Investigating the determinants of immunotherapy response in the primary tumor of clear cell renal cell carcinoma (RCC).

e16507 Background: While previous studies have examined the determinants of a patient’s overall response to immune checkpoint inhibitor (ICI) in advanced RCC, the factors that specifically influence ICI response in the primary tumor remain unclear. This has implications in both the neoadjuvant setting and the metastatic setting where the primary tumor is still in place. To deepen our understanding of ICI response in the primary tumor and to understand the evolution of RCC on ICI, we conducted a comprehensive genomic analysis of paired pre- and post-treatment primary RCC tumors treated with ICI. Methods: 46 RCC tissue samples comprised of 15 pre-treatment (biopsy of kidney) and 31 post-treatment nephrectomy samples (n = 33 patients, n = 14 with paired samples) were analyzed. 17 samples were from responders ( > 30% radiographic shrinkage or ypT0, n = 4 pre-treatment, n = 13 post-treatment) and 29 samples were from non-responders ( < 30% shrinkage, n = 11 pre-treatment, n = 18 post-treatment). Whole-exome and RNA-seq were performed at Caris Life Science. Wilcoxon rank-sum test was used to compare total mutation burden (TMB), loss of heterozygosity (LOH), and HLA evolutionary divergence (HED). Fisher’s exact test was used to assess the prevalence of driver genes for genes mutated in more than 3 samples. Comparisons were made between pre- and post-treatment specimens, and between response and nonresponse in pre-treatment samples. Ranked gene set enrichment analysis (GSEA) was performed using the 50 hallmark gene sets. Published immune signatures were quantified using single-sample GSEA. P-values were FDR adjusted, with a significance threshold at < 0.05. Results: There was no association between TMB, LOH, HLA class I zygosity, HED, or frequency of RCC driver gene mutations and ICI response in pre- or post-treatment samples. Analysis of pre- and post-treatment samples revealed enriched metabolic pathways, including fatty acid metabolism and heme metabolism, and suppressed genomic repair mechanisms, such as DNA repair and G2M checkpoint, in post-treatment samples, reflecting the evolution of RCC biology while on ICI (independent of response). Pre-treatment tumors that responded to ICI were significantly enriched for gene expression signatures of immune response, including interferon-alpha and interferon-gamma. Conversely, pre-treatment tumors resistant to ICI were significantly enriched for pathways associated with hypoxia, epithelial-to-mesenchymal transition, and metabolic activity. Further, a signature of immune suppression decreased in ICI-responsive tumors and increased in ICI-non-responsive tumors. Conclusions: Our findings demonstrate pathways associated with genomic repair are upregulated following therapy, and pre-treatment primary RCC tumors with enriched immune response and depleted metabolic activity are more responsive to ICI therapy.