Abstract 5885: A pathway-centric approach to characterising tumour heterogeneity and cell diversity across multiple cancer types

Abstract

Abstract Biological tissues are composed of heterogeneous populations intricately organized in 3D architectures, yet cell type composition and spatial organization remain largely unknown for most tissue types. Single-cell sequencing analysis provides a systematic and quantitative approach to identify cell types and determine their composition in tissues. However, the spatial organization of cells and cell-cell interactions that are critical for tissue function are lost when cells are dissociated from tissue. This is especially important when working with the complex microenvironment of a tumour where such information is needed to fully understand patient prognosis. Complex diseases such as cancer can be better understood from the perspective of dysregulated pathways, rather than as a disease resulting from alterations of individual genes. Recently, spatially resolved, single-cell imaging platforms have provided a necessary solution to bridge the spatial information gap evident in previous transcriptomic methodologies. Vizgen’s MERSCOPE platform, built on multiplexed error robust in situ hybridisation (MERFISH) technology, enables the direct profiling of the spatial organisation of intact tissue with subcellular resolution. Here, we present a Pan-Cancer approach to characterise various cancers in human clinical samples with MERSCOPE. Using a 500-gene panel that includes the canonical signalling pathways of cancer, cancer type-specific genes, key immune genes, proto-oncogenes, and tumour-suppressor genes we demonstrate MERSCOPE’s ability to spatially profile gene expression across multiple tumour samples, including breast, colon, prostate, ovarian, lung, and skin cancer. To confirm the specificity and sensitivity of our pathway-focused gene panel, we assessed how tumour cell clusters in each dataset expressed different cancer type-specific markers. As expected, we found specific cancer types enriched in corresponding marker genes. To further evaluate how individual cell types in each tumour are dysregulated by cancer, we compared non-tumour cell clusters to similarly annotated clusters derived from single-cell RNA-sequencing data of normal tissue. Notably, the non-tumour cells in these datasets exhibit higher expression of stress and inflammation pathways, while immune cell clusters demonstrate higher immune activity and higher immune exhaustion compared to those in normal tissue. These results demonstrate the power of the MERSCOPE platform to generate individualised, accurate cell atlases from patient-derived tumours and to enable further insights into the relationship between genomic profiles, dysregulated pathways, and disease phenotype. Such network-centric approaches are critical for identifying genotypic causes of diseases, classifying disease subtypes, and identifying drug targets. Citation Format: Leiam Colbert, Benjamin Patterson, Cheng Yi Chen, Nicolas Fernandez, Jiang He. A pathway-centric approach to characterising tumour heterogeneity and cell diversity across multiple cancer types [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5885.