Kirolous S. Hanna1,2, Fatima Dhalla1, Andreas Tarcevski1, Mary E. Deadman1, Pietro Zara2, Bernard Liem2, Rhona Taberham3, Antonios Kourliouros3, Dionisios Stavroulias3, Adam E. Handel2, M. Isabel Leite2, Georg A. Hollander1
1Department of Paediatrics and the Institute of Developmental and Regenerative Medicine, University of Oxford, Oxford, UK;
2Nuffield Department of Clinical Neurosciences, University of Oxford, Oxford, UK;
3Department of Cardiothoracic Surgery, Oxford University Hospitals, Oxford, UK
Correspondence to: Kirolous S. Hanna, MD. Department of Paediatrics and the Institute of Developmental and Regenerative Medicine, University of Oxford, Level 2, IDRM, IMS-Tetsuya Nakamura Building, Old Road Campus, Roosevelt Dr, Headington, Oxford OX3 7TY, UK; Nuffield Department of Clinical Neurosciences, University of Oxford, Oxford, UK. Email: kirolous.hanna@ndcn.ox.ac.uk.
Background: Thymomas are malignancies originating from the thymus, the primary lymphoid organ crucial for T cell development and selection. Normal thymopoiesis relies on physical and functional interactions between T-cell precursors and thymic stromal cells. Abnormalities in the frequency of T-cell developmental stages within dissociated thymoma tissue suggest that the tumour microenvironment (TME) substantially impacts thymopoiesis. Yet a detailed understanding of the precise cellular and spatial irregularities of thymoma is incomplete. This study aims to map the thymoma TME to analyse the cellular organization and interactions associated with T-cell development in the tissue.
Methods: Multiplexed immunofluorescence imaging detecting 28 distinct epitopes was employed to identify lymphoid and stromal cells in type B2 thymoma tissue and healthy human thymus of young and adult individuals. A computational pipeline for spatial profiling of microenvironments was developed that includes the accurate segmentation of complex-shaped stromal cells and the quantitative and qualitative analysis of cellular neighbourhoods and interactions. This pipeline was designed and validated to allow comparison of extracted spatial proteomic data across multiple tissue groups.
Results: Simultaneous characterisation of 57 cell-types and states was achieved for T-cells at various developmental stages, epithelial cells, hematopoietic antigen-presenting cells, and non-epithelial stromal cells. Detailed analyses of healthy tissue have been conducted on the cell-type composition and density, parenchymal distribution, tissue organization, and cell-cell interactions relevant to normal thymopoiesis. The density of T-cells as well as composition and protein expression of stromal populations in the TME substantially deviates from that observed in healthy tissue, particularly in the cortical compartments. Novel tissue niches have been identified within thymic histological compartments for both healthy and thymoma tissue, where well-defined regions of healthy thymi are either absent or reduced, in the TME, and where abnormalities in cellular interactions have been observed.
Conclusions: Our study established analytical methods allowing the mapping of type B2 thymoma TME at single-cell resolution as well as for high-throughput and in-depth analysis of spatial proteomic datasets. This investigation reveals region-specific spatial irregularities and abnormal cellular interactions within novel functional niches in the TME, contributing to an understanding of abnormalities of T-cell development in thymomas.
Keywords: Thymic epithelial tumors (TETs); thymopoiesis; Hyperplex; PhenoCycler; proteomics
