D-Dimer Kinetics, Chemotherapy and Risk of Bleeding in Acute Promyelocytic Leukemia

Abstract Introduction: Acute promyelocytic leukemia (APL) is a medical emergency with a high incidence of early mortality, mainly due to hemorrhage induced by coagulopathy. Despite All-Trans Retinoic Acid (ATRA) and cytotoxic chemotherapy (CC) therapy that rapidly normalizes coagulopathyin APL, there is concern for worsening prior to this anticipated improvement. At our center, we have observed that patients who receive CC as part of their induction regimen often develop worsening coagulopathyresulting in clinically significant bleeding. Our impression was that this exceeds that seen in patients who receive ATRA without CC. We retrospectively reviewed patients with APL to identify a differential effect of CC on the coagulation system that could explain this excessive bleeding. Methods: Following IRB approval, the medical records of 27 consecutive newly diagnosed APL patients presenting between July 2007 and 2014 to Froedtert Hospital /Medical College of Wisconsin were reviewed. D-dimerlevels, fibrinogen and platelet counts before, during and after administration of systemic therapy were analyzed. Hemorrhage during induction was analyzed as occurring post-CC vs. ATRA with or without Arsenic Trioxide (ATO). This was correlated with any clinically significant bleeding and thrombosis. Results: Of the 27 patients identified, the median age was 50, and 13 were males. Ten patients(37%) were induced without CC. APL risk category did not differ between those induced with and without CC. Overall, nine (30%) patients had a bleeding complication and of those, seven (78%) received CC (Table 1). The two patients that did not receive CC had bleeding as their initial presentation. Of the seven that received CC and bled, only one had bleeding as the initial presentation and this patient did not bleed again after induction. Use of CC therapy did not statistically increase bleeding rates, but this was likely influenced by sample size (p=0.24, Fisher's exact). The D-dimer kinetics also differed between patients treated with and without CC (fig1&2). 26 of 27(96%) patients presented with high D-dimer levels initially. These levels improved over the next 72 hours after ATRA administration in patients without CC. A second rise in D-Dimer levels after CC was seen in 16 out of 17(94%) patients. This second spike correlated with clinically significant bleeding (per International Society of Thrombosis and Hemostasisdefinition) in 6 out of the 7 patients (85%). The platelet count, risk category or presence of clot did not seem to have a confounding effect on the risk of bleeding in these patients. Discussion: This is the first report of bimodal D-Dimer kinetics and associated bleeding diathesis in patients with APL whose induction therapy includes CC. Although the sample size is small, this increase in clinically significant bleeding risk among patients treated with ATRA + CC highlights the importance of close hemostatic monitoring after starting CC. If confirmed in larger studies, prophylactic hemostatic treatments like anti-fibrinolytic therapy could be considered in patients with second elevation in D-Dimer. 100% of the patients in the low risk group that received CC experienced hemorrhage. Hence, physicians should consider induction regimens for low risk APL that donot contain CC to limit the risk of hemorrhage given published trials that CC does not improve outcomes in low-risk disease over ATRA plus ATO. Table1: Difference in incidence of bleeding based on treatment type Bleeding (%) No bleeding (%) Total (%) Chemotherapy 7(41) 10(59) 17(100) No chemotherapy 2(20) 8(80) 10(100) Figure 1: D-dimer kinetics in a patient who received chemotherapy Figure 1:. D-dimer kinetics in a patient who received chemotherapy Figure 2: D-dimer kinetics in a patient who did not receive chemotherapy Figure 2:. D-dimer kinetics in a patient who did not receive chemotherapy Disclosures No relevant conflicts of interest to declare.