Patient and tumor characteristics of histological subtypes of renal cell carcinoma and its risk of upstaging to ≥pT3

Recurrence after surgery for clear cell and papillary renal cell carcinoma: Head-to head comparison of validated risk scores

INTRODUCTION

The American Urology Association (AUA) recently introduced in their guidelines a subtype-agnostic, 4-tiered risk classification score to assess oncologic outcomes after surgery in patients with localized renal cell carcinoma (RCC). We provide a head-to-head comparison of the AUA score with 3, internationally validated and EAU recommended, histological-specific models.

MATERIALS AND METHODS

We retrieved from a prospectively-maintained database 2,560 surgically-treated patients with localized RCC in a single high-volume European center (1987-2023). Patients were classified into risk categories, according to 3 histology-specific prognostic models (i.e., Leibovich 2003, Leibovich 2018 and VENUSS). Kaplain-Meier (KM) analyses estimated 5-year progression-free (PFS) and cancer-specific (CSS) survival. Area under curves (AUCs) of each prognostic model to predict clinical progression or cancer-specific mortality (CSM) were calculated, according to histological subtypes.

RESULTS

Overall, 2,145 (84%) patients harbored clear-cell (ccRCC), whereas 415 (16%) harbored papillary (pRCC) RCC. Within ccRCC, proportions of low vs. intermediate vs. high/very-high risk patients ranged from 57 to 58% vs. 17 to 28% vs. 6.3 to 19%, respectively. In pRCC, proportions of low vs. intermediate vs. high/very-high risk patients ranged from 54 to 73% vs. 21 to 24% vs. 4.8 to 12%, respectively. Leibovich 2018 and VENUSS exhibited the highest AUCs for clinical progression (0.805 vs. 0.884) and CSM (0.847 vs. 0.862) prediction within ccRCC vs. pRCC, respectively. The AUA model demonstrated AUCs of 0.783 vs. 0.864 for clinical progression and 0.797 vs. 0.826 for CSM in ccRCC vs. pRCC, respectively.

CONCLUSION

Within a European cohort of RCC patients, the AUA prognostic stratification exhibited reliable performance. Nevertheless, histology-specific models continue to demonstrate a substantial advantage in predicting RCC outcomes.