Comparison of preoperative nutritional scoring systems for predicting survival in upper tract urothelial carcinoma

Zejing GUO; Changhai SUN; Shicong LAI; Hanlin GAO; Yiqing DU; Tao XU

doi:10.19723/j.issn.1671-167X.2026.04.018

Journal of Peking University(Health Sciences) >

2026 , Vol. 58 >Issue 4: 808 - 817

DOI: https://doi.org/10.19723/j.issn.1671-167X.2026.04.018

Comparison of preoperative nutritional scoring systems for predicting survival in upper tract urothelial carcinoma

Zejing GUO ¹ ,
Changhai SUN ¹^,² ,
Shicong LAI ¹ ,
Hanlin GAO ¹ ,
Yiqing DU ^,¹^,* ,
Tao XU ¹

Expand

1. Department of Urology, Peking University People's Hospital, Beijing 100044, China
2. Department of Urology, Peking University People's Hospital Qingdao Hospital, Qingdao University Affiliated Women and Children's Hospital, Qingdao 266034, Shandong, China

DU Yiqing, e-mail, duyiqing@bjmu.edu.cn

Received date: 2026-02-27

Online published: 2026-05-27

Supported by

the Noncommunicable Chronic Diseases-National Science and Technology Major Project(2024ZD0525700)

the National Natural Science Foundation of China(82471866)

the National Natural Science Foundation of China(82472912)

the Beijing Natural Science Foundation(L252190)

Copyright

Fold

Abstract

Objective: To investigate the association between preoperative nutritional indicators and survival outcomes in patients with upper tract urothelial carcinoma (UTUC) undergoing radical nephroureterectomy (RNU), and to compare the predictive performance of different nutritional scoring systems. Methods: A retrospective analysis was conducted on UTUC patients who underwent RNU. Nutritional scores [geriatric nutritional risk index (GNRI), prognostic nutritional index (PNI), controlling nutritional status (CONUT), triglycerides-total cholesterol-body weight index (TCBI)] were calculated. Variables with P < 0.10 in univariable Cox regression were entered into multivariable analysis using backward stepwise selection based on the Akaike Information Criterion (AIC) to identify independent predictors and construct a nomogram. Kaplan-Meier curves with log-rank tests and time-dependent ROC curves were applied for survival analysis. A two-sided P < 0.05 was considered statistically significant. Results: A total of 481 UTUC patients were included. Multivariable analysis identified pathological T3/T4 stage, N1/N2 stage, high World Health Organization/International Society of Urological Pathology (WHO/ISUP) grade, estimated glomerular filtration rate (eGFR), blood transfusion, ureteral management, GNRI, and PNI as independent predictors of overall survival (OS) (P < 0.05). Independent predictors of cancer-specific survival (CSS) included pathological T2-T4 stage, N1/N2 stage, laparoscopic surgical approach, blood transfusion, GNRI, and PNI (P < 0.05). The nomogram constructed based on these independent predictors demonstrated good calibration. Patients with GNRI < 98 or PNI < 52 had significantly shorter OS and CSS (Log-rank P < 0.05). At the end of 60 months, the area under the curve (AUC) values of GNRI and PNI for predicting OS were 0.623 and 0.632, respectively, and for predicting CSS were 0.632 and 0.645, respectively. Time-dependent ROC analysis showed that both indicators had stable predictive performance for CSS, while their predictive ability for OS showed a mild increasing trend over the follow-up time. Conclusion: Preoperative GNRI and PNI are independently associated with OS and CSS in patients with UTUC after RNU, demonstrating stable predictive performance across different follow-up periods.

Key words： Upper tract urothelial carcinoma; Preoperative nutrition; Overall survival; Cancer-specific survival

Cite this article

Zejing GUO , Changhai SUN , Shicong LAI , Hanlin GAO , Yiqing DU , Tao XU . Comparison of preoperative nutritional scoring systems for predicting survival in upper tract urothelial carcinoma[J]. Journal of Peking University(Health Sciences), 2026 , 58(4) : 808 -817 . DOI: 10.19723/j.issn.1671-167X.2026.04.018

上尿路尿路上皮癌(upper tract urothelial carcinoma，UTUC)是一种起源于肾盂或输尿管尿路上皮组织的恶性肿瘤，约占所有尿路上皮癌的5%~10%^[1]。对于局限性UTUC，根治性肾输尿管切除术(radical nephroureterectomy，RNU)是当前标准的一线外科治疗方式。由于UTUC本身具有较强的侵袭性，其总体预后较膀胱尿路上皮癌更差，术后患者面临较高的疾病进展及死亡风险^[2]。目前临床风险分层主要依赖肿瘤分期、分级及淋巴结状态等病理因素，但此类信息多需在术后方可明确，限制了其在术前治疗决策中的应用价值。因此，寻找能够在术前早期、便捷且可靠地预测UTUC患者复发及生存风险的生物学标志物，对于指导个体化治疗决策和优化随访策略具有重要的临床意义。

