Molecular and clinical predictors of venous thromboembolism in acute myeloid leukemia and myelodysplastic syndrome.

e19008 Background: Malignancy is a well-described risk factor for venous thromboembolism (VTE). Although widely recognized and studied in solid tumors, VTE is also an appreciable cause of morbidity in patients with hematologic malignancies. Many validated scoring systems exist to predict risk of VTE based on cancer type and other clinical characteristics. However, hematologic malignancies are poorly represented in these models. The purpose of this study was to determine risk factors for the development of VTE in a large cohort of patients with acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS) with the goal of creating a risk stratification model to predict VTE risk in this patient population. Methods: Patients presenting to the University of Wisconsin-Madison with newly-diagnosed AML or MDS from 1/1/2010 through 12/31/2021 were included. Data were collected retrospectively, including patient demographics, clinical characteristics, molecular data, treatment, and complications. Univariate and multivariable logistic and log-binomial regression was used to assess whether variables of interest had any association with VTE. Poisson regression was used to estimate the mortality rate. Results: 547 patients were included in the analysis, including 304 patients with AML and 243 patients with MDS. The overall prevalence of VTE was 14% (95% CI 11.4-17.2%). Those with AML were more likely to experience VTE compared to those with MDS (19.1% vs. 7.8%, RR = 2.4, 95% CI 1.5-4.1, p < 0.001). Higher body mass index (BMI, p = 0.022), induction chemotherapy (p = 0.001), and younger age (p = 0.001) were also associated with development of VTE. Within a subgroup of patients with molecular data available (154 patients), 18.2% experienced VTE. Harboring a mutation in a chromatin modification gene ( ASXL1/2, EED, EZH2, KMT2A) was associated with an 87% reduction in the risk of developing VTE (RR = 0.13, 95% CI 0.01-0.58, p = 0.002). When adjusted for diagnosis and age, the mortality rate for those who experienced VTE was only half the rate of those who never experienced VTE (95% CI 35-69%, p < 0.001). Conclusions: At our institution, the total incidence of VTE was higher than previously described, which underscores the importance of elucidating risk factors of VTE in order to prevent this complication. AML diagnosis, elevated BMI, younger age, and induction chemotherapy were associated with increased risk of VTE, while mutations in chromatin modification genes were associated with a reduction in risk. Surprisingly, the development of VTE was associated with decreased mortality, perhaps reflecting the association of curative-type chemotherapy regimens with VTE in this cohort. In summary, this study demonstrates interesting associations that can ultimately be used to create a risk prediction model of VTE in myeloid malignancies in order to provide optimal supportive care for this patient population.