Squamous cell carcinoma of the head and neck (HNSCC) is a complex disorder that contains multiple cell populations, making it is difficult to study the nature and cellular function and delineate relationship among subpopulations. The analysis of global gene expressions in single cell level has evolved at an astounding pace in the past few years, and now reaching a sophisticated level to solve heterogeneity in complex organs. To understanding the in vivo tumor microenvironment, a large scale (~10,000 cells) single cell gene expression assay can elucidate transitional states and delineate relationships among subpopulations. The goals of this proposal are to characterize a distinct population of HNSCC stromal progenitors, to explore cellular mechanisms by which these cells regulate tumor progression, and to ultimately translate the findings into clinical cancer therapies. Using genetic tools and SCC animal models, data has been generated to demonstrate that: 1) a unique postnatal Gli1+ stromal population is found in the craniofacial region, 2) Gli1+ cells behave as progenitors under homeostatic and disease conditions in vivo, 3) tumor progression is supported by stromal components that are elevated in HNSCC microenvironment and serve as a chemo- attractant for tumor invasion, and 4) 3D culture of primary tumor keratinocytes with cancer associated fibroblasts (CAFs) showed that the stratified growth, cell proliferation, and differentiation are comparable between co-cultures and their respective native tissues, while they largely differed in cultures without CAFs. We hypothesize that Gli1+ cells contain a distinct subpopulation of stromal progenitors in the craniofacial region, and that Gli1+ cells play a niche role in supporting HNSCC progression through remodeling epithelial morphogenesis. During this proposal, we will explore whether Gli1+ cells as a niche support cancer progression and metastasis through lineage tracing in Gli1CreERT2;TdTomato mice and diphtheria-mediated loss-of-function in Gli1CreERT2;DTA mice (Aim 1). Subsequently, this proposal will use transcriptomic and spatial molecular analysis in single-cell level combined with computational analysis and functional validation to identify unique in vivo stromal populations in mouse and human HNSCC and elucidate their molecular and pathway signatures. We will validate these unique cancer stromal subpopulations and investigate their features, regulatory mechanism, function, and spatial and molecular characteristics. Based on the findings, by 3D organoid modeling with novel vascularized organ-on-a chip technique, newly identified tumor specific stromal progenitor populations will be utilized to generate 3D vascularized SCC organoids and discover novel therapeutic avenues for HNSCC management (Aim 2). Successful completion of the proposal will advance our understanding of the nature of in vivo stromal effects in tumor microenvironment and will develop new treatments for HNSCC.