近年来，越来越多的研究证据表明，全身营养状态与肿瘤患者的治疗耐受性及长期预后密切相关^[3-7]。老年营养风险指数(geriatric nutritional risk index，GNRI)基于血清白蛋白水平及实际体重与理想体重的比值评估患者的营养储备状况，已经在多种恶性肿瘤中被证实具有良好的预后预测价值^[8]。在UTUC等泌尿系统肿瘤研究中，低GNRI水平与术后并发症发生率升高以及癌症特异性生存期(cancer-specific survival，CSS)和总生存期(overall survival，OS)缩短显著相关^[4]。预后营养指数(prognostic nutritional index，PNI)结合血清白蛋白和外周血淋巴细胞计数，在其他肿瘤领域已被证明是预后危险因素^{[5, 9]}，且在UTUC中低PNI水平与较差的OS和CSS之间的相关性亦有报道^[10-11]。此外，控制营养状态评分(controlling nutritional status，CONUT)通过综合白蛋白、淋巴细胞计数及胆固醇水平评估营养及免疫状态，已被证实与多种恶性肿瘤预后显著相关^[12-13]。甘油三酯-总胆固醇-体重指数(triglycerides-total cholesterol-body weight index，TCBI)主要反映脂质代谢与体重储备状态，其评估侧重与前述营养评分存在差异，在多种慢性疾病相关研究中显示出一定的预测价值^[14-15]，TCBI可能从代谢层面覆盖更多风险信息。

目前，关于营养指标在UTUC预后评估中应用的现有研究存在一定局限性，多数既往研究仅关注单一营养指标，且在结局指标选择上不一致，难以对不同评分体系的预测效能进行直接比较和综合评价，尤其缺乏在同一UTUC患者队列中，系统比较多种常用营养指标在不同生存终点上预测能力的研究。因此，本研究以OS和CSS为主要结局指标，系统评估GNRI、PNI、CONUT及TCBI在同一人群中对UTUC预后的预测能力，旨在识别具有临床推广价值的术前营养评估工具。

1 资料与方法

1.1 一般资料

连续选择2007年1月至2021年4月在北京大学人民医院接受治疗的UTUC患者的病例资料进行回顾性分析。纳入标准：(1)成年UTUC患者；(2)接受根治性肾输尿管切除术；(3)术后组织病理学诊断为尿路上皮癌；(4)随访资料完整，术前检查资料完整。排除标准：(1)确诊即合并其他肿瘤者；(2)合并严重的心、肺、肝、肾功能障碍者；(3)合并慢性感染或自身免疫性疾病者。最终481例患者被纳入，本研究开始前已经北京大学人民医院医学伦理委员会审查批准(批件号：2023PHB194-001)，所有研究程序均符合《赫尔辛基宣言》的要求。

1.2 手术及随访

根治性肾输尿管切除术根据入路不同分为腹腔镜手术和开放手术，通常采用经腹膜后入路，切除范围涵盖患侧肾、全程输尿管以及输尿管膀胱开口周围的部分膀胱壁。对于术前评估提示肿瘤具有侵袭性特征、影像学检查发现可疑淋巴结转移，或术中探查见淋巴结肿大的患者，在根治性肾输尿管切除术的基础上同期行淋巴结清扫术。所有患者术前均完成血常规、尿常规、血生化、凝血功能、胸部X线片及心电图等常规检查，在确认无手术禁忌证后实施手术。术后第3个月进行第一次随访，术后两年内每半年随访一次，两年后每年随访一次，随访采用住院复查、门诊就诊及电话联系的方式。根治性手术后前两年内，每半年一次膀胱镜检查；两年后每年一次膀胱镜检查。每次随访均同步进行血液学、尿液分析以及计算机断层扫描(computed tomography，CT)或磁共振成像(magnetic resonance imaging，MRI)检查。对于临床怀疑存在肿瘤远处转移的患者，酌情增加全身性检查，如骨扫描或正电子发射计算机断层显像(positron emission tomography-computed tomography，PET-CT)。本研究的主要观察终点为OS与CSS，其中，OS定义为从手术日期至因任何原因死亡或末次随访日期的时间间隔；CSS定义为从手术日期至明确因UTUC进展导致死亡或末次随访日期的时间间隔。

