In China, elderly patients with non-small cell lung cancer (NSCLC) accounts for the highest percentage of lung cancer patients, and most of them are found to have no surgical possibilities at the time of diagnosis. Moreover, these patients are increasing as aging advances. Increasing their survival rate will help to reduce the overall mortality of lung cancer patients. So identifying effective prognostic indicators in NSCLC patients with non-surgical treatment is of great significance in prognostic stratification, which also contributes to clinical studies aiming at improving the survival rate of such patients via prognostic stratification-based treatments.

To investigate the relationship between systemic immune-inflammatory-nutritional index (SIINI) and survival in non-surgically treated elderly patients with NSCLC.

Patients (n=231, ≥65 years old) with first treatment for NSCLC were retrospectively recruited from Hebei General Hospital from January 1, 2014 to June 30, 2018. Clinical characteristics were collected, mainly including age, sex, prevalence of smoking, baseline diseases, BMI, pathology, differentiation, and clinical stage of NSCLC. Some calculated data based on baseline routine blood test parameters, and/or serum albumin, and/or BMI using different approaches were also collected, including neutrophil to lymphocyte ratio (NLR) , derived NLR (dNLR) , platelet to lymphocyte ratio (PLR) , prognostic nutrition index (PNI) , systemic immune-inflammation index (SII) , advanced lung cancer inflammatory index (ALI) and SIINI 〔using a formula proposed in clinical retrospective studies, in which all variables are measured before treatment: (neutrophil count×platelet count×hemoglobin level) / (lymphocyte count×BMI×serum albumin level) 〕. Post-treatment follow-up was conducted till February 1, 2020 through outpatient reexamination, telephone or text messages with death as the endpoint. For assessing prognostic values of NLR, dNLR, PLR, PNI, SII, ALI and SIINI, ROC analysis was performed with defined optimal cut-off value and the area under the curve (AUC) for each indicator (if the AUC value is less than 0.5, then the optimal cut-off value is defined using the median value, by which the AUC value is defined as large or small when it is greater or less than the value) . The survival curves were comparatively analyzed by different patient characteristics. Cox regression analysis was applied to identify the influencing factors of survival. The survival rate curve was visualized using GraphPad Prism 8.0.2.

The optimal cut-off values using NLR, dNLR, PLR, PNI, SII, ALI and SIINI in assessing the prognosis were 3.30, 2.51, 179.99, 273.65, 736.54, 46.05 and 102.89, respectively. The survival curves varied significantly by age, sex, prevalence of smoking, pathology, differentiation, and clinical stage of NSCLC, NLR, dNLR, PLR, ALI, SII, PNI and SIINI (P<0.05) . Further analysis indicated that the difference between the survival curves of 65-70-year-olds and 76-and-over-year-olds was statistically significant (P<0.05) . The survival curves between those with low or moderate differentiation and those with high differentiation were significantly different (P<0.05) . The survival curves of patients with stageⅠ NSCLC were different from those of patients with stage Ⅱ, Ⅲ or Ⅳ NSCLC (P<0.05) . Cox regression analysis revealed that ≥76 years old (P<0.001) , highly differentiated NSCLC (P<0.001) , stage Ⅲ NSCLC (P=0.012) and Ⅳ NSCLC (P<0.001) and SIINI (P=0.001) were prognostic factors of patients. Moreover, there existed significant differences in survival curves by NLR, dNLR, PLR, ALI, SII, PNI, and SIINI (P<0.05) .

We found that SIINI, a new indictor calculated based on immunity, inflammation and nutrition factors, is effective in predicting the overall survival in non-surgically treated elderly patients with NSCLC, and it may be superior to NLR, dNLR, PLR, PNI, SII, ALI in terms of survival prediction-related application and in-depth research.

老年非小细胞肺癌（NSCLC）患者数量居于我国肺癌患者数量的首位，其中大多数老年NSCLC患者在确诊时已经失去手术机会。随着我国人口老龄化进程加剧，该类患者数量逐年增加。提高这部分患者的生存率将会降低整体肺癌患者的死亡率。探索识别接受非手术治疗患者的预后指标对患者预后分层具有重要作用，并为后续根据分层采取不同诊疗策略以提高患者生存率的临床研究奠定基础。

探讨系统免疫炎症营养指数（SIINI）与非手术治疗的老年NSCLC患者预后的关系。

回顾性分析2014-01-01至2018-06-30于河北省人民医院初治的、年龄≥65岁的231例NSCLC患者的临床资料。收集患者的临床特征，主要包括年龄、性别、吸烟、基础疾病、体质指数（BMI）、病理类型、分化程度、临床分期。根据治疗前血常规、血清白蛋白、BMI计算出中性粒细胞/淋巴细胞比值（NLR）、衍生的中性粒细胞/淋巴细胞比值（dNLR）、血小板/淋巴细胞比值（PLR）、预后营养指数（PNI）、系统免疫炎症指数（SII）、晚期肺癌炎症指数（ALI）和SIINI〔基于临床回顾性数据探索发现，将治疗前中性粒细胞计数、淋巴细胞计数、血小板计数、血红蛋白水平、血清白蛋白水平、BMI以中性粒细胞计数×血小板计数×血红蛋白水平/（淋巴细胞计数×BMI×血清白蛋白水平）的计算方式得出〕。绘制各指标判断老年NSCLC患者预后的受试者工作特征（ROC）曲线，确定临界值并应用ROC曲线下面积（AUC）评价其诊断价值，若AUC<0.5，则根据中位数取临界值，根据各指标临界值分为高、低水平。治疗结束后通过门诊复查、电话或短信等形式进行随访，随访终点事件为死亡，随访时间截至2020-02-01。比较不同特征患者的生存曲线，采用多因素Cox比例风险回归模型分析探讨患者生存的影响因素，并采用Graphpad 8.0.2绘制生存曲线。

NLR、dNLR、PLR、PNI、SII、ALI和SIINI的临界值分别为3.30、2.51、179.99、273.65、736.54、46.05和102.89。不同年龄、性别、吸烟情况、病理类型、分化程度、临床分期、NLR、dNLR、PLR、ALI、SII、PNI、SIINI患者的生存曲线比较，差异有统计学意义（P<0.05）；65~70岁与≥76岁患者生存曲线比较，差异有统计学意义（P<0.05）。低分化、中分化患者与高分化患者生存曲线比较，差异有统计学意义（P<0.05）；临床分期Ⅱ期、Ⅲ期、Ⅳ期患者生存曲线与Ⅰ期患者比较，差异有统计学意义（P<0.05）。多因素Cox比例风险回归模型分析结果显示，年龄≥76岁（P<0.001）、高分化（P<0.001）、临床分期Ⅲ期（P=0.012）、临床分期Ⅳ期（P<0.001）和SIINI（P=0.001）是患者预后的影响因素。不同NLR、dNLR、PLR、ALI、SII、PNI、SIINI患者生存曲线比较，差异有统计学意义（P<0.05）。

基于临床回顾性数据探索发现综合免疫、炎症、营养等形成的新指标SIINI是预测非手术治疗的老年NSCLC患者预后的有效指标，相较于NLR、dNLR、PLR、PNI、SII、ALI，其在预后判断方面可能更具有应用价值及深层次的研究价值。