multiplex images in ihc research

How multiplex images transform current IHC research

Written by:

Elisa Opriessnig and Fanny Dobrenova

Learn about the latest uses of multiplex images, immunohistochemistry staining and multiplexed imaging technology in current research practice. We bring the various benefits of this method of image acquisition within the context of IHC research into focus. Find out how the image acquisition workflow with multiplex techniques works. Get the practical advice you need in order to efficiently handle multichannel images in the IKOSA Platform

The benefits of multiplexed imaging in life science research  

With novel therapeutic concepts emerging in the treatment of cancer as well as in biological sample analysis in clinical research has become more and more complex. The primary reason for that is the need for simultaneous assessment of multiple bio-morphological structures based on the same tissue specimen. At the same time researchers have to deal with the small tissue area available in biopsy specimens. 

Multiplex images, multichannel human tonsil epithelial cells image

Multiplex IF image of a human tonsil, composite and individual channels. A: composite image of all channels. B: Nuclear counterstain with blue DAPI dye. C: Cytotoxic T- ”killer cells'' in green (CD8, FITC). D: Scavenger cells or macrophages in yellow (CD68, TRITC). E: cells using a special mechanism to “escape” from a potential attack by immune cells in red (PD-L1, Cy5), and F: Epithelial cells in light blue (Cytokeratin, Cy7). This image is used with kind permission from Ultivue, Inc.]

Multiparametric or multiplexed methods addressing the complex interaction between tumor cells and the tumor microenvironment provide deep insights into cancer immunotherapy response prediction and help improve cancer classification (Bruni et al., 2020). 

Multiplexed staining and imaging techniques, also referred to as multichannel or multispectral imaging techniques, depending on the exact technological setup used thus provide a reliable method to handle the increasing complexity and the sample scarcity issues in IHC research. This approach allows researchers to assign labels to an entire panel of biomarkers on a single tissue section. 

At the same time using conventional methods would have required repeating the procedure on multiple individual sections (Huss, 2021). Thus, advanced multichannel imaging techniques contribute to additional depth of IHC image analysis. This enables researchers to gain a more comprehensive view of marker distribution in order to examine pathogenic processes more efficiently.

A schematic overview of the multiplex imaging workflow from the staining to the final image analysis is shown in the figure below. 

A schematic overview of the workflow from staining to final image analysis

A schematic overview of the workflow from staining to final image analysis. Using an automated IHC staining device (1), tissue sections are incubated with fluorescently labeled antibodies, or fluorescence is achieved by enzymatic conversion. The stained slides are then transferred into a slide-scanning system (2). The light emitted by the light source is split by special filters, allowing only a defined fraction of the spectrum to excite the respective fluorescent label. Upon excitation, light of a slightly longer wavelength is emitted by the fluorophore. This emitted light is captured by the camera inside the device (not shown) and converted into an image, which can then be analyzed by specialized software (3). Image created with Biorender.

The relevance of multiplex imaging and image analysis in immunohistochemistry  

The current clinical pipeline for cancer diagnosis involves expert knowledge and evaluation skills of large sections of tissue stained with dyes using a variety of molecular markers. A tissue section offers the possibility to detect a patient’s condition and thus, improve diagnostics, make better prognostic evaluations and choose the best treatment option. However, due to the limited choice of tools and methods available, important information in tissue sections can sometimes only be partially accessed (Stack et al., 2014). 

Multiplex immunohistochemistry (mIHC) and multiplex immunofluorescence (mIF) are methods used to detect multiple targets in a single histological section using improved standard staining techniques. These novel techniques take advantage of different colored chromogens such as DAB, AES, BCIP for mIHC or fluorophores for mIF. 

Tips and tricks: With the help of state-of-the-art multiplex imaging methods you can extract complex quantitative data on multiple biomarkers at the same time. 

While traditional IHC uses a single antibody for each tissue section, mIHC allows to increase the number of markers which can be assessed within a single tissue sample (Fassler et al., 2020). Consequently, mIHC and mIF allow the observation of distinct cell types (e.g. immune cells, tumor cells) and biomarkers of interest. However, in order to classify each cell type the detection of uniquely colored chromogens of cells expressing biomarkers of interest is required.

What is a multiplex immunoassay? 

Compared to Hematoxylin & Eosin and other, rather nonspecific special stain techniques, IHC staining is targeted at a specific protein marker representative for a target cell type (McKay et al., 2017). Originally developed in the late 1970s (Sternberger, 1979; Jasani et al., 1981), IHC has become a standard practice in pathology institutions and laboratories worldwide in combination with light microscopy evaluation by a pathologist. 

The main concept of the methodology has not changed since then. Antibodies with their almost unlimited diversity and ability to bind their respective cognate antigen are used to specifically mark a particular molecular structure in the tissue. This mark is then visualized, or stained, in a second step, mostly by enzymatically catalyzed reactions resulting in a precipitating dye reaction product at the site of antigen-antibody interaction (Mescher, 2021). 

