Prospective evaluation of pro/anti-inflammatory cytokines during TKI treatment in chronic myeloid leukemia patients

Chronic myeloid leukemia (CML) treatment with tyrosine kinase inhibitors (TKIs) has been associated to an increased risk of Arterial Occlusive Events (AOEs), mainly with nilotinib, but the mechanisms underlying these events have been not clarified yet. Previously, we confirmed in our retrospective cross-sectional study a higher cardiovascular (CV) risk in nilotinib treated patients, particularly if harboring the unfavorable OLR1 polymorphism and we found a nilotinib-associated pro-inflammatory effect. We started a multicenter prospective study of tyrosine Kinase Inhibitors induced pro-AtherothROmbotic (KIARO) status in CML patients to furtherly investigate a possible pro-atherothrombotic nilotinib-induced status in a cohort of chronic phase (CP)-CML patients treated with first-line imatinib, nilotinib and dasatinib. In particular, our intents were: to assess any changes in the inflammation status during TKI treatment by measuring pro/anti-inflammatory cytokines (IL-6, IL-10, TNFα and ox-LDL) plasma levels; 2) to record AOEs after applying CV SCORE and evaluate its predictive role; 3) to correlate AOEs with altered inflammation status. A total of 186 CP-CML patients were enrolled in this study of which 89/186 (48%) were treated with imatinib, 59/186 (32%) with nilotinib and 38/186 (20%) with dasatinib. Results from biochemical analyses performed by enzyme-linked immunosorbent assay (ELISA) test showed higher IL-10 levels at 6 and 12 months in imatinib (p=0.012 and p=0.009, respectively) and dasatinib (p=0.032 and p=0.014, respectively) cohorts compared to nilotinib, while ox-LDL levels increased at 12 months in the nilotinib cohort (p=0.041) in contrast to imatinib and dasatinib. Consequently, IL-6/IL-10 and TNFα/IL-10 ratios were higher in nilotinib cohort compared to imatinib (p=0.042, p=0.044 at 6 months; p=0.040, p=0.041 at 12 months) and dasatinib (p=0.049, p=0.040 at 6 months; p=0.041, p=0.044 at 12 months), suggesting a TKI-driven pro-inflammatory status in nilotinib treated patients. We recorded an AOE only in 5% of patients and, due to the small number of events detected, it was not possible to establish a correlation between AOEs and pro/anti-inflammatory cytokine levels. Although we applied the SCORE chart in all CML patients enrolled to better identify patients with high risk to experience AOEs, this parameter was not predictive for our cohort. This result may be explained by the strategic choice of TKI at diagnosis, as documented also by the lower median age of the nilotinib treated patients compared to the other two TKI subgroups and by the higher number of traditional CV risk factors in imatinib cohort respect to nilotinib and dasatinib. Considering that our results showed a pro-inflammatory status in nilotinib subgroup during the first year of treatment and that AOEs occurred after a median treatment duration of 19,1 months, we believe that a further evaluation of pro/anti-inflammatory cytokines at longer treatment follow-up should be performed to better investigate the correlation between pro-atherothrombotic status and AOEs. In conclusion, we suggest a careful selection of the TKI treatment, according to the presence of baseline CV risk factors and/or previous CV history, in order to offer the best and safe long-term TKI treatment for CML patients, considering as ultimate goal the possibility of reaching a safe TFR and reducing AOEs associated comorbidities.