Development of a Single Molecule Counting Assay to Differentiate Chromophobe Renal Cancer and Oncocytoma in Clinics.

Malignant chromophobe renal cancer (chRCC) and benign oncocytoma (RO) are two renal tumor types difficult to differentiate using histology and immunohistochemistry-based methods because of their similarity in appearance. We previously developed a transcriptomics-based classification pipeline with "Chromophobe-Oncocytoma Gene Signature" (COGS) on a single-molecule counting platform. Renal cancer patients (n = 32, chRCC = 17, RO = 15) were recruited from Augusta University Medical Center (AUMC). Formalin-fixed paraffin-embedded (FFPE) blocks from their excised tumors were collected. We created a custom single-molecule counting code set for COGS to assay RNA from FFPE blocks. Utilizing hematoxylin-eosin stain, pathologists were able to correctly classify these tumor types (91.8%). Our unsupervised learning with UMAP (Uniform manifold approximation and projection, accuracy = 0.97) and hierarchical clustering (accuracy = 1.0) identified two clusters congruent with their histology. We next developed and compared four supervised models (random forest, support vector machine, generalized linear model with L2 regularization, and supervised UMAP). Supervised UMAP has shown to classify all the cases correctly (sensitivity = 1, specificity = 1, accuracy = 1) followed by random forest models (sensitivity = 0.84, specificity = 1, accuracy = 1). This pipeline can be used as a clinical tool by pathologists to differentiate chRCC from RO.

Oncocytoma-Related Gene Signature to Differentiate Chromophobe Renal Cancer and Oncocytoma Using Machine Learning.

Publicly available gene expression datasets were analyzed to develop a chromophobe and oncocytoma related gene signature (COGS) to distinguish chRCC from RO. The datasets GSE11151, GSE19982, GSE2109, GSE8271 and GSE11024 were combined into a discovery dataset. The transcriptomic differences were identified with unsupervised learning in the discovery dataset (97.8% accuracy) with density based UMAP (DBU). The top 30 genes were identified by univariate gene expression analysis and ROC analysis, to create a gene signature called COGS. COGS, combined with DBU, was able to differentiate chRCC from RO in the discovery dataset with an accuracy of 97.8%. The classification accuracy of COGS was validated in an independent meta-dataset consisting of TCGA-KICH and GSE12090, where COGS could differentiate chRCC from RO with 100% accuracy. The differentially expressed genes were involved in carbohydrate metabolism, transcriptomic regulation by TP53, beta-catenin-dependent Wnt signaling, and cytokine (IL-4 and IL-13) signaling highly active in cancer cells. Using multiple datasets and machine learning, we constructed and validated COGS as a tool that can differentiate chRCC from RO and complement histology in routine clinical practice to distinguish these two tumors.