Evaluation of ISTH DIC Score to Predict Significant Bleeding and Thrombosis Events in Pediatric Acute Promyelocytic Leukemia; A Report from the Children’s Oncology Group AAML0631 Trial

Early deaths and morbidity due to coagulopathy (both hemorrhage and thrombosis) remain a significant problem in APL. Some studies have shown that various clinical and laboratory parameters (including elevated WBC count, low platelet count, decreased fibrinogen, and elevated PT) are associated with risk for hemorrhage and thrombosis, but individually these features are neither sensitive nor specific predictive tools. The ISTH DIC score was developed as a set of objective criteria for the diagnosis and severity of DIC with a score of ≥5 correlating with overt DIC (Taylor, et al, Thromb Haemost, 2001). When applied to coagulopathy in APL, one retrospective study showed that an ISTH DIC score of ≥6 independently correlated with hemorrhagic early death (Mitrovic, Med Onc, 2013). Here we evaluate the predictive value of the ISTH DIC score for pediatric patients with newly diagnosed APL treated on Children’s Oncology Group study AAML0631.

The ISTH DIC score is calculated using platelet count (≥100,000 = 0; 50,000-99,999 = 1; <50,000 = 2), fibrinogen level (≥100 mg/dL = 0; <100 mg/dL = 1), prothrombin time prolongation above upper limit of normal (ULN) (<3 seconds = 0, 3-6 seconds = 1, >6 seconds = 2), and D-dimer (<2 times ULN = 0, 2-4 times ULN = 2, >4 times ULN = 3). Out of the 102 eligible patients on AAML0631, complete data on coagulation parameters and clinically significant adverse events (AEs) related to coagulopathy (grade III-V bleeding or thrombosis) during induction were available on 76 patients. Among 31 pts with ISTH DIC score ≥6, 10 patients had at least 1 AE related to coagulopathy in induction. Four of these were high risk patients (WBC >10,000 at diagnosis) who experienced grade V early deaths. Among 45 pts with ISTH DIC score <6, there were only 4 patients with at least 1 coagulopathy AE and no grade V events. The ISTH DIC score ≥6 was significantly associated with a higher rate of grade V events compared to score <6 (13% vs. 0%, P=0.025). When considering all AEs (grade III-V) due to coagulopathy, patients with ISTH DIC score ≥6 compared to those with score <6 also had a significantly increased rate of at least 1 AE (grade III-V) related to coagulopathy (32% vs. 9%, P=0.015). However, the sensitivity of the ISTH DIC score in predicting patients experiencing at least 1 coagulopathy event during induction was only 66.7% and the specificity was 65.6%.

The coagulopathy of APL is unique as there is activation of the coagulation cascade due to expression of tissue factor and other procoagulants with concomitant increase in primary and secondary fibrinolysis due to expression of Annexin II on the APL blasts. Traditional coagulation tests such as those used to calculate the ISTH DIC score cannot measure the unique contribution of fibrinolysis or the balance between coagulation and fibrinolysis. The current study demonstrates that ISTH DIC score is significantly correlated with AEs related to coagulopathy, and yet it has low sensitivity and specificity as a predictive model in pediatric APL patients. Other biomarkers and coagulation assays such as thrombomodulin levels and thrombin generation ought to be studied for their predictive value and to inform potential intervention strategies. Such prospective studies are planned for the next APL clinical trial in the Children’s Oncology Group.