Background The effect of racial/ethnic group on survival in upper tract urothelial carcinoma (UTUC) is unknown. We tested this concept in non-metastatic UTUC patients treated with radical nephroureterectomy (RNU) and hypothesized that important differences may exist according to racial/ethnic groups.

Material and Methods We relied on the Surveillance Epidemiology and End Results database (2004-2016). We relied on Propensity-score matching (ratio 1:4). Subsequently, cumulative incidence plots and multivariable competing risks regression models (CRR) addressed cancer-specific mortality (CSM).

Results Of 9129 assessable patients, 7454 (81.7%) were Caucasian vs. 665 (7.3%) Hispanic vs. 584 (6.4%) Asian vs. 426 (4.7%) African-American. No statistically significant differences were recorded for tumor grade or T-stage, between all racial/ethnic groups. However, within patents who received lymph-node dissection (n = 2694, 29.5%), Asians exhibited the highest rate of more than 2 positive lymph nodes at RNU (19.0%, followed by 17.1% African-Americans, 15.0% Caucasians and 12.6% Hispanics, P < 0.001). After PS-matching and multivariable CRR, Asian race/ethnicity independently predicted higher CSM, relative to Caucasians (Hazard ratio: 1.29, P < 0.01). No statistically significant differences according to CSM was recorded in the remaining races/ethnicities comparisons (all P ≥ 0.1)

Conclusion Important CSM differences may exist according to race/ethnicity in non-metastatic UTUC patients treated with RNU. However, these differences only apply to Asian patients, who account for 6% of the overall non-metastatic UTUC cohort treated with RNU. In consequence, in clinical practice Asian patients should be given particular attention with the intent of reducing the CSM disadvantage that cannot be clearly explained by stage and/or grade disadvantage at diagnosis.

背景种族/族裔群体对上尿路上皮癌 (UTUC) 生存率的影响尚不清楚。我们在接受根治性肾输尿管切除术 (RNU) 治疗的非转移性 UTUC 患者中测试了这一概念，并假设根据种族/族裔群体可能存在重要差异。

材料和方法我们依赖于监测流行病学和最终结果数据库（2004-2016）。我们依靠倾向得分匹配（比率 1:4）。随后，累积发病率图和多变量竞争风险回归模型 (CRR) 解决了癌症特异性死亡率 (CSM)。

结果在 9129 名可评估患者中，7454 名 (81.7%) 为白种人，665 名 (7.3%) 西班牙裔与 584 名 (6.4%) 亚裔与 426 名 (4.7%) 非洲裔美国人。在所有种族/民族组之间，没有记录到肿瘤分级或 T 阶段的统计学显着差异。然而，在接受淋巴结清扫的患者中（n = 2694, 29.5%），亚洲人在 RNU 的淋巴结阳性率最高（19.0%，其次是 17.1% 的非洲裔美国人、15.0% 的白种人和 12.6 % 西班牙裔，P < 0.001）。在 PS 匹配和多变量 CRR 之后，相对于白种人，亚洲种族/民族独立预测更高的 CSM（风险比：1.29，P <0.01)。在剩余的种族/民族比较中，根据 CSM 没有记录到统计学上的显着差异（所有P ≥ 0.1）

结论在接受 RNU 治疗的非转移性 UTUC 患者中，根据种族/民族可能存在重要的 CSM 差异。然而，这些差异仅适用于亚洲患者，他们占接受 RNU 治疗的整个非转移性 UTUC 队列的 6%。因此，在临床实践中，应特别关注亚洲患者，以减少诊断时无法通过分期和/或等级劣势清楚解释的 CSM 劣势。