Background Our ability to predict tumor response to immunotherapeutic (IO) strategies is limited by the relatively poor predictive power of currently available biomarkers. The comprehensive characterization of the tumor immune microenvironment through spatial omic approaches holds significant promise to stratify the ability or inability of the tumor microenvironment (TME) to respond to IO. Here we developed and tested a hyperplex (25-plex) TME panel on the novel Lunaphore COMET multiplexed immunofluorescence (mxIF) platform for performing high-throughput TME characterization studies.

Methods We curated a panel of 25 analytes that represent a broad range of diverse immune and tumor targets. The final panel consisted of the following targets: CD3, CD4, CD8, FoxP3, Granzyme B, Ki67, PD-1, PD-L1, CK, CD39, CD103, ICOS, LAG-3, Tbet, HLA-ABC, HLA-DR, CD68, CD163, CD11c, CD66b, CD14, Trem2, CD20, CD34, Transgelin. Initial testing on formalin-fixed paraffin-embedded (FFPE) tonsil tissue served as a positive control for assessing staining efficiency and identifying optimal conditions. We were able to rapidly determine optimal antibody conditions on the platform thanks to its ability to simultaneously assess multiple dilutions for multiple antibodies in a single run. Noise-to-signal ratios were evaluated using image viewing software, enabling the selection of optimal conditions. Epitope stability was assessed on additional tissue sections to determine the positioning of each antibody in the panel.

Results All biomarkers were successfully optimized and validated using FFPE tonsil sections. Out of the 25 biomarkers, 20 epitopes demonstrated high stability after 20 cycles, guiding the selection of the staining order. The optimized panel was successfully transferred to head and neck squamous cell carcinoma neoadjuvant IO clinical trial samples, and comprehensive images containing all 25 biomarkers were acquired.

Conclusions Here we demonstrate the rapid development of a comprehensive hyperplex IF panel enabling the high-throughput characterization of the TME in FFPE tissues. We are currently applying this panel across several large studies to assess the impact on the TME of pre- and post-neoadjuvant IO. Antibody clone information, optimal dilutions and run order reported herein may guide the assessment of the TME in IO settings and may help minimize efforts and facilitate the successful development of similar panels in other labs.

Acknowledgements We acknowledge Providence Foundations of Oregon for support.

Ethics Approval This research was conducted on deidentified specimens under a protocol approved by the Providence IRB (#06–108)