Thrombosis in ALL: a risk without clear mitigation

In this issue of Blood, Orvain and colleagues, on behalf of the GRAALL, the French cooperative group leukemia study group, present data that help to define more clearly the incidence and risk factors for thromboembolism in adults receiving treatment on a pediatric-inspired regimen, GRAALL-2005.^1,2  In recent years, a number of prospective trials have demonstrated improved survival rates for young adults treated with pediatric-inspired regimens that contain frequent dosing of asparaginase and glucocorticoids, both known to be associated with an increased risk of thromboembolic events.^3-6  One of the challenges to further improving treatment outcomes in adult patients is to understand (and, one hopes, to modify) the toxicities that occur more frequently in the adult population treated with these pediatric-inspired regimens.

The GRAALL-2005 enrolled 813 adults with untreated acute lymphoblastic leukemia (ALL) from ages 18 to 60 years old between 2006 and 2014, and 784 of these patients were included in this analysis. Overall, the incidence rate of venous thromboembolism was 16% during the intensive part of chemotherapy, which is considerably higher than the cumulative rate reported recently of 7.9% by Rank et al in the NOPHO ALL2008 study that also employed an intensive pediatric regimen and included patients 1 to 45 years old.⁷  Orvain et al identified several risk factors for thrombosis, including older age, female sex, obesity, and a higher platelet count at diagnosis as associated with increased thrombotic risk. The majority of these thrombotic events (84; 69%) occurred during induction therapy; importantly, the most serious of these events, cerebral venous thrombosis, occurred almost exclusively during induction therapy. Of note, neither oral contraceptive use prior to diagnosis nor history of smoking was associated with thrombotic rate.

These are interesting, valuable, and provocative data that identify for whom and when it may be most important to intervene with some kind of prophylaxis. The investigators in GRAALL-2005 provided guidelines (without prospective randomization) for treatment of coagulopathy and prevention of thromboembolism that included heparin (recommended unfractionated heparin), antithrombin supplementation to prevent thrombosis, and fibrinogen concentrates or plasma for patients at risk of bleeding. Although the investigators describe the results of these suggested interventions, this part of the study has some major limitations because these interventions were not prospectively randomized, not all patients received the recommended treatments, and many patients received a combination of therapies. The authors found some mitigation of thrombosis when a combination of antithrombin and heparin prophylaxis was used without risk of significant bleeding but found that fibrinogen concentrates increased the risk of thrombosis. Although these findings provide important hypothesis-generating data for a prospective study, they are clearly not definitive and are difficult to interpret, a fact acknowledged by the authors. An important prospective trial, THROMBOTECT, randomized pediatric patients ages 1 to 18 with ALL to low-molecular-weight heparin, antithrombin concentrates, or unfractionated heparin during induction therapy.⁸  Overall, the thrombotic rate was 4.4% in this pediatric population, and patients randomized to antithrombin and low-molecular-weight heparin had significantly lower thrombotic risk than those randomized to unfractionated heparin (which was used primarily by patients enrolled in GRAALL-2005). The study demonstrated safety of the thromboprophylactic regimens (8/949 patients with major hemorrhage).

In conclusion, the study by Orvain and colleagues is a welcome step in defining and highlighting the problem of thrombosis in adults with ALL, identifying patients who might be at greatest risk, and alerting us to the highest risk period for thrombotic complications. The authors also provide hypothesis-generating information on potentially useful and not useful interventions for the coagulopathy associated with ALL and the asparaginase-intensive therapies that we are giving to improve survival rates that can inform the design of future studies. As we move forward in the brave new world of intensified pediatric regimens for adults with ALL, this study highlights the need for a carefully designed prospective study of thromboprophylaxis that might be focused on those early weeks of treatment when the thrombotic risks are highest. Currently, the Children’s Oncology Group is taking the next step by performing a randomized study (#NCTO2369653) to examine the potential benefit of thromboprophylaxis with an oral anticoagulant, Apixaban, during induction therapy for children with ALL. The adult ALL investigators should follow suit!