Related Experiment Video
Updated: Jun 13, 2025
07:32
Author Spotlight: Investigating Immune Cell Dynamics in the Tumor Microenvironment — Challenges and Innovations in Cancer Prognosis
Published on: April 12, 2024
1.2K
Hepatocellular Carcinoma Immune Microenvironment Analysis: A Comprehensive Assessment with Computational and
Caner Ercan1, Salvatore Lorenzo Renne2,3, Luca Di Tommaso2,3
1Institute of Medical Genetics and Pathology, University Hospital Basel, University of Basel, Basel, Switzerland.
Summary
Deep learning accurately analyzes the hepatocellular carcinoma (HCC) tumor immune microenvironment (TIME). This computational pathology tool aids in prognosis and understanding immune cell distribution, with needle biopsies showing 75% accuracy in predicting immunophenotypes.
Area of Science:
- Computational pathology
- Immunohistochemistry
- Deep learning in oncology
Background:
- The tumor immune microenvironment (TIME) in hepatocellular carcinoma (HCC) has spatial variability and unclear clinical relevance.
- Understanding immune cell infiltration is crucial for HCC prognosis and treatment strategies.
Purpose of the Study:
- To develop a deep learning (DL)-based image analysis model for spatial analysis of immune cell biomarkers in HCC.
- To microscopically evaluate the distribution of immune infiltration within the HCC TIME.
Main Methods:
- Histological classification of 92 HCC resections and 51 needle biopsies into inflamed, immune-excluded, and immune-desert immunophenotypes.
- Design of a multistage DL algorithm (IHC-TIME) to detect and localize immune cells on immunohistochemistry (IHC)-stained slides.
- Training of models for immune cell detection (98% accuracy) and tumor-stroma segmentation (91% accuracy).
Main Results:
- Patients with inflamed HCC tumors demonstrated improved recurrence-free survival compared to those with immune-excluded or immune-desert tumors.
- Needle biopsies achieved 75% accuracy in representing the immunophenotypes of the primary tumor.
- An automated algorithm for immunophenotype definition based on IHC-TIME analysis achieved 80% accuracy.
Conclusions:
- A DL-based tool accurately quantifies immune cells on HCC IHC slides, enabling microscopic TIME classification for patient prognostication.
- Needle biopsies offer valuable insights for TIME-related prognostic prediction, despite limitations.
- This computational pathology tool offers a novel approach to studying the HCC TIME.