1.3 统计学分析

使用R 4.3.2(R Foundation for Statistical Computing, Vienna, Austria)进行统计分析，对于患者的基线特征，分类变量以n (%)表示，符合正态分布的连续变量以${\bar x}$±s表示。采用单因素和多因素Cox回归模型探索影响OS、CSS的独立危险因素，结果以风险比(hazard ratio, HR)及其95%置信区间(confidence interval, CI)表示。采用Cox比例风险模型进行单因素和多因素生存分析。首先对各个变量分别进行单因素Cox回归分析，将单因素分析中P < 0.10的变量以及临床认为重要的基线变量纳入多因素Cox回归模型，基于每个营养指标单独建立模型。多因素分析采用向后逐步回归法进行变量筛选，以赤池信息准则(Akaike Information Criterion, AIC)最小化作为变量保留和剔除的标准。采用Kaplan-Meier法绘制生存曲线，并使用Log-rank检验比较组间生存差异。应用时间依赖性受试者工作特征(receiver operating characteristic, ROC)曲线评估各营养指标对OS和CSS的预测效能，并计算曲线下面积(area under the curve, AUC)。基于多因素Cox回归模型筛选出的独立预后因素，构建列线图以可视化呈现预测模型，所有检验均为双侧检验，P < 0.05认为差异有统计学意义。

2 结果

2.1 病例基本信息

共纳入481例UTUC患者，中位随访时间为56个月(范围：0.7 ~ 198.5个月)，随访期间癌症特异性死亡人数占总体死亡人数的58.9%(96/163)，患者男性248例(51.6%)，女性233例(48.4%)。平均年龄为(66.6±9.78)岁，年龄≥65岁者共289例(60.1%)。肿瘤位于肾盂227例(47.2%)，位于输尿管192例(39.9%)，肾盂和输尿管同时发生62例(12.9%，表 1)。

表1 UTUC患者的临床资料

Table 1 Clinical characteristics of patients with UTUC

Indicator	Total (n=481)
Age/years	66.55±9.78
Gender
Male	248 (51.6)
Female	233 (48.4)
Smoking	73 (15.2)
Previous bladder cancer	30 (6.2)
Concomitant bladder cancer	41 (8.5)
T
Ta/T1	119 (24.7)
T2	121 (25.2)
T3/T4	241 (50.1)
N
Nx/N0	464 (96.5)
N1/N2	17 (3.5)
Location
Ureter	192 (39.9)
Renal pelvis	227 (47.2)
Ureter + Renal pelvis	62 (12.9)
Multifocality	112 (23.3)
WHO/ISUP grade
Low grade	86 (17.9)
High grade	395 (82.1)
eGFR/[mL/(min·1.73 m²)]	64.44±25.42
Surgical approach
Open	39 (8.1)
Laparoscopic	442 (91.9)
Surgical access
Transabdominal	80 (16.6)
Retroperitoneal	401 (83.4)
Distal ureter management
Total laparoscopic	129 (26.8)
Bladder opening	321 (66.7)
Endoscopic-assisted	31 (6.4)
Lymph node dissection	88 (18.3)
Intravesical instillation	31 (6.4)
Blood transfusion	96 (20.0)

Data are n(%) or ${\bar x}$±s UTUC, upper tract urothelial carcinoma; WHO/ISUP, World Health Organization/International Society of Urological Pathology; eGFR, estimated glomerular filtration rate.

2.2 Cox回归分析

单因素和多因素Cox回归分析发现，病理分期T3/T4期、病理分期N1/N2、WHO/ISUP高级别、eGFR、输血、输尿管末端处理方式(打开膀胱及内镜辅助)、GNRI、PNI是UTUC患者术后OS的独立危险因素(P < 0.05，表 2)，病理分期T2~T4期、病理分期N1/N2、术式、输血、GNRI、PNI是术后CSS的独立危险因素(P < 0.05，表 3)。

表2 影响UTUC术后OS的单因素和多因素Cox回归分析

Table 2 Univariate and multivariate Cox analysis of factors influencing OS after UTUC surgery

Variable	Univariate analysis			Multivariate analysis
Variable	HR	95%CI	P	HR	95%CI	P
Age/years	1.027	1.010-1.044	0.002	1.008	0.990-1.026	0.401
Gender^a
Male	Reference
Female	0.863	0.636-1.172	0.346
Smoking index (cigarettes·years)	1.000	0.999-1.000	0.334
Previous bladder cancer^a	1.779	1.045-3.026	0.034	1.396	0.806-2.420	0.234
Concomitant bladder cancer^a	1.552	0.951-2.533	0.079	1.180	0.712-1.958	0.520
T^a
Ta/T1	Reference
T2	2.250	1.235-4.102	0.008	1.681	0.908-3.112	0.098
T3/T4	4.906	2.907-8.278	< 0.001	3.716	2.166-6.374	< 0.001
N^a
Nx/N0	Reference
N1/N2	3.298	1.783-6.098	< 0.001	2.003	1.071-3.748	0.030
Location^a
Renal pelvis	Reference
Ureter	1.022	0.734-1.422	0.899
Ureter + renal pelvis	0.999	0.615-1.622	0.997
Multiple vs. single^a	0.967	0.669-1.396	0.857
WHO/ISUP grade^a
Low grade	Reference
High grade	2.399	1.467-3.923	< 0.001	1.827	1.095-3.046	0.021
eGFR/[mL/(min·1.73 m²)]	0.985	0.978-0.991	< 0.001	0.993	0.986-1.000	0.038
Surgical approach^a
Open	Reference
Laparoscopic	0.571	0.363-0.898	0.015	0.687	0.431-1.096	0.115
Surgical access^a
Transabdominal	Reference
Retroperitoneal	0.932	0.608-1.429	0.748
Distal ureter management^a
Total laparoscopic	Reference
Bladder opening	0.479	0.348-0.660	< 0.001	0.574	0.412-0.798	< 0.001
Endoscopic-assisted	0.431	0.214-0.867	0.018	0.465	0.228-0.950	0.036
Lymph node dissection^a	0.983	0.640-1.509	0.937
Intravesical instillation^a	0.992	0.538-1.831	0.980
Blood transfusion^a	2.117	1.521-2.948	< 0.001	1.807	1.275-2.562	< 0.001
GNRI	0.969	0.953-0.985	< 0.001	0.979	0.963-0.996	0.017
PNI	0.946	0.921-0.972	< 0.001	0.969	0.941-0.997	0.032
CONUT	1.145	1.028-1.274	0.014	1.045	0.932-1.171	0.453
TCBI	1.000	0.999-1.001	0.899

