Background: Patients with acute myeloid leukemia (AML) are at increased risk of vein thromboembolic events (VTE). However, thromboprophylaxis is largely underused, mainly due to increased risk of bleeding and limited number of studies analizing risk factors for VTE development. Aims: To assess potential risk factors for the VTE development and develop a simple predictive model, based on patients and disease-related parameters. Methods: We conducted a retrospective cohort study of adult patients with newly diagnosed AML who were treated in our institution between January 2009 and December 2021. Patients with AML-M3 were excluded. Primary end point of our study was imaging-confirmed VTE including simptomatic central venous catheter -related thrombosis (CVC). Participants were followed from the AML diagnosis to death or 6 months after the last cycle of therapy. We analyzed demographic factors (age, gender), Eastern Cooperative Oncology Group performance status (ECOG PS), body mass index (BMI), smoking status, comorbidities, Hematopoietic Cell Transplantation-specific Comorbidity Index (HCT-CI), baseline laboratory findings (complete bood counts, fibrinogen, PT, PTT, D dimer, LDH) leukemia related paremeters (cytogenetics, molecular genetics (FLT3, NPM), flow cytometry), Khorana score (KRS) as potential risk factors for the VTE development. Univariate and multivariate logistic regression model was used as the method for analyzing binary outcome (VTE development) and potential predictors. Based on our final model we cretaed a nomogram to enable calculalation of probability of VTE development. Results: The study population included 608 eligible AML patients, mean age 55 (range: 18-81) years, male/female 341/267 (56.1%/43.9%). During the follow-up period 81/608 (13.3%) patients developed VTE: CVL 61/608 (10.0%), deep vein thrombosis 16/608 (2.63%), pulmonary embolism 4/608 (0.01%). Before therapy, during induction, consolidation, transplantation and relapse VTE were developed in 5/81 (6.2%), 36/81 (44.4%), 30/81 (37%), 4/81 (4.9%), 6/81 (7.4%) patients respectively. Among all tested parameters five which are independent predictictors were included in predictive model: male sex (p=0.020, OR 1,92 95% CI: 1.11-3.31), prior history of TE (p=0.001, OR 3.83, 95% CI: 1.72-8.55), PT (p=0.107, OR 1.01, 95% CI: 0.99-1.03), ECOG PS (p=0.038, OR 0.72, 95% CI: 0.53-0.98), HCT-CI (p=0.032, OR 0.79, 95% CI: 0.64-0.98). All other patient and disease related paremeters, including Khorana score were not predictive. Our nomogram provides a visual depiction of the relative contribution of each prognostic factor to the total point score and, thus, the weight of factors regarding probability for VTE development (Picture 1). Image:Summary/Conclusion: Our study point out that: 1. A KRS score has no predictive value in AML; 2. Male patients and patients with history of prior thrombosis have 1.92 and 3.83 higher odds of developed VTE; 3. Increasing of ECOG and HCT-CI for one point decrease probability of develope thrombosis for 28% and 21%; 4. Thromboprophylaxis should be considered in patients with prior history of TE, especially with male sex, good PS and low HCT-CI. Our nomogram may help clinicians in clinical decision making. However, validation of our predictive model through prospective study with a larger number of patients are required.
Read full abstract