Abstract 1202: Spatially-resolved transcriptome analysis of renal tumors with sarcomatoid/rhabdoid dedifferentiation uncovers underlying biology and new biomarkers relevant to these tumors.

Abstract The presence of sarcomatoid (Sarc) and rhabdoid (Rhab) dedifferentiation in renal cell carcinoma (RCC) is linked to poor clinical outcomes and higher risk of metastatic disease. Previous molecular studies on renal tumors with these dedifferentiated features lacked spatial resolution to distinguish these aggressive regions from classical RCC histology, leaving the biology underlying these aggressive phenotypes poorly understood. We applied spatial whole transcriptome profiling (GeoMx DSP) to 370 regions exhibiting Sarc, Rhab, clear cell, papillary, or benign kidney from 54 RCC tumors. We combined these spatial transcriptome profiles with our single-cell RNA-sequencing atlas consisting of different RCC subtypes to investigate Sarc/Rhab biology and verified clinical significance of our findings using public databases such as The Cancer Genome Atlas (TCGA) and Clinical Proteomic Tumor Analysis Consortium (CPTAC). Deconvolution of the spatially resolved transcriptome data revealed substantial gene expression reprogramming in Sarc/Rhab regions. Notably, Sarc/Rhab areas featured a high fibroblast-like gene expression pattern and a loss of cytokeratin expression as a defining hallmark of Sarc and Rhab dedifferentiation, reflecting their histological nature. We established cancer cell-intrinsic gene expression signatures of Sarc and Rhab tumors and found they are highly correlated with higher tumor grade in the ccRCC cohorts from both TCGA (n=485) and CPTAC (n=103). Importantly, these signatures significantly predicted poor disease-free survival in clinically low-risk (Stage 1-2) ccRCC patients. In the ccRCC cohort of TCGA, tumors with high Sarc gene signature levels were enriched for mutations in SETD2, PTEN, MTOR, and KDM5C, as well as specific copy-number variations, notably loss of 14q, 9, 18, and 6p and gain of 20, 12, and 8q, compared to other tumors. Pathway enrichment analyses suggested that genes related to extracellular matrix organization, coagulation, and epithelial-mesenchymal transition were highly enriched in Sarc/Rhab regions compared to classical ccRCC regions, while pathways for oxidative phosphorylation, adipogenesis, and amino acid metabolism were decreased.These findings provide novel insights into the biology of Sarc/Rhab dedifferentiation in RCC, paving the way for the development of improved prognostic markers and targeted therapeutic strategies. Citation Format: Minjun Kim, Mustafa Soytas, Burge Ulukan, Tamiko Nishimura, Senthilkumar Kailasam, Ariel Madrigal, Zohreh Mehrjoo, Kate Glennon, Eleonora Scarlata, Madeleine Arseneault, Morag Park, Hamed Najafabadi, Fadi Brimo, Ozgur Sahin, Simon Tanguay, Yasser Riazalhosseini. Spatially-resolved transcriptome analysis of renal tumors with sarcomatoid/rhabdoid dedifferentiation uncovers underlying biology and new biomarkers relevant to these tumors abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 1202.