8 Single Cell Transcriptomic Investigation of Renal Cell Carcinoma (RCC) Reveals Tissue Resident Memory Exhausted CD8+ T Cell Signature Associated with Resistance to Immune Checkpoint Inhibition (ICI)
Abstract Background The current standard of care for advanced RCC is ICI-based combination therapies. However, most patients with advanced RCC develop disease progression despite ICI treatment, suggesting a lack of durable immune response. Although a lack of T cell infiltration or the presence of non-tumor-reactive “bystander” T cells are hypothesized mechanisms of ICI resistance across tumor types, therapeutic resistance in RCC may still occur in the presence of abundant infiltration of tumor-specific CD8+ T cells. We therefore investigated whether CD8+ T cell phenotype in the RCC tumor microenvironment (TME) impacts ICI response or resistance. Methods 70 tumor samples from 63 RCC patients were collected before (n = 48) or after (n = 22) therapies (VEGFi, n = 9; ICI monotherapy, n = 20; ICI combination, n = 26; others, n = 15). 11 samples were collected from patients without tumors. RCC variants included 59 clear cell and 11 non-clear cell samples. 18 were labeled as clinical benefit and 11 as no-clinical benefit. Single-cell RNA sequencing (10x Genomics) was performed on these samples to generate a transcriptome of the RCC TME. Graph-based clustering identified cell type populations, which were annotated with known lineage genes. Non-negative matrix factorization (NMF) identified gene programs within exhausted CD8+ T cells (Tex). Differential gene expression analysis determined the most differentially expressed genes between resident memory Tex and other cell populations. Results Within CD8+ T cells, Tex cells were identified through expression of TOX, PDCD1 (PD-1), and HAVCR2 (TIM-3). NMF generated 4 gene programs within Tex cells, expressing markers for immediate early genes (JUNB, FOS), exhaustion/activation (GZMK, CD74, LAG3), tissue residency (GZMH, ITGAE, IL7R), and stress response (HSPA1A, HSPA6). The tissue residency program was associated with resistance to ICI therapy (p = 0.05); this association was only found in samples with abundant tumor-specific CD8+ T cells. Differential expression between resident memory Tex (Tex-RM) and other cell types generated a signature of 10 markers that were most highly expressed in Tex-RM. Response and survival data of external bulk RNA-seq cohorts were analyzed. A signature score subtracting for Tex-RM signature was calculated (normalized to overall abundance of Tex cells by signature analysis), which was significantly higher in patients with progressive disease than those with complete/partial response (p = 0.0046), specifically for patients receiving ICI-based therapies. Additionally, survival analysis revealed that ICI-based patients with a higher (top 25%) signature score had significantly worse progression free survival (PFS; p = 0.0048) as well as overall survival (p = 0.0069) with ICI. For ICI-treated patients, the Tex-RM signature score was associated with worse PFS, with a hazard ratio of 2.1 (90% CI [1.3, 3.25]). There was no significant impact on patients receiving TKI monotherapy. Conclusions Through scRNA-seq analysis, we identify a tissue residency gene program in Tex cells associated with non-response to immunotherapy. A signature derived from this program was additionally shown to predict significantly worse response and outcomes for patients receiving ICI-based therapies within a group of bulk RNA-seq clinical trial cohorts. This study provides a framework for using scRNA-seq to identify mechanisms of ICI resistance in RCC and nominates resident memory exhausted CD8+ T cells as a targetable subset of cells to improve CD8+ T cell-mediated anti-tumor immunity. DOD CDMRP Funding yes