Abstract 3181: Single cell characterization and anti-tumor efficacy of single & multi-armored CAR Ts against CLDN18.2

Abstract Adoptive cell therapies such as CAR T have struggled to translate to solid tumors due to challenges posed by the solid tumor microenvironment (TME) and the low-throughput of in vivo testing. Pooled CAR T screens are typically restricted to single readouts (e.g. proliferation) and cannot measure the cell-cell and cell-TME interactions that are critical for CAR T success in solid tumors. We performed in vivo spatial pooled screening to profile >250 different CAR T designs to identify efficacious constructs based on spatial-dependent CAR T properties and CAR T-TME interactions. Mouse xenograft models of pancreatic cancer were each injected with hundreds of unique CAR T constructs, including variations in CAR designs and armors. For each library, we prepared 30-40 FFPE slides, harvested from different mice across time points and treated with CAR T cells from different donors. Each slide was stained for 20-30 protein markers, and we also performed in situ genotyping to determine CAR T construct identities in the same microscope images. We obtained image-based phenotypes of ∼20M total single cells and used custom computer vision pipelines to define single cell spatial phenotypes for each CAR T design, including cytotoxicity, exhaustion, local-neighborhood phenotypes, tumor infiltration, on-target/off-tumor toxicity, proliferation, and activity in immunosuppressive tumor regions. We evaluated selected screen hits in single-construct studies, and trained a random forest model with cross-validation to identify the combination of single cell spatial phenotypes for CAR T designs that were most predictive of tumor shrinkage in the efficacy studies. Overall, we identified multiple single- and double-armored constructs that had superior spatial single cell-level phenotypes (e.g. improved infiltration, increased persistence, decreased PD1/LAG3, etc.) compared to both unarmored 2nd gen CLDN18.2 CAR Ts and TGFβRIIDN-armored CLDN18.2 CARs. We then confirmed superior anti-tumor efficacy of these CAR modifications and armors in multiple xenograft models, at multiple doses ranging from 0.5x106 to 2x106 cells. Combinations of these top armors will be further tested in WP-001, a CAR T construct for solid tumors expressing CLDN18.2. Citation Format: Megha Sah, Liang He, Gundula Povysil, Cassandra Kontur, Ashley Chui, Stephen Pu, Annie Dyatel, Malavika Sabu, Alice HK Kwon, Xinchen Wang, David Phizicky. Single cell characterization and anti-tumor efficacy of single & multi-armored CAR Ts against CLDN18.2 [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 3181.

Authors

Institutions

Publication Details

Journal
Cancer Research
Published
2025-04-21
DOI
https://doi.org/10.1158/1538-7445.am2025-3181
Primary Topic
CAR-T cell therapy research
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Abstract 3181: Single cell characterization and anti-tumor efficacy of single & multi-armored CAR Ts against CLDN18.2

Stephen Pu, Cassandra Kontur, David V. Phizicky, Malavika Sabu et al.
Cancer Research
CAR-T cell therapy research
article

Abstract 3181: Single cell characterization and anti-tumor efficacy of single & multi-armored CAR Ts against CLDN18.2

Stephen Pu, Cassandra Kontur, David V. Phizicky, Malavika Sabu, Amy Chui, Gundula Povysil, Annie Dyatel, Liang He, Xinchen Wang, Megha Sah, A.-H. Kwon
article en

Abstract

Abstract Adoptive cell therapies such as CAR T have struggled to translate to solid tumors due to challenges posed by the solid tumor microenvironment (TME) and the low-throughput of in vivo testing. Pooled CAR T screens are typically restricted to single readouts (e.g. proliferation) and cannot measure the cell-cell and cell-TME interactions that are critical for CAR T success in solid tumors. We performed in vivo spatial pooled screening to profile >250 different CAR T designs to identify efficacious constructs based on spatial-dependent CAR T properties and CAR T-TME interactions. Mouse xenograft models of pancreatic cancer were each injected with hundreds of unique CAR T constructs, including variations in CAR designs and armors. For each library, we prepared 30-40 FFPE slides, harvested from different mice across time points and treated with CAR T cells from different donors. Each slide was stained for 20-30 protein markers, and we also performed in situ genotyping to determine CAR T construct identities in the same microscope images. We obtained image-based phenotypes of ∼20M total single cells and used custom computer vision pipelines to define single cell spatial phenotypes for each CAR T design, including cytotoxicity, exhaustion, local-neighborhood phenotypes, tumor infiltration, on-target/off-tumor toxicity, proliferation, and activity in immunosuppressive tumor regions. We evaluated selected screen hits in single-construct studies, and trained a random forest model with cross-validation to identify the combination of single cell spatial phenotypes for CAR T designs that were most predictive of tumor shrinkage in the efficacy studies. Overall, we identified multiple single- and double-armored constructs that had superior spatial single cell-level phenotypes (e.g. improved infiltration, increased persistence, decreased PD1/LAG3, etc.) compared to both unarmored 2nd gen CLDN18.2 CAR Ts and TGFβRIIDN-armored CLDN18.2 CARs. We then confirmed superior anti-tumor efficacy of these CAR modifications and armors in multiple xenograft models, at multiple doses ranging from 0.5x106 to 2x106 cells. Combinations of these top armors will be further tested in WP-001, a CAR T construct for solid tumors expressing CLDN18.2. Citation Format: Megha Sah, Liang He, Gundula Povysil, Cassandra Kontur, Ashley Chui, Stephen Pu, Annie Dyatel, Malavika Sabu, Alice HK Kwon, Xinchen Wang, David Phizicky. Single cell characterization and anti-tumor efficacy of single & multi-armored CAR Ts against CLDN18.2 [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 3181.

Cancer ResearchVol. 85(8_Supplement_1)
Bridgepoint (United States) (US)
Openalex Percentile: Top 88%
CAR-T cell therapy research
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.