肿瘤学杂志

快速检索			高级检索

基于营养-免疫-炎症评分构建晚期胰腺癌化疗患者预后模型

Development and Validation of a Prognostic Model Based on Nutritional-Immune-Inflammatory Score in Advanced Pancreatic Cancer Patients Undergoing Chemotherapy

投稿时间：2