Introduction A thrombotic event is seen in 10% of newly diagnosed acute myeloid leukemia (AML) patients and has been shown to be associated with disseminated intravascular coagulation (Libourel et al. Blood. 2016). As of yet, the association of developing thrombosis during intensive chemotherapy with AML risk models such as the Medical Research Council (MRC) cytogenetics-based assessment and the European LeukemiaNet (ELN) 2017 molecular based risk model has not been rigorously evaluated. In addition, there is a paucity of data pertaining to the long-term prognostic impact of thrombosis in AML patients. Aiming to investigate these prognostic facets of thrombosis in AML, we performed an analysis of patients with AML diagnosed with thrombosis during treatment with intensive chemotherapy. Methods For this analysis, we collected data on baseline clinical, laboratory, cytogenetic and molecular characteristics of patients with newly diagnosed AML who were treated with intensive chemotherapy and were diagnosed with venous thrombosis. These data were compared with intensively treated patients who did not experience thrombosis to assess for clinical parameters associated with a higher likelihood of developing thrombosis, and to determine whether overall survival was affected by thrombosis. Results The analyzed cohort consisted of 335 newly diagnosed AML patients diagnosed and treated between 2007 through 2021 at the Sheba Medical Center. Median patient age at diagnosis was 55 years. At the time of data analysis 162 patients were alive (48%). Two hundred and eighty-three patients were classified as de novo AML (85%) and the remainder were categorized as secondary disease (15%). Ninety patients harbored the FLT3-ITD mutation (28%) and 117 patients harbored NPM1 mutations (39%). Of evaluable patients, 7 were found to have biallelic CEBPA mutations (8%). In terms of MRC cytogenetic risk, 35 patients (11%) were classified as favorable risk, 219 (66%) patients as intermediate risk, 58 patients (17%) as adverse risk and 21 patients had technically insufficient cytogenetic studies. Using the ELN 2017 risk stratification model, 132 patients (40%) had favorable risk disease, 122 patients (36%) intermediate risk, and 80 patients (24%) adverse risk disease. Venous thrombosis was seen in 33 patients (9.9%) with 28 patients diagnosed with central catheter related thrombosis and 5 patients with lower extremity deep vein thrombosis. In 23 patients (70%) thrombosis occurred during the initial induction chemotherapy phase, in 5 patients (15%) during the first consolidation treatment phase, in 3 patients (9%) during the second consolidation treatment phase, and in 2 patients (6%) during admission for salvage chemotherapy. Following diagnosis of thrombosis, the central catheter was taken out in 9 patients (28%), and 3 patients (8.8%) experienced a recurrence of thrombosis. At the time of thrombosis diagnosis, 14 patients (42%) were found to have a concurrent blood stream bacterial infection. Nineteen patients (58%) were started on anticoagulation following the diagnosis of thrombosis. In univariate analysis, patient age, initial levels of white blood cells, hemoglobin, platelets, creatinine, uric acid, lactate dehydrogenase, INR, albumin, and liver function tests were not significantly different between patients who developed thrombosis and those who did not. Disease type (de novo versus secondary), FLT3-ITD and NPM1 mutational status, achievement of remission, and ELN 2017 risk category also not did not differ to a significant degree between groups. However, MRC intermediate risk group patients were significantly more likely to experience thrombosis compared to favorable risk and adverse risk patients (12.8% versus 5.7% and 1.7%, respectively; p=0.049). Overall survival was not significantly impacted by the diagnosis of thrombosis with an estimated median overall survival of 3.7 years seen in patients without thrombosis compared with 2.2 years in patients with thrombosis (p=0.47). Leukemia-free survival was also comparable between groups (2.4 years versus 1 year; p=0.15). Conclusion Venous thrombosis is seen mostly during the initial induction phase of AML patients receiving intensive therapy with a tight association with the MRC intermediate risk category, however long-term survival and leukemia related outcomes are not significantly impacted by the diagnosis of thrombosis.
Read full abstract