Personalized Medicine in Kidney Cancer: Learning How to Walk Before We Run

Abstract

Advances in targeted therapy have resulted in an expanded menu of treatment options for metastatic kidney cancer patients. This necessitated the search for additional biomarkers for accurate classification, prognostic assessment, and prediction of treatment efficiency. Accumulated experience has shown that morphology cannot be relied on as a sole indicator of tumor behavior. Tumors that look the same do not necessarily behave the same way. The introduction of molecular profiling approaches that enable simultaneous analysis of thousands of molecules marked a new milestone in patient management in kidney and other cancers [1]. This development will lead to substantial improvement of outcome by customizing the management plan for each patient based on the molecular biology of the tumor, as manifested by the molecular expression profile. In this month’s issue of European Urology, Buttner et al identified a molecular signature, named the S3-score, that enables prognostic risk stratification of clear cell renal cell carcinoma (ccRCC) [2]. In this elegant study, the authors hypothesized that the degree of divergence of ccRCC from its cell of origin in the nephron correlates with cancerspecific survival. Patients with high scores, indicating higher similarity to gene expression of the third region of the proximal tubules—the presumed tissue of origin of ccRCC—had longer cancer-specific survival, whereas those with lower scores more frequently had advanced stage tumors, necrosis, and metastasis. The authors compared their panel with two established gene expression signatures: ccA/ccB [3] and ClearCode34 [4]. The S3-score was able to significantly improve the ccA/ccB signature and was superior to ClearCode 34. The innovative approach of this study highlights the great impact that understanding tumor biology has on patient management. Mining different levels of molecular alterations in renal cell carcinoma (RCC) also promises to uncover the biological drivers of tumor progression. In addition to gene expression analysis, other molecular approaches including quantitative proteomics by mass spectrometry [5] and epigenetic changes have been used to identify unique signatures associatedwith aggressive tumor behavior. Other molecules including microRNAs and long noncoding RNAs were also shown to have potential prognostic utility [6]. More recently, the concept of integrated genomics, which compiles multiple levels of molecular alterations, has emerged as a strong tool for the identification of RCC prognostic biomarkers [7]. As shown in recent studies, several events might be required to establish an aggressive phenotype. Studying the interplay between different levels of molecular aberrations and their related biological pathways is key to success. Research is moving a step further by exploring the novel dimension of genotype–phenotype associations in human cancers through the integration of advanced molecular platforms and computational analysis of whole-slide imaging [8]. This approach can substantively improve our understanding of the biology of tumor by correlating molecular changes withmorphologic characteristics including cell–stroma relationships and tumor microenvironment. As much as the paper by Buttner et al [2] highlights the great promise of molecular pathology, it also emphasizes that the journey from bench to bedside is still a work in progress. As noted by the authors, validation with large EU RO P E AN URO LOG Y 6 8 ( 2 0 1 5 ) 1 0 2 1 – 1 0 2 2