A modified version of the EASIX (Endothelial Activation and Stress Index) score, when calculated prior to and soon after chimeric antigen receptor (CAR) T-cell infusion, predicted severe cytokine release syndrome (CRS) and immune effector cell–associated neurotoxicity syndrome (ICANS) before their occurrence. This is according to study findings published in Blood Advances.
Corresponding study author Miguel-Angel Perales, MD, of Memorial Sloan Kettering Cancer Center (MSKCC) in New York, explained that the EASIX score represents a reliable endothelial damage marker that is correlated with allogeneic hematopoietic cell transplantation outcomes. The formula for calculating the EASIX score includes lactate dehydrogenase (LDH) multiplied by creatinine, the product of which is divided by platelet count. “We know from prior studies that elevated LDH and low platelets have been associated with severe CRS and ICANS as has C-reactive protein (CRP),” Dr. Perales told ASH Clinical News. “As a result, we were interested in exploring the EASIX in patients treated with CAR T-cell therapies.”
To explore the role of EASIX in predicting outcomes in CAR T-cell recipients, Dr. Perales and colleagues developed a modified EASIX formula, which replaces creatinine with CRP. The researchers applied this modified formula in 53 patients with B-cell acute lymphocytic leukemia (B-ALL) and 65 patients with diffuse large B-cell lymphoma (DLBCL). Patients with B-ALL in the retrospective study were enrolled at MSKCC in a phase I trial of 1928z CAR T-cell therapy from 2010 through 2016. Patients with DLBCL were consecutively treated at the same center with axicabtagene ciloleucel or tisagenlecleucel from 2018 through 2019.
The overall median age of the study population was 58 years (44 years for patients with B-ALL and 64 years for patients with DLBCL). Pre-infusion Eastern Cooperative Oncology Group scores were 0 to 1 in 100% of patients with B-ALL and 86% of patients with lymphoma. Approximately 66% of the study population presented with a high disease burden prior to CAR T-cell infusion, as evidenced by bone marrow blasts percentage >5% in patients with B-ALL (62%) and stage 3 to 4 and/or bulky disease in patients with DLBCL (69%).
The primary objective of the study was to evaluate the modified EASIX formula’s ability to predict either severe CRS or ICANS. Secondary objectives included exploring the association between the modified EASIX and rates of complete response and best overall response rate to CAR T-cell therapy.
Both the original and modified EASIX formulas demonstrated similar predictive power for severe CRS and ICANS, Dr. Perales and colleagues reported. Low platelets and higher CRP values, however, were the only variables individually correlated with CRS and ICANS. Higher platelet counts were significantly associated with lower odds of severe CRS (odds ratio [OR] = 0.87; 95% CI 0.80-0.94; p<0.001) and of severe ICANS (OR=0.85; 95% CI 0.78-0.92; p<0.001). However, there was no association between levels of LDH or creatinine and severe CRS and ICANS.
Only the modified EASIX significantly predicted disease response and was able to predict severe CRS prior to the onset of severe symptoms (area under the curve [AUC] at lymphodepletion = 80.4%; day –1 = 73.0%; day +1 = 75.4%). At day +3, the modified scoring system also demonstrated ability to identify risk for severe ICANS (AUC = 73%).
“From a very practical point of view, we now have a tool that is readily accessible at all centers and that allows the clinician to predict severe CAR T-cell–associated toxicity before it occurs,” said Dr. Perales. “This may determine whether the patient should be treated outpatient or inpatient, or when the patient should be admitted after infusion of the cells.” He suggested that the modified score may also be useful in designing pre-emptive trials that seek to mitigate severe toxicity based on risk stratification.
A limitation of the study included its single-center, retrospective design. “We are currently working on addressing this limitation with a consortium of centers in the United States,” explained Dr. Perales. “In addition, we will need to design prospective clinical trials that stratify patients based on their modified EASIX score to demonstrate the utility in the setting of pre-emptive interventions to reduce severe CRS and ICANS.”
The study authors reported no relevant conflicts of interest.
Pennisi M, Sanchez-Escamilla M, Flynn JR, et al. Modified EASIX predicts severe cytokine release syndrome and neurotoxicity after chimeric antigen receptor T cells. Blood Adv. 2021;5(17):3397-3406.