dl) was significantly lower in those treated with imatinib [34/186 (18.2%) vs 20/56 (35.7%) treated with 2G-TKIs, p=0.006]: at the 12^th month, no difference was observed [15/186 (8.0%) in patients treated with imatinib vs 3/56 (5.3%) in patients treated with 2G-TKIs, p=0.498], while at the 60^th months the rate of patients with mild to moderate anemia became significantly higher in those treated with imatinib [29/186 (15.6%) vs 2/56 (3.6%) treated with 2G-TKI, p=0.018].

Conclusions Present data highlight that long-lasting treatment with imatinib can have a negative late effect on erythropoiesis, with incomplete recovery of hemoglobin levels and occurrence of late anemia in about 15% of patients at the 60^th month of therapy.This possible adverse event, which is still unrecognized and seems very rare with 2G-TKIs, could affect quality of life and should be recognized in the long-term management of CML patients.