Flow diagram detailing patient eligibility, recruitment, randomization, and attrition for the Trial of Enoxaparin vs Aspirin in Patients With Cancer and Stroke (TEACH). SI conversion factors: to convert ALT and AST to microkatals per liter, multiply by 0.0167; creatinine to micromoles per liter, multiply by 88.4; hemoglobin to grams per liter, multiply by 10; platelet count to ×109 microliters, multiply by 1. ALT indicates alanine transferase; AST, aspartate transferase; ICA, internal carotid artery; INR, international normalized ratio; MRI, magnetic resonance imaging; PTT, partial thromboplastin time.
aSome patients had multiple exclusion criteria.
bExclusion criteria were removed after study initiation to increase patient recruitment.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Navi BB, Marshall RS, Bobrow D, et al. Enoxaparin vs Aspirin in Patients With Cancer and Ischemic Stroke: The TEACH Pilot Randomized Clinical Trial. JAMA Neurol. 2018;75(3):379–381. doi:10.1001/jamaneurol.2017.4211
Patients with cancer face a heightened risk of acute ischemic stroke (AIS).1,2 Since many of these strokes are attributed to cancer-mediated hypercoagulability, cancer-associated AIS is often managed with anticoagulation, especially low-molecular weight heparins.2,3 However, hypercoagulable stroke mechanisms are rarely definitively diagnosed in patients with cancer antemortem,2 while atherosclerosis, which is generally managed with antiplatelets, is common in patients with cancer.2 Furthermore, patients with cancer are predisposed to bleeding, and many randomized clinical trials of anticoagulant therapy have concluded that increased bleeding offsets reductions in stroke risk.4 Despite these countervailing considerations, to our knowledge, no prospective study has compared antithrombotic strategies in this population.
We conducted a pilot, open-label, randomized clinical trial comparing enoxaparin with aspirin in patients aged 18 to 85 years with active solid or hematological cancer and magnetic resonance imaging–confirmed AIS within 4 weeks (clinicaltrials.gov NCT01763606) (Supplement). Stroke mechanisms were adjudicated after standardized evaluations. Exclusion criteria are listed in the Figure. Patients were recruited from a comprehensive cancer center and 2 comprehensive stroke centers whose institutional review boards approved this study. Enrolled participants/surrogates provided written consent.
Patients were randomized electronically to receive subcutaneous enoxaparin (1 mg/kg twice daily) or oral aspirin (81-325 mg/d) for 6 months. Stratified block randomization was performed to ensure similar numbers of adenocarcinomas in each group. Antithrombotic therapy beyond 6 months was determined by treating physicians. Clinical visits occurred at 1, 3, and 6 months, and electronic health records were reviewed until 12 months.
The primary outcome was feasibility, defined as an enrollment rate among 100 eligible patients for which the lower-bound 95% CI exceeded 30%. This was based on the prespecified determination of the investigators that enrolling 40 of 100 eligible participants (40%; 95% CI, 30.3%-50.3%) would indicate feasibility for larger clinical trials. Because of funding constraints, recruitment was halted after 20 enrollments. Descriptive statistics and Kaplan-Meier survival analyses were used to compare feasibility, safety, and efficacy outcomes between intention-to-treat groups.
From January 2013 to April 2016, 469 patients with cancer with suspected AIS were screened; 49 (10.4%) met eligibility criteria. Leading exclusion criteria were clear indications for anticoagulation (n = 128, 30%) and inactive cancer (n = 83, 20%). Investigators enrolled 20 of 49 eligible patients (41%; 95% CI, 27%-55%), indicating that on termination, the study was on pace to meet the prespecified feasibility end point of 30% enrollment. Enrollment failures occurred because of the aversion of patients to receiving injections (n = 11, 38%), patient/physician preferences for anticoagulation (n = 9, 31%), patients declining clinical trial participation (n = 7, 24%), and patient/physician preferences for antiplatelets (n = 2, 7%).
Ten patients were randomized to each arm, and the median (interquartile range) age was 71 (57-76) years (Table). Six patients (60%) who were randomized to receive enoxaparin crossed over to use aspirin during follow-up because of discomfort with receiving injections (n = 4) or drug costs (n = 2). The median time to crossover was 6 days (interquartile range, 1-22).
At 1 year after enrollment, 3 patients who were randomized to receive aspirin had nonfatal gastrointestinal bleeding, while 1 patient who was randomized to receive enoxaparin had nonfatal pulmonary hemorrhage. One patient who was randomized to receive aspirin had nonfatal myocardial infarction, while 1 patient who was randomized to receive enoxaparin had fatal recurrent AIS. The cumulative rates of major bleeding, thromboembolic events, and survival were not significantly different between the groups.
In what to our knowledge is the first randomized clinical trial comparing anticoagulation vs antiplatelet therapy in patients with cancer and AIS, approximately 10% of participants were eligible and approximately 40% of these were enrolled, in line with prior stroke clinical trials.5 The leading reason for enrollment failure was a patient’s aversion to receiving injections, and 40% of patients who were randomized to receive enoxaparin crossed over to using aspirin because of discomfort with receiving injections. Larger blinded clinical trials to determine the optimal antithrombotic strategy for these high-risk patients appear feasible and safe. Comparing aspirin with direct oral anticoagulants instead of injectable heparins should be considered for future clinical trials, assuming the confirmation of preliminary data that suggest that these medicines may be safe and effective for varying manifestations of cancer-associated thrombosis.6
Corresponding Author: Babak B. Navi, MD, MS, Department of Neurology, Weill Cornell Medicine, 525 E 68th St, F610, New York, NY 10065 (email@example.com).
Accepted for Publication: November 5, 2017.
Published Online: January 8, 2018. doi:10.1001/jamaneurol.2017.4211
Author Contributions: Dr Navi had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Navi, Singer, Stone, DeAngelis.
Acquisition, analysis, or interpretation of data: Navi, Marshall, Bobrow, Singer, Stone, DeSancho.
Drafting of the manuscript: Navi.
Critical revision of the manuscript for important intellectual content: All authors.Statistical analysis: Navi, Bobrow.
Obtained funding: Navi.
Administrative, technical, or material support: Navi, Bobrow, DeAngelis.
Supervision: Navi, Marshall, Singer, DeAngelis.
Conflict of Interest Disclosures: None reported.
Funding/Support: This randomized clinical trial was funded by grants KL2TR000458-06 (Drs Navi and DeAngelis), K23NS091395 (Dr Navi), and P30CA008748 (Dr DeAngelis) from the National Institutes of Health and by the Florence Gould Endowment for Discovery in Stroke.
Role of the Funder/Sponsor: The funding organizations had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Meeting Presentation: This study was presented in part at the 2017 International Stroke Conference; February 21, 2017; Houston, TX.
Additional Contributions: We thank Hooman Kamel, MD, and Costantino Iadecola, MD, from Weill Cornell Medicine; Mitchell S. V. Elkind, MD, MS, Columbia University, and Katherine S. Panageas, DrPH, Memorial Sloan Kettering Cancer Center, for their study supervision and administrative support. We also thank Monica Chen, BA, Weill Cornell Medicine, for copyediting and clerical support. These individuals did not receive compensation for their contributions.
Create a personal account or sign in to: