Abstract:Since the approval of the first checkpoint inhibitor, ipilimumab (α-CTLA4), there continues to be enthusiasm around leveraging the immune system to eradicate solid tumors. However, unknowns about the cellular mechanism(s) that cause some patients to durably respond while others relapse, continue to stymie the field. To support ongoing efforts elucidating cell subsets responsible for differences in patient responses, we aim to evaluate the functional heterogeneity of T-cell subsets responding to checkpoint inhibitors in mouse models of solid tumors. Single-cell RNA-seq has enabled discoveries of transcriptionally defined cell subsets, but understanding their function require inference from transcription, flow-based validation biased by pre-selection of subsets with specific markers, or reliance on surrogate markers for cell-extrinsic functions (e.g., cytotoxicity). Thus, functional heterogeneity could be obscured. To address this, we leveraged a novel technology that converges live, cellular functions with transcriptomics at the resolution of single cell-cell interactions. This technology captures tens of thousands of individual cells or defined combinations of groups of cells in bio-compatible hydrogel compartments, enabling longitudinal imaging over days to weeks. Throughout this period, functions are characterized via bright-field or immunofluorescence readouts, including antibodies, cytokine capture beads, and dyes. Cells are then lysed within compartments to generate barcoded libraries for mRNA-sequencing and gene signatures from individual cells computationally deconvoluted. With this technology, we developed a novel assay that enables longitudinal, unbiased functional characterization of individual T cells mapped to single-cell transcriptomics. Single mouse primary T cells were encapsulated with one or more MC38 target cells and imaged over time. We validated combinations of different readouts can be measured from the same T-cell, including cell-counts for proliferation, CD8/3 expression to differentiate T cells from target cells, activation markers, and / or target cell death via Sytox. Thus, single T cells can be multi-functionally evaluated over time, enabling the study of attributes inaccessible at high scale with prior technologies, including kinetics of killing or the number of target cells killed by a single T-cell. For assays combining cytotoxicity and surface receptors, we generated barcoded libraries from functionally characterized T cells for sequencing and developed an analytical pipeline that deconvolutes mRNA from single T cells mapped to their multi-functional profiles, to identify novel pathways that regulate anti-tumor functions. We will leverage this workflow to functionally characterize single T cells in tumor-bearing mice treated with checkpoint inhibitors and map the data to underlying transcriptomics. In doing so, we aim to discover functional subsets, potentially with identical transcriptional profiles, to support hypothesis generation around mechanisms governing anti-tumor efficacy of checkpoint inhibitors. Citation Format: Jacqueline Yee, Praveer Sharma, Yunmin Li, Isaac Thomas, Pier F Gherardini, Teresa Ai, Mostafa Ronaghi, Gary Schroth, Matthew H Spitzer. Novel workflow for characterizing T-cell functional heterogeneity in response to checkpoint inhibitors [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Tumor Immunology and Immunotherapy; 2024 Oct 18-21; Boston, MA. Philadelphia (PA): AACR; Cancer Immunol Res 2024;12(10 Suppl) nr A025.

Understanding TCR T cell knockout behavior using interpretable machine learning

Cellular behavior analysis from live-cell imaging of TCR T cell-cancer cell interactions

Functional genomic landscape of T-cell mediated cytotoxicity.

Disease associated human TCR characterization by deep-learning framework TCR-DeepInsight

Abstract LB289: Single-cell CRISPR screens at scale to understand T cell biology

T-Cell-Based Platform for Functional Screening of T-Cell Receptors Identified in Single-Cell RNA Sequencing Data Sets of Tumor-Infiltrating T-Cells

Integrating T-cell Receptor and Transcriptome for Large-Scale Single-Cell Immune Profiling Analysis

Abstract 16: CD4+ and CD8+ TCRαβ-deficient bioluminescent reporter T cells for screening and characterization of neoantigen-specific TCRs

TCRpred: incorporating T-cell receptor repertoire for clinical outcome prediction

Abstract 8: Deep Immunomics pipeline for discovery and validation of novel cancer-specific T cell receptors

Prediction of tumor-reactive T cell receptors from scRNA-seq data for personalized T cell therapy

Transfer learning between preclinical models and human tumors identifies a conserved NK cell activation signature in anti-CTLA-4 responsive tumors

Perturbomics of tumor-infiltrating NK cells

Abstract A025: Novel workflow for characterizing T-cell functional heterogeneity in response to checkpoint inhibitors

Abstract 14: Identification of tumor-reactive T cell receptors through functional single cell interaction analyses for personalized T cell therapy

Abstract LB533: TEtor: Improved Machine Learning Approach Unveils the Clues of Cancer Lurking in Individual TCR Repertoire

TCRfinder: Improved TCR virtual screening for novel antigenic peptides with tailored language models

Abstract 5371: Multimodal CRISPR activation screens reveal new mechanisms to sensitize cancer cells to targeted T cell mediated elimination

TCR-H: explainable machine learning prediction of T-cell receptor epitope binding on unseen datasets

Targeting Precision in Cancer Immunotherapy: Naturally-Occurring Antigen-Specific TCR Discovery with Single-Cell Sequencing

TCR-L: an analysis tool for evaluating the association between the T-cell receptor repertoire and clinical phenotypes