Related Experiment Video
Updated: Jan 12, 2026
Patient-Derived Tumor Explants As a "Live" Preclinical Platform for Predicting Drug Resistance in Patients
Published on: February 7, 2021
AI-powered spatial cell phenomics enhances risk stratification in non-small cell lung cancer
Simon Schallenberg1, Gabriel Dernbach1,2, Sharon Ruane2
1Institute of Pathology, Charité - Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin and Berlin Institute of Health, Berlin, Germany.
Artificial intelligence and multiplex imaging enhance risk stratification for non-small cell lung cancer (NSCLC) patients. This approach identifies immune cell patterns in the tumor microenvironment, improving therapy selection and potentially benefiting high-risk patients.
Area of Science:
- Computational pathology
- Cancer immunology
- Artificial intelligence in oncology
Background:
- Accurate risk stratification is crucial for selecting optimal therapies in non-small cell lung cancer (NSCLC).
- The tumor microenvironment's complex cellular interactions significantly influence cancer progression and patient outcomes.
- Current staging methods may not fully capture the prognostic information embedded within the tumor microenvironment.
Purpose of the Study:
- To develop an artificial intelligence (AI)-powered spatial cellomics approach for NSCLC risk stratification.
- To characterize complex cellular relationships and identify prognostic cell niches within the tumor microenvironment.
- To improve risk stratification and treatment selection by integrating AI-driven imaging analysis with conventional staging.
Main Methods:
- Utilized a retrospective cohort of 1168 NSCLC patients from two German cancer centers.
- Combined histology with multiplex immunofluorescence imaging to analyze 43 distinct cell phenotypes.
- Employed multimodal machine learning to model spatial cell relationships and identify survival-associated cell niches.
Main Results:
- The AI model identified specific cell niches within the tumor microenvironment linked to patient survival.
- Risk stratification accuracy improved by 14% for lung adenocarcinoma and 47% for squamous cell carcinoma when combining niche patterns with conventional staging.
- Complex immune cell niche patterns effectively identified high-risk NSCLC patients who may benefit from adjuvant therapy.
Conclusions:
- AI-powered spatial cellomics and multiplex imaging offer a powerful tool for understanding the tumor microenvironment in NSCLC.
- The identified immune cell niche patterns can significantly enhance prognostic accuracy and guide treatment decisions.
- This approach holds promise for identifying undertreated high-risk patients, optimizing therapy selection, and improving outcomes in NSCLC.