CI, confidence interval; HR, hazard ratio; UTUC, upper tract urothelial carcinoma; OS, overall survival; WHO/ISUP, World Health Organization/International Society of Urological Pathology; eGFR, estimated glomerular filtration rate; GNRI, geriatric nutritional risk index; PNI, prognostic nutritional index; CONUT, controlling nutritional status; TCBI, triglyceride-total cholestorol-body mass index. a, included in the regression model as a categorical variable; reference category as listed.

表3 影响UTUC术后CSS的单因素和多因素Cox回归分析

Table 3 Univariate and multivariate Cox analysis of factors influencing CSS after UTUC surgery

Variable	Univariate analysis			Multivariate analysis
Variable	HR	95%CI	P	HR	95%CI	P
Age/years	1.009	0.989-1.030	0.376
Gender^a
Male	Reference
Female	0.806	0.541-1.200	0.288
Smoking index (cigarettes×years)	0.999	0.998-1.000	0.081	0.999	0.998-1.000	0.104
Previous bladder cancer^a	1.132	0.495-2.585	0.769
Concomitant bladder cancer^a	1.730	0.945-3.166	0.075	1.294	0.690-2.426	0.422
T^a
Ta/T1	Reference
T2	6.232	1.826-21.266	0.003	4.159	1.194-14.484	0.025
T3/T4	16.667	5.259-52.828	< 0.001	11.988	3.708-38.759	< 0.001
N^a
Nx/N0	Reference
N1/N2	4.867	2.523-9.386	< 0.001	2.986	1.530-5.826	0.001
Location^a
Renal pelvis	Reference
Ureter	1.184	0.770-1.822	0.441
Ureter+renal pelvis	1.142	0.613-2.128	0.676
Multiple vs. single^a	1.137	0.722-1.791	0.580
WHO/ISUP grade^a
Low grade	Reference
High grade	3.554	1.644-7.682	0.001	1.838	0.831-4.063	0.133
eGFR/[mL/(min·1.73 m²)]	0.988	0.980-0.996	0.005	0.999	0.989-1.009	0.837
Surgical approach^a
Open	Reference
Laparoscopic	0.475	0.269-0.839	0.010	0.515	0.287-0.925	0.026
Surgical access^a
Transabdominal	Reference
Retroperitoneal	0.637	0.392-1.034	0.068	0.901	0.529-1.534	0.700
Distal ureter management^a
Total laparoscopic	Reference
Bladder opening	0.509	0.334-0.777	0.002	0.646	0.416-1.005	0.053
Endoscopic-assisted	0.672	0.299-1.510	0.335	0.789	0.341-1.824	0.579
Lymph node dissection^a	1.323	0.807-2.168	0.267
Intravesical instillation^a	0.913	0.400-2.087	0.829
Blood transfusion^a	2.024	1.309-3.131	0.002	1.835	1.169-2.880	0.008
GNRI	0.966	0.946-0.987	0.002	0.971	0.950-0.991	0.005
PNI	0.946	0.913-0.980	0.002	0.959	0.924-0.995	0.025
CONUT	1.165	1.017-1.335	0.027	1.122	0.974-1.292	0.110
TCBI	1.000	0.998-1.001	0.511

CI, confidence interval; HR, hazard ratio; UTUC, upper tract urothelial carcinoma; CSS, cancer-specific survival; WHO/ISUP, World Health Organization/International Society of Urological Pathology; eGFR, estimated glomerular filtration rate; GNRI, geriatric nutritional risk index; PNI, prognostic nutritional index; CONUT, controlling nutritional status; TCBI, triglyceride-total cholesterol-body mass index. a, included in the regression model as a categorical variable; reference category as listed.

