Biochemical risk factors and outcomes of acute promyelocytic leukemia patients with thrombotic events: a matched pair analysis

J Thromb Thrombolysis. 2024 May 3. doi: 10.1007/s11239-024-02988-x. Online ahead of print.

Abstract

Acute promyelocytic leukemia (APL) stands out as a distinctive form of acute leukemia, exhibiting a higher occurrence of thrombotic events when contrasted with other leukemia subtypes. Since thrombosis is a relatively rare but unfavorable condition with poor prognostic implications, it is crucial to determine the risk factors for thrombotic events in APL(thrombosis in large venous or arterial from onset to differentiation therapy in 30d). We performed a retrospective study involving 950 APL patients between January 2000 and October 2022, from which 123 were excluded by younger than 16 years of age, 95 were excluded by incomplete data, and 6 were excluded by thrombosis related to CVC or PICC. A total of 23 APL patients with thrombosis for inclusion in our analysis were performed a 1:5 ratio matching based on sex (perfect match) and age (within 5 years) to patients without thrombosis. These patients were continuously monitored in the outpatient department over a period of 5 years. We meticulously examined clinical and laboratory data to pinpoint the risk factors related to thrombotic events in APL. Our primary clinical endpoints were all-cause mortality and achieving complete remission, while secondary clinical outcomes included APL relapse. Thrombotic events were observed in 2.4% (23/950) of APL patients. Compared to patients without thrombosis, patients with thrombosis had higher lactate dehydrogenase (LDH) [313 (223, 486) vs. 233 (188, 367) U/L, p = 0.020], higher indirect bilirubin [11.2 (7.4, 18.6) vs.8.3 (6.0, 10.7) umol/L, p = 0.004], higher creatinine [72 (62, 85) vs. 63 (54, 74) umol/L, p = 0.026], higher CD2 expression (65.2 vs. 15.2%, p < 0.001), higher CD15 expression (60.9 vs. 24.3%, p = 0.001), and PML/RARαisoforms (p < 0.001). Multivariate-logistic-regression analysis revealed several factors that were markedly related to thrombosis, including LDH (OR≈1.003, CIs≈1.000-1.006, p = 0.021), indirect bilirubin (OR≈1.084, CIs≈1.000-1.188, p = 0.043), CD2 expression positive (OR≈16.629, CIs≈4.001-62.832, p < 0.001), and CD15 expression positive (OR≈7.747, CIs≈2.005-29.941, p = 0.003). The S-type (OR≈0.012, CIs≈0.000-0.310, p = 0.008) and L-type (OR≈0.033, CIs≈0.002-0.609, p = 0.022) PML/RARα isoforms were negatively associated with thrombosis. Kaplan-Meier curves indicated that the survival rates were remarkably varied between APL patients with and without thrombosis (HR:21.34, p < 0.001). LDH and indirect bilirubin are variables significantly associated with thrombosis in APL, S-type and L-type PML/RARαisoforms exhibit a negative association with thrombotic events. The thrombotic events of APL can predict the subsequent survival of thrombosis. The findings of our study have the potential to facilitate early detection of thrombosis and enhance the prognosis for individuals with APL who develop thrombosis. Further validation of our findings will be essential through future prospective or multicenter studies.

Keywords: Acute promyelocytic leukemia; Biochemical risk factor; Prognosis; Thrombotic events.