Quantification of Molecular Markers in 2-plex IHC
Immunohistochemistry as a method for visualization and characterization of protein and biomarker expression patterns, as well as specific phenotypes, has been employed for decades in various fields of biomedical research and clinics.
Both of its subtypes, chromogenic and fluorescent, are employed in current research, which are often demanding more advanced visualization techniques and multiplexing options. While immunofluorescence offers a variety of multiplexing methods, it is often a tedious and resource-demanding process. Chromogenic IHC, on the other hand, can be performed and visualized with a standard light microscope, available in most labs, and samples can be stored longer without losing signal intensity/bleaching.
Depending on the scope of the research, a standard microscopy setting may not be sufficient to visualize several markers and differentiate between them efficiently enough to produce reliable data which can be later statistically analyzed. Furthermore, manual image analysis may be complicated, as color distinguishing by the user is largely subjective and may falsify the follow-up statistical analysis.
Within the following two case studies, TissueGnostics´ image analysis solution, StrataQuest, was used to quantify 2-plex IHC processed tissue sections. It is also important to mention that correct color separation - unmixing multicolor images and transferring them into marker-associated greyscale images - is the most fundamental step of the analysis process. Additionally, the presented analysis workflows need minimal user input to optimize settings and are fully automated.
Project 1: Assessment of the spatial distribution of NCAM and Ki67 within kidney cancer.
The aim of this project was to:
- Perform nuclei segmentation based on hematoxylin;
- Detect Ki67+ (marker for proliferating cells), NCAM+ (marker for cell-to-cell interactions) as well as Ki67+NCAM+ positive cells;
- Generate density heatmap of Ki67 and NCAM positive cells and compute measurements (percentage of positive and double positive cells, area), presenting output as scatterplots.
(a) Overview picture of a kidney sample stained for hematoxylin (blue), Ki67 (red), and NCAM (brown). Red arrows indicate Ki67+ cells.
The first step of the analysis was the generation of grey images for the markers, utilized automatically by TG´s color separation algorithm, based on the marker color and intensity: (b) shows the original image, (c) is the nuclei grey image, (d) and (e) represent unmixed gray channels for Ki67 and NCAM correspondingly.
Next, these channels are used as the basis for cell segmentation (nuclei) and marker detection (Ki67, NCAM): (f) is the original image, (g) shows Ki67+ cells in turquoise nuclei mask, and (h) combined overall cell detection in pink, with Ki67+ cells highlighted in turquoise.
Following the cell segmentation and marker detection, the whole NCAM positive cell area (i) was detected, highlighted by a green mask, and further, a density mask (j) was applied to find NCAM high-density spots in the sample.
In StrataQuest, FACS-like scattergrams (k) are used to define and further segment specific cellular phenotypes. In this case study, the exact number of Ki67+ NCAM+ cells within the sample was found to be 404. The backward connection allows the visualization of double-positive cells in turquoise from the upper right quadrant or switch to Ki67+NCAM- cells in upper left.
The example below (l) shows fields of views (FOVs) - separate acquired images which are stitched together during slide scanning to a whole-slide image - with a density map and corresponding scattergram, which indicates the FOV where Ki67+NCAM+ double-positive cells have the highest density.
Project 2: IHC 3 App
Another example of 2-plex IHC is the IHC 3 App, available in StrataQuest as a standard App. This App can be applied to any 2-plex staining that also includes additional nuclei staining. Project 2 was similar to project 1: nuclei segmentation and the detection of cells positive for each of the markers.
In order to complete this, first, the color separation algorithm was used to create gray channels of each marker (n – marker 1, o – marker 2) from the original image (m). Then the individual nuclei were identified based on the hematoxylin gray channel, shown by the green mask (p).
To assess the percentage of marker-positive cells, scattergrams are generated, and a cut-off on marker intensity was set to distinguish between marker+ and marker- cells. To visualize these marker+ cells directly in the image (q), they can be selected by the backward connection feature from the corresponding scattergram (red mask). Additionally, if the scattergram is presented as a heatmap, it will give an indication where most cells are concentrated, which can help to adjust the cut-off.
Overall, image analysis empowered by StrataQuest allows for quick, automatic, and reproducible data output, suitable for publishing. With StrataQuest’s streamlined workflow, multiplex IHC digital images can be easily analyzed due to its color separation and nuclei segmentation algorithms, and then the software delivers numerical values, which can be later statistically evaluated. Combing it with high-grade visualization, image analysis with StrataQuest can be a powerful tool to significantly enhance your research.
TissueGnostics has decade-long expertise in imaging and image analysis solutions, made by researchers, for researchers. To help you in promoting your research, we offer the development of Apps tailored to your specific needs. Contact our team today and discover more about our products.
Resource:
- StrataQuest App Center. https://tissuegnostics.com/products/contextual-image-analysis/strataquest-apps
