Pre-infusion ADAMTS13 and VWF levels are associated with clinical outcomes of patients receiving CAR-T cell therapy

Abstract Background: Chimeric antigen receptor T (CAR-T) cell therapy has been a breakthrough in the management of relapsed/refractory B cell malignancies with unprecedented efficacy but is associated with significant early toxicity including cytokine release syndrome (CRS), immune effector cell-associated neurotoxicity syndrome (ICANS), infections and pancytopenia. Pre-infusion chemotherapy lymphodepletion increases the levels of interleukin-15 and interleukin-7 cytokines which enhance CAR-T expansion and persistence. However, chemotherapy can be toxic to vascular endothelial cells. This endothelial priming leads to the release of biomarkers such as von Willebrand factor (VWF) and synecan-1 (SDC-1). Persistent or severe endothelial damage (with ongoing VWF and SDC-1 release) may outpace the availability of ADAMTS13, resulting in relatively reduction of ADAMTS13 to VWF ratio and possible thrombotic microangiopathy, which can be manifested in severe inflammation as in sepsis and severe COVID-19. Previous studies have shown the correlation between increased levels of VWF antigen and development of CRS and ICANS. Nonetheless, data are scarce exploring the effect of VWF and ADAMTS13 activity and outcomes related to the efficacy and toxicity of CAR-T cell therapy. In this study, we longitudinally assessed the changes of plasma ADAMTS13 activity, VWF antigen levels, at days 0, 3, 14, and 28 post-CAR-T infusion and correlation with outcomes post CAR-T cell therapy. Methods We reviewed charts of 62 patients who received CAR-T cells from January 2018 to December 2023. Plasma samples for ADAMTS13 activity, VWF, and SDC-1 antigens were collected on Day 0 from 62 patients, and from 43 of these patients on Days 3, 14, and 28 post–CAR-T infusion. Plasma ADAMTS 13 activity was analyzed by FRETS-VWF73 and VWF antigen by ELISA. Citrated plasma samples from health donors (n=30) were used as controls. Data analyses were conducted using GraphPad Prism and statistical significance was considered at p<0.05. Nonparametric Mann-Whitney test was used to compare experimental and control groups. Results: 62 patients were treated with CAR-T cell therapy and the median age was 66 years (26–82), and a male predominance (62.9%). Most patients (80.6%) were diagnosed with diffuse large B-cell lymphoma (DLBCL). Regarding to stem cell transplantation status, 29% had undergone a hematopoietic stem cell transplant, of which 83.3% were autologous and 16.7% allogeneic. The CAR-T products administered included axicabtagene ciloleucel (67.7%), tisagenlecleucel (22.5%), and other products (9.6%). All patients received 3-day lymphocyte-depleting chemotherapy prior to CAR-T therapy consisting of fludarabine and cyclophosphamide. Plasma ADAMTS13 activity prior to CAR-T infusion on day 0 was significantly lower than that in healthy controls, and were further reduced on days 3, 14, and 28 post CAR-T cell infusion. Plasma levels of VWF antigen prior to CAR-T were significantly higher than that in the healthy controls, but the levels did not increase further post CAR-T cell infusion, suggesting endothelial exhaustion. Patients who achieved complete response (CR) at day 30 post CAR-T cell infusion had higher ADAMTS13 activity on day 0 compared with non-responders (96 Vs. 81%; p= 0.0095). Patient who developed CRS grade II or higher had a trend towards lower ADMATS13 activity level at day 0 (68 vs 83%; p=0.08). Patients who developed ICANS at any grade had lower levels of ADMATS13 (62 vs 83%; p=0.13) and higher VWF antigen activity (280 vs 244%) but did not reach statistical significance (p=0.11). Patients who developed graded severe thrombocytopenia had a trend towards higher VWF activity at Day 0 (293 vs 225; p=0.089). In addition, patients who developed infections in the first 30 days post CAR-T cell infusion demonstrated an increase in plasma ADAMTS13 activity (83 vs. 64%; p=0.0364) and higher VWF activity (346 vs 202%; p=0.0028) compared to those with no infections. Conclusions: Our data demonstrate the association of ADAMTS13/VWF activity and clinical outcomes following CAR-T cell therapy. Our results reveal that lymphodepleting chemotherapy prior to CAR-T cell likely affects ADAMTS13/VWF balance. This balance if tilted severe enough may associate with decreased responses to CAR-T cell therapy and increased toxicity in the form of CRS, ICANS, infections and thrombocytopenia.