IMMUNOLOGY2026™ Conference Recordings For Attendees-The 1000+ mouse project: large-scale spatiotemporal parametrization and modeling of preclinical cancer immunotherapies

The 1000+ mouse project: large-scale spatiotemporal parametrization and modeling of preclinical cancer immunotherapies

Video Transcription

Video Summary

The speaker presented a recently published study from the Gregoire Altenier Lab analyzing over 1,000 mouse CAR T-cell therapy images. The project aimed to move beyond simple visual inspection of bioluminescent IVIS images by creating a Python package and GUI to automatically quantify tumor radiance over time. This enabled detailed tracking of tumor growth, decay, and relapse patterns across many mice. The method validated known differences between CD28 and 4-1BB CAR T-cell designs: CD28 showed stronger initial killing, while 4-1BB gave better long-term persistence and relapse control. The analysis also revealed a non-linear relationship between CAR T-cell dose and relapse, slower-growing tumors were more likely to evade killing, and relapse growth rates were generally slower than initial growth rates. The platform could also distinguish patterns in a few anatomical regions, though 2D imaging limits tissue-specific precision.

Meta Tag

Date April 19, 2026 8:13 AM - 8:26 AM

Room 102

Session Development of Statistical, Machine Learning, and Genetic Models

Speaker Sooraj Achar

Track Computational and Systems Immunology (COMP)

Year 2026

Keywords

CAR T-cell therapy

bioluminescent imaging

tumor radiance quantification

CD28 vs 4-1BB

tumor relapse dynamics

April 19, 2026 8:13 AM - 8:26 AM

102

Development of Statistical, Machine Learning, and Genetic Models

Sooraj Achar

Computational and Systems Immunology (COMP)

2026