Abstract B030: Spatially-resolved cell-cell interactions between tumor microenvironment and immunologically “hot” and “cold” locally advanced prostate tumors

Abstract

Abstract Background: Immune-checkpoint inhibitors (ICI) such as anti-PD-L1 have shown promise in solid tumors with high mutational burden and immune infiltration such as melanoma. However, ICI have shown have only shown mild success in prostate cancer. One likely explanation is that majority of prostate cancer have low mutational burden leading to poor immune infiltration. However, exceptions to the immune infiltration have been observed in a subset of locally advanced prostate cancer specimens. We hypothesized that specific immune subsets, particular myeloid progenitor cells, drive microenvironmental differences between hot and cold tumors. Methods: A total of four formalin-fixed paraffin-embedded tissue punch samples were obtained from biobank at UCLA and were sectioned at 5µm thickness. The sections were deparaffinized and de-crosslinked per manufacture’s instructions, followed by spatial transcriptome profiling using the Visium (10x Genomics) platform. Libraries were constructed from RNA and sequenced on Illumina platform. The reads were processed using Spaceranger. Spot level read counts were first filtered by gene detection rate, total gene count and mitochondrial content. Spots passing QC were normalized and clustered based on the top 2000 most highly variable genes. Cell-cell communication was assessed using the LIANA ligand-receptor analysis framework. To identify regions of inflammation, cell type deconvolution was performed using SpatialDWLS (dampened weighted least squares) and gene-set enrichment was performed using ssgsea from GSVA package. All analysis and statistics were perform using R version 4.1. Results: The number of spots passed QC ranged from approximately 3400 – 4000. Clustering by gene expression resulted in 14 – 23 clusters that showed high concordance with expert pathologist annotation of tumor and tumor subtype mapping to intraductal carcinoma and invasive carcinoma. Cell-type deconvolution by SpatialDWLS found that different tissue regions harbored approximately 1.5-4.5% of immune cells. The myeloid fraction was 0.8%- 3% and T-cells of total immune cells, representing 0.4% - 0.9% of the total cell population. These regions with myeloid cell enrichment were associated with elevated interferon alpha and gamma response, as well as IL2-STAT5 signaling in geographically proximal regions. Conclusion: The presence of myeloid cells in the microenvironments of immunologically “hot” tumors may represent a major factor contributing to immunosuppression in tumors predicted to respond to ICI. Ongoing efforts profiling “hot” tumors will assess the feasibility of modulating myeloid cells in order to recruit cytotoxic T cells. Citation Format: Anson Ku, Rong Rong Huang, Huihui Ye, Adam Sowalsky. Spatially-resolved cell-cell interactions between tumor microenvironment and immunologically “hot” and “cold” locally advanced prostate tumors [abstract]. In: Proceedings of the AACR Special Conference: Advances in Prostate Cancer Research; 2023 Mar 15-18; Denver, Colorado. Philadelphia (PA): AACR; Cancer Res 2023;83(11 Suppl):Abstract nr B030.