Abstract 73: Spatial profiling of pediatric rhabdomyosarcoma to elucidate their immunosuppressive tumor microenvironment

Abstract

Abstract Background: Pediatric rhabdomyosarcoma (RMS) is a mesenchymal soft tissue malignancy. Despite intensive chemotherapy regimens, there is currently a gap in improving treatment outcomes for RMS. Chemotherapy can modulate the tumor microenvironment (TME) by altering the immune cell composition or their functions. However, the impact of chemotherapy in pediatric RMS TME is unclear. RMS tumors are also considered as “cold” tumors due to low immune infiltration, which could be a determinant of why immunotherapy is unlikely to work. Additional investigations are required to determine the reasons of why RMS is “cold” and whether a combined immunotherapy and chemotherapy approach should be administered for this hard-to-cure cancer. Methods We conducted for the first time spatial GeoMx whole transcriptome atlas and 51-plex Akoya CODEX profiling for a tissue microarray (TMA) from a cohort of pediatric RMS patients (n = 12) with pre-chemotherapy (n = 5) and post-chemotherapy (n = 5) samples. Regions of interest (ROIs) were selected based on tumor areas with immune cell infiltration, with the lymphocytes, tumor cells, and nuclei visualized using fluorescent morphological markers CD45, Desmin, and DAPI. ROIs were segmented into tumor stroma compartments and immune cell compartments. Computational analyses were conducted on the transcriptomics data using R package standR including quality control, normalization, differential expression analysis, gene set enrichment analysis, and cell type deconvolution. For CODEX data, we preformed cell segmentation, followed by further quality control, cell phenotyping, cellular neighborhood and cellular interaction analysis. Results: From differential expression analysis of the immune compartment, we have identified type I interferon-related genes (DTX3L, ADAP2, RNF135) and reactive oxygen species (ROS)-induced gene (NCF1) are upregulated, and an antioxidant enzyme gene (PRDX2) is downregulated in post-chemotherapy samples compared to pre-chemotherapy samples. Gene set enrichment analysis of the immune compartment revealed an enriched antigen presentation pathway in post-chemotherapy samples. In the tumor stroma compartment, we identified increased TGF-β binding pathway in post-chemotherapy samples. Cellular deconvolution revealed a higher proportion of endothelial cells in post-chemotherapy samples, which is validated via CODEX imaging analysis. Cellular neighborhoods and cell-cell interaction analysis for CODEX data also reveal how chemotherapy affect functional units and cellular interactions in RMS TME with spatial context preserved. Conclusion: By applying spatial transcriptomic and multiplexed imaging to RMS samples, we have revealed key genes, pathways and cell types involved in chemotherapy-induced alterations in the TME. This will provide a new perspective for developing novel chemoimmunotherapy for RMS patients. Citation Format: Cui Tu, Chin Wee Tan, Ning Liu, Arutha Kulasinghe, Fernando Souza-Fonseca-Guimaraes. Spatial profiling of pediatric rhabdomyosarcoma to elucidate their immunosuppressive tumor microenvironment [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 73.