ADC target profiling in NSCLC: Generalizable AI separates TROP-2 and cMET phenotypes
Purpose:
Antibody–drug conjugates (ADC) targeting trophoblast cell surface antigen 2 (TROP-2) and cMET are entering clinical trials in non–small cell lung cancer (NSCLC). Their translation depends on reliable biomarker assessment, a task still dominated by subjective visual scoring and inconsistent reproducibility.
Experimental Design:
We built a modular Artificial Intelligence (AI) pipeline that detects cells, classifies carcinoma cells, and quantifies membranous and cytoplasmic expression. A membranous scorer trained on TROP-2 was applied zero-shot to cMET, human epidermal growth factor receptor 2 (HER2), and PD-L1. For TROP-2, cMET, and HER2, expression was quantified using the H-score, defined as (1× % weakly positive cells) + (2× % moderately positive cells) + (3× % strongly positive cells), with negative cells excluded (range, 0–300), and categorized as negative (0–50), weakly positive (50–100), moderately positive (100–200), or strongly positive (200–300). PD-L1 expression was assessed using both the H-score and the tumor proportion score (TPS), defined as the percentage of viable tumor cells showing membranous PD-L1 staining relative to all viable tumor cells, multiplied by 100. The analysis covered 1,142 resected NSCLCs, integrating expression maps with clinicopathologic, molecular, and tumor microenvironment (TME) features.
Results:
The AI scorer recapitulated pathologist annotations with near-perfect correlation [Pearsons's correlation (r) = 0.98–0.99, Spearman's rho (ρ) = 97–98, Kendall's tau (τ) = 0.88–0.89, Lin's concordance correlation coefficient (CCC) = 0.97–0.98] for TROP-2 and generalized to other markers [cMET r = 0.99, ρ = 0.96, τ = 0.91, CCC = 0.96; HER2 r = 0.93, ρ = 0.72, τ = 0.60, CCC = 0.92; and PD-L1 (TPS) r = 0.85, ρ = 0.84, τ = 0.68, CCC = 0.82/(H-score) r = 0.87, ρ = 0.86, τ = 0.70, CCC = 0.85]. Its agreement with six pathologists matched interobserver variability (0.86–0.96). Expression maps revealed contrasting spatial and cellular patterns: TROP-2 dominated lung squamous cell carcinoma [LUSC; mean H-Scores 141.3/103.2 vs. 74.5/45.7 in lung adenocarcinoma (LUAD) for membrane/cytoplasm] and marked immune-deserted tumors. cMET prevailed in LUAD (mean 50.4 vs. 20.6 in LUSC), colocalized with fibroblast-rich, immune-active TME, and KRAS mutations.
Conclusions:
Foundation model–based scoring produces expert-level, scalable biomarker quantification. The resulting TME phenotypes—TROP-2–high immune-deserted versus cMET-high, immune-active—reveal therapeutic implications for combining ADCs with immunotherapies or kinase inhibitors.