2.3 GNRI和PNI的ROC曲线

以60个月生存状态为结局，绘制ROC曲线(图 1)，AUC预测OS，GNRI为0.623，PNI为0.632；AUC预测CSS，GNRI为0.632，PNI为0.645。

显示原图|下载原图ZIP|生成PPT

图1 GNRI和PNI预测60月OS(A)和CSS(B)生存情况的ROC

Figure 1 ROC curves of GNRI and PNI in predicting OS (A) and CSS (B) survival at 60 months

GNRI, geriatric nutritional risk index; PNI, prognostic nutritional index; OS, overall survival; CSS, cancer-specific survival; ROC, receiver operating characteristic; AUC, area under the curve.

2.4 GNRI和PNI的Kaplan-Meier生存曲线

参考GNRI分级标准分为两组(≥98 vs. ＜98)，另外参考60个月生存状态对PNI作ROC曲线得出最佳截断值，将OS(≥51.25 vs.＜51.25)和CSS(≥51.82 vs.＜51.82)生存数据各分为两组，分别作Kaplan-Meier生存曲线(图 2)，Log-rank检验提示两组间OS和CSS差异均有统计学意义(P < 0.05)。

显示原图|下载原图ZIP|生成PPT

图2 GNRI(A和B)和PNI(C和D)二分类的生存曲线，风险集人数小于10截断

Figure 2 Kaplan-Meier survival curves for GNRI (A, B) and PNI (C, D) dichotomization, truncated when risk set size < 10

GNRI, geriatric nutritional risk index; PNI, prognostic nutritional index.

2.5 GNRI和PNI的时间依赖性ROC线下面积

自术后6个月起每隔6个月，基于生存状态对GNRI和PNI绘制ROC曲线，其AUC变化趋势见图 3，预测OS时，GNRI和PNI均在30、72、96、150个月达到阶段峰值；预测CSS时，GNRI和PNI均在18、30、72、96个月达到阶段峰值。

显示原图|下载原图ZIP|生成PPT

图3 GNRI(A和B)和PNI(C和D)预测生存AUC随时间变化曲线，AUC阶段峰值标注

Figure 3 Time-dependent AUC curves for GNRI (A, C) and PNI (B, D) predicting survival, with stage peak AUC values marked

GNRI, geriatric nutritional risk index; PNI, prognostic nutritional index; OS, overall survival; CSS, cancer-specific survival; AUC, area under the curve.

2.6 GNRI和PNI的多因素列线图

基于向后逐步回归法多因素Cox回归筛选出的预测因素，构建了整合术前营养指标(GNRI与PNI)的四种列线图见图 4，通过500次Bootstrap重抽样进行内部验证，基于GNRI和PNI构建的列线图在预测患者术后3年及5年OS和CSS时均表现出优异的校准度，经偏差校正后的预测概率与实际随访观察概率高度一致，未见显著的风险高估或低估现象。

显示原图|下载原图ZIP|生成PPT

图4 GNRI预测OS(A)、PNI预测OS(B)、GNRI预测CSS(C)和PNI预测CSS(D)的列线图

Figure 4 Nomograms for GNRI predicting OS (A), PNI predicting OS (B), GNRI predicting CSS (C) and PNI predicting CSS (D)

GNRI, geriatric nutritional risk index; PNI, prognostic nutritional index; OS, overall survival; CSS, cancer-specific survival; pT, pathological tumor stage; pN, pathological node stage; WHO/ISUP, World Health Organization/International Society of Urological Pathology; eGFR, estimated glomerular filtration rate; LG, low grade; HG, high grade.

3 讨论

本研究基于单中心回顾性UTUC根治性肾输尿管切除术患者队列，系统评估了多种临床病理特征及术前营养相关指标与患者预后的关系，结果显示，在校正肿瘤分期、淋巴结状态及肾功能等已知相关因素后，术前GNRI和PNI均为UTUC患者术后OS和CSS的独立危险因素，提示营养状态在UTUC预后评估中具有重要意义。GNRI和PNI在临床工作中易于计算，数据来源客观可靠，在多因素模型和时间依赖的ROC曲线中表现出较为稳定的预测效能，可视化的列线表便于它们应用于临床风险分层。

本研究分别从OS和CSS两个结局指标维度评估了GNRI和PNI的时间依赖性预测能力，结果显示，两者对CSS的预测效能在整个随访期间保持相对稳定，对OS的预测效能则随随访时间延长呈轻度增强趋势。这一差异可能反映了两类结局在生物学基础上的不同，CSS主要受肿瘤侵袭性及机体抗肿瘤免疫状态影响，术前营养状况在一定程度上已整合了肿瘤相关消耗与免疫储备水平，其对CSS的影响在疾病早期即已体现。相比之下，OS除肿瘤因素外，还受感染、心血管事件及慢性肾功能进展等非肿瘤因素影响，营养不良可通过削弱免疫功能、降低抗感染能力影响整体生理状态，表现为对OS预测能力的时间累积效应。总体而言，GNRI和PNI在不同时间节点均表现出相对稳定的预测能力，提示其在短期与长期风险分层中均具有潜在临床价值。在具体应用时，可结合患者年龄结构、炎症状态和基础疾病等情况选择更可靠的指标。

