Related Experiment Video
Updated: Aug 12, 2026
06:08
Assessment of Chimeric Antigen Receptor T Cell-Associated Toxicities Using an Acute Lymphoblastic Leukemia Patient-Derived Xenograft Mouse Model
Published on: February 10, 2023
Predictive Models for Toxicities after CAR T-cell Therapy: Challenges and Opportunities
Julie Ma1,2, August Culbert1, Tian-Gen Chang3
1Pediatric Oncology Branch, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, Maryland.
Blood Cancer Discovery
|August 11, 2026
Summary
Predictive models for chimeric antigen receptor (CAR) T-cell therapy toxicities show promise but face challenges. Improving model accuracy and data sharing is crucial for wider clinical adoption and patient safety.
Area of Science:
- Oncology
- Immunotherapy
- Biostatistics
Background:
- Chimeric antigen receptor (CAR) T-cell therapy is expanding beyond blood cancers.
- Predictive models are being developed to forecast CAR T-cell therapy toxicities.
Purpose of the Study:
- To review existing and emerging predictive models for CAR T-cell therapy toxicities.
- To identify strengths, challenges, and future directions for predictive modeling in this field.
Main Methods:
- Comprehensive literature review of predictive models for CAR T-cell therapy toxicities.
- Analysis of model components: discrimination, calibration, biomarker integration, and validation.
- Discussion of challenges like overfitting and data quality.
Main Results:
- Existing models offer potential for risk stratification but face limitations.
- Small sample sizes, overfitting, and poor data quality hinder reproducibility.
- Heterogeneity in patients and products poses modeling challenges.
Conclusions:
- Further development requires identifying new biomarkers and context-specific models.
- Standardizing guidelines for emerging toxicities is essential.
- Federated learning can enhance collaborative data sharing for improved model development.
