Abstract PO-078: Single-cell transcriptomic analysis of highly-multiplexed IHC images via antigen mapping

Abstract

Abstract Recently developed technologies for digital imaging and highly multiplexed immunohistochemistry (mIHC) are enabling complex quantitative descriptions of tissue architecture. Imaging mass cytometry, multiplexed ion beam imaging, and co-detection by indexing can be used to profile the expression of dozens of proteins in a tissue section with single-cell resolution. Despite this progress, the amount of cell types and states that can be simultaneously identified by mIHC in a tissue section is still limited. Standard methods for identifying cell populations cluster cells according to expression similarities of the profiled antigens. These clusters are then manually annotated using previous knowledge of cell population markers. However, this is generally a partial, subjective, and biased process. Clusters often differ little in their antigenic profile and the interpretation of those differences is unclear. Moreover, the design of comprehensive antibody panels that include specific markers for every cell type and state present in the tissue is usually unfeasible. To overcome these limitations and increase the level of annotation of mIHC data, we propose an approach for enriching mIHC slides with single-cell RNA-seq data. Currently available single-cell RNA-seq technologies profile the expression of thousands of genes in each cell, allowing researchers to finely classify cells based on their gene expression profile. Some of the most recent approaches, like CITE-seq, REAP-seq, and Ab-seq, allow augmenting single-cell transcriptomes with concurrent protein measurements by staining dissociated cells with oligo-tagged antibodies. These approaches can be therefore used to determine the quantitative relation between gene and antigen expression levels. Here, we build upon CITE-seq and computational methods for the integration of single-cell omics data to identify and annotate cell populations in mIHC images in an automated way based on single-cell gene expression data from the same tissue. Our method, Spatially-resolved Transcriptomics via Epitope Anchoring (STvEA), performs transcriptome-guided annotation of mIHC datasets. It increases the phenotypic resolution of histological analyses by enabling annotation of subtle cell populations, spatial patterns of transcription, and interactions between cell types. We use the murine spleen as a test system to benchmark the stability and performance of STvEA and demonstrate its utility by using it to annotate published CODEX and CyTOF datasets of this organ. Citation Format: Kiya W. Govek, Emma C. Troisi, Zhen Miao, Steven Woodhouse, Pablo G. Camara. Single-cell transcriptomic analysis of highly-multiplexed IHC images via antigen mapping [abstract]. In: Proceedings of the AACR Virtual Special Conference on Tumor Heterogeneity: From Single Cells to Clinical Impact; 2020 Sep 17-18. Philadelphia (PA): AACR; Cancer Res 2020;80(21 Suppl):Abstract nr PO-078.