GNRI最初是针对老年群体改良的营养风险指数，多项研究已经证实，GNRI在中青年患者中同样具有适用性和预后预测价值。Qiu等^[16]基于NHANES(National Health and Nutrition Examination Survey)数据库对3 253名40岁及以上的癌症患者分析显示，GNRI在40~60岁和60岁以上两个年龄组均表现出良好的稳定性和适用性。在UTUC和其他多种肿瘤的研究中^{[4, 8, 17-19]}，GNRI也被证实在不同年龄层中与预后存在显著的相关性，这与本文研究结果一致。因此，GNRI的应用不应严格受限于年龄，而应综合考虑体重、白蛋白等指标的可靠性来选择营养评估工具。

除GNRI和PNI外，本研究亦评估了CONUT和TCBI两个常用的营养评分，结果显示，CONUT在单因素分析中与OS和CSS均存在显著相关性，提示其在一定程度上能够反映UTUC患者的整体营养及免疫状态；而TCBI在单因素及多因素分析中均未显示出显著的预测价值。在多因素模型中，CONUT的独立预测效应未能保持，可能与其分级的形式以及各级病例数量不均的队列特征有关^[8]。GNRI和PNI以连续性评分形式被构建，白蛋白水平占据主导因素，反映患者更长期和相对稳定的营养状态，这可能是其在多因素分析中仍保持独立预测价值的重要原因。上述结果提示，不同营养评分在UTUC预后评估中的侧重点和稳定性存在差异，应结合具体人群和研究目的进行选择。对于高龄患者可优先考虑采用GNRI评估，而对于伴有明显炎症状态或体重数据可靠性存疑(如合并严重水肿)的患者，则更适合应用PNI，二者均可作为术前营养评估的一线工具。

既往研究已证实，营养不良及免疫功能受损与多种实体肿瘤的不良结局密切相关^[3-7]。Zheng等^[4]报道，GNRI可作为UTUC患者根治性手术后的独立预后指标；Zheng等^[5]和Xue等^[11]分别报道PNI与OS和CSS显著相关。GNRI基于血清白蛋白和体重指数，反映患者营养储备及消耗状态；PNI结合白蛋白与外周血淋巴细胞计数，综合体现营养和免疫系统情况。白蛋白下降是肿瘤相关慢性炎症对机体影响的综合体现^[20]。促炎因子可通过激活急性期反应，抑制肝白蛋白合成并促进全身蛋白分解代谢^[21-23]。炎症信号可同时通过STAT3和NF-κB信号通路促进细胞增殖、血管生成及上皮-间质转化，增强肿瘤侵袭与转移潜能^[24]。针对尿路上皮癌的研究进一步验证了这一机制。尿路上皮癌细胞被证实可诱导或分泌白介素(interleukin, IL)-1、IL-6及肿瘤坏死因子(tumor necrosis factor, TNF)-α等促炎细胞因子，引发外周脂肪组织的褐变和骨骼肌异常分解, 导致全身性恶病质^[25]，也可以直接刺激肝急性期蛋白质的合成，从而竞争性抑制白蛋白的生成^[26]。同时，泌尿系统肿瘤微环境中的肿瘤相关巨噬细胞与调节性T细胞会构建免疫屏障，导致外周和局部抗肿瘤淋巴细胞的耗竭与功能障碍^[27]。因此，低白蛋白水平既反映宿主营养储备不足，也提示炎症驱动的肿瘤生物学恶性特征。基于炎症与营养对癌症反应的交互作用，C反应蛋白和白蛋白比值等基于白蛋白的复合指标被引入肿瘤预后评估^[28]。UTUC患者常伴有高龄、慢性肾功能损害及反复尿路感染等基础状态^[29]，可进一步加剧系统炎症反应及低白蛋白^[30-31]。近期研究在临床队列中证实，低GNRI水平与UTUC患者组织学高分级和深层浸润显著相关^[4]。GNRI和PNI同时整合了肿瘤负荷、慢性炎症及宿主储备功能等多重信息，使其在UTUC中的预后意义更为全面而敏感。

本研究存在一定局限性。首先，本研究为单中心回顾性分析，结局事件数相对有限，部分指标可能存在Ⅱ类误差、选择偏倚和信息偏倚风险，其结论需要多中心及前瞻性研究的进一步验证；其次，对于不同亚组患者，如年龄分层、合并基础疾病或不同肿瘤分期，可能存在不同的最佳预测模型，本研究尚未进行充分验证；再次，本研究仅基于术前单次指标评估营养状态，未能动态反映围手术期及随访过程中的变化情况；最后，本研究患者均未接受辅助治疗，尽管排除了辅助化疗作为主要混杂因素对生存结局的影响，也限制了将研究结论直接外推至接受辅助化疗的患者群体。未来研究可结合动态营养评估、细化患者分类分层，评估加入营养指标后是否可增强临床病理指标对预后预测的准确性，进一步优化UTUC患者的个体化风险评估。

