Boquet I, Kassambara A, Lui A, Tanner A, Latil M, Lovera Y, Arnoux F, Hermitte F, Galon J, Catteau A. Comparison of Immune Response Assessment in Colon Cancer by Immunoscore (Automated Digital Pathology) and Pathologist Visual Scoring. Cancers. 2022. Download the PDF
Summary of the Study
Accurate assessment of tumor immune infiltration is crucial for predicting colon cancer prognosis. This study compared Immunoscore (IS), a standardized and automated digital pathology assay, with manual immune scoring performed by pathologists.
- 50 early-stage colon cancer specimens were evaluated for CD3+ and CD8+ T-cell density
- Four expert pathologists scored immune infiltration as High/Low, pre- and post-training
- Their assessments were compared to the reference Immunoscore
Key Findings
- 48% disagreement between pathologists and Immunoscore
- Inter-pathologist agreement was low (Kappa: 0.34 before training; 0.57 after)
- Training had limited impact on improving accuracy
- Immunoscore delivered standardized and reproducible results, outperforming manual scoring
In this study, Alboukadel Kassambara performed all statistical analyses and data visualizations, supporting the conclusion that Immunoscore provides a more robust and reproducible evaluation of immune response in colon cancer than manual visual assessment.
Citation
Publication: In Cancers
Date: February 24, 2022
Type: Journal Article
PDF: Download the PDF
Scientific Contributions
Here are more scientific abstracts authored or co-authored by Alboukadel Kassambara. These contributions span computational biology, bioinformatics, biostatistics, machine learning, and multi-omics, with a focus on immuno-oncology and translational research.
