Quantitative Image Analysis Guidelines: An Open letter from Liron Pantanowitz
Posted 20th October 2017 by Jane Williams
There has been a steady increase in the adoption of Quantitative Image Analysis (QIA) for diagnostic purposes, especially for breast biomarker testing.
This is not surprising because QIA has been shown to improve the consistency and accuracy of interpretation compared to manual reads by pathologists. In particular, Digital Image Analysis (DIA) for breast cancer biomarkers (ER, PR, HER2 and Ki67) not only offers better sensitivity and specificity than manual assessments but also shows better concordance with gene expression assays (Stålhammar G et al 2016).
Unfortunately, there is lack of standardisation when it comes to QIA. The College of American Pathologists (CAP) has accordingly developed evidence-based recommendations. This attempts to improve the reproducibility, precision, and accuracy in the interpretation of HER2 immunohistochemistry for breast cancer where QIA is employed. This guideline was established by a panel of pathologists, histotechnologists, and computer scientists who all have expertise in this area. A recent open comment period encouraged feedback from the public (see here for the feedback).
I plan on going over these new CAP guidelines with participants at the upcoming Digital Pathology Congress in London on the 30th Nov -1st Dec 2017. I hope to see you at the meeting and look forward to your feedback.
Professor of Pathology and Biomedical Informatics, University of Pittsburgh, USA
Register now for the 4th Digital Pathology Congress: Europe.
Stålhammar G et al. Digital image analysis outperforms manual biomarker assessment in breast cancer. Mod Pathol. 2016 Apr;29(4):318-29.
Leave a Reply