Abstract
The ccrcc1-4 transcriptomic subtypes were previously identified in fresh-frozen clear-cell renal cell carcinoma (ccRCC) samples and have proven their prognostic value after nephrectomy/metastasectomy and predictive value for first-line vascular endothelial growth factor receptor-tyrosine kinase inhibitors (VEGFR-TKI). We aimed to create a consensus molecular classification approach called Clearseq1-4 on formalin-fixed, paraffin-embedded (FFPE) tissues and to evaluate its prognostic and predictive value across the RCC treatment landscape. RNA sequencing of tumoral FFPE tissue was performed. A classifier called Clearseq was designed to determine ccrcc1-4 subtypes. Subtypes were correlated with outcomes after surgical and systemic therapies. External validation and characterization at the single-cell transcriptomic level were pursued. A total of 668 tumoral samples (337 primary tumors and 331 metastases) from 364 patients were assigned to ccrcc1, ccrcc2, ccrcc3, and ccrcc4 tumors. The angiogenic ccrcc2 subtype had a favorable prognosis after nephrectomy in a localized setting, after cytoreductive nephrectomy, and upon metastasectomy with curative intent and was correlated with improved outcomes on first-line VEGFR-TKIs. Ccrcc4 tumors were enriched for sarcomatoid features and had the largest treatment benefit from immune checkpoint blockade (ICB) treatment in any line, resulting in overall survival outcomes comparable with those of less aggressive subtypes. These findings were corroborated in a post-nephrectomy cohort and external cohorts of metastatic patients treated with ICB and/or angiogenesis inhibitors. We created a consensus molecular classification approach, called Clearseq1-4, in order to predict the ccrcc1-4 molecular subtype on FFPE tissues and confirmed its performance with respect to previous biomarker findings for both surgical and systemic treatment approaches. SIGNIFICANCE: Clear-cell kidney cancers display a more indolent or aggressive clinical behavior after surgery and different sensitivities to currently available medical therapies: immune therapy or angiogenesis inhibitors. We developed an easy-to-use molecular classification that divides these tumors into four subgroups predicting outcomes after surgery or upon medical therapy.