Prospective Evaluation of Thromboembolic Events During Chemotherapy of Childhood Acute Lymphoblastic Leukemia: Results of Randomized Trial ALL-BFM 2000.

Abstract 990

Poster Board I-12

Thromboembolic events (TE) are common complications in association with acute lymphoblastic leukemia (ALL). The incidence, as reported in the literature, is 3.2% on average, but varies from 1.1% to over 30%, which might be explained by different definitions of the events (symptomatic/asymptomatic), diagnostic methods, study designs (prospective/retrospective) and treatment protocols. Pathogenesis and etiology of ALL-related TE have not yet been fully clarified so far and the existing data on the relevance of potentially influencing risk factors are quite inconclusive. The higher risk of TE in patients with ALL and under ALL treatment may be caused by the ALL itself and the applicated chemotherapy, first of all asparaginase but potentially in combination with other drugs, e.g. steroids. The chemotherapy leads to an imbalance of coagulation factors inducing a hypercoagulable state in the patient. Other modulating factors might be the preexistence of hereditary thrombophilia, the existence of an implanted central venous line (CVL) and the potential treatment with antithrombotic prophylaxis.

Data on TE were prospectively collected from 2419 patients with ALL (age 1-18 years) enrolled from August 1999 to September 2005 into study ALL-BFM 2000 of the Berlin-Frankfurt-Münster group (Germany, Austria, Switzerland). Patients participating in the “Thrombotect” study (randomized study for evaluation of antithrombotic therapies in ALL) were excluded from the analysis. Among the analyzed patients, 104 TE were documented (4.3%); 41 TE were localized in the CNS (mainly sinus venous thromboses), 63 TE outside of the CNS (lower deep venous system, n=14; upper deep venous system n=40; cardiac, n=6; pulmonary embolisms, n=3). Nine TE were classified as life-threatening (7 cerebral, 2 non-cerebral), 1 patient died from a cerebral infarct with concomitant serious bleeding. About 62% of the TE (n=64) were diagnosed during the 5-week induction treatment with prednisone or dexamethasone, vincristine, daunorubicin, asparaginase, and intrathecal methotrexate, 14% (n=15) during the subsequent consolidation phase, and 24% during later treatment phases. Adolescent patients had a significantly higher risk to develop TE with a rate of 3.4% in patients <10 years and 6.6% in patients ≥10 years of age (p=0.001). There was a trend to a higher TE rate in male patients (4.8% vs. 3.4% in females, p=0.077), which was evident in the younger age group only. TE rate in T-ALL patients was significantly higher than in non-T-ALL (7.0% vs. 3.8%, p=0.006), which was independent of the age group. After a 7-day prednisone prephase, patients were randomized to receive dexamethasone or prednisone for three weeks plus tapering during induction treatment. Incidence rate of TE was 5.2% in randomized patients treated in the prednisone arm (n=968) and 3.7% in the dexamethasone arm (n=857) (p=0.14). The impact of the time point of CVL implantation was analyzed in 1439 patients with available data on CVL implantation. Of these patients, 43% received their CVL during the first week of induction treatment, 19% during the second week, and 38% after the second week. Time point of CVL implantation did not influence the TE incidence with a rate of 4.0%, 5.2%, and 5.2% in the above mentioned three groups, respectively (p=0.59). However, nearly all patients (97%) had received their CVL by start of consolidation phase (protocol day 36). Therefore, no statement could be made about the TE incidence in patients without CVL throughout therapy.

Several parameters which have potential impact on TE incidence are unknown in this study. These include anticoagulatory prophylaxis by institutional practice, preexisting hereditary thrombophilia as well as the coagulation status at time of diagnosis of ALL, under chemotherapy, and at diagnosis of TE.

TE are a rather rare but potentially serious complication of contemporary ALL treatment. Risk factors are numerous. So far, besides age, immunophenotype, and treatment phases robust predictive factors are still missing. Prospective randomized studies are underway to identify risk factors and to develop preventive strategies.