Premium
Automated image analysis for high‐throughput quantitative detection of ER and PR expression levels in large‐scale clinical studies: The TEAM Trial Experience
Author(s) -
Faratian Dana,
Kay Charlene,
Robson Tammy,
Campbell Fiona M,
Grant Margaret,
Rea Dan,
Bartlett John M S
Publication year - 2009
Publication title -
histopathology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.626
H-Index - 124
eISSN - 1365-2559
pISSN - 0309-0167
DOI - 10.1111/j.1365-2559.2009.03419.x
Subject(s) - tissue microarray , clinical trial , immunohistochemistry , medicine , computational biology , pathology , bioinformatics , computer science , biology
Aims: Routine immunohistochemistry is regarded as a semiquantitative method for the evaluation of in situ protein expression. Analysis of tissue biomarkers in large clinical trials is central to the development of novel targeted approaches to therapy, requires the analysis of tens of thousands of data points, and frequently makes use of high‐throughput analysis of tissue microarrays (TMAs). The aim of this study was to investigate the potential of image analysis for accurate and reproducible quantitative evaluation of biomarkers. Methods and results: We showed, in 397 cases of breast cancer from the Phase III TEAM clinical trial, excellent correlations between semiautomated image analysis of TMAs and manual scoring of oestrogen receptor (ER) and progesterone receptor (PR) levels (interclass correlation coefficients 0.93 and 0.96 respectively). Two or more TMA cores were excellently correlated with manual scores, and using more than three cores increased the number of cases available for analysis to >92%. TMAs are confirmed as representative of whole sections for immunohistochemical analysis of the tissue biomarkers ER and PR. Conclusions: Semiautomated image analysis is appropriate for the analysis of tissue biomarkers within large clinical trials. These data provide support for the use of TMAs and image analysis in translational research.