综上所述，肿瘤病理T分期、N分期、输血以及术前营养状态与上尿路尿路上皮癌患者根治性肾输尿管切除术后的生存结局显著相关。校正病理分期、手术方式及肾功能后，GNRI和PNI仍是OS和CSS的独立预测因素，在不同随访阶段保持相对稳定的区分能力。作为简便、可重复的术前营养评估工具，GNRI和PNI可为UTUC患者的术前风险分层和个体化管理提供有价值的补充信息。

利益冲突 所有作者均声明不存在利益冲突。

作者贡献声明 郭泽菁：分析数据，撰写文章；孙长海、赖世聪：收集数据, 撰写文章；高汉林：整理数据；杜依青、徐涛：指导研究，总体把关和审定论文，经费支持。所有作者均参与论文修改，并对最终文稿进行审读和确认。

References

Publishing order | Descend order by publishing year | Descend order by cited within

1	Ma Z , Yi Y , Qiu Z . Recent trends in incidence, mortality, survival, and treatment of upper tract urothelial carcinoma[J]. Fr J Urol, 2024, 34 (2): 102573. DOI

2	Peyronnet B , Seisen T , Dominguez-Escrig JL , et al. Oncological outcomes of laparoscopic nephroureterectomy versus open radical nephroureterectomy for upper tract urothelial carcinoma: An European association of urology guidelines systematic review[J]. Eur Urol Focus, 2019, 5 (2): 205- 223. DOI

3	Teoh JY , Ng CF , Eto M , et al. Radical nephroureterectomy for UTUC conferred survival benefits irrespective of age and comorbidities[J]. World J Urol, 2022, 40 (11): 2657- 2665. DOI

4	Zheng L , Ye J , Wu Q , et al. Geriatric Nutritional Risk Index as a prognostic marker for predicting survival outcomes in patients with UTUC after radical nephroureterectomy[J]. Sci Rep, 2025, 15, 8836. DOI

5	Zheng Y , Yu D , Yu Z , et al. Association of preoperative systemic Immune-inflammation index and prognostic nutritional index with survival in patients with upper tract urothelial carcinoma[J]. J Cancer, 2020, 11 (19): 5665- 5677. DOI

6	Yu Z , Xiong Z , Ma J , et al. Prognostic and clinicopathological significance of systemic immune-inflammation index in upper tract urothelial carcinoma: A meta-analysis of 3 911 patients[J]. Front Oncol, 2024, 14, 1342996. DOI

7	Marra A , Bondesan A , Caroli D , et al. Complete blood count (CBC)-derived inflammation indexes are useful in predicting metabolic syndrome in adults with severe obesity[J]. J Clin Med, 2024, 13 (5): 1353. DOI

8	Huang YC , Chen SW , Chiang YS . Prognostic role of geriatric nutritional risk index (GNRI) and controlling nutritional status (CONUT) on outcomes in patients with head and neck cancer: A systematic review and meta-analysis[J]. BMC Cancer, 2025, 25 (1): 242. DOI

9	Pinato DJ , North BV , Sharma R . A novel, externally validated inflammation-based prognostic algorithm in hepatocellular carcinoma: The prognostic nutritional index (PNI)[J]. Br J Cancer, 2012, 106 (8): 1439- 1445. DOI

10	尹健, 赵华才, 薛苗新, 等. 术前预后营养指数在上尿路上皮癌根治性手术患者中的临床应用价值[J]. 临床医学研究与实践, 2024, 9 (16): 59- 63.

11	Xue W , Tan P , Xu H , et al. Impact of the preoperative prognostic nutritional index on survival outcomes in upper tract urothelial carcinomas[J]. Cancer Med, 2019, 8 (6): 2971- 2978. DOI

12	Kuroda D , Sawayama H , Kurashige J , et al. Controlling Nutritional Status (CONUT) score is a prognostic marker for gastric cancer patients after curative resection[J]. Gastric Cancer, 2018, 21 (2): 204- 212. DOI

13	Tokunaga R , Sakamoto Y , Nakagawa S , et al. CONUT: A novel independent predictive score for colorectal cancer patients undergoing potentially curative resection[J]. Int J Colorectal Dis, 2017, 32 (1): 99- 106. DOI

14	Sun L , Yang Y , Yan R , et al. Higher Controlling Nutritional Status (CONUT) score indicates increased risk of sarcopenia in elderly hospitalized patients: A single institution study in China[J]. Front Nutr, 2025, 12, 1669225. DOI

15	Xu Y , Yan Z , Li K , et al. Association between nutrition-related indicators with the risk of chronic obstructive pulmonary disease and all-cause mortality in the elderly population: Evidence from NHANES[J]. Front Nutr, 2024, 11, 1380791. DOI

