Safety and effectiveness of standard versus extended duration of anticoagulation in children with acute lymphoblastic leukemia and venous thrombo-embolism

Abstract

Abstract Background: Venous thrombo-embolism (VTE) occurs in up to 10% of pediatric patients with acute lymphoblastic leukemia (ALL). While anticoagulation is the mainstay of treatment, the optimal duration of therapeutic and/or prophylactic anticoagulation remains unclear. Aims: To compare VTE progression/recurrence and bleeding rates based on duration of anticoagulation among pediatric patients treated for ALL with a history of central venous catheter (CVC)-related VTE, with duration of anticoagulation categorized as standard (i.e. until the end of asparaginase's expected effect or earlier) or extended (later than asparaginase's expected effect). Methods: We performed a retrospective cohort study from three hospitals. Permission from the research ethics board was obtained at each institution. The cohort consisted of pediatric patients aged 1-21 years old with newly diagnosed ALL (2010-2023) receiving asparaginase-containing chemotherapy regimen who experienced a radiologically proven catheter-related VTE requiring medical intervention. Both symptomatic and clinically unsuspected VTEs were considered. The primary effectiveness outcome was time to VTE progression/recurrence. Safety outcomes were time to major bleeding and to clinically relevant non-major (CRNM) bleeding. All outcomes were defined using ISTH criteria (Mitchell, J Thromb Haemost, 2023). Asparaginase's expected effect was based on asparaginase formulation. Patients' characteristics are presented descriptively. We performed Cox proportional hazard models, with death considered a competing event, to compare rates of a) VTE progression/recurrence and b) CRNM bleeding, based on duration of anticoagulation. Results: We included 109 patients (median age at ALL diagnosis: 9 years, interquartile range [IQR]: 4-13, 58% male) with ALL of B-cell (78%), T-cell (20%), or mixed/bilineage phenotype (2%). Most patients (104/109, 95%) received Dana-Farber Cancer Institute-based chemotherapy regimens. Patients had a median of 2 CVCs (IQR: 2-3); peripherally inserted central catheters were typically the first CVC (96/109, 88%) followed by totally implanted catheters as second CVC (57/88, 65%). Pegaspargase was the most common asparaginase used (91%). Index VTE occurred at a median of 75 days (IQR: 25-182) following ALL diagnosis. Anticoagulation was used in 106/109 (97%) of patients. The preferred initial agent was enoxaparin (101/106, 95%) and median duration of anticoagulation was 6.4 months (IQR: 3.2-9.5). Anticoagulation was stopped before and after the end of asparaginase's expected effect in 46 (43%) and 40 (38%) of patients, respectively, while 20 patients (19%) had extended anticoagulation, typically until CVC removal. Overall, 19 patients (17%) sustained a VTE recurrence/progression, at a median of 63 days (IQR: 32-280) following index VTE. Eight events were progression (5/8 symptomatic, all while on anticoagulation), and 11 were recurrences (7/11 symptomatic, 7/11 while on anticoagulation). Of 15 patients on anticoagulation at time of VTE recurrence/progression, 10 were receiving therapeutic doses and 5 were receiving prophylactic doses. Duration of anticoagulation did not predict VTE recurrence/progression (extended vs. standard: HR: 1.76, 95% CI: 0.57-5.39, p=0.323). Major bleed was rare, occurring in 3 (2.8%) patients. Two major bleeds were intracranial hemorrhage occurring within the first week of therapeutic anticoagulation in patients with systemic fungal infections; one was an intracranial bleeding following a stroke. CRNM bleed occurred in 2/20 patients on extended anticoagulation and 3/86 patients on standard anticoagulation (HR: 3.01, 95% CI: 0.55-16.60, p=0.205).Conclusions: VTE progression/recurrence following a first VTE occurred in 17% of children and adolescents with ALL. In this retrospective analysis, extended anticoagulation was not associated with a reduced risk of VTE progression/recurrence. This could suggest that an extended anticoagulation approach is not more effective or could reflect confounding by indication where patients considered at higher risk of progression/recurrence received longer antithrombotic treatment. Overall, in our cohort, the risk of VTE progression/recurrence appeared higher than the risk of major and/or CRNM bleeding. Further investigation into alternative approaches to reduce VTE progression/recurrence is warranted in children and adolescents with ALL receiving asparaginase-containing regimens.