Association between molecular changes and perioperative immunotherapy in patients with advanced clear cell renal cell carcinoma undergoing cytoreductive nephrectomy.

Abstract

573 Background: Immunotherapy (IO) combinations are standard first-line treatments for advanced clear cell renal cell carcinoma (ccRCC). However, identifying patients who will respond to IO remains a challenge, complicating treatment selection. Previous studies on biomarkers for IO response or resistance often used unmatched biopsies, limiting insights into therapy-induced changes due to patient heterogeneity. This study aims to evaluate transcriptomic changes in matched pre- and post-IO patient samples to detect treatment-induced gene and pathway alterations. Methods: A retrospective analysis was performed on patients with advanced ccRCC receiving IO-based therapies who underwent cytoreductive nephrectomy (Nx) at UT Southwestern/Simmons Comprehensive Cancer Center (SCCC). Patient-level data were extracted from an IRB-approved registry, and samples were collected from the SCCC biorepository. The primary endpoint was to identify IO-induced transcriptomic changes using paired pre- and post-treatment tissue sections. Regions of interest (ROIs) were selected based on pan-cytokeratin (PanCK) expression using the GeoMx Digital Spatial Profiler (DSP, NanoString). Whole-transcriptome analysis was performed, and data were analyzed using a two-sample t-test. Results: We identified 58 patients with advanced ccRCC who received perioperative IO regimens and deferred Nx. Of these, 7 patients had paired pre- and post-treatment samples sequenced (3 received ipilimumab/nivolumab, 3 nivolumab, and 1 pembrolizumab/axitinib). Individual analysis of each matched pair revealed an average of 92 DE genes (log fold-change >1.3, p<0.01). Gene set enrichment analysis (GSEA) showed differential enrichment in pathways such as immune response (interleukin signaling, JAK/STAT signaling, innate immune system activation), translational machinery, MHC-mediated tumor antigen presentation, axon guidance, and cell death pathways. Notably, a key post-IO downregulated pathway was GPCR signaling, which is implicated in tumor growth and metastasis. Taken in aggregate, bulk analysis revealed only 9 differentially expressed (DE) genes in PanCK+ tumor regions, and GSEA did not identify commonly enriched pathways. Conclusions: While bulk analysis detected few changes, individual pair analysis highlighted significant transcriptomic shifts in RCC tumors post-IO. These RCC tumors upregulated gene expression in several immune pathways, including interleukin, JAK/STAT, and innate immune pathways. This study underscores the importance of matched samples to capture patient-specific changes after IO combinations.