Spatially informed gene signatures for response to immunotherapy in Melanoma

Gene signatures have been shown to predict the response/resistance to immunotherapies but with only modest accuracy. Reduction in this precision might be due to the lack of spatial information which prevents the ability from distinguish tumor from tumor-microenvironment (TME) genes. Here we collected gene expression data spatially from three compartments (CD68+macrophages, CD45+leukocytes and S100+tumor cells) of 59-immunotherapy-treated melanoma specimens using Digital Spatial Profiling-Whole Transcriptome Atlas. We developed a computational pipeline to discover compartment-specific gene signatures and determine if adding spatial information can improve patient stratification. We achieved AUC=0.90 for CD45, AUC=0.94 for CD68, and AUC=0.86 for S100B signatures, whereas AUC=0.70 for pseudo-bulk () signature. Cross-testing in different compartments (e.g., CD45 signature in CD68 and S100B compartments) showed poor performance indicating compartment-specificity. Our novel spatial S100B signature showed the best performance with AUC=0.80 in the validation cohort (N=46). Testing our signatures in computationally deconvolved pseudo-compartments revealed lower AUCs. We conclude that the spatially defined compartment signatures utilize tumor and TME-specific information, leading to more accurate prediction of treatment outcome, and thus merit prospective clinical Overall design: Using DSP-WTA, we generated the first spatial transcriptomic map from pretreatment samples of 59 melanoma patients (discovery cohort) treated with ICIs (pembrolizumab; PEMBRO, nivolumab; NIVO, or ipilimumab plus nivolumab; IPI+NIVO), who had unresectable stage III or IV melanoma at the time of treatment. Our aim was to understand the spatial gene expression enrichment within the tumor and TME and to develop spatially defined compartment-specific gene signatures that would enable us to predict immunotherapy outcomes. We built a computational pipeline to develop robust compartmentalized signature models for analyzing compartment/cell type-specific WTA data that uniquely provided differentially expressed (DEX) genes in spatially defined molecular compartments compared to bulk RNA-seq observations.