ADC target profiling in NSCLC: Generalizable AI separates TROP-2 and cMET phenotypes

Key Points

• This research aims to evaluate how generalizable AI can distinguish between TROP-2 and cMET phenotypes in non-small cell lung cancer (NSCLC).
• Utilized foundation model-based scoring for biomarker quantification.
• Analyzed TME phenotypes to categorize as TROP-2-high immune-deserted and cMET-high immune-active.
• Evaluated therapeutic implications of identified phenotypes.
• AI effectively distinguished TROP-2 and cMET phenotypes in NSCLC.
• Identified therapeutic combinations involving ADCs and immunotherapies or kinase inhibitors.
• Results demonstrate significant variability in immune activity between the two phenotypes.