Time-Series Clustering Captures Patterns of Early Immune Effector Cell–Associated Hematotoxicity That Are Predictable Using Tree-Based Models
JCO Clinical Cancer Informatics · 2026
DOI: 10.1200/cci-25-00148 ↗Kurz gefasst
Fachartikel in der Zeitschrift JCO Clinical Cancer Informatics, 2026. Verfasst von Emily C Liang und 16 weiteren Autor:innen. Beteiligte Einrichtung: Fred Hutch Cancer Center, Seattle, WA. Hauptthemen laut MEDLINE: Neoplasms. Die Arbeit nennt 18 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NCI NIH HHS, NIGMS NIH HHS, NHLBI NIH HHS.
Abstract (Originalsprache)
PURPOSE Immune effector cell–associated hematotoxicity (ICAHT) is a major cause of nonrelapse mortality after chimeric antigen receptor (CAR) T-cell therapy. We hypothesized that unsupervised time-series clustering could better identify archetypal patterns of early hematotoxicity compared to the early ICAHT (eICAHT) grading system. METHODS We applied unsupervised k-means time-series clustering based on Euclidean distances to longitudinal absolute neutrophil count (ANC) data from days +0 through +30 post–CAR T-cell infusion in 691 patients treated at our center (training set: n = 483, 70%; test set: n = 208, 30%). RESULTS Within our training set, we identified an optimal cluster solution based on four ANC recovery clusters, which were labeled as very good, good, poor, and very poor. We trained a random forest (RF) model including the top five most important features (day +3, +4, +5, +26, and +27 ANC values) to predict the cluster assignments. Within our test set, we applied the RF model to predict cluster assignments. Compared with the eICAHT criteria, the RF-predicted clusters were more compact and better separated (Dunn index: 0.078 v 0.034; average silhouette width: 0.12 v 0.010). In addition, the RF model identified patients in the good recovery cluster with intermediate overall survival (hazard ratio [HR], 1.70 [95% CI, 1.05 to 2.74]; P = .029; reference, very good), which was not captured by grade 2 eICAHT (HR, 1.37 [95% CI, 0.80 to 2.35]; P = .25; reference, grade 0-1). CONCLUSION Unsupervised time-series clustering identified distinct and clinically relevant patterns of hematotoxicity after CAR T-cell therapy. We trained and tested an RF model that accurately predicted cluster assignments using only five features. Predictions can be generated using our online web application.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Fred Hutch Cancer Center, Seattle, WA.
- University of Washington, Seattle, WA.
ORCID-Profile: Emily C Liang, Yein Jeon, Yang Qiao, Xiancheng Wu, Jennifer J Huang, Andrew J Portuguese, Aiko Torkelson, Delaney Kirchmeier, Kristina Braathen, Andrew J Cowan, Mazyar Shadman, Alexandre V Hirayama, Erik L Kimble, Qian Wu, Jordan Gauthier
Förderung
- NCI NIH HHS
- NIGMS NIH HHS
- NHLBI NIH HHS
Zitieren
APA
Liang, E. C., Jeon, Y., Qiao, Y., Wu, X., Huang, J. J., Portuguese, A. J., Basom, R., Torkelson, A., Kirchmeier, D., Braathen, K., Cowan, A. J., Shadman, M., Hirayama, A. V., Till, B. G., Kimble, E. L., Wu, Q., & Gauthier, J. (2026). Time-Series Clustering Captures Patterns of Early Immune Effector Cell–Associated Hematotoxicity That Are Predictable Using Tree-Based Models. JCO Clinical Cancer Informatics. https://doi.org/10.1200/cci-25-00148
Vancouver
Liang EC, Jeon Y, Qiao Y, Wu X, Huang JJ, Portuguese AJ, et al. Time-Series Clustering Captures Patterns of Early Immune Effector Cell–Associated Hematotoxicity That Are Predictable Using Tree-Based Models. JCO Clinical Cancer Informatics. 2026. doi:10.1200/cci-25-00148
BibTeX
@article{Liang2026,
author = {Liang, Emily C and Jeon, Yein and Qiao, Yang and Wu, Xiancheng and Huang, Jennifer J and Portuguese, Andrew J and Basom, Ryan and Torkelson, Aiko and Kirchmeier, Delaney and Braathen, Kristina and Cowan, Andrew J and Shadman, Mazyar and Hirayama, Alexandre V and Till, Brian G and Kimble, Erik L and Wu, Qian and Gauthier, Jordan},
title = {Time-Series Clustering Captures Patterns of Early Immune Effector Cell–Associated Hematotoxicity That Are Predictable Using Tree-Based Models},
journal = {JCO Clinical Cancer Informatics},
year = {2026},
doi = {10.1200/cci-25-00148}
}
EUVIMED