Quantifying Estrogen and Progesterone Receptor Expression in Breast Cancer by Digital Imaging

Abstract

Developments in digital imaging and fluorescent microscopy provide a new method and opportunities for quantification of protein expression in human tissue. Archived collections of paraffin-embedded tumors can be used to study the relationship between quantitative differences in protein expression in tumors and patient outcome. In this report we describe the use of a DeltaVision Restoration deconvolution microscope, combined with fluorescent immunohistochemistry, to obtain reproducible and quantitative estimates of protein expression in a formalin-fixed paraffin-embedded tissue. As proof of principle, we used antibodies to the estrogen and progesterone receptors in a hormone receptor-positive breast cancer specimen. We provide guidelines for control of day-to-day variability in camera and microscope performance to ensure that image acquisition leads to reproducible quantitative estimates of protein expression. We show that background autofluorescence related to formalin fixation can be controlled and that for proteins that are expressed in nearly every cell, multiplexing two primary antibodies on the same slide does not significantly affect the results obtained. We demonstrate that for proteins whose expression varies markedly from cell to cell, data reproducibility, as assessed by imaging successive tissue sections, is more difficult to determine.

Chip-Based Comparison of the Osteogenesis of Human Bone Marrow- and Adipose Tissue-Derived Mesenchymal Stem Cells under Mechanical Stimulation

“…1.44P, NIH) was used to quantify the mean fluorescent intensity and the area (%) occupied by positive staining, following immunohistology of osteogenic ECMs and CD markers. [56], [57] Each gray scale image for green fluorescent staining was separated from the RGB channels and normalized to remove background staining. To measure the mean background fluorescence intensity for each slide, two boxes were placed in background areas in which there was no binding by primary antibody.…”

Quantification of immunohistochemistry—issues concerning methods, utility and semiquantitative assessment II

“…Not only can one label physically block the successful labelling of the next antigen due to steric hindrance, but the various labelling procedures can be chemically incompatible. Suffice it to say that the performance of multiple labellings on a single specimen increases the demands for appropriate controls 43 . Assuming that the labelling procedures have been performed satisfactorily, unmixing of three or more chromogens is entirely feasible 38,44 (Levenson, submitted).…”

Quantification of immunohistochemistry—issues concerning methods, utility and semiquantitative assessment I

“…Image analysis systems assess the amount of staining by measuring absorption, so the non‐linear relationship that occurs at higher levels between amount of antigen and intensity can result in inaccurate readings. Recent approaches have used antibody‐conjugated fluorophores and fluorescent microscope systems, 8,34 but are still within the research rather than the diagnostic arena. Image analysis has been proposed as a more reliable method for assessing HER‐2 immunostaining, 35–37 but seems feasible only if assays are centralized to a small numbers of centres due to cost of the systems.…”