Development of ALL-Hematotox: predicting post-CAR T-cell hematotoxicity in B-cell acute lymphoblastic leukemia

Monica S. Nair, Sara K. Silbert, Kai Rejeski, Karilynn A. Wilson, Adam J. Lamble, Yannis Valtis, Bonnie Yates, Alexa Morales Arana, Roni Shouval, Kevin Curran, Rebecca A. Gardner, Haneen Shalabi, Colleen Annesley, Jae H. Park, Marion Subklewe, Nirali N. Shah

Research output: Contribution to journalArticlepeer-review

5 Scopus citations

Abstract

Immune effector cell–associated hematotoxicity (ICAHT) is a major B-cell targeted chimeric antigen receptor (CAR) T-cell related toxicity. Although ICAHT incidence and severity is documented in large B-cell lymphoma (LBCL), mantle cell lymphoma (MCL), and multiple myeloma (MM), ICAHT has not been described in B-cell acute lymphoblastic leukemia (B-ALL). Similarly, the CAR-HEMATOTOX (CAR-HT) model, designed to predict severe prolonged neutropenia (≥14 days of absolute neutrophil count [ANC] <500/μL), has been validated in LBCL, MCL, and MM, but not in B-ALL. As B-ALL bone marrow (BM) infiltration can impact cytopenias, we sought to describe ICAHT and assess CAR-HT for predicting hematotoxicity in B-ALL. In a cohort of 156 children and young adults with relapsed/refractory B-ALL, the median duration of severe neutropenia (ANC <500/μL) was 13 days (95% confidence interval, 10-16 days), with 83 (53%) experiencing grade ≥3 ICAHT. Applying CAR-HT, nearly 90% were classified as high risk, demonstrating limited discriminative power and prompting further development. Using the association identified between BM disease burden and postinfusion neutropenia (r = 0.64, P < .0001), we developed the ALL-Hematotox (ALL-HT) score, which substitutes BM disease burden for ferritin in CAR-HT. The ALL-HT score associated with severe prolonged neutropenia (area under the curve = 0.84, P < .0001), and appropriately discriminated high-risk patients (47%) who had more cumulative days of neutropenia (26 vs 4 days; P < .0001), fewer rates of complete response (88% vs 98%; P = .03), and shorter median overall survival (9.8 vs 24 months; log-rank P = .0002). ALL-HT was also validated in 2 independent cohorts. The ALL-HT score refines a widely accepted predictive model of postinfusion hematotoxicity, which is applicable in B-ALL.

Original languageEnglish
Pages (from-to)1136-1148
Number of pages13
JournalBlood
Volume145
Issue number11
DOIs
StatePublished - 13 Mar 2025
Externally publishedYes

Bibliographical note

Publisher Copyright:
© 2025 American Society of Hematology

Fingerprint

Dive into the research topics of 'Development of ALL-Hematotox: predicting post-CAR T-cell hematotoxicity in B-cell acute lymphoblastic leukemia'. Together they form a unique fingerprint.

Cite this