ABSTRACT: This dataset contains CODEX multiplexed tissue imaging data generated from a human head and neck squamous cell carcinoma (HNSCC) tissue microarray. The dataset was produced as part of a study investigating spatial heterogeneity in HNSCC, with emphasis on malignant epithelial states and their interactions with stromal and immune microenvironmental compartments. The deposited data include the full TMA image, individual OME-TIFF images for quality-control-retained laryngeal HPV-negative tumor cores, cell and nuclei segmentation annotations generated using InstanSeg, tumor-boundary annotations, and cell-level quantitative tables.
The original TMA contained 44 cores, of which 28 were retained for downstream analysis after exclusion of cores with artifacts, low staining quality, sparse tissue, or related quality-control issues. For the retained cores, this submission provides per-cell morphology measurements, marker intensity values, and per-marker Nimbus scores for a filtered panel of 31 high-quality CODEX markers. These data support downstream analyses of epithelial, stromal, endothelial, myeloid, and lymphoid compartments, including spatially resolved assessment of tumor architecture and microenvironmental organization.
All files are de-identified. Core identifiers are consistent across image files, segmentation masks, tumor-boundary annotations, marker-intensity tables, Nimbus-score tables, and metadata files.
