Deep Learning-Based Spatial Profiling Identifies MET-Associated Immune Exclusion and Therapeutic Vulnerability Underlying Racial Disparities in Head and Neck Cancer

Higher predicted spatial MET expression is linked to reduced CD8 T cell infiltration and improved pembrolizumab plus cabozantinib outcomes, especially in Black/African American head and neck squamous cell carcinoma patients.

• Why it matters: Black/African American patients with HNSCC have more aggressive disease and poorer survival, but the tumor immune microenvironment features driving these racial disparities are not well defined.
• What they did: Deep learning-based spatial profiling compared White- versus Black/African American–derived HNSCC tumors, then tested MET expression associations with CD8 infiltration across independent cohorts and with response in a phase 2 pembrolizumab plus cabozantinib trial cohort.
• The result: Across cohorts, higher MET expression is consistently associated with reduced CD8 T cell infiltration, and in the phase 2 trial higher predicted spatial MET expression is strongly associated with treatment response and improved clinical outcomes, with a particularly pronounced effect among Black/African American patients.