16	Qiu X , Wu Q , Zhang Y , et al. Geriatric nutritional risk index and mortality from all-cause, cancer, and non-cancer in US cancer survivors: NHANES 2001-2018[J]. Front Oncol, 2024, 14, 1399957. DOI

17	Yamamoto S , Hashimoto Y , Takahashi F , et al. Association between muscle quality and GNRI in patients with type 2 diabetes[J]. Nutrients, 2026, 18 (2): 275. DOI

18	Zhou H , Lv D , Cui F , et al. Prognostic value of the geriatric nutritional risk index in patients with non-metastatic clear cell renal cell carcinoma: A propensity score matching analysis[J]. Nutr J, 2024, 23 (1): 114. DOI

19	Minami S , Miyoshi N , Fujino S , et al. The geriatric nutritional risk index as a prognosis predictor in patients with rectal cancer receiving neoadjuvant chemotherapy[J]. Anticancer Res, 2022, 42 (7): 3759- 3766. DOI

20	McMillan DC . The systemic inflammation-based glasgow prognostic score: A decade of experience in patients with cancer[J]. Cancer Treat Rev, 2013, 39 (5): 534- 540. DOI

21	Argilés JM , Busquets S , Stemmler B , et al. Cancer cachexia: Understanding the molecular basis[J]. Nat Rev Cancer, 2014, 14 (11): 754- 762. DOI

22	Almasaudi AS , Dolan RD , Edwards CA , et al. Hypoalbuminemia reflects nutritional risk, body composition and systemic inflammation and is independently associated with survival in patients with colorectal cancer[J]. Cancers, 2020, 12 (7): 1986. DOI

23	Al-Shaiba R , McMillan DC , Angerson WJ , et al. The relationship between hypoalbuminaemia, tumour volume and the systemic inflammatory response in patients with colorectal liver metastases[J]. Br J Cancer, 2004, 91 (2): 205- 207. DOI

24	Mantovani A , Allavena P , Sica A , et al. Cancer-related inflammation[J]. Nature, 2008, 454 (7203): 436- 444. DOI

25	Nagai T , Naiki T , Sugiyama Y , et al. A narrative review of peripheral blood parameters for urothelial carcinoma treated with systemic antitumor drugs[J]. Transl Androl Urol, 2023, 12 (5): 790- 801. DOI

26	O'Brian D , Prunty M , Hill A , et al. The role of C-reactive protein in kidney, bladder, and prostate cancers[J]. Front Immunol, 2021, 12, 721989. DOI

27	Peng L , Du C , Meng C , et al. Controlling nutritional status score before receiving treatment as a prognostic indicator for patients with urothelial cancer: An exploration evaluation methods[J]. Front Oncol, 2021, 11, 702908. DOI

28	Matsumoto T , Itoh S , Yoshizumi T , et al. C-reactive protein: Albumin ratio in patients with resectable intrahepatic cholangiocarcinoma[J]. BJS Open, 2020, 4 (6): 1146- 1152. DOI

29	Rouprêt M , Babjuk M , Burger M , et al. European association of urology guidelines on upper urinary tract urothelial carcinoma: 2020 update[J]. Eur Urol, 2021, 79 (1): 62- 79. DOI

30	Stenvinkel P , Heimbürger O , Paultre F , et al. Strong association between malnutrition, inflammation, and atherosclerosis in chronic renal failure[J]. Kidney Int, 1999, 55 (5): 1899- 1911. DOI

31	Kalantar-Zadeh K , Ikizler TA , Block G , et al. Malnutrition-inflammation complex syndrome in dialysis patients: Causes and consequences[J]. Am J Kidney Dis, 2003, 42 (5): 864- 881. DOI

Options

Outlines

模态框（Modal）标题

Abstract

Cite this article

1 资料与方法

1.1 一般资料

1.2 手术及随访

1.3 统计学分析

2 结果

2.1 病例基本信息

表1 UTUC患者的临床资料

2.2 Cox回归分析

表2 影响UTUC术后OS的单因素和多因素Cox回归分析

表3 影响UTUC术后CSS的单因素和多因素Cox回归分析

2.3 GNRI和PNI的ROC曲线

图1 GNRI和PNI预测60月OS(A)和CSS(B)生存情况的ROC

2.4 GNRI和PNI的Kaplan-Meier生存曲线

图2 GNRI(A和B)和PNI(C和D)二分类的生存曲线，风险集人数小于10截断

2.5 GNRI和PNI的时间依赖性ROC线下面积

图3 GNRI(A和B)和PNI(C和D)预测生存AUC随时间变化曲线，AUC阶段峰值标注

2.6 GNRI和PNI的多因素列线图

图4 GNRI预测OS(A)、PNI预测OS(B)、GNRI预测CSS(C)和PNI预测CSS(D)的列线图

3 讨论

References