Back

Age-associated transcriptional programs in tumor-associated macrophages are associated with tumor progression in HPV-negative oral cancer

Gujar, H.; Hanache, S.; Aggarwal, S.

2026-10-08 immunology
10.64898/2026.10.02.756061 bioRxiv
Show abstract

Aging and tobacco exposure are major determinants of risk and outcome in HPV-negative oral cavity squamous cell carcinoma (OSCC), yet the cellular programs linking host factors to tumor immunity remain poorly defined. We reprocessed and reanalyzed previously published single-cell RNA-seq data from 19 HPV-negative OSCC tumors and we identified a robust age- and smoking-associated expansion of SPP1high tumor-associated macrophages (TAMs), accompanied by a shift toward a myeloid-dominant and lymphocyte-reduced tumor microenvironment. SPP1high TAMs exhibit a distinct transcriptional state enriched for extracellular-matrix remodeling and metabolic programs and reduced antigen-presentation pathways. They also express inflammatory chemokines such as CXCL2 and CXCL3 and down-regulate the cytotoxic mediator TNFSF10, suggesting a broadly protumor TAM phenotype. Ligand-receptor modeling suggests that SPP1high TAMs act as key communication hubs, with dominant SPP1-CD44 signaling to T-cell subsets. Transcription-factor analysis further implicates a POU2F1/POU2F2/SP1 module in maintaining this state. In the TCGA HNSCC cohort, the association between SPP1 expression and overall survival did not reach statistical significance after adjustment for age and tumor purity. These findings demonstrate that aging and tobacco exposure converge to expand an SPP1high TAM population that remodels the extracellular matrix and suppresses antitumor immunity, highlighting the SPP1-CD44 axis and its upstream regulators as potential therapeutic targets in high-risk OSCC.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.