3,5-Diaminobenzidine (DAB) with its classic brown reaction product and FastRed were the only chromogenic dyes used in the early days of IHC staining. Back then it was  possible to detect up to two different targets on a tissue slide. However, in the past years those stains have been complemented by several chromogenic and multiple fluorescent dyes. 

This opened the door for the simultaneous detection of multiple biomarkers on a single tissue section, nowadays called multiplex immunohistochemistry (mIHC). In relation to chromogenic dyes, fluorescent stains and light microscopy evaluation these methods fall into the multiplex immunofluorescence (mIF) category (Mescher, 2021). 

How does multiplex tissue imaging work?  

Multispectral (fluorescence) imaging technologies (MSI) find utilization in medical and biological research for the visualization of whole-organisms and cell morphology. Various studies discuss the benefits of this imaging mode in an array of scientific disciplines like histology, pathology, spatial biology, high content screening (HCS), immunohistochemistry (Oth et al., 2014; Alshammari et al., 2016; Kang et al., 2018).

Current research aims to develop automated workflows to detect numerous markers and create individual cancer profiles for patients with the use of image analysis algorithms to better predict therapy responses (Allam et al., 2020). Multiplexed image data is acquired through multispectral microscopy methods utilizing optical filters which enable the capturing of image data within a specific wavelength range across the electromagnetic spectrum. 

Multispectral imaging allows the separation of multiple fluorophores even with overlapping emission spectra and thus increases accuracy by minimizing “bleed-through” of signals to neighboring channels. Each signal gets visualized separately. This process is often accompanied by microscopy image analysis. MSI is predominantly used to accomplish complex biomarker multiplexing in brightfield, and fluorescence imaging of a tissue sample, as well as in in-vivo preclinical imaging (Gao et al., 2004; Taylor et al., 2006; Fereidouni et al., 2018). 

The use of MSI comes with certain challenges as the process involves complex and expensive instrumentation including single bandpass or liquid crystal tunable filters (LCTF) and other precisely controlled optical components. It can be inefficient in its use of light as longer exposure times are required compared to monochrome or simple color imaging methods (Fereidouni et al., 2018). This might affect temporal resolution and also photosensitivity. 

Ongoing advances have introduced new experimental capabilities to multispectral fluorescence imaging. A fluorescence microscope picks up light signals from specific protein compounds within tissues and cells.

The underlying process of this technique involves the absorption of light by an indicator/fluorescent molecule (called fluorophore) followed by the emission of some of this light a bit later. The emission allows detecting the fluorescence signal as fluorophores are excited at a specific wavelength whereby their emission intensity can be measured. The wavelength of a light signal corresponds to an individual channel. 

An image generated through multispectral fluorescence imaging is typically made of multiple layers/channels. A channel refers to the number of colors displayed within the context image processing. For example, an RGB image has three channels (Red, Green and Blue). The color of each pixel represents an intensity value (0-255) in each of the three channels which constitutes the respective pixel. During the analysis you can view the image as a composite channel image containing all or selected channels or as an individual channel image (Sanderson et al., 2014, Chen et al., 2021). 

Tips and tricks: Before you submit your multiplex image data for analysis, select a number of channels corresponding to the biomarkers you are interested in.

mIHC Image Acquisition

In order to perform multiple stainings at the same time and on the same tissue section, color staining for several markers can be used.  Due to the increased complexity such stainings can be detected by means of light microscopy or of digital image analysis algorithms. In either case, it is inevitable to develop strategies to stain with contrasting colors (Bolognesi et al., 2017). 

Multiplex immunohistochemistry followed by an evaluation of scanned images using quantitative digital pathology software is commonly used in medical research. The application of this method as an addition to the standard pathology portfolio is also widely discussed (Aeffner et al., 2019). 

The IKOSA image analysis application Sparkfinder is specifically designed to assist in the study of multiplex images. With its help you can easily measure channel intensities in multiplexed fluorescent images

instance segmentation of cell nuclei with specialized software

Bring multiplex image analysis research to the next level with Sparkfinder.

Quantitative parameters of interest yielded with automated image analysis methods include e.g. lymphocyte density, spatial distribution and cellular content of different tissue compartments. As an example, IHC staining was found to be more accurate in the detection of tumor-associated tertiary lymphoid structures than H&E staining (Sautez-Fridman et al., 2019). 

Standard biomarkers used in pathology and immuno-oncology research include e.g. ki-67 as a biomarker for tumor and immune cell proliferation and growth, or CD4 and other biomarkers for immune cell classification.

A common practice in the field of mIHC assays and staining techniques is the use fluorescent dyes, which can be applied to directly label antibodies with a fluorophore or can also be converted into a more stable product by enzymatic reactions.

Fluorescent stains, also known as reactive dyes or fluorophores, are chemical compounds that upon light excitation re-emit light in a defined range of the visible region of the spectrum. This offers a greater degree of spectral separation than chromogenic dyes (Shakya, 2020). The table below provides an overview of fluorescent stains commonly used in multichannel fluorescence microscopy.  

