Thrombosis and outcomes in hematologic neoplasia: Experience from a specialized cancer center in a developing country

Abstract

Abstract Introduction: Cancer is a hypercoagulable state due to host, tumor, or treatment-related factors, with a 9-fold increased risk of venous thromboembolism (VTE) and a 2.5% increased risk of arterial thrombosis (AT) compared to the general population. Patients with hematologic malignancies are at particularly high risk but are underrepresented in thrombotic risk assessments in oncology. This gap is further exacerbated in low-income settings due to a lack of clinical trials, descriptive studies, and limited access to anticoagulation therapy and medical services. To address this, we conducted a study at a reference cancer center in Colombia, aiming to identify associations between thrombotic episodes, hematologic malignancies, and outcomes such as re-thrombosis, bleeding, or death. Methods This retrospective cohort study was conducted in Bogotá, Colombia, between January 2015 and December 2022. Inclusion criteria were adult patients (≥18 years) with a diagnosis of a hematologic malignancy who presented a thrombotic episode after diagnosis and received anticoagulation therapy based on the institutional protocol (Dalteparin 200 IU/kg during the first month, followed by 150 IU/kg thereafter). Electronic medical records were independently reviewed by two physicians to collect clinical information, and data accuracy was confirmed by an additional reviewer. Thrombotic events were classified as typical (deep vein thrombosis [DVT] of the lower extremities or pulmonary embolism) or atypical (other venous or arterial thromboses). Hematologic malignancies were grouped into four categories: lymphoma, myeloma, acute leukemias, and other neoplasms. Primary outcomes included bleeding: any hemorrhagic event after thrombosis; re-thrombosis: recurrent thrombotic event despite anticoagulation; and death associated with thrombotic event: mortality attributed by a physician to the thrombotic episode. Descriptive statistics were used to summarize patient characteristics. Univariate associations with outcomes were assessed using Cramér's V and chi-square tests. To evaluate multivariable associations and avoid overfitting, least absolute shrinkage and selection operator (LASSO) logistic regression was performed. Results A total of 123 patients were included. The median age was 59 years (IQR, 39–79), and 69 of these (56.1%) were female. Sixty-one (49.6%) patients had a BMI over 25 kg/m², with 15 (12.2%) of them being greater than 30 kg/m². Lymphoma was the most common diagnosis (n = 69, 55.3%), which included Hodgkin and non-Hodgkin. Typical thrombosis occurred in 88 (71%) patients, most commonly as DVT (n = 62, 50.4%). Fifteen patients (12.2%) experienced re-thrombosis; LASSO regression showed that female patients had reduced odds of re-thrombosis (OR, 0.15; 95% CI, 0.03–0.54; p = 0.007), while a BMI ≥ 30 kg/m² was associated with increased odds (OR, 12.53; 95% CI, 2.10–89.53; p = 0.007). For death associated with the thrombotic event analysis, 54 cases were excluded due to follow-up loss; among the remaining 69 patients, 9 (13%) died following a thrombotic event. LASSO regression did not show any significant associations between clinical variables and patient mortality. Regarding bleeding outcomes, 8 patients (6.5%) experienced at least one bleeding event. LASSO regression showed that age was significantly associated with bleeding (OR, 0.95; 95% CI, 0.90–0.99; p = 0.028), indicating that younger age was associated with increased bleeding risk. Conclusions In this study, obesity and female gender were identified as independent predictors of recurrent thrombosis despite guideline-based LMWH therapy. Younger age was associated with a slightly higher risk of bleeding events. No clinical characteristic was significantly associated with death following thrombotic episodes, though statistical power may have been limited due to small sample size resulting from loss of follow-up. These findings suggest that additional clinical variables should be considered when assessing thrombotic risk in patients with hematologic malignancies in the setting of low-resource countries. Future studies with larger cohorts are needed to validate these findings, seeking to improve outcomes in this underrepresented population.