There is a pressing need for tests that can accurately predict response to immunotherapies, writes Cliff Hoyt, our VP of Translational & Scientific Affairs, in a review article for Frontiers in Molecular Biosciences. Established methods are limited in the amount of information they can pull out of formalin- fixed, paraffin-embedded (FFPE) tissue sections. Conventional immunohistochemistry (IHC), for example, is constrained to one or two protein markers, which cannot capture the full breadth of cellular interactions in the tumor microenvironment.
Multispectral imaging of multiplex immunofluorescence (mIF) has emerged as a tool with significant predictive capabilities for immunotherapy response. Multispectral mIF enables deeper interrogation into the spatial biology of the tumor microenvironment, including the capture of spatially resolved, cell-to-cell interactions. There is growing evidence that spatial biology is key to understanding immunotherapy response and clinical outcomes, and thus it is critical to develop mIF assays that can be translated to the clinic.
Below, we outline some of the key points from Cliff’s article, including guiding principles for multispectral mIF and the requirements for translating this approach to clinical practice.
An accessible, standardized workflow for multispectral mIF
Multispectral mIF combines the benefits of conventional immunohistochemistry (IHC) and immunofluorescence with advances in multiplex staining, high-throughput slide imaging, and computer vision. This approach eliminates fluorophore crosstalk and increases signal-to-noise ratio by capturing each fluorescent spectral signal while isolating the tissue autofluorescence to allow for true signal quantification.
However, for mIF to support translational and clinical research, writes Cliff, we must develop analytical performance standards suitable for cancer diagnostic testing. There are three guiding principles for current multispectral mIF. First, mIF assays must match the analytical sensitivity of chromogenic IHC, the current gold standard for clinical testing. Second, multispectral mIF should be supported by fast process workflows that enable consistent and accurate image analysis. Finally, the assay workflow must be practical, economical, and accessible to the entire research community, to produce actionable and reliable data that improves our chances of discovering effective biomarkers.
At Akoya, we’ve developed an end-to-end workflow based on the Phenoptics™ mIF platform, which includes antibodies and detection reagents for manual and automated staining, image acquisition instruments that support field-of-view and whole-slide multispectral imaging, and software for image analysis, data reduction, and cloud-based image storage and sharing.
Fundamentals of the MOTiF workflow for imaging and spectral unmixing. A tissue is stained with Opal fluorophores and imaged using the Vectra Polaris slide scan protocol, Using the inForm software, the exact spectral signature of each fluorophore is isolated to properly unmix each whole-slide composite image, as well as isolate and remove tissue autofluorescence.
The road to the clinic
In immuno-oncology, mIF has proven to be a powerful tool to characterize cellular interactions in the tumor microenvironment and uncover underlying biological mechanisms of cancer. In fact, mIF has been shown to outperform other biomarker testing modalities in predicting response to anti-PD-1/PD-L1 therapies.
Bringing multispectral mIF into the clinic will require a platform that is open, flexible, and practical for researchers to implement into existing laboratory workflows. In the context of mIF, this means the ability to freely select antibodies, design panels, capture expression levels of markers in line with clinical parameters via signal amplification, and perform thorough data analysis to explore the biology behind treatment response in immuno-oncology.
The new frontier of biomarker discovery based on spatial biology has a practical path toward the clinic, based on practically, economically, and analytically robust workflows, which promises to have material benefit for cancer patients.
It is also vital to define standards that guide biomarker development and validation on multispectral mIF platforms and streamline mIF workflows. While there are several multi-institutional groups currently working towards mIF standardization, platform providers must focus on designing robust, automated assays and simple laboratory workflows with minimal operator dependencies. The platform and assays must also support the rigorous regulatory requirements of clinical trials and will need to be run in laboratories compliant with CLIA/GCP/GCLP standards.
As high-plex discovery platforms become more prevalent in cancer research, it’s necessary to pull out meaningful, actionable insights from the high volume of data generated. The nature of these approaches – low throughput and, in most cases, low sensitivity – make them ill-suited for translation to the clinic. However, biomarkers discovered with these approaches can be reduced to the most informative markers for rapid whole-slide analysis with a multispectral mIF workflow for use in clinical trials and eventually, standard of care.
“The new frontier of biomarker discovery based on spatial biology has a practical path toward the clinic,” writes Cliff, “based on practically, economically, and analytically robust workflows, which promises to have material benefit for cancer patients.”
Check out the full article in Frontiers in Molecular Biosciences for an in-depth look into the future of multispectral mIF in the clinic.