Contributions
Abstract: S308
Type: Oral Presentation
Session title: Malignancies and thrombosis
Background
Thrombosis is one of the most frequent complication in cancer. Although recent studies suggest that patients with blood cancers may be at a similar or even higher thrombotic risk than those with solid tumors, the frequency of thrombosis in acute myeloid leukemia (AML) has been evaluated in a few reports only and no validated predictive model is currently available.
Aims
In this study we aimed to evaluate the frequency of thrombosis in AML patients treated with intensive chemotherapy and to assess the ability of genetic and clinical factors to predict the risk of thrombosis, including the Khorana risk score (KRS) or disseminated intravascular coagulation (DIC) score.
Methods
We performed a retrospective observational study including 222 newly diagnosed adult AML patients treated with intensive chemotherapy between January 2013 and February 2020, stratified by European Leukemia Net (ELN) 2010 genetic/cytogenetic risk assessment. The diagnosis of thrombosis was confirmed by imaging modalities including venous ultrasonography, computer tomography or magnetic resonance imaging and clinical, enzymatic and electrocardiographic criteria for myocardial infarction.
Results
222 patients (52% males) were included, with a median age of 59.5 years (range 20-78), 20.5% in the ELN2010 adverse risk category. With a median follow-up of 43.5 months, we observed 50 thrombotic events and 90% were venous (VTE): twenty-eight patients (62.2% of VTE) had catheter-related thrombosis (CRT), 93% of which were peripherally inserted central catheter (PICC)-related; in 6 and 5 case VTE occurred in lower and upper extremity, respectively, while 3 patients experienced pulmonary embolism. Among arterial thrombosis, 2 myocardial infarctions and 3 cerebral vascular accidents occurred. The prevalence of thrombosis was 22.5% and 6-months, 1-year and 2-year cumulative incidence was 10%, 22% and 25%, respectively. The median time to thrombosis was 84 days and 52% of the events occurred within 100 days from AML diagnosis, during induction/consolidation chemotherapy or just before starting treatment. History of a thrombotic event (p=0.043), especially VTE (p=0.0053), and baseline platelet count higher than 100x10^9/L (p=0.036) significantly increased the risk of thrombosis while AML genetic profile did not affect its occurrence with statistical significance. A sub-analysis considering only early thromboses (i.e. occurring within day 100) confirmed these results. Conversely, if only no-CRT VTE were considered, ELN Intermediate-1 risk group was significantly associated with thromboses (p=0.039), especially FLT3-ITD/NPM1 mutated patients. KRS and DIC score failed to stratify patients according to their thrombotic risk, but DIC score at diagnosis was independently associated with reduced survival (p=0.004) by multivariate analysis. No impact of thrombosis on survival was observed.
Conclusion
Thromboses are a frequent complication in AML, especially during chemotherapy, and most of them are VTE. However, no impact on survival was observed. Previous VTE and baseline platelet count could predict thrombotic risk, while AML genetic profile did not significantly affect thrombosis occurrence. We confirmed that KRS is not a robust tool in this setting, and we could not validate the association of DIC score with thromboses, warranting further studies on the subject to better predict thrombosis occurrence, hopefully allowing to design therapeutic measures to prevent it.
Keyword(s): Acute myeloid leukemia, Disseminated intravascular coagulation, Thrombosis
Abstract: S308
Type: Oral Presentation
Session title: Malignancies and thrombosis
Background
Thrombosis is one of the most frequent complication in cancer. Although recent studies suggest that patients with blood cancers may be at a similar or even higher thrombotic risk than those with solid tumors, the frequency of thrombosis in acute myeloid leukemia (AML) has been evaluated in a few reports only and no validated predictive model is currently available.
Aims
In this study we aimed to evaluate the frequency of thrombosis in AML patients treated with intensive chemotherapy and to assess the ability of genetic and clinical factors to predict the risk of thrombosis, including the Khorana risk score (KRS) or disseminated intravascular coagulation (DIC) score.
Methods
We performed a retrospective observational study including 222 newly diagnosed adult AML patients treated with intensive chemotherapy between January 2013 and February 2020, stratified by European Leukemia Net (ELN) 2010 genetic/cytogenetic risk assessment. The diagnosis of thrombosis was confirmed by imaging modalities including venous ultrasonography, computer tomography or magnetic resonance imaging and clinical, enzymatic and electrocardiographic criteria for myocardial infarction.
Results
222 patients (52% males) were included, with a median age of 59.5 years (range 20-78), 20.5% in the ELN2010 adverse risk category. With a median follow-up of 43.5 months, we observed 50 thrombotic events and 90% were venous (VTE): twenty-eight patients (62.2% of VTE) had catheter-related thrombosis (CRT), 93% of which were peripherally inserted central catheter (PICC)-related; in 6 and 5 case VTE occurred in lower and upper extremity, respectively, while 3 patients experienced pulmonary embolism. Among arterial thrombosis, 2 myocardial infarctions and 3 cerebral vascular accidents occurred. The prevalence of thrombosis was 22.5% and 6-months, 1-year and 2-year cumulative incidence was 10%, 22% and 25%, respectively. The median time to thrombosis was 84 days and 52% of the events occurred within 100 days from AML diagnosis, during induction/consolidation chemotherapy or just before starting treatment. History of a thrombotic event (p=0.043), especially VTE (p=0.0053), and baseline platelet count higher than 100x10^9/L (p=0.036) significantly increased the risk of thrombosis while AML genetic profile did not affect its occurrence with statistical significance. A sub-analysis considering only early thromboses (i.e. occurring within day 100) confirmed these results. Conversely, if only no-CRT VTE were considered, ELN Intermediate-1 risk group was significantly associated with thromboses (p=0.039), especially FLT3-ITD/NPM1 mutated patients. KRS and DIC score failed to stratify patients according to their thrombotic risk, but DIC score at diagnosis was independently associated with reduced survival (p=0.004) by multivariate analysis. No impact of thrombosis on survival was observed.
Conclusion
Thromboses are a frequent complication in AML, especially during chemotherapy, and most of them are VTE. However, no impact on survival was observed. Previous VTE and baseline platelet count could predict thrombotic risk, while AML genetic profile did not significantly affect thrombosis occurrence. We confirmed that KRS is not a robust tool in this setting, and we could not validate the association of DIC score with thromboses, warranting further studies on the subject to better predict thrombosis occurrence, hopefully allowing to design therapeutic measures to prevent it.
Keyword(s): Acute myeloid leukemia, Disseminated intravascular coagulation, Thrombosis