Tranexamic Acid to Prevent Bleeding in Patients with Hematologic Malignancies and Severe Thrombocytopenia (TREATT trial). a Randomized Placebo-Controlled Trial

LJE & ZM and EMW & SJS contributed equally Patients undergoing treatment for hematologic malignancies continue to have a high burden of clinically significant bleeding (WHO grade 2 or above) despite receiving prophylactic platelet transfusions (43% in TOPPS trial, NEJM 2013). One potential approach to reduce the burden of bleeding is antifibrinolytic therapy, which has been shown to reduce bleeding and transfusion needs in patients undergoing surgery. The Trial to EvaluAte Tranexamic acid therapy in Thrombocytopenia (TREATT) is a double-blind randomized controlled trial evaluating the safety and efficacy of tranexamic acid (TXA) in patients with hematologic malignancies with severe thrombocytopenia (EudraCT number 2014-001513-35). We enrolled adults with hematologic malignancies having intensive chemotherapy or a hematopoietic stem cell transplant (HSCT) who were expected to have severe thrombocytopenia (platelet count ≤10x10 9/L for ≥5 days) at 16 UK and 11 Australian sites. We stratified randomization by site. We randomized participants to receive TXA (1g iv or 1.5g orally) or placebo (saline or capsule/tablet) 8-hourly from when platelet count ≤30x10 9/L or as soon as possible if the platelet count was already ≤ 30x10 9/L (no later than 72 hours after start of chemotherapy or HSCT). We stopped treatment if the unsupported platelet count reached ≥ 30x10 9/L, thrombosis occurred, or patient had received 30 days treatment. Patients received red cell and platelet transfusions according to national transfusion guidelines. Bleeding assessments were performed by trained research staff (inpatients) or via a patient diary, with significant bleeding reported to clinicians (outpatients), using validated tools. The primary outcome was the proportion of participants who died or had bleeding of WHO Grade 2 or above from study day 1 ([D1], first day of treatment) to D30. We used marginal Cox proportional hazards regression modelling (adjusted for site to account for clustering) to compare the two arms. Secondary outcomes included: safety outcomes (thrombotic events to D120) and use of blood components. Recruitment was affected by the COVID pandemic. We randomized 616 participants between June 2015 and February 2022 and included 597 in the modified intention to treat (mITT) analysis (platelet count ≤ 30x10 9/L for ≥ 1 day and no exclusion criteria). 16 participants were lost to follow-up prior to D120 (8 each arm); we included them in analyses if possible. Participant