Design and Development of Ultrahigh-Plex CODEX Panel to Visualize FFPE Tissue from Genetically Reprogrammed Murine Melanoma

Introduction: The tumor immune microenvironment (TIME) is highly complex, composed of various immune cells, fibroblasts, and other supporting cells communicating in close proximity. Several techniques, such as flow cytometry and single-cell sequencing, are used to study the complexity of the TIME. However, these techniques lose any spatial context, and most current applications of immunofluorescence/immunohistochemistry (IF/IHC) imaging are limited to ~3 markers. Co-detection by indexing (CODEX) is an emerging high-plex technique based on DNA barcoded technology that allows the visualization of biomarkers in a tumor region while maintaining tissue architecture. Unfortunately, while more options are available for fresh-frozen (FF) samples, very few validated antibodies are available for murine formalin-fixed paraffin-embedded FFPE tissues. FFPE offers several advantages over FF tissue, particularly concerning ease of handling/storage and the ability to design tissue microarrays. To address this gap in ultrahigh-plex imaging, we designed and developed a FFPE CODEX panel to study the TIME in a murine melanoma model and applied it to samples of murine melanoma treated with nanoparticle-based genetic reprogramming. The panel can be adapted to study different murine immune organs and other tumor types.

Materials and

Methods: B16F10 murine flank tumors were injected with poly(beta-amino ester)-based NPs delivering control plasmids (luciferase) or reprogramming plasmids on days 9, 11, 16, and 18 following tumor implantation. Tumors were harvested and fixed on day 20, and formalin-fixed paraffin-embedded (FFPE) samples were sectioned onto poly-L-lysine-treated coverslips and slides (Figure 1Ai). BSA-free antibody clones were selected and individually validated with the help of immunofluorescence (IF) and double-IF imaging using a conventional primary antibody and fluorophore-tagged secondary antibody approach. Based on the imaging results, antibodies were selected for further conjugation with CODEX barcodes. Fluorophores were assigned based on the abundance and expression of each biomarker to design the optimal imaging panel (Figure 1Aii). Barcodes were conjugated based on manufacturer instructions, and antibody-barcode conjugation was confirmed with the help of gel electrophoresis. Next, a small CODEX antibody panel (6-9 biomarkers) was visualized on coverslips using a PhenoCycler device attached to a Keyence microscope, and antibody dilutions and exposure times were recorded (Figure 1Aiv). After obtaining the QPTIFF file, images were opened in QuPath, and a quality control step was performed for each marker; antibodies were qualified for the final ultrahigh-plex panel based on this quality control step. A final panel of more than 30 biomarkers was designed, and tumor specimens from treated and control samples were visualized with a PhenoCycler-Fusion 2.0 device by adjusting the exposure times (Figure 1Av).

Results, Conclusions, and Discussions: Antibody validation was performed with IF imaging; representative examples of these validation experiments are shown in Figure 1B. Antibodies expressing positive signals were chosen for barcode conjugation, with further confirmation through gel electrophoresis. The developed panel allowed for the visualization of control and treated tumor samples on coverslips and slides. Our CODEX panel (representative treated and untreated tumors are shown in Figure 1C) optimized and run on the PhenoCycler fusion enabled the simultaneous labeling of cancer cells (SOX10), T cell subsets (CD3, CD4, CD8), B cells (CD20), NK cells (NK1.1), macrophages (F4/80) and macrophage polarization (CD86, CD206), and endothelial cells (CD31). In addition, it allows for the evaluation of proliferation state (Ki67), T cell functional status (granzyme B, T-bet, EOMES), immunologic checkpoints (LAG-3, TIM-3, PD-L1), and antigen presentation (MHC-I, MHC-II, beta-2 microglobulin, LMP2). Advantages of this approach include the ability to identify cells defined by multiple markers such as regulatory T cell (CD3+CD4+FoxP3+) populations in their spatial context in treated and untreated tumor environments. Additionally, this approach allows for the investigation of architectural features such as the formation of tertiary lymphoid structures in treated tumors. Such features are difficult to detect with flow cytometry.
In conclusion, we designed and extensively validated a murine FFPE CODEX panel and provided a step-by-step strategy to design the custom-conjugated antibody panel. This developed panel can be applied to help study the spatial proteomics of pre-clinical samples, allowing for improved characterization of how different immunotherapies alter the TIME.

Acknowledgements (Optional):