Abstract B015: Understanding cell heterogeneity in single-cell renal cell carcinoma datasets

Abstract

Abstract Studying the heterogeneity in RCC may provide insights for improved diagnosis and treatment. Cell heterogeneity may be reflected at the mRNA level, and cell deconvolution tools like CIBERSORT have been used to decipher cell-specific signals in bulk RNA interrogation. Single-cell RNA seq technologies (scRNA-seq) enable dissecting cell heterogeneity in tumor samples further. An exploratory scRNA-seq study was conducted on three subtypes of RCC: 2 samples of clear cell RCC (ccRCC1 and ccRCC2), 1 sample of type-2 papillary RCC (pRCC), and 1 sample of chromophobe RCC (chRCC) (Su et al., 2021). This study provided valuable insights, such as delineating new tumor markers and predicting drug responses. As we advance our understanding of scRNA-seq, additional tools allow annotating of more cell types and exploring ploidy levels. Here we reanalyze this dataset further to illustrate the tumor microenvironment heterogeneity in RCC samples. We downloaded the datasets (GSE152938) from Gene Expression Omnibus and reprocessed the data using Seurat. The cellular composition and ploidy levels were explored using Azimuth and CopyKAT. A total of 30261 high-quality single-cell transcriptomic information of the RCC samples were obtained for the downstream analysis. The RCC subtypes showed both intra and inter-tumoral heterogeneity. Within the clear cell carcinomas, the ccRCC2 sample showed about 63.57% immune cell infiltration compared to 45.68% in ccRCC1. Rare subpopulations of 0.86% fibroblast and 0.81% mast cells were observed only in ccRCC1. In contrast, 1.56% of B cells were observed only in ccRCC2. Using an improved cell deconvolution algorithm, we identified natural killer T cells, neutrophils, and plasma cells in the ccRCC samples that were not reported in the original publication. We also identified five subtypes of endothelial cells (instead of the two reported in the original article) which may further our understanding of angiogenesis and immune regulation in these ccRCC samples. Our analysis showed that the macrophages in the ccRCC and pRCC samples are predominantly M2 macrophages. In the chRCC sample, we also identified 0.2% B cells and 2.19% M2 macrophage subpopulations, which may further illuminate the immune landscape of the chRCC tumor microenvironment. We found the ccRCC sample to be the most immune-rich with about 56.49% immune infiltration, followed by the pRCC (46.88% infiltration) and chRCC (8.86% infiltration) samples. In the combined ccRCC samples, three of the four ccRCC cell clusters were predominantly diploids (85.6%, 97.8%, and 60%, respectively), whereas only one showed a higher proportion of aneuploidy (95.8%). Further studies with a larger sample are required to validate these observations. This reanalysis of previously published data may enrich the results of the original manuscript and provide additional information about the tumor microenvironment of these RCC subtypes. We will integrate these findings with our more extensive collection of scRNA-seq at our institution, for further interrogating and validating these results. Citation Format: Sadia Islam Kana, Lucas A. Salas. Understanding cell heterogeneity in single-cell renal cell carcinoma datasets [abstract]. In: Proceedings of the AACR Special Conference: Advances in Kidney Cancer Research; 2023 Jun 24-27; Austin, Texas. Philadelphia (PA): AACR; Cancer Res 2023;83(16 Suppl):Abstract nr B015.