ASCO GU 2018: U-SMART: A Novel Scoring System of Preoperative Predictors to Stratify Oncologic Risk of Small Renal Mass

San Francisco, CA ( Dr. Kendrick Yim from UC San Diego presented their institution’s data assessing U-SMART, a novel scoring system of preoperative predictors to stratify oncologic risk of small renal mass. Small renal masses (< 4 cm in diameter) are heterogeneous, with significant proportions of benign as well as high-grade malignancy. The objective of this study was to develop a scoring system incorporating patient factors, serum markers, and morphometric characteristics to elucidate benign and high-grade pathology and guide decision making.

For this study, the authors performed a retrospective analysis of the UC San Diego database of surgically treated small renal masses from 2003-2017, identifying 312 patients that fit inclusion criteria. Patients were categorized into three groups: benign, low-grade, or high-grade disease. Demographic and clinical factors collected included gender, ALT (U/L, RENAL score, and tumor diameter (cm). Univariable and multivariable logistic regression models were used to screen for an association between potential parameters and the three groups. Each significant variable was analyzed by risk group and broken into quartiles. The 75th percentile of the high-grade group was assigned a value of 3, and below the 75th percentile of the benign group was assigned a value of 1. Values that fell between these cutoffs were assigned 2 points. Tumor diameter was weighted twice that of other factors. Receiver-operating-characteristic (ROC) analysis was used to assess for predictive capability.

Among these 312 patients, there were 65 benign (20.8%), 204 low-grade (65.4%), and 43 high-grade (13.8%) lesions. Factors associated with increased risk of high-grade tumors were male sex (OR 1.868, p = 0.045), higher ALT (OR 1.036, p = 0.022), higher RENAL score (OR 1.318, p = 0.002), and larger tumor diameter (OR 2.415, p < 0.001):


Furthermore, patients with low (5-8), intermediate (9-11) and high (12-14) scores had 32.8%, 5.2%, and 0% frequency of benign pathology. Patients with low, intermediate, and high scores had 7.7%, 18.6%, and 34.9% frequency of high-grade tumors. ROC analysis revealed AUC of 0.767.

The authors concluded that preoperative clinical parameters incorporated into a model significantly predicted benign and high-grade pathology for small renal masses. This risk stratification may provide a non-invasive method to aid in clinical decision making. As the authors aptly point out, this model requires external validation.

Presented by: Kendrick Yim, University of California San Diego, San Diego, CA

Co-Authors: Ahmet Bindayi, Stephen Ryan, Madhumitha Reddy, Ryan Nasseri, Fang Wan, Christopher Long, Zachary Hamilton, Ithaar Derweesh; University of California San Diego, San Diego, CA; University of California San Diego, La Jolla, CA; Saint Louis University, St Louis, MO

Written by: Zachary Klaassen, MD, Urologic Oncology Fellow, University of Toronto, Princess Margaret Cancer Centre, @zklaassen_md at the 2018 American Society of Clinical Oncology Genitourinary (ASCO GU) Cancers Symposium, February 8-10, 2018 - San Francisco, CA