Highlights d Dysfunctional CD226-negative T cells accumulate in mouse and human tumors d CD155 induces ubiquitination via CBL-B and proteasomal degradation of CD226 d A mutation of Y319 maintains CD226 expression and improves anti-tumor immunity d ICB-therapy in melanoma patients relies on the presence of CD226 + CD8 + T cells

Profiling the tumour microenvironment (TME) has been informative in understanding the underlying tumour–immune interactions. Multiplex immunohistochemistry (mIHC) coupled with molecular barcoding technologies have revealed greater insights into the TME. In this study, we utilised the Nanostring GeoMX Digital Spatial Profiler (DSP) platform to profile a non-small-cell lung cancer (NSCLC) tissue microarray for protein markers across immune cell profiling, immuno-oncology (IO) drug targets, immune activation status, immune cell typing, and pan-tumour protein modules. Regions of interest (ROIs) were selected that described tumour, TME, and normal adjacent tissue (NAT) compartments. Our data revealed that paired analysis (n = 18) of matched patient compartments indicate that the TME was significantly enriched in CD27, CD3, CD4, CD44, CD45, CD45RO, CD68, CD163, and VISTA relative to the tumour. Unmatched analysis indicated that the NAT (n = 19) was significantly enriched in CD34, fibronectin, IDO1, LAG3, ARG1, and PTEN when compared to the TME (n = 32). Univariate Cox proportional hazards indicated that the presence of cells expressing CD3 (hazard ratio (HR): 0.5, p = 0.018), CD34 (HR: 0.53, p = 0.004), and ICOS (HR: 0.6, p = 0.047) in tumour compartments were significantly associated with improved overall survival (OS). We implemented both high-plex and high-throughput methodologies to the discovery of protein biomarkers and molecular phenotypes within biopsy samples, and demonstrate the power of such tools for a new generation of pathology research.

BackgroundImmune checkpoint inhibitors (ICI) have shown durable and long-term benefits in a subset of head and neck squamous cell carcinoma (HNSCC) patients. To identify patient-responders from non-responders, biomarkers are needed which are predictive of outcome to ICI therapy. Cues in the tumor microenvironment (TME) have been informative in understanding the tumor-immune contexture.MethodsIn this preliminary study, the NanoString GeoMx™ Digital Spatial Profiling (DSP) technology was used to determine the immune marker and compartment specific measurements in a cohort of HNSCC tumors from patients receiving ICI therapy.ResultsOur data revealed that markers involved with immune cell infiltration (CD8 T-cells) were not predictive of outcome to ICI therapy. Rather, a number of immune cell types and protein markers (CD4, CD68, CD45, CD44, CD66b) were found to correlate with progressive disease. Cross platform comparison with the Opal Vectra (Perkin Elmer) for a number of markers across similar regions of interest demonstrated concordance for pan-cytokeratin, CD8, and PD-L1.ConclusionThis study, to our knowledge, represents the first digital spatial analysis of HNSCC tumors. A larger cohort of HNSCC will be required to orthogonally validate the findings.