Study reveals lower predictive ability for renal cancer models

Study reveals lower predictive ability for renal cancer models

(HealthDay)—Prospective validation of existing renal cell carcinoma (RCC) prediction models demonstrates a decrease in their predictive ability, according to a study published online June 19 in the Journal of Clinical Oncology.

Andres F. Correa, M.D., from the Fox Chase Cancer Center in Philadelphia, and colleagues validated eight RCC recurrence models and the Tumor, Node, Metastasis (TNM) staging system using data from 1,647 patients with resected, localized, high-grade or locally advanced disease from the ASSURE trial cohort. Model predictive performance was quantified by assessing discriminatory and calibration abilities.

The researchers identified a in each model's predictive ability in prospective validation compared with their original and externally validated discriminatory estimates. The SSIGN model performed best and the UISS model performed worst (C-statistics, 0.688 and 0.556, respectively). Most of the models marginally outperformed standard staging compared with the 2002 TNM staging system (C-index, 0.60). Significant variability was seen for the predictive ability of all models over time, and they were most useful within the first two years of diagnosis.

"Medicine has come to rely on prognostic models and has put a lot of time and resources into developing them, but they are far less robust than we would have hoped," a coauthor said in a statement. "In some cases, using these models to predict future events is not much better than the flip of a coin."

Two authors disclosed financial ties to the pharmaceutical industry.

Explore further

AI model uses serial imaging to predict lung cancer therapy response

More information: Abstract/Full Text (subscripti … ent may be required)
Journal information: Journal of Clinical Oncology

Copyright © 2019 HealthDay. All rights reserved.

Citation: Study reveals lower predictive ability for renal cancer models (2019, June 24) retrieved 27 May 2020 from
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.

Feedback to editors

User comments