primary thromboprophylaxis using direct oral anti-coagulants (DOAC) in malignant pediatric tumors with vascular compression

Abstract

Abstract INTRODUCTION Cancer-associated thrombosis (CAT) is a recognized complication of pediatric malignancies. Vascular compression increases the risk of thrombotic events due to impaired blood flow. Prophylactic anticoagulation has been suggested across various oncologic malignancies, though the studies are at best small, contradictory, and not widely applicable. Sarangi previously demonstrated a reduction in thrombotic events for patients with mediastinal masses and tumor-related vascular compression (TRVC) who received prophylactic enoxaparin (Sarangi et. al. Pediatric Blood Cancer 2021). Although not statistically significant, it highlighted an important target group for prophylaxis. Other studies also suggest prophylaxis benefit for >25% vascular compression (Gartrell et al. Pediatric Blood & Cancer 2019). Data on the use of direct oral anticoagulants (DOACs) as prophylaxis in pediatric malignancy is lacking. PREVAPIX-ALL, the only pediatric study on DOAC prophylaxis in cancer, detected a subgroup where apixaban led to a statistically significant reduction in thrombotic events (O'Brien et. al. Lancet Hematology 2024). Our study aims to characterize the safety of DOACs as primary thromboprophylaxis in another proposed high-risk subgroup of pediatric patients with TRVC. METHODS A combined retrospective chart review and prospective cohort study was conducted from January 2023 to July 2025 of patients (0-21 years) with a malignant mass and radiologically confirmed vascular compression. Patients who received a DOAC as primary thromboprophylaxis were included. Demographic data including age, sex, type of cancer, central venous line (CVL), DOAC type, bleeding, and thrombotic events were recorded. Descriptive statistics (median and interquartile range (Q1-Q3) for continuous variables; frequencies and proportions for categorical variables) were calculated. RESULTS A total of 19 patients with TRVC and DOAC prophylaxis are included (17 retrospective, 2 prospective); 57.9% (n=11) were diagnosed with lymphoma, 26.3% (n=5) neuroblastoma, and 15.8% (n=3) were mixed (sarcoma, desmoid fibromatosis, or paraganglioma). Fourteen patients (≥12 years and >40 kg) received apixaban (2.5 mg twice a day) and 5 patients received rivaroxaban (weight-based dosing). Four patients initially on enoxaparin were switched to a DOAC (after 9, 17, 35, and 80 days). The median (Q1-Q3) duration of DOAC prophylaxis (n=16) was 55 (32-82) days. There were 3 bleeding (15.8%) and 2 thrombotic events (10.5%). One patient died due to a procedural complication (day 9 of prophylaxis), 1 was lost to follow-up (day 98), and 3 remain on prophylaxis for continued compression (>100 days). DISCUSSION Children with cancer exhibit a 600-fold increased risk of developing a venous thromboembolism (VTE). Studies on the use of heparin-based products for thromboprophylaxis are conflicting. While PREVAPIX-ALL did not find a statistically significant reduction in VTE among those receiving apixaban, a subgroup of obese patients did have significant reduction. This highlights the importance of identifying higher-risk groups, such as ours, that may benefit from prophylaxis. In our cohort, primary thromboprophylaxis was tolerated with minimal adverse effects. Although three bleeding events occurred (2 minor and 1 major), the direct impact of DOACs on bleeding cannot be concluded. On apixaban, 1 patient had rapid self-resolution of epistaxis, and another had hematuria with concurrent ifosfamide administration. A major intra-tumoral bleeding event occurred on day 6 of prophylaxis (rivaroxaban) in a newly diagnosed neuroblastoma, known to have a risk for this complication. Thrombotic events included a superficial PICC-associated thrombus and a questionable intracardiac thrombus incidentally found only after TRVC resolution. This highlights the notion that DOACs may be effective in preventing clinically relevant thrombotic events with minimal bleeding risk for this subgroup during peak risk. CONCLUSION Though limited by its mostly retrospective design, small sample size, and single-institution population, our study highlights an at-risk subgroup of pediatric oncology patients with TRVC that may benefit from universal primary thromboprophylaxis with a DOAC. Larger studies are needed to evaluate this further.