Abstract PO-012: Spatial transcriptomic analysis of HPV-related and HPV-unrelated head and neck squamous cell carcinoma
Thomas F. Barrett,Dor Simkin,Alissa R. Greenwald,Anuraag Parikh,Hiram Gay,Anthony Apicelli,Douglas Adkins,Wade Thorstad,Jason T. Rich,Randal C. Paniello,Paul Zolkind,Patrik Pipkorn,Ryan S. Jackson,Rebecca Chernock,Itay Tirosh,Sidharth V. Puram
DOI: https://doi.org/10.1158/1557-3265.aacrahns23-po-012
IF: 13.801
2023-09-16
Clinical Cancer Research
Abstract:Background: Recent work characterizing head and neck squamous cell carcinoma (HNSCC) using single-cell RNA-sequencing (scRNA-seq) has highlighted intra- and inter-tumoral heterogeneity in both HPV-related and HPV-unrelated disease. These studies have identified previously undescribed gene expression programs associated with poor survival and adverse features. Unfortunately, these approaches necessarily forego the spatial localization of diverse cell types in the tumor microenvironment (TME). Spatial transcriptomics (ST) is a novel platform that allows for the unbiased detection of thousands of genes with near single-cell resolution, while retaining the physical, geographic source of these mRNA transcripts. Interactions between malignant cell populations and the surrounding TME influence tumor behavior, yet no study to date has characterized HNSCC with ST. Methods: We characterized 27 HNSCC patient tumors (19 HPV-unrelated and 8 HPV-related tumors) with the 10X Visium ST platform, 11 of which had paired scRNA-seq data. Cell phenotypes were identified by defining consensus meta-programs through a combined use of Leiden clustering and non-negative matrix factorization (NMF). Malignant regions were computationally defined through adaptation of previously employed inferred copy number alteration (CNA) algorithms. Relationships between cell types were characterized using distance-based metrics. Using consensus meta-program assignments and CNA scores, we defined 3 zones: tumor stroma, tumor-stroma interface (TSI), and tumor nests. Within each of these zones, we characterized enrichment for cell types and cell-to-cell interactions. Results: After quality control filtering, we retained 77,604 capture spots with a mean depth of 83,519 reads/spot and a median of 3,729 genes/spot for downstream analysis. In total, 13 consensus meta-programs were defined, including immune cell populations (e.g. T cells, macrophages), stromal cells (e.g. fibroblasts, endothelial cells), and multiple epithelial cell populations. Inferred CNA analysis suggested that these epithelial populations consisted of both normal epithelium and multiple malignant cell populations, including cells expressing markers typical of the previously described partial epithelial mesenchymal transition (p-EMT) gene signature as well as a hypoxia signature. The p-EMT signature was enriched at the TSI and the hypoxia signature was enriched in the center of the tumor nests in HPV-unrelated tumors, while HPV-related tumors had distinct gene expression at the TSI. Conclusions: Our study represents the first unbiased ST analysis of the HNSCC TME. Our results suggest that heterogeneity of intra-tumoral malignant cell states are consistent across multiple lesions and strongly associated with HPV etiology, particularly at the TSI. These findings suggest that HPV-related and HPV-unrelated tumors may have unique modes of local invasion. Citation Format: Thomas F. Barrett, Dor Simkin, Alissa R. Greenwald, Anuraag Parikh, Hiram Gay, Anthony Apicelli, Douglas Adkins, Wade Thorstad, Jason T. Rich, Randal C. Paniello, Paul Zolkind, Patrik Pipkorn, Ryan S. Jackson, Rebecca Chernock, Itay Tirosh, Sidharth V. Puram. Spatial transcriptomic analysis of HPV-related and HPV-unrelated head and neck squamous cell carcinoma [abstract]. In: Proceedings of the AACR-AHNS Head and Neck Cancer Conference: Innovating through Basic, Clinical, and Translational Research; 2023 Jul 7-8; Montreal, QC, Canada. Philadelphia (PA): AACR; Clin Cancer Res 2023;29(18_Suppl) nr PO-012.
oncology