Predicting gastric cancer outcome from resected lymph node histopathology images using deep learning

Area of Science:

• Oncology
• Digital Pathology
• Artificial Intelligence

Background:

• N-staging is crucial for cancer prognosis and treatment, relying on manual lymph node analysis by pathologists.
• Current methods for assessing metastatic lymph nodes (MLNs) can be time-consuming and may not fully capture prognostic variability.
• Patient outcomes can differ significantly even within the same N stage.

Purpose of the Study:

• To develop and validate a deep learning framework for analyzing whole-slide images (WSIs) of lymph nodes.
• To identify lymph node and tumor regions for calculating the tumor-area-to-MLN-area ratio (T/MLN).
• To evaluate T/MLN as a novel prognostic factor in cancer.

Main Methods:

• A deep learning model was trained to analyze lymph node WSIs.
• The model identified lymph nodes and tumor regions to calculate the T/MLN ratio.
• Model performance was assessed against experienced pathologists and validated on independent gastric cancer cohorts.

Main Results:

• The deep learning model demonstrated tumor detection performance comparable to experienced pathologists.
• The model achieved similar performance on two independent gastric cancer validation cohorts.
• The study identified T/MLN as an interpretable and independent prognostic factor.

Conclusions:

• Deep learning models can effectively assist pathologists in detecting lymph node metastases.
• The T/MLN ratio, derived from AI analysis, offers a new, interpretable prognostic factor for cancer patients.
• This approach facilitates the exploration of novel prognostic indicators that are challenging for manual calculation.