New Test Indicates Likelihood of Renal Cell Carcinoma Recurrence

By Brian Rini, MD

A deeper understanding of the molecular basis of renal cell carcinoma (RCC) is required to better assess a patient’s risk of recurrence after nephrectomy for localized disease.

Several markers have been explored in RCC, but none has clinical utility. Our research group sought to identify and validate a novel prognostic gene signature to improve prediction of the recurrence risk of clear cell (cc) RCC, the most common subtype of kidney cancer. Approximately 30 percent of patients treated for localized ccRCC relapse.

Methodology and validation

By analyzing RNA-based tumor gene expression, we characterized a large cohort of patients with localized ccRCC who underwent curative-intent nephrectomy. We created an observational cohort consisting of 942 patients with clinical stage I-III ccRCC who underwent nephrectomy between 1985 and 2003 at Cleveland Clinic and had fixed, paraffin-embedded tumor blocks available.

RNA was extracted from the tumor blocks and expression of 732 genes was analyzed using quantitative reverse transcription-polymerase chain reaction. Primary analysis was to evaluate the degree of association between gene expression and recurrence-free interval and was conducted using Cox proportional hazards models.

These data resulted in a final list of 11 cancer-related and five reference genes most strongly associated with recurrence. Predominant gene families significantly associated with recurrence (after adjustment for the clinical/pathologic factors and accounting for false discovery) included those responsible for angiogenesis and immune response, in addition to cell cycle and cell adhesion.

Subsequently, we performed a validation study of this gene signature using an independent sample set of 645 nephrectomy samples from Institute Gustave Roussy. The recurrence score (derived from a formula of weighted gene expression results) was associated with disease recurrence (hazard ratio 3.91 for each 25-unit increase in score; p < 0.0001. This association remained significant in multivariate analysis accounting for known clinicopathologic parameters.

Figure 1. Risk profiles of continuous recurrence score (RS) versus five-year recurrence risk by stage in the validation study. The continuous curves showing the association between RS and five-year risk of recurrence were generated with the use of a log-hazardratio model stratified by stage (green for stage I and blue for stages II and III) using a 2-degree-of-freedom spline. The dashed curves indicate 95 percent confidence intervals. The dots in the box below the x-axis indicate the distribution of RS by stage.

Figure 2. Forest plot illustrating the performance of gene groups for development and validation studies. Standardized hazard ratios for each group were calculated by dividing the gene expression by the standard deviation (SD) across all patients. The squares indicate standardized hazard ratio point estimates for each gene group, and whiskers are 95 percent confidence intervals.

(Reprinted from, Vol. 16, Lancet Oncology. Rini B, Goddard A, Knezevic D, et al. A 16-gene assay to predict recurrence after surgery in localised renal cell carcinoma: development and validation studies. 2015 Jun;16(6):676-685. Used with permission from Elsevier.)

An aid for ongoing treatment decisions

This prognostic test in RCC could be highly useful in treatment decisions for localized RCC, especially regarding the management of small renal masses.

Additional large-scale efforts are required to further extend such data from prognostic to predictive — that is, investigating whether the identified genes are not only prognostic for recurrence, but also predictive for outcome with currently available systemic therapies in the advanced setting.

Large-scale adjuvant trials are pending with targeted agents and likely in the future with checkpoint inhibitors, and such a test, if predictive, may allow for enhanced patient selection in the adjuvant setting to best balance risk and benefit.

Dr. Rini is the Director of Cleveland Clinic’s Genitourinary Cancer Program and a staff member of the Department of Hematology and Medical Oncology and the Department of Urology. He is a Professor of Medicine at Cleveland Clinic Lerner College of Medicine.