Background Interactions between tumor and neighboring immune cells are dynamic and complex within the tumor microenvironment (TME), requiring the use of spatial biology techniques to characterize them, and thereby identify potential targets for immunotherapy. While technologies supporting deep dives into spatial biology have rapidly evolved across modalities, including multiplex immunohistochemistry,1 2 a limited number of protein markers can be stained sequentially. Therefore, as the demand for studying different cell phenotypes increases, the number of protein markers required for staining grows exponentially too. Here, we demonstrate the use of an automated hyperplex staining and imaging system to stain 40-plex on various cancer tissues using off the shelf antibodies. A comparison of current available platforms can be found at (table 1).
Methods Heat-mediated antigen retrieval was performed on formalin-fixed paraffin embedded (FFPE) patient tissues: colorectal cancer (CRC), nasopharyngeal cancer (NPC), lung cancer and gastric cancer tissue microarray (GC TMA). Fresh frozen patient NPC, lung cancer and oropharyngeal squamous cell carcinoma (OSCC) tissues were post-fixed and permeabilized. Some of these cancers were immuno-oncotherapy (IO)-treated. Tonsil tissue was used for antibodies optimization and validation. Sequential immunofluorescence protocol was performed using the COMET platform (Lunaphore Technologies SA, Switzerland) with microfluidic chips for the spatial detection of 40 markers with off the shelf primary antibodies (table 2) and fluorophore-conjugated secondary antibodies.3
Results The 40 markers were optimized on human tonsil sections within 5 weeks (figure 1). Two 40-plex slides were concurrently stained and imaged under 24 hours at a time. 9mm x 9mm high resolution whole slide images acquired then provide an overview of the TME including tumor nest and stroma. The display of several immune cell and tumor markers allows for easy visualization of the spatial proximity of lymphoid aggregates to the tumor (figures 2 and 3). Furthermore, the expression of selected markers in the GC TMA can be viewed and cores of interest can be easily identified (figure 4). The images were hosted on ImmunoAtlas (https://immunoatlas.org/) for interactive visualization (table 3).
Conclusions Using this technology, complex cellular phenotypes can be studied with higher plexes. Coupled with spatial data, in-depth analyses can be conducted to stratify patients’ status of treatment response and disease progression. Moreover, preserved tissue integrity allows for downstream stains like immunohistochemistry to be conducted on the same tissue section. Work in progress includes downstream analysis of the images on the VisioPharm platform and hosting of more 40-plex images in the ImmunoAtlas site.