Predictive models for the practical management of renal cell carcinoma

Abstract

The expanding availability of multiple therapeutic strategies and sequencing options for patients with renal cell carcinoma (RCC) has increased the importance of skilled individualized outcome estimation for patients. This need has driven the development of statistical models to guide patient management in a variety of common clinical settings, including the management of small renal masses, identification of patients with high-risk localized RCC requiring systemic therapy and selection of suitable targeted therapies in metastatic disease. With an increasing number of different predictive models described in the literature, identifying those models most relevant for practical use is challenging. In addition to statistical models based on clinical data, there has also been an evolution towards incorporation of molecular markers into predictive algorithms. These models also serve as important benchmarks for the researchers developing novel prognostic and predictive molecular biomarkers.