Quantitative image analysis for immunohistochemistry

Abstract

Precision medicine currently demands precision diagnostics. As a result, anatomical pathology has transitioned from qualitative to more quantitative reporting of immunohistochemistry (IHC) results. The most widely employed quantitative image analysis (QIA) performed clinically by pathology laboratories is for breast biomarkers (ER, PR, HER2). Compared to manual assessment, QIA offers (1) better precision and accuracy of quantitative measurements, (2) standardisation and more reproducible results, especially for intermediate categories and complex scoring systems, and (3) automation which reduces the time consumption for pathologists, especially for performing mundane tasks like counting. However, QIA results can be affected by numerous variables. Pre-analytical variables include tissue handling, slide preparation, stain variation, and image acquisition (e.g., whole slide scanner differences). Analytical variation may be attributed to different image file formats and compression, tumour heterogeneity, analysing different regions of interest (e.g., hot spots), artifacts (e.g., tissue folds, air bubbles, crushed tissue, overlapping cells), and counting errors (e.g., cells between frames). Post-analytical variables include human interpretation error and result discrepancies. Recent QIA guidelines from the College of American Pathologists1 have helped address some of these concerns by providing recommendations for improving the reproducibility, precision, and accuracy of QIA for HER2 by IHC. These guidelines assist pathologists with appropriate algorithm selection, validation for clinical use, calibration using controls, training of laboratory personnel, reporting of results, performance monitoring, and retention requirements for image and computed test results. These guidelines can be extrapolated for similar QIA lab tests, and will hopefully enhance the adoption of QIA in pathology. Reference1.Bui MM, Riben MW, Allison KH, et al. Quantitative image analysis of human epidermal growth factor receptor 2 immunohistochemistry for breast cancer: Guideline From the College of American Pathologists. Arch Pathol Lab Med 2019; 143: 1180–95.