**Background:** Upper tract urothelial carcinoma (UTUC) accounts for 5–10% of urothelial carcinomas, with a higher incidence in Taiwan (40.2% of all UCs) due to arsenic water contamination and other factors. Radical nephroureterectomy (RNU) with bladder cuff excision is the standard treatment, but oncologic outcomes have remained unchanged. Malnutrition is common in cancer patients and may lead to cachexia, poor treatment response, and worse prognosis. The Geriatric Nutritional Risk Index (GNRI), calculated from serum albumin and the ratio of actual to ideal body weight, is a simple screening tool initially designed for hospitalized elderly patients. While GNRI has been studied in various malignancies, no study had examined its association with localized UTUC.
**Methods:** This retrospective chart review included 488 patients with pathologically confirmed UTUC who underwent RNU at Taichung Veterans General Hospital from January 2001 to December 2015. GNRI was calculated as: 1.489 × serum albumin (g/L) + 41.7 × (actual body weight/ideal body weight), with ideal body weight defined as [height(m)]² × 22. Malnutrition was defined as GNRI < 92. Patients were followed with periodic monitoring including laboratory studies, urine cytology, imaging, and cystoscopy. Endpoints were disease-free survival (DFS), cancer-specific survival (CSS), and overall survival (OS). ROC curve analysis, Kaplan-Meier survival curves, and Cox proportional hazards models were used.
**Key Results:** Of 488 patients, 102 (20.9%) were in the malnutrition group (GNRI < 92) and 386 (79.1%) in the normal nutrition group (GNRI ≥ 92). The malnutrition group had significantly lower median albumin (3.1 vs 4.0 g/dL, p<0.001), lower BMI (22.9 vs 24.2 kg/m², p<0.001), and were older (median age 70.0 vs 67.0 years, p=0.023). The malnutrition group also had more advanced disease features: higher pathological N stage (p<0.001), positive surgical margins (16.7% vs 8.5%, p=0.026), and lymphovascular invasion (34.3% vs 18.7%, p=0.001). The 5-year survival rates in the normal vs malnutrition groups were: DFS 48.6% vs 28.0%, CSS 80.5% vs 53.2%, and OS 80.5% vs 40.0%. Median OS was 30.16 months in the malnutrition group vs not reached in the normal group (p<0.001). Median DFS was 10.97 vs 52.93 months (p<0.001). On multivariate analysis, GNRI < 92 was an independent risk factor for OS (HR=3.94, 95% CI 2.70–5.74, p<0.001), DFS (HR=1.90, 95% CI 1.42–2.54, p<0.001), and CSS (HR=5.42, 95% CI 3.24–9.06, p<0.001). Other independent risk factors for OS included age (HR=1.04, 95% CI 1.02–1.06), positive surgical margin (HR=1.78, 95% CI 1.13–2.82), pathological T3 (HR=2.54, 95% CI 1.53–4.21), pathological T4 (HR=6.75, 95% CI 3.17–14.37), and lymphovascular invasion (HR=1.81, 95% CI 1.16–2.81). The ROC analysis showed an AUC of 0.69 for GNRI predicting OS. Perioperative complications did not differ significantly between groups.
**Clinical Implications:** Preoperative GNRI is a simple, objective, and independent prognostic marker for UTUC patients undergoing RNU. Low GNRI may reflect both patient frailty and sequelae of advanced disease, as lower GNRI scores were associated with higher pathological T stage, N stage, positive surgical margins, and lymphovascular invasion. The GNRI could be incorporated into preoperative risk stratification to identify high-risk patients who may benefit from nutritional intervention, closer postoperative surveillance, or more aggressive multimodal therapy. The study is limited by its retrospective design, changes in surgical technique over time (including inconsistent template lymph node dissection before 2008), and the fact that only 23.9% of patients received adjuvant chemotherapy. Prospective studies are needed to validate these findings and assess the impact of nutritional optimization on outcomes.