Hematopoietic cells were stained with fluorochrome-conjugated antibodies against human CD45, CD3, CD19 and CD14 and stromal cells with fluorochrome-conjugated antibodies against human CD45 and CD235a. Live/dead cell discrimination was performed by adding 7-amino-actinomycin D (7AAD; Calbiochem) prior to acquisition. CD45– CD235a– stromal cells, CD45+ CD3+ T cells and CD45+ CD19+ B cells were sorted with a BD FACS Melody cell sorter (BD Biosciences) and run on the 10x Chromium analyzer (10X Genomics). cDNA library generation was performed following the established commercial protocol for Chromium Single Cell 3’ Reagent Kit (v3 Chemistry). Libraries were run via Novaseq 6000 for Illumina sequencing at the Functional Genomic Center Zurich. A total of 20 samples were collected from 9 patients and processed in 6 batches. All samples from the same patient were processed in the same batch. Gene expression estimation from sequencing files was done using CellRanger (v3.0.2) count with Ensembl GRCh38.9 release as reference to build the index for human samples. Next, quality control was performed in R v.4.0.0 using the R/Bioconductor package scater (v.1.16.0) and included removal of damaged and contaminating cells based on (1) very high or low UMI counts (>2.5 median absolute deviation from the median across all cells), (2) very high or low total number of detected genes (>2.5 median absolute deviation from the median across all cells) and (3) high mitochondrial gene content (> 2.5 median absolute deviations above the median across all cells). In addition, contaminating cells expressing any of the markers CD3E, PTPRC, CD79A or GYPA were removed from stromal cell samples. Downstream analysis was performed using the Seurat R package (v.4.0.1) and included normalization, scaling, dimensionality reduction with PCA and UMAP, graph-based clustering and calculation of unbiased cluster markers as well as dimensionality reduction with diffusionmap as implemented in the scater R/Bioconductor package (v.1.16.0). Clusters were characterized based on the expression of calculated cluster markers and canonical marker genes as reported in previous publications. For the extended stromal cell analysis, two contaminating clusters with 50 cycling cells and 150 cells expressing both fibroblast and endothelial marker genes (indicative of doublets) were removed. For high resolution FRC analysis, FRC subsets were re-embedded and two clusters containing 256 cells with high levels of endothelial or mitochondrial/non-coding genes, respectively, were excluded. Comparative analysis included determination of cell type-, subset- and condition-specific gene signatures. Thereby differentially expressed genes were calculated running the FindAllMarkers function from Seurat R package.

IntroductionNadofaragene firadenovec (Ad-IFNα/Syn3) is now approved for BCG-unresponsive bladder cancer (BLCA). IFNα is a pleiotropic cytokine that causes direct tumor cell killing via TRAIL-mediated apoptosis, angiogenesis inhibition, and activation of the innate and adaptive immune system. We established an immunocompetent murine BLCA model to study the effects of murine adenoviral IFNα (muAd-Ifnα) gene therapy on cancer cells and the tumor microenvironment using a novel murine equivalent of Nadofaragene firadenovec (muAd-Ifnα).MethodsTumors were induced by instilling MB49 cells into the bladders of mice; luciferase imaging confirmed tumor development. Mice were treated with adenovirus control (Ad-Ctrl; empty vector), or muAd-Ifnα (3x1011 VP/mL), and survival analysis was performed. For single-cell sequencing (scRNAseq) analysis (72h), bladders were harvested and treated with collagenase/hyaluronidase and TrypLE for cell dissociation. Single cells were suspended in PBS/1% FBS buffer; viability was assessed with Vicell cell counter. scRNAseq analysis was performed using 10X genomics 3’ sequencing. Raw RNAseq data were pre-processed using Cell Ranger single-cell software. Seurat (R package) was used to normalize and cluster the scRNA data. Pooled differential gene expression analysis in specific cell clusters was performed with DESeq2.ResultsWe identified 16 cell clusters based on marker expression which were grouped into epithelial (tumor), uroplakin-enriched, endothelial, T-cells, neutrophils, and macrophage clusters. Top differentially expressed genes between muAd-Ifnα and Ad-Ctrl were identified. Within the specific cell clusters, IPA analysis revealed significant differences between muAd-Ifnα and control. IFNα signaling and hypercytokinemia/chemokinemia were upregulated in all clusters. Cell death pathways were upregulated in tumor and endothelial clusters. T-cells demonstrated upregulation of the immunogenic cell death signaling pathway and a decrease in the Th2 pathway genes. Macrophages showed upregulation of PD1/PD-L1 pathways along with downregulation of macrophage activation pathways (alternate and classical). Multiplex immunofluorescence confirmed increased infiltration with macrophages in muAd-Ifnα treated tumors compared to controls. PD1/PD-L1 expression was reduced at 72h.DiscussionThis single-cell analysis builds upon our understanding of the impact of Ad-IFNα on tumor cells and other compartments of the microenvironment. These data will help identify mechanisms to improve patient selection and therapeutic efficacy of Nadofaragene firadenovec.