Case Studies

Comprehensive Assessment of the Inflammatory Tumor Microenvironment

Combat or Surveillance? Why morphology and spatial relationships are key to describe the functional state of the immune system.


Understanding the cellular and molecular mechanisms of the inflammatory tumor microenvironment is becoming increasingly important as cancer immunotherapies are the focus of almost all pharmaceutical companies.

This study presents a novel way of comprehensively characterizing the immune cell-tumor cell interaction, with implications for biomarker discovery and increased disease understanding.


  • Proof of concept study using automated image analysis to assess immune cell subtypes (T cells, cytotoxic T cells and NK cells) and their distribution in invasive breast cancer (n=33 cases)
  • Development of a scoring system that describes and quantifies immune cell interaction with tumor cells and goes beyond manual assessment of cell counts
As a long-term benefit of this approach, we envision important contributions to further clarifying the role of inflammatory host reactions in breast cancer by adding comprehensive contextual information to conventional scoring approaches, and supporting further standardization of histopathological immune cell evaluation.


  • Consistency of readouts: Robust data and excellent correlation (CCC>0.9 for T cells) with manual assessment
  • Improved information retrieval from tissue:
    • Approach enables systematic access to data that cannot be captured manually on a routine basis
    • Analysis of spatial relationships between immune cells and tumor cells allows distinguishing between surveillance and combat – different functional states of the immune system that cannot always be discriminated with overall cell counts but likely have prognostic or predictive value
  • Better data: The approach enables a systematic evaluation of tumor heterogeneity


  • The study indicates that advanced automated image analysis can provide a systematic, comprehensive and quantitative assessment of the inflammatory tumor microenvironment
  • The approach helps to increase our understanding of the role of immune cells in cancer, and helps in developing diagnostic tests:
    • The immune cell signatures can serve as prognostic biomarkers ( Immunoscore)
    • Immune cell profiles might evolve into predictive, stratifying biomarkers for immunotherapy drug candidates that require an already active/primed immune system
  • Automated image analysis can help in the development of scoring guidelines in the developing field of immunoprofiling which currently lacks standardization

Additional Information

The findings have been published in the Journal of Pathology: Krüger et al.; J Pathol. 2013;229:569-578 (

Case Study Visuals

Automatic Quantification

Automatic quantification of CD3-positive T cells (DAB+), CD3-Perforin positive cytotoxic T cells (DAB+ PRD+), and Peroforin-positive CD3-negative Natural Killer cells (PRD+). Tumor cells were classified based on morphological criteria. All other cells are marked in black.

A) Original Image at 40x (250µmx250µm). B) Segmentation of cell nuclei. C) Classification of cells.
Scoring System Describing Immune Cell Interaction

Scoring system to describe immune cell interaction with the tumor cells as surveillance (green) or combat (red-pink). Single points of interaction appear yellow (left image). Blue regions denote clusters of immune cells without contact to tumor cells.