Prehospital Tranexamic Acid Use for Traumatic Brain Injury

1. Overview This multi-center, Phase II trial is designed to determine if Tranexamic Acid
(TXA) initiated in the prehospital setting improves long-term neurologic outcome
compared to placebo in patients with moderate to severe TBI who are not in shock. This
study protocol will be conducted as part of the Resuscitation Outcomes Consortium (ROC)
at trauma centers in the United States and Canada. ROC is funded by the National Heart
Lung and Blood Institute (NHLBI) in partnership with the US Army Medical Research and
Materiel Command (USAMRMC), Canadian Institutes of Health Research, the Heart & Stroke
Foundation of Canada, the American Heart Association (AHA), and the Defense Research and
Development Canada. ROC is a clinical trials network focusing on research primarily in
the area of prehospital cardiopulmonary arrest and severe traumatic injury. The mission
of ROC is to provide infrastructure and project support for clinical trials and other
outcome-oriented research in the areas of cardiopulmonary arrest and severe traumatic
injury that lead to evidence-based change in clinical practice.

2. Specific Aims/Hypothesis Statement

2.1 Clinical Hypotheses and Aims

Specific aim 1: To compare 6-month neurologic outcome between subjects who are randomly
assigned to TXA to subjects who are randomly assigned to placebo by evaluating the
Glasgow Outcome Scale Extended score (GOS-E) at 6 months post-injury.

Primary Hypotheses: We will perform a one-sided test of the following null hypothesis:
The proportion of subjects who have a favorable neurologic outcome (GOS-E > 4) at six
months post injury who are randomly assigned to TXA is not different from the proportion
of subjects who have a favorable neurologic outcome (GOS-E > 4) who are randomly
assigned to placebo. This hypothesis will be tested versus the alternative that the
proportion of subjects with a favorable neurologic outcome who are randomly assigned to
TXA is higher than in subjects who are randomly assigned to placebo at the .1 level and
versus the alternative that the proportion of subjects with a favorable neurologic
outcome who are randomly assigned to TXA is lower than it is in the placebo group at the
.025 level

Specific aim 2: To assess differences in morbidity and mortality measured from
randomization through 28 days or initial hospital discharge and differences in
neurologic outcomes at 6 months between subjects in the bolus/maintenance arm, bolus
only arm, and placebo arm.

Secondary Hypotheses: The null hypotheses are that there will be no difference between
subjects who are randomly assigned to TXA and subjects who are randomly assigned to
placebo in the following: both absolute and relative volume of intracranial hemorrhage
(ICH) progression, proportion of subjects with ICH progression, frequency of
neurosurgical interventions, GOS-E measured at discharge and 6 months, Disability Rating
Scale score (DRS) measured at discharge and 6 months, 28-day survival, and
ventilator-free, intensive care unit (ICU)-free, and hospital-free days.

Specific aim 3: To assess differences in adverse events measured from randomization to
initial hospital discharge between subjects in the bolus/maintenance arm, bolus only
arm, and placebo arm.

Tertiary Hypotheses: The null hypotheses are that there will be no difference between
subjects who are randomly assigned to TXA and subjects who are randomly assigned to
placebo in the following: proportion of subjects experiencing seizures, cerebral
ischemic events, myocardial infarction (MI), deep venous thrombosis (DVT), or pulmonary
thromboembolism (PE) post randomization through 28 days or discharge, whichever occurs
first.

2.2 Laboratory Hypotheses and Aims

Specific aim 1: To compare coagulation profiles over time using kaolin activated
thrombelastography (TEG) results between subjects who are randomly assigned to TXA and
subjects who are randomly assigned to placebo.

Primary hypothesis: The null hypothesis is that there will be no difference in the
degree of fibrinolysis as assessed by percentage of clot lysis determined 30 minutes
after the maximum amplitude is reached (LY30) between subjects who are randomly assigned
to TXA and subjects who are randomly assigned to placebo.

Specific aim 2: To explore the underlying mechanism of TXA by comparing fibrinolytic
pathway mediator activity between subjects who are randomly assigned to TXA and subjects
who are randomly assigned to placebo.

Secondary hypothesis: The null hypothesis is that there will be no change in
fibrinolytic pathway mediators between subjects who are randomly assigned to TXA and
subjects who are randomly assigned to placebo.

Specific aim 3: To estimate the association between the degree of fibrinolysis based on
kaolin activated TEG results and fibrinolytic pathway mediators on primary and secondary
clinical outcomes.

Tertiary hypothesis: The null hypothesis is that no association will exist between the
degree of fibrinolysis and fibrinolytic pathway mediators and primary and secondary
clinical outcomes.

3. Study Enrollment

EMS agencies will carry blinded sealed study drug kits. Once the seal is broken in the
presence of the patient, the patient is randomized. The EMS study drug kit will contain
a vial of either 1 gram TXA, 2 grams TXA, or placebo. EMS will mix the study drug in a
250 mL bag of 0.9% sodium chloride and administer the bolus infusion as soon as
life-saving interventions are performed. After randomization, EMS will provide the study
drug kit ID# to the receiving pharmacy. The hospital pharmacist will obtain the
randomization assignment from the coordinating center and prepare the appropriate drug
to be administered in the hospital.

4. Sample Size and Statistical Analysis

The total sample size is 963 (321 per group) starting treatment, which will allow for
80% power to detect a 7.1% absolute difference in favorable long-term neurological
outcome as determined by the GOS-E 6 months after injury comparing the combined TXA
treatment groups to placebo, using a one-sided, level 0.1 test.

Statistical analysis of primary hypothesis: Modified intention-to-treat analysis using
logistic regression to test for association and estimate the strength of the association
of treatment group with a favorable 6-month outcome (defined as a GOS-E > 4), after
adjustment for study site.

5. Human subjects protection

This study qualifies for the exception from informed consent (EFIC) required for emergency
research outlined in FDA regulation 21CFR50.24. EFIC applies because of life-threatening
situation, intervention must be administered before consent is feasible, no reasonable way to
identify prospectively individuals at risk, patients have the prospect of benefit from the
treatment, and the research could not practically be carried out without the waiver of
consent.

Inclusion Criteria:

1. Blunt or penetrating traumatic mechanism consistent with traumatic brain injury

2. Prehospital Glasgow Coma Score (GCS) score ≤ 12 at any time prior to randomization and
administration of sedative and/or paralytic agents

3. Prehospital systolic blood pressure (SBP) ≥ 90 mmHg prior to randomization

4. Prehospital intravenous (IV) or intraosseous (IO) access

5. Estimated Age ≥ 15 (or estimated weight > 50 kg if age is unknown)

6. Emergency Medicine System (EMS) transport to a participating trauma center

Exclusion Criteria:

1. Prehospital GCS=3 with no reactive pupil

2. Estimated time from injury to hospital arrival > 2 hours

3. Unknown time of injury - no known reference times to support estimation

4. Clinical suspicion by EMS of seizure activity or known history of seizures, acute
myocardial infarction (MI) or stroke

5. Cardio-pulmonary resuscitation (CPR) by EMS prior to randomization

6. Burns > 20% total body surface area (TBSA)

7. Suspected or known prisoners

8. Suspected or known pregnancy

9. Prehospital TXA given prior to randomization

10. Subjects who have activated the "opt-out" process when required by the local
regulatory board