Mega Multiplexing for Exploring Diseased Tissues

Share this post on:


Researchers Francesca Bosisio and Frederik De Smet fro KU Leuven pose outside of the multiplex immunohistochemistry facility that they co-chair.

Francesca Bosisio and Frederik De Smet co-chair the KU Leuven mIHC facility, which specializes in the MILAN method.

Rob Stevens, KU Leuven

An appreciation of tissue complexity in disorders such as cancer has led researchers to develop multiplex immunohistochemistry methods (mIHC), where dozens of markers are assayed on a single slide. These methods lend insights into disease states by highlighting tissue architecture at the single-cell level within spatial context.

Dermatopathologist Francesca Bosisio and bioengineer Frederik De Smet are co-chairs of the KU Leuven facility for mIHC. Their flagship high-plex mIHC technology, Multiple Iterative Labeling by Antibody Neodeposition (MILAN), is a cyclic staining method that uses conventional fluorescent antibodies to stain up to 80 markers on the same slide.1 In their facility, they offer end-to-end mIHC services for scientists at KU Leuven and beyond. In this interview with The Scientist, Bosisio and De Smet discuss MILAN’s origin and the benefits of mIHC.

How was the MILAN method developed?

Francesca Bosisio: When I was a medical resident at the University of Milano-Bicocca. I was working with Giorgio Cattoretti, and we thought it was time to have more than ten markers on a pathology slide. To develop an mIHC protocol, we started comparing different antibody stripping methods. We spent quite some time developing an efficient stripping solution that was gentle on tissue. 

Then I came to KU Leuven for graduate school, and the staining method was ready to be tested. I had a panel of 39 phenotypical and functional markers to investigate the microenvironment in primary melanoma tissues. Suddenly, I had a huge number of multiplex images with no way to analyze them. It took us two years to develop our own pipeline from scratch, to go from the single cell signals to drawing inferences about cellular relationships in the tissue. That led us to develop our own analysis software and to understand how the activation status of T lymphocytes changes in different regions of primary melanomas.2

How does MILAN compare to other high-plex mIHC methods?

Frederik De Smet: Most other technologies are confined to a one centimeter or smaller area that can be stained. With MILAN, we analyze the entire surface of a regular histology slide.

FB: We do not use engineered antibodies; we can use any antibody of interest for the project after we have validated it. If a researcher has developed their own antibody, we can implement it immediately. Additionally, MILAN uses secondary antibodies to amplify signals. This allows the capture of lower intensity signals, so the range of a protein’s expression that can be detected is broader than when using primary antibodies without an amplification system.

How does your facility optimize the MILAN method? 

FDS: We curate the whole process from start to finish. When researchers bring their samples to our facility, we have pathologists involved. It is a challenge to integrate the spatial and single cell data. It is important to have somebody with years of knowledge and training looking at the stained tissues. Pathologists can determine if certain signals are relevant before a researcher dives into the interactions they identified.

See also “Traditional Techniques, New Applications”

Also, we have in-house instrumentation that can do a lot of the work automatically, which is beneficial for standardization and data quality. The data analysis is usually challenging, so we are working on a solution where someone without bioinformatics skills can understand the expression levels that are present in different phenotypes. 

You both are cancer researchers. How does mIHC benefit this field?

FB: Cancer has a complex immune ecosystem. There are different types of inflammatory cells fighting the tumor, and the tumor itself is heterogenous, assuming different cell states. To characterize every cell on the slide, researchers need to stain specific markers in each of these cancer cells. This way we can show a complex picture for each patient in terms of their specific cancer and the inflammatory population composition. If we can see the patient-specific heterogeneity, we can find a better drug that addresses the patient’s particular cancer. 

In a recent study, we mapped the immune phenotypes of patients with melanoma before receiving immunotherapy.The spatial component of mIHC is what made the difference in our results. To determine if a patient would respond to immunotherapy, it was important to see the expression of markers on certain cell types that were located in a specific part of the tumor.

FDS: For glioblastoma, the currently available immunotherapy approaches are not working well. To explore new strategies, we need to know where certain cells are located, how are they interacting, and the tissue architecture. We are using multiplex technologies to understand a spectrum of patient profiles and determine how we could intervene in a personalized way.

What is on the horizon for the MILAN method?

FDS: The next stage is to combine MILAN with other technologies, such as performing mIHC and superimposing transcriptomics on top of individual cells on the same slide. This adds another level of complexity and understanding of disease processes that would not be possible with these methods on their own. That’s the ultimate goal. 

This interview has been condensed and edited for clarity.

References

  1. Bolognesi MM, et al. Multiplex staining by sequential immunostaining and antibody removal on routine tissue sections. J Histochem Cytochem. 2017;65(8):431-444. 
  2. Bosisio FM, et al. Functional heterogeneity of lymphocytic patterns in primary melanoma dissected through single-cell multiplexing. eLife. 2020;9:e53008. 
  3. Antoranz A, et al. Mapping the immune landscape in metastatic melanoma reveals localized cell-cell interactions that predict immunotherapy response. Cancer Res. 2022;82(18):3275-3290. 
Share this post on:

Leave a Reply

Your email address will not be published. Required fields are marked *