AI assessment of tumor-infiltrating lymphocytes on routine H&E-slides as a predictor of response to neoadjuvant therapy in breast cancer-a real-world study.

This study evaluates tumor-infiltrating lymphocytes (TILs) as predictive and prognostic biomarkers in triple-negative (TNBC) and HER2-positive breast cancer patients treated with neoadjuvant chemotherapy using an AI-based deep learning approach.

AI analyzed TILs on hematoxylin and eosin-stained slides from multi-institutional cohorts, showing a strong correlation with pathologists’ assessments.

Key findings include:

• Higher AI-quantified stromal and intraepithelial TIL levels were independently associated with better pathological complete response in both subtypes.
• Elevated TILs correlated with improved disease-free and overall survival in TNBC but not in HER2-positive patients.

The findings support incorporating AI-driven TIL quantification into digital pathology workflows to enhance risk stratification and treatment response prediction.