High risk of bleeding following CAR T-cell therapy: Insights from the dutch “Follow that CAR!” registry

Abstract Background: Chimeric antigen receptor (CAR) T-cell therapy has revolutionized treatment and outcomes for patients with relapsed/refractory aggressive large B-cell lymphoma (R/R LBCL). Cytokine release syndrome (CRS), immune effector cell-associated neurotoxicity syndrome (ICANS) and infections are commonly reported CAR T-cell-related toxicities. Recently, thrombosis and bleeding complications have also been described but their clinical relevance and risk factors are unclear, due to heterogeneity between reporting studies. Objectives: To investigate the incidence of thrombosis and bleeding in a real-world cohort of patients undergoing CAR T-cell therapy, and to explore associations with CRS/ICANS, anticoagulant therapy and laboratory parameters. Methods: For this observational multicenter study, we used patients from the Dutch “Follow that CAR!” registry combined with additional data from medical records. We included adults aged ≥18 years with R/R LBCL, treated between May 2020 and December 2023 with standard lymphodepleting chemotherapy before infusion of commercially available axicabtagene ciloleucel (axi-cel). Data was collected from start of lymphodepleting chemotherapy until 1 year after axi-cel infusion. We recorded all objectified thrombosis (venous or arterial) and bleeding events (classified as major bleeding or clinically relevant non-major bleeding (CRNMB), according to International Society on Thrombosis and Hemostasis criteria), CRS/ICANS and death. Complete blood count and coagulation parameters were collected up to 90 days post-infusion. We calculated 1-year cumulative incidences of thrombosis and bleeding with 95% confidence intervals (CI95). Subgroup analyses were performed by anticoagulant use around CAR T-cell infusion and outcomes were compared using hazard ratios (HRs). Multivariable Cox proportional hazard regression models were used to evaluate risk factor associations. Results: We included 240 consecutive patients with R/R LBCL who received axi-cel therapy in ≥3rd line. Median age was 61 years, 67% were male. 32% of patients had prior thrombosis, 9% atrial fibrillation. CRS/ICANS grade ≥2 were observed in 51% and 40%, respectively. Thrombosis occurred in 15 patients, 9 venous and 6 arterial events, with a 1-year cumulative incidence of 6.6% [CI95 3.8-10.4]. 25 patients experienced bleeding (32% gastrointestinal, 20% tumor-related), of which 11 major (1 fatal) and 14 CRNMB, with a 1-year cumulative incidence of 11.0% [CI95 7.3-15.5]. Median time to thrombosis and bleeding was both 28 days. 1-year overall survival was 62.4% [CI95 56.5-69.0], most patients died due to disease progression. Overall survival was similar irrespective of thrombosis (60.0% vs 62.6%; HR 1.08, p=0.89), but bleeding was associated with poorer overall survival (38.1% vs 65.2%; HR 2.43, p=0.002). 55 (23%) patients received therapeutic anticoagulation around axi-cel infusion, 41 (17%) prophylactic and 144 (60%) antiplatelets or no anticoagulation. Cumulative thrombosis incidence did not differ between patients receiving therapeutic or prophylactic anticoagulation, and those who did not. However, cumulative bleeding incidence was significantly higher in patients on therapeutic anticoagulation than in those without anticoagulation (21.8% vs 7.0%; HR 3.54, p=0.003). Bleeding with prophylactic anticoagulation was similar to no anticoagulation (7.2% vs 7.0%; HR 1.07, p=0.92). In multivariable analyses, ICANS grade ≥2 was associated with a higher risk of thrombosis (HR 3.2, p=0.045) and bleeding (HR 2.4, p=0.036). Thrombocytopenia grade ≥3 (platelets <50x109/L; HR 3.4, p=0.035) and therapeutic anticoagulation (HR 4.8, p=0.001) were also associated with a higher bleeding risk. Coagulation parameters were not routinely measured and associations could not be analyzed. Conclusions: In this homogenous large real-world cohort study, we confirm that patients with R/R LBCL are at considerable risk of thrombosis and high risk of bleeding following axi-cel therapy, especially in the first 30 days following infusion. Bleeding risk was higher with ICANS grade ≥2, thrombocytopenia grade ≥3 and therapeutic anticoagulant therapy, and bleeding was associated with poorer overall survival. These findings suggest caution with anticoagulant therapy around axi-cel infusion and support further research on thrombosis, bleeding and anticoagulation in the context of CAR T-cell therapies for LBCL and other indications.