Identification of a claudin-low subtype in clear cell renal cell carcinoma with implications for the evaluation of clinical outcomes and treatment efficacy

Abstract

BackgroundIn bladder and breast cancer, the claudin-low subtype is widely identified, revealing a distinct tumor microenvironment (TME) and immunological feature. Although we have previously identified individual claudin members as prognostic biomarkers in clear cell renal cell carcinoma (ccRCC), the existence of an intrinsic claudin-low subtype and its interplay with TME and clinical outcomes remains unclear.MethodsTranscriptomic and clinical data from The Cancer Genome Atlas (TCGA)- kidney clear cell carcinoma (KIRC) cohort and E-MTAB-1980 were derived as the training and validation cohorts, respectively. In addition, GSE40435, GSE53757, International Cancer Genome Consortium (ICGC) datasets, and RNA-sequencing data from local ccRCC patients were utilized as validation cohorts for claudin clustering based on silhouette scores. Using weighted correlation network analysis (WGCNA) and multiple machine learning algorithms, including least absolute shrinkage and selection operator (LASSO), CoxBoost, and random forest, we constructed a claudin-TME related (CTR) risk signature. Furthermore, the CTR associated genomic characteristics, immunity, and treatment sensitivity were evaluated.ResultsA claudin-low phenotype was identified and associated with an inferior survival and distinct TME and cancer immunity characteristics. Based on its interaction with TME, a risk signature was developed with robust prognostic prediction accuracy. Moreover, we found its association with a claudin-low, stem-like phenotype and advanced clinicopathological features. Intriguingly, it was also effective in kidney chromophobe and renal papillary cell carcinoma. The high CTR group exhibited genomic characteristics similar to those of claudin-low phenotype, including increased chromosomal instability (such as deletions at 9p) and risk genomic alterations (especially BAP1 and SETD2). In addition, a higher abundance of CD8 T cells and overexpression of immune checkpoints, such as LAG3, CTLA4 and PDCD1, were identified in the high CTR group. Notably, ccRCC patients with high CTR were potentially more sensitive to immune checkpoint inhibitors; their counterparts could have more clinical benefits when treated with antiangiogenic drugs, mTOR, or HIF inhibitors.ConclusionWe comprehensively evaluated the expression features of claudin genes and identified a claudin-low phenotype in ccRCC. In addition, its related signature could robustly predict the prognosis and provide guide for personalizing management strategies.

Identification and validation of a glycolysis‐related taxonomy for improving outcomes in glioma

“…Subclass mapping (SubMap) methodology was used to assess the consistency of glioma clustering of NTP and PAM methodologies 16 . Silhouette scores were determined using the specified clusters 17 . Enrichment pathways were identified using the R package clusterProfiler and Gene Set Variation Analysis (GSVA) 18 .…”

“…16 Silhouette scores were determined using the specified clusters. 17 Enrichment pathways were identified using the R package clus-terProfiler and Gene Set Variation Analysis (GSVA). 18 Enrichment scores were assessed using the GSVA R package's single-sample Gene Set Enrichment Analysis (ssGSEA) function.…”