Abstract: Background　Gastric cancer is a common digestive tract cancer worldwide that seriously endangers the life and health of residents. Nutritional status has been reported to be closely related to cancer patients' prognosis，which is assessed by prognostic nutritional index（PNI） in increasing studies to predict long-term prognosis of cancer patients. Objective　To evaluate the prognostic value of PNI in gastric cancer patients with lung metastasis，and to establish a PNI-based prognostic nomogram. Methods　Participants were 245 gastric cancer patients with lung metastasis who were admitted to the First Affiliated Hospital of Xinjiang Medical University from July 2010 to July 2018. By reviewing medical records and test reports，demographic and clinicopathological data were collected，including age，sex，primary lesion，histological grade，T stage，N stage，type of lung metastasis，timing of lung metastasis，chemotherapy status，tumor diameter，prevalence of pleural effusion and intravascular tumor thrombus，primary lesion surgery，serum CEA，serum CA199，number of peripheral lymphocytes and albumin level one week before operation，and calculated PNI. Follow-up data were collected through a telephone or clinic follow-up till April 2019. The influencing factors of PNI and its association with the prognosis were analyzed. A prognostic nomogram was constructed based on potential prognostic factors of gastric cancer patients with lung metastasis. Results　The optimal cut-off value of PNI for predicting 3-year survival in gastric cancer patients with lung metastasis was 46.1 with 71.7% sensitivity and 67.8% specificity. Multivariate Logistic regression analysis showed that age >65 years old〔OR=2.123，95%CI（1.201，3.752）〕，advanced T stage〔OR=2.104，95%CI（1.336，3.313）〕，advanced N stage〔OR=1.876，95%CI（1.286，2.735）〕 and tumor diameter> 5 cm 〔OR=1.839，95%CI（1.054，3.208）〕were independent risk factors for PNI<46.1（P<0.05）. The 1-year，2-year and 3-year survival rates of gastric cancer patients with lung metastasis were 33.3%，15.1% and 4.3%，respectively. The 3-year survival rate of low PNI group（n=114，PNI<46.1） was lower than that of high PNI group（n=131，PNI>46.1）（P=0.001）. Multivariate Cox regression analysis showed that histological grade〔HR=1.414，95%CI（1.059，1.887）〕，type of lung metastasis〔HR=1.647，95%CI（1.213，2.237）〕，chemotherapy〔HR=0.740，95%CI（0.559，0.981）〕，primary lesion surgery 〔HR=0.649，95%CI（0.475，0.886）〕 and PNI〔HR=0.733，95%CI（0.550，0.978）〕 were independent prognostic factors for gastric cancer patients with lung metastasis（P<0.05），and all of which were used to construct the prognostic nomogram to predict the 1-year，2-year and 3-year overall survival with a C-index of 0.755. Conclusion　PNI may be a major prognostic factor for gastric cancer patients with lung metastasis. And the prognostic nomogram based on PNI is helpful for individualized prognosis analysis.

Key words: Stomach neoplasms, Nutrition assessment, Prognostic nutrition index, Lung metastasis of gastric cancer, Nomogram

摘要： 背景　胃癌是全球范围内常见的消化道恶性肿瘤之一，严重危害居民的生命和健康。有研究表明肿瘤患者的预后与营养状态具有较高的相关性。近年来越来越多的研究者将预后营养指数（PNI）纳入肿瘤的预后研究中，并通过PNI评价患者的营养状态及预测患者的远期预后。目的　探讨PNI对胃癌肺转移患者预后的评估价值，并建立列线图预测模型。方法　收集2010年7月至2018年7月新疆医科大学第一附属医院诊治的245例胃癌肺转移患者的住院病历及检测报告，分析其临床病理特征，包括年龄、性别、原发灶、组织学分级、T分期、N分期、肺转移类型、肺转移时间、化疗情况、肿瘤直径、血清癌胚抗原（CEA）水平、血清CA199水平、胸腔积液发生情况、脉管癌栓发生情况、原发灶手术情况和术前1周外周血淋巴细胞总数、血清白蛋白水平等，并计算PNI。对患者进行电话或门诊随访，随访截止时间为2019年4月。分析PNI的影响因素及其对胃癌肺转移患者预后的影响，根据患者预后的影响因素建立列线图预测模型。结果　PNI预测胃癌肺转移患者生存结局的最佳截断值为46.1〔依据截断值，将患者分为低PNI值组（n=114）和高PNI值组（n=131）〕，灵敏度为71.7%，特异度为67.8%。多因素Logistic回归分析结果显示，年龄>65岁〔OR=2.123，95%CI（1.201，3.752）〕、T分期高〔OR=2.104，95%CI（1.336，3.313）〕、N分期高〔OR=1.876，95%CI（1.286，2.735）〕、肿瘤直径>5 cm〔OR=1.839，95%CI（1.054，3.208）〕是导致胃癌肺转移患者PNI低的独立危险因素（P<0.05）。胃癌肺转移患者1年、2年、3年生存率分别为33.3%、15.1%、4.3%。低PNI值组3年生存率为1.0%，低于高PNI值组的11.6%（P=0.001）。多因素Cox回归模型分析结果显示，组织学分级〔HR=1.414，95%CI（1.059，1.887）〕、肺转移类型〔HR=1.647，95%CI（1.213，2.237）〕、化疗〔HR=0.740，95%CI（0.559，0.981）〕、原发灶手术〔HR=0.649，95%CI（0.475，0.886）〕、PNI〔HR=0.733，95%CI（0.550，0.978）〕为胃癌肺转移患者预后的独立影响因素（P<0.05）。绘制列线图，纳入多因素Cox回归模型中有统计学意义的影响因素，对胃癌肺转移患者1、2、3年总体生存率进行预测，一致性指数（C-index）为0.755。结论　PNI对胃癌肺转移患者的预后具有重要的预测价值，以此为基础构建的列线图预测模型有利于临床上对患者的预后进行个体化分析。

关键词: 胃肿瘤, 营养评价, 预后营养指数, 胃癌肺转移, 列线图