Disseminated Intravascular Coagulation and Early Thrombo-Hemorrhagic Complications in Acute Promyelocytic Leukemia

Introduction: Acute promyelocytic leukemia (APL) patients have the best long term survival outcomes among patients with acute leukemia. However, early thrombo-hemorrhagic complications and mortality within the first 30 days present a challenge in their care. There are no risk models to predict bleeding and venous thromboembolism (VTE) in APL patients, limiting our ability to identify high risk patients. We sought to examine the incidence and factors associated with early bleeding and VTE in APL. Methods: We performed a retrospective analysis of adults (≥18 years) diagnosed with APL between 2010 and 2024 at the University of Alabama at Birmingham (UAB), US. Demographic, clinical and treatment related variables were abstracted from medical records. Outcomes included major and clinically-relevant non-major bleeding (CRNMB) events per International Society of Thrombosis and Haemostasis (ISTH) criteria and VTE (deep and superficial vein thrombosis at any site), occurring ±30 days of the diagnosis of APL. Disseminated intravascular coagulation (DIC) score was calculated based on ISTH criteria; a score ≥5 was used to define overt DIC. Fine Gray sub-distribution hazards model was used to identify VTE and bleeding risk factors, with death as a competing risk. Results: The study included 99 patients (median age: 51 years; 54.5% male and 65.7% non-Hispanic white individuals). All patients were treated with all-trans retinoic acid (ATRA). There were 34 major bleeding episodes (including 16 intracranial hemorrhage events) and 22 CRNMB. The 30-day cumulative incidence of bleeding was 49.5% (standard error [SE] ±5.1%). Univariate analysis showed that elevated white blood cell count ≥20,000 cells/μL (hazard ratio [HR]=2.55, 95% confidence interval [CI]=1.47-4.43; ref:<20,000 cells/μL), lactate dehydrogenase ≥2 times upper limit of normal (ULN) (HR=2.79, 95%CI=1.60-4.89; ref: <2x ULN), creatinine ≥1.0 mg/dL (HR=2.07, 95%CI=1.21-3.55; ref: <1 mg/dL), peripheral blood blasts ≥66% by flow cytometry (HR=1.98, 95%CI=1.07-3.66; ref: <66%), increased body mass index (BMI ≥30 kg/m2) (HR=2.45, 95%CI=1.23-4.85; ref: BMI<30 kg/m2) and overt DIC (HR=2.65, 95%CI=1.21-5.78; ref: non-overt DIC) were associated with bleeding risk. Low platelet count, even <20,000 cells/μL was not associated with bleeding risk. Multivariable analysis showed that obesity (BMI ≥30 kg/m2: HR=2.29, 95%CI=1.16-4.51; ref: BMI<30 kg/m2) and overt DIC (HR=2.52, 95%CI=1.15-5.50; ref: non-overt DIC) were associated with bleeding risk. The cumulative incidence of bleeding at 30 days was 59.4% (SE±6.0%) in patients with overt DIC vs. 28.0% (SE±9.2%) in those without. VTE occurred in 17 patients and included 5 deep vein thrombosis in extremities, 3 pulmonary emboli and 9 superficial venous thrombosis; 12 of these events were catheter related VTEs. Cumulative incidence of VTE was 17.19% (SE±3.81%) at 30 days. Platelet count ≥50,000 cells/μL (HR=2.75, 95%CI=1.07-7.08; ref:<50,000 cells/μL) was associated with increased VTE risk and overt DIC with decreased VTE risk (HR=0.36, 95%CI=0.14-0.93; ref: non-overt DIC) on univariate analysis. Both variables remained significant in independent multivariable models after adjusting for age, sex and race. The cumulative incidence of VTE at 30 days was 13.0% (SE±4.1) among patients with DIC and 32.0% (SE±9.6) among patients without DIC. Of the 11 patients that died within 30 days of diagnosis, seven patients had a major bleeding event (5 intracranial hemorrhage and 2 pulmonary hemorrhage). Conclusions: Early bleeding and VTE incidents remain high in APL patients in the ATRA era. ISTH DIC score is associated with a higher risk of bleeding and lower risk of VTE. We did not see an association with low platelet count and bleeding but platelet count ≥50,000 cells/μL was associated with increased VTE risk. Majority of VTE events were catheter related and future studies examining the role of novel anticoagulants for thromboprophylaxis in patients without bleeding and platelets ≥50,000 cells/μL may help in prevention of VTE.