Assessing the Predictive Value of Pre-Lymphodepletion and Infusion Endothelial Activation and Stress Index in CD19-Directed CAR-T Therapy for B-Cell Lymphoma

• Published in: Transplantation and cellular therapy 2024-02, Vol.30 (2), p.S223-S224
• Main Authors: Upreti, Harsh Vardhan, Lee, Jimmy, Sannareddy, Aishwarya, Reves, Heather, Turer, Laura, Abraham, Pearl, Liu, Yulun, Yilmaz, Elif, Wolfe, Heather, Kaur, Gurbakhash, Khan, Adeel, Madanat, Yazan F., Thiebaud, Julio Alvarenga, Velarde, Alejandro Marino, Collins, Robert H., Chung, Stephen, Geethakumari, Praveen Ramakrishnan, Anderson, Larry D., Awan, Farrukh T, Afrough, Aimaz, Anderson, Julia Marie
• Format: Article
• Language: English
• ISSN: 2666-6367; 2666-6367
• DOI: 10.1016/j.jtct.2023.12.290

The Endothelial Activation and Stress Index (EASIX) formula (creatinine [mg/dL] × lactate dehydrogenase [U/L])/platelets [109 cells/L]) measures endothelial activation and is relevant in CAR T-related toxicities. Prior studies linked pre-lymphodepletion (pre-LD) EASIX and ferritin (EASIX-F) to CRS grade 2-4 and EASIX combined with ferritin and CRP (EASIX-FC) to ICANS grade 2-4 in lymphoma patients (pts) treated with Axi-cel. We aimed to assess their predictive value at pre-LD and CAR-T infusion in B cell lymphoma pts receiving CD19-directed CAR T therapy. We retrospectively studied 97 pts with B cell lymphoma who received CD19 CAR T from Dec 2018 to Mar 2023. EASIX-F and EASIX-FC at pre-LD and infusion were calculated per original publication (PMID: 34264268). We estimated the cumulative incidence of CRS and ICANS starting from the infusion date, considering death before toxicity as a competing risk. Univariate Cox-regression was conducted using SPSS v.29. Baseline patient characteristics in Table 1. CRS occurred in 88% (30% grade 2-4), while ICANS was recorded in 44% (24% grade 2-4). Pre-LD and infusion EASIX-F and EASIX-FC were calculable in 97-99% of pts. Pre-LD EASIX-F predicted CRS grade 2-4: 20% in low (reference(ref)), 41% (Hazard ratio (HR) 2.2; 95% Confidence interval (CI) 1.04-5; p=.03) in intermediate, and 75% (HR 4.36; 95% CI 1.2-15.4; p=.02) in high risk. Infusion EASIX-F showed no correlation with CRS grade 2-4 (p=.6) (Fig 1). EASIX-FC was not associated with ICANS grade 2-4 at pre-LD or infusion. Cumulative incidence of ICANS grade 2-4 was 26%, 17%, and 40% at pre-LD and 30%, 19%, 25% at infusion, in low, intermediate, high risk, respectively. Findings differ from prior studies, possibly due to limited sample size and lower severe ICANS rate. Response data were available for 93% of pts. No significant difference in 30-day complete response (CR) was observed according to EASIX-F and EASIX-FC at various time points. However, high risk pts had a significantly lower overall response rate (≥PR) compared to intermediate and low risk. This difference was statistically significant in pre-LD EASIX-F (p=.02), infusion EASIX-F (p=.003), and infusion EASIX-FC (p=.002). Median follow-up: 8 months (mon) (range, 5-49). We observed a significant association between infusion EASIX-F and PFS (p=.003). In the high risk, median PFS was 2.8 mon (HR 3.05; 95% CI 1.16-8.04; p=.024), while in intermediate, was not reached (HR 0.69; 95% CI 0.34-1.3; p=.2), and was 7