Recurrence Risk Stratification in Clear Cell Renal Cell Carcinoma Using TNM 2017 and WHO/ISUP Grade

Background/Aim: To re-evaluate established recurrence risk models for clear cell renal cell carcinoma (ccRCC) using the tumor, node, metastasis (TNM) 2017 classification and the World Health Organization/International Society of Urological Pathology (WHO/ISUP) grading system, and to perform an exploratory analysis of whether these updates improve the predictive accuracy of recurrence risk models.

Patients and Methods: We retrospectively analyzed 295 patients who underwent surgery for non-metastatic ccRCC at Osaka Metropolitan University Hospital between 2013 and 2020. Pathological features were re-evaluated using both the TNM 2009 and TNM 2017 classifications, as well as the Fuhrman and WHO/ISUP grading systems. We examined the frequency of pathological upstaging to pT3a and compared recurrence predictions across grading systems. In addition, we assessed recurrence risk stratification using the Leibovich 2003 and 2018 models with updated pathological classifications. Recurrence-free survival (RFS) was assessed using Kaplan–Meier analysis, and prognostic accuracy was evaluated using Harrell’s concordance index (C-index).

Results: Among cT1 cases, pathological upstaging to pT3a occurred more frequently according to TNM 2017 (18.9%) than TNM 2009 (16.0%). TNM 2017 showed superior prognostic performance for predicting recurrence in cT1 tumors (C-index, 0.748 vs. 0.726). The WHO/ISUP grade demonstrated greater predictive accuracy than the Fuhrman grade (C-index, 0.817 vs. 0.737). The C-index of the Leibovich 2003 model improved from 0.812 (TNM 2009 plus Fuhrman classification) to 0.834 (TNM 2017 plus the WHO/ISUP classification). The highest predictive accuracy was achieved using the Leibovich 2018 model (C-index, 0.870).

Conclusion: Re-evaluation using TNM 2017 and WHO/ISUP grading was associated with improved predictive accuracy of recurrence risk models for localized ccRCC and may aid postoperative risk stratification.