Soluble Urokinase-Type Plasminogen Activator Receptor (suPAR) and Growth Differentiation Factor-15 (GDF-15) Levels Are Significantly Associated with Endothelial Injury Indices in CAR-T Cell Recipients

Introduction: Endothelial dysfunction indices, such as the Endothelial Activation and Stress Index (EASIX) and the modified EASIX (mEASIX) have been previously associated with CAR-T (Chimeric Antigen Receptor T) cell-related toxicities, mainly cytokine release syndrome (CRS) and neurotoxicity (ICANS), and poor outcomes. Other markers of endothelial injury, including soluble urokinase-type plasminogen activator receptor (SuPAR) and growth differentiation factor-15 (GDF-15) have been measured in patients who receive allogeneic hematopoietic stem cell transplantation (alloHCT) and have been associated with endothelial injury syndromes post alloHCT. We hypothesized that suPAR and GDF-15 would reflect endothelial injury in CAR-T cell recipients.

Methods: We enrolled consecutive adult patients who received CAR-T cell products in our JACIE-accredited Unit. Plasma was collected and stored immediately at -80oC before admission to the Unit and SuPAR and GDF-15 levels were measured using immunoenzymatic techniques. Calculation of EASIX [lactate dehydrogenase (U/L) × creatinine (mg/dL)/thrombocytes (10⁹ cells per L)] was performed at baseline, at the day of CAR-T infusion and 14 days after the administration of the CAR-T therapy. Furthermore, mEASIX score, in which the creatinine value is replaced by the C-reactive protein value, was calculated at baseline. All patients received lymphodepleting therapy before CAR-T cell infusion with cyclophosphamide and fludarabine. Tisagenlecleucel and axicabtagene ciloleucel (tisa-cel and axi-cel) were administered since 2020, and brexucabtagene autoleucel (brexu-cel) since 2022. For patient monitoring as well as for prophylaxis and treatment of side effects, the EBMT and MD Anderson guidelines were adopted. Levetiracetam prophylactic treatment was administered. The data cut-off date was the end of June 2024, with at least 1 month of follow-up post-infusion. The following factors were analyzed: age, disease, CAR-T cell product infused, EASIX and mEASIX scores, laboratory values, suPAR, GDF-15 levels, overall survival, and progression-free survival.

Results: We studied 45 CAR-T cell recipients (24 axi-cel, 14 tisa-cel, and 7 brexu-cel), and 20 healthy controls. We found significantly higher suPAR (median: 3.7 ng/ml, range: 2-31.2) and GDF-15 (median: 2807.5, range: 569.2-8655) levels in CAR-T patients compared to healthy controls (p<0.001). Then, we further analyzed the characteristics of CAR-T cell recipients.

Severe CRS (grade> 3) was recorded in 6 patients, while severe ICANS in 3. SuPAR levels at baseline were associated with the mEASIX score, calculated at the same time point (r=0.295, p=0.052). SuPAR and GDF-15 concentrations were associated with EASIX score at day 14 post-infusion (r=0.885, p<0.001 and r=0.557, p<0.001, respectively). Among the laboratory values used to calculate EASIX (LDH, creatinine, platelets), suPAR was associated with LDH levels during infusion (r=0.312, p=0.039) and platelets at day 14 (r=0.295, p=0.052). Furthermore, GDF-15 was associated with ferritin levels at baseline (r=-0.305, p=0.044). In the multivariate regression analysis, suPAR levels were correlated with mEASIX at baseline (p<0.001), EASIX at day 14 (p=0.017), and LDH values at infusion (p<0.001), while GDF-15 levels were associated with baseline ferritin values (p=0.036) and EASIX score at day 14 (p<0.001). SuPAR and GDF-15 levels did not correlate with other studied factors.

Conclusion: Our study shows for the first time that suPAR and GDF-15 reflect endothelial injury in CAR-T cell recipients. In accordance with other patient populations, suPAR emerges as a marker of endothelial dysfunction, characterizing high-risk in endothelial injury syndromes and in particular, CRS. However, prior to their clinical usefulness, rigorous validation in multiple cohorts is warranted.