Single-Cell and Bulk RNA Sequencing Reveal Tumor Cell Characteristics and Communication Features of Primary and...

Lymph node metastasis (LNM) is strongly associated with poor prognosis in hypopharyngeal squamous carcinoma (HPSCC). Identifying key drivers of LNM and potential therapeutic targets in HPSCC is therefore essential for the early detection of high-risk patients and for informing personalized treatment strategies.

Single-cell RNA sequencing data were used to characterize malignant epithelial cells (maECs) in HPSCC primary tumors (PT) and LNM, as well as differences in cell-to-cell communication. Concurrently, combined with bulk RNA sequencing data, a ligand receptor pairs (LRs) model was developed to predict the prognosis of HPSCC patients.

PT and LNM maECs have different gene expression characteristics, with genes involved in interferon signaling and TGF-β response pathways enriched in LNM maECs, suggesting potential immunosuppressive reprogramming. Cell communication analysis revealed distinct interactions and signaling features in PT and LNM microenvironments. Subsequently, a 4-LRs model was constructed to stratify HPSCC patients into low-or high-risk groups, with the high-risk group demonstrating significantly worse overall survival (OS) outcomes compared with the low-risk group in the training (p < 0.0001), testing (p = 0.0021), and entire (p < 0.0001) cohorts. Receiver operating characteristic curves showed that this risk model can effectively predict the 1-, 3-, and 5-year OS of HPSCC patients. Notably, the risk score effectively discriminated LNM status (area under the curve [AUC] = 0.927) among HPSCC patients, highlighting its potential as a HPSCC metastasis prediction tool.

These results provide biomarkers of LNM maECs as well as potential mechanisms of HPSCC metastasis, which may help with the precision treatment, diagnosis, and prognosis of HPSCC.