https://orcid.org/0000-0003-4281-4910
In this issue of Blood Advances, Zandaki et al1 validated that the endothelial activation and stress index (EASIX) and modified EASIX (m-EASIX) scores predict the development of cytokine release syndrome (CRS) and immune cell–associated neurotoxicity syndrome (ICANS) in pediatric and adolescent/young adult (AYA) patients after CD19-directed chimeric antigen receptor (CAR) T cells.
The EASIX score (lactate dehydrogenase [LDH] × creatinine/platelet) and a derivative score, m-EASIX (LDH × C-reactive protein/platelet), have been previously demonstrated to be predictive for the development of CRS and ICANS in adult patient populations treated with CD19-directed CAR T-cell therapy.2,3 Zandaki et al extended the use of EASIX and m-EASIX scores for predicting CAR T-cell–related complications to pediatric/AYA patients. This retrospective analysis included 76 heavily pretreated pediatric/AYA patients with relapsed/refractory (R/R) B-cell acute lymphoblastic leukemia (B-ALL) treated at 2 institutions (St. Jude Children’s Research Hospital, n = 51; Johns Hopkins Hospital, n = 25) with either an investigational CD19-directed CAR T-cell product (NCT03573700; n = 22) or tisagenlecleucel (n = 54). At all evaluated time points (days –5, 0, +3), median EASIX scores were higher for patients who developed severe CRS and any-grade ICANS than those with no or mild CRS/ICANS. Median m-EASIX scores at the same time points were also higher in patients who developed severe CRS and severe ICANS than those without these toxicities. The authors’ analysis demonstrates the m-EASIX uniformly outperformed EASIX, except for predicting any-grade ICANS. This study is limited by the small sample size and data spanning both investigational and commercially available cellular therapy products. Further validation in a larger cohort of homogeneous patients is warranted, and developing strategies to implement the use of these tools in a prospective manner would be instrumental in validating the clinical impact of the authors’ findings.
The clinical outcomes for pediatric/AYA patients with R/R B-ALL have been improved with the use of CD19-directed CAR T-cell therapy.4 Broad application of this cellular therapy can be hindered by the potential for severe complications such as CRS, ICANS, and immune effector cell–associated hemophagocytosis syndrome. “Real-world” evidence has now demonstrated lower rates of severe complications after the use of a commercially available product (tisagenlecleucel) in pediatric/AYA patients.5 However, treating institutions have developed complex and comprehensive toxicity management algorithms that typically lack predictive models. Comprehensive tools for the early detection of postinfusion toxicities have the potential to enable prophylactic interventions that reduce the incidence and/or severity of toxicities. Predictive tools that enable the identification of patients at risk for severe complications may also widen the network of treating centers able to offer novel cellular therapies.
Future studies of these predictive tools should incorporate a comprehensive algorithm including other established risk factors of CRS (pretreatment disease burden, early CAR T-cell activation [CD28 costimulatory domains], and higher CAR T-cell dose) and ICANS (pretreatment disease burden, high-grade CRS, peak CAR T-cell expansion, extramedullary disease/central nervous system involvement, higher CAR T-cell dose, preexisting neurologic comorbidity, and high inflammatory cytokine concentrations within 3 days of treatment).4,6-9 This comprehensive algorithm could help stratify patients at risk for postinfusion complications, identify patients requiring inpatient treatment, and/or distinguish those patients eligible to receive prophylactic interventions.10 Beyond CD19-directed CAR T cells, the field of cellular therapy would benefit from a predictive model of toxicity as we extend treatment to other novel targets, solid malignancies, and off-the-shelf allogeneic cellular therapy products.
Conflict-of-interest disclosure: K.J.C. has received research support from Juno Therapeutics, Novartis, and Celgene, and has consulted, participated in advisory boards, or participated in educational seminars for Novartis. S.S. declares no competing financial interests.