Fluorescent Dye

Target

Color (approx. peak emission wavelength)

DAPI / 4',6-diamidino-2-phenylindole

DNA, adenine–thymine-rich   regions in DNA

Blue (457 nm)

Nuclear yellow /Hoechst S769121

DNA

Green (504 nm)

cell-permeant SYTO 59/SYTO Red Fluorescent Nucleic Acid Stain Sampler Kit (S-11340)

DNA

Red (659 nm)

SYTO 9 green fluorescent nucleic acid stain

DNA

Green (503 nm)

FITC

any, depends on antigen (dye conjugated to primary antibodies)

Green (516 nm)

TRITC

any, depends on antigen (dye conjugated to primary antibodies)

Yellow (570 nm)

Cy5

any, depends on antigen (dye conjugated to primary antibodies)

Red (670 nm)

Cy7

any, depends on antigen (dye conjugated to primary antibodies)

(far-) red (780 nm)

Table: Overview of fluorescent dyes used in IHC research

Each of the staining and imaging technologies discussed has different abilities, requirements and limitations. This must be considered already at the stage of IHC, mIHC or mIF assay validation. 

Brightfield scanning emulates standard brightfield microscopy and is currently the most common approach in digital pathology labs. Fluorescent scanning resembles fluorescent microscopy and is therefore used to digitize fluorescently labeled slides. Multispectral imaging captures information across the spectrum of light and can be applied to both brightfield and fluorescent settings (Zarella et al., 2019), however, at the cost of speed and resulting in large image files. 

Approaches for the analysis of multiplex immunohistochemistry image data 

Brightfield scanning of mIHC tissue sections yields RGB images and requires color decomposition with specialized (viewing) software to separate each single signal. Each pixel in these images is defined by a certain value (0-255) within the color channels red, green and blue. Consequently, similar colors leading to similar pixel values are harder to separate in image analysis. 

Images produced by fluorescence scanners are in principle stacked black-and-white images per fluorescence channel. In each channel, each pixel is defined by a different value (0-65536) representing the intensity of the emitted light at a certain wavelength. Within whole slide viewing software applications, this information is converted into a pseudocolor referring to the respective emission wavelength for visualization purposes only. 

Multiplex images in histology

Multiplex IF image of a human lung cancer , composite and individual channels. A: composite image of all channels. B: Nuclear counterstain with blue DAPI dye. C: Cytotoxic T- ”killer cells'' in green (CD8, FITC). D: Scavenger cells or macrophages in yellow (CD68, TRITC). E: cells using a special mechanism to “escape” from a potential attack by immune cells in red (PD-L1, Cy5), and F:  tumor cells in light blue (Cytokeratin, Cy7). Tumor cells expressing PD-L1 are “invisible” to the immune system, but this can be reverted with targeted anti-tumor therapy. This image is used with kind permission from Ultivue, Inc.]

With the help of current bioimage analysis tools available on the market users can perform different processing tasks like: 

  • check
    splitting and merging channels, 
  • check
    merging RGB colors,
  • check
    creating composite images and
  • check
    exporting multichannel image results.

How to handle multiplex images in the IKOSA Platform? 

The IKOSA Platform allows you to work with multiplex images using the multichannel feature available. The multichannel feature is supported on the IKOSA Platform for image viewing purposes and currently available within the context of some of our image analysis applications like Sparkfinder.

Multichannel images can easily be recognized and selected in the image library.

handling multiplex images in IKOSA

Select multiplex images for analysis in the IKOSA image library.

When viewing multichannel images, you can effortlessly enable the visibility of each channel, and select the colors you want to view the image with. 

handling multiplex images in IKOSA

Select individual color channels in the IKOSA image viewer.

This can be done either directly in the IKOSA image viewer or by adjusting the multichannel settings in the individual image analysis application. These functionalities allow you to conveniently fine-tune intensity dynamics across different channels for the best possible image view.

The multichannel feature within IKOSA

Adjusting channel intensity in a preview window of IKOSA viewer.

Further, you can create a combined view of multiple images by stacking them on top of one another. Composite images allow you to see multiple channels at the same time. 

When working on a multichannel image (hyperstack) in the composite mode, the composite view displays the maximum intensity of all channels. This enables you to quickly check the intensity ranges across all channels (e.g. multichannel image stacks). Furthermore, it is possible to add annotations on all channels of an image at once. 

handling multiplex images in IKOSA

View individual channel intensities in composite mode.

The multichannel feature within IKOSA is a fast and user-friendly tool for multichannel viewing and analysis and you can use the channels you need in order to increase certain feature visibility while viewing and/or annotating.

The multiplex image analysis in IKOSA

Specialized applications available in the IKOSA Prisma portfolio like enable you to effortlessly analyze multiplex images. With the help of Sparkfinder you can extract reliable quantitative measurements of individual channel intensities. You can further utilize the gathered intensity data for conducting colocalization analysis and effectively localize biomarkers on your multichannel images.

handling multiplex images in IKOSA

Quantify channel intensities with Sparkfinder.

Test the multiplex viewing and analysis features in IKOSA

With the IKOSA Platform multichannel image viewing and analysis are easy tasks. Try the multichannel functionalities within IKOSA yourself. If you have questions, contact us at office@kmlvision.com.

See references