Concordance between clinical and pathologic staging of T2a–b and T3a renal cell carcinoma.

600 Background: Among patients with renal cell carcinoma (RCC), 5–23% of those with cT1 and smaller cT2 tumors may be upstaged to pT3a disease after surgery. The pathologic restaging rate of larger or clinically more invasive tumors is understudied and has implications for perioperative systemic therapy clinical trial enrollment. We examined rates of pathologic restaging for cT2a-b and cT3a RCC after surgery. Methods: Using the National Cancer Database, we identified adult patients with cT2a, cT2b, and cT3a RCC undergoing partial or radical nephrectomy. We compared pathologic restaging rates between the clinical stage groups using a Chi-square test. Subgroup analysis of restaging rates was performed for histology (clear cell vs non–clear cell), clinical nodal status (cN1 vs cN0), and clinical metastatic status (cM1 vs cM0). Sensitivity, specificity, positive predictive value, and negative predictive value were calculated for the overall cohort and subgroups. Multivariable logistic regression was performed to assess predictors of pT3a upstaging among patients with cT2a and cT2b tumors. Results: We identified 31,912 patients who met inclusion criteria (13,840 cT2a, 8,079 cT2b, and 9,993 cT3a). he overall rate of restaging was 47.4% for cT2a disease (10.7% downstaged and 36.7% upstaged) for a sensitivity and specificity of 89% and 72%, respectively. The overall rate of restaging was 52.5% for cT2b disease (10.6% downstaged and 41.9% upstaged), for a sensitivity and specificity of 85% and 96%, respectively. The overall rate of restaging was 9.5% for cT3a disease (5.1% downstaged and 4.4% upstaged), for a sensitivity and specificity of 55% and 94%, respectively (p < 0.001). The positive predictive value was 52.5%, 47.5%, and 90.6%, and the negative predictive value was 95%, 97%, and 67% for cT2a, cT2b, and cT3a disease, respectively. On multivariable analysis, among patients with cT2 tumors, the presence of clinically node-positive (OR 2.8 [95% CI 2.43–3.17]) metastatic (OR 2.4 [95% CI 2.18–2.65]) or cT2b disease (OR 1.18 [95% CI 1.10–1.26]) was associated with pathologic upstaging (p < 0.001). Conclusions: Restaging is common for patients with cT2a and cT2b tumors, but the specificity of cT3a staging is high. These findings suggests that opportunities exist to improve the accuracy of clinical staging for cT2 tumors, which will have implications for future adjuvant/perioperative RCC clinical trials.