Related Experiment Video
Updated: Jun 11, 2026
08:40
Quantitation of Protein Expression and Co-localization Using Multiplexed Immuno-histochemical Staining and Multispectral Imaging
Published on: April 8, 2016
Reliability of stromal markers multiplex immunofluorescent staining: pathologist assessment compared to quantitative
Lusine Yaghjyan1, Yujing J Heng2, Yaileen D Guzman-Arocho2
1Department of Epidemiology, College of Public Health and Health Professions and College of Medicine, University of Florida Gainesville, FL, USA.
American Journal of Cancer Research
|June 10, 2026
Summary
This study compared manual and automated analysis of stromal markers in breast tissue from cancer-free women. Automated analysis showed good correlation for most markers, supporting its use in large studies with careful validation.
Area of Science:
- Oncology
- Biopathology
- Computational Pathology
Background:
- The tumor microenvironment, including stromal components, is crucial in breast cancer development.
- Expression data for key stromal markers (αSMA, FAP, MMP14, TNC, s100a6) in normal breast tissue is lacking.
- Multiplex immunofluorescence (IF) is a powerful technique for assessing protein expression in tissue.
Purpose of the Study:
- To compare manual expert pathologist assessment with automated image analysis for five stromal markers (αSMA, FAP, MMP14, TNC, s100a6) in normal breast tissue.
- To evaluate the homogeneity of marker expression across tissue cores from cancer-free women.
- To assess the reliability of computational platforms for large-scale epidemiologic studies.
Main Methods:
- Multiplex immunofluorescence (IF) was performed on breast tissue samples from 73 cancer-free women.
- Manual scoring by an expert pathologist was compared with automated analysis using inForm software.
- Spearman correlation and sensitivity/specificity analyses were used to evaluate agreement between methods.
- Intra-class correlation (ICC) was used to assess marker homogeneity across cores.
Main Results:
- Automated analysis showed strong correlation with manual assessment for FAP, MMP14, and s100a6 (correlation coefficients 0.70-0.78).
- Moderate correlations were observed for αSMA (0.37) and TNC (0.42).
- Marker expression homogeneity was strong for FAP, MMP14, and s100a6 (ICC 0.63-0.72), moderate for αSMA (0.35), and poor for TNC (0.21).
- Sensitivity varied by marker and cut-off, with TNC showing the lowest sensitivity at a 1% cut-off.
Conclusions:
- Computational assessment of stromal markers using IF shows variable but often strong agreement with manual evaluation.
- Automated image analysis is suitable for most stromal markers in large epidemiologic studies, but TNC requires further investigation.
- Pilot studies are essential to determine appropriate cut-offs for defining staining positivity in automated analyses.
