Related Experiment Video
Updated: May 5, 2026
Multiplexed Immunofluorescence Analysis and Quantification of Intratumoral PD-1+ Tim-3+ CD8+ T Cells
Published on: February 8, 2018
AI-driven spatial analysis of tumor-infiltrating lymphocytes predicts chemo-immunotherapy response in triple-negative
Hai Liang1, Hejie Zhu1, Liwei Wang1
1Shanghai Qilu Pharmaceutical R&D Center Ltd., Shanghai, China.
Introduction:
Triple-negative breast cancer (TNBC) is an aggressive subtype with limited therapeutic options due to lack of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) expression. While neoadjuvant immunotherapy shows promise, the impact of spatial organization of tertiary lymphoid structures (TLS) and tumor-infiltrating lymphocytes (TILs) on treatment efficacy remains incompletely characterized. This exploratory study aims to establish an AI-based quantitative framework for analyzing tumor-immune spatial interactions and develop a predictive model for chemo-immunotherapy response.
Methods:
We developed an integrated AI pipeline for automated analysis of hematoxylin-eosin (HE)-stained breast cancer samples. The framework incorporates: 1) TLS detection and classification, 2) TIL quantification, and 3) spatial relationship mapping between lymphocytes and tumor cells. Biomarkers associated with Miller-Payne (MP) response grades were identified using multivariate statistics, enabling construction of a random forest prognostic model.
Results:
The TLS recognition model demonstrated substantial agreement with pathologists (κ=0.73), while the TIL classifier achieved 0.92 accuracy. Analysis of 32 HE images from triple-negative breast cancer (TNBC) cases treated with neoadjuvant chemo-immunotherapy revealed significant associations: stromal lymphocyte density and percentage positively correlated with MP grades (p<0.05); average cell counts in 2-cell and 3-cell lymphocyte aggregates showed no significant correlations; shorter mean minimum distances between lymphocytes and tumor cells within 10-30 μm radius range were inversely associated with MP grades (p<0.05). The spatial-feature-based prediction model achieved an AUC of 0.81 (95% CI: 0.76-0.86).
Conclusions:
This study established an AI-driven HE analysis pipeline that precisely quantified spatial determinants of tumor-immune interactions. Lymphocyte spatial organization, particularly proximity to tumor cells and lymphocyte aggregation patterns, serves as a critical predictor of chemo-immunotherapy response beyond density metrics. The validated MP-grade prediction model demonstrates translational potential for clinical decision-making in TNBC management, although external validation in larger cohorts is warranted.
Insights
An AI framework accurately predicts triple-negative breast cancer (TNBC) treatment response by analyzing immune cell spatial organization in tumor samples. This approach enhances understanding of tumor-immune interactions for better therapeutic strategies.
Area of Science:
- Oncology
- Computational Biology
- Immunology
Background:
- Triple-negative breast cancer (TNBC) lacks targeted therapies, necessitating novel treatment strategies like neoadjuvant chemo-immunotherapy.
- The spatial arrangement of tertiary lymphoid structures (TLS) and tumor-infiltrating lymphocytes (TILs) significantly influences treatment efficacy but is not fully understood.
- Predicting response to chemo-immunotherapy in TNBC remains a challenge.
Purpose of the Study:
- To develop an AI-based quantitative framework for analyzing tumor-immune spatial interactions in TNBC.
- To create a predictive model for chemo-immunotherapy response based on spatial features.
- To investigate the association between TLS, TILs, and treatment outcomes.
Main Methods:
- An integrated AI pipeline was developed for automated analysis of hematoxylin-eosin (HE)-stained breast cancer samples.
- The pipeline included TLS detection/classification, TIL quantification, and spatial relationship mapping.
- A random forest prognostic model was constructed using multivariate statistics to identify biomarkers associated with Miller-Payne (MP) response grades.
Main Results:
- The AI models achieved high accuracy in TLS recognition (κ=0.73) and TIL classification (0.92).
- Analysis of 32 TNBC cases revealed that stromal lymphocyte density and proximity to tumor cells correlated with MP grades (p<0.05).
- The spatial-feature-based prediction model demonstrated strong predictive performance with an AUC of 0.81.
Conclusions:
- An AI-driven HE analysis pipeline effectively quantifies spatial determinants of tumor-immune interactions.
- Lymphocyte spatial organization is a critical predictor of chemo-immunotherapy response in TNBC.
- The validated MP-grade prediction model shows potential for clinical decision-making in TNBC management.
More Related Videos
07:36Tumor Transplantation for Assessing the Dynamics of Tumor-Infiltrating CD8+ T Cells in Mice
Published on: June 12, 2021
06:05Author Spotlight: Multiplex Immunofluorescence Combined with Spatial Image Analysis for the Clinical and Biological Assessment of the Tumor Microenvironment
Published on: June 2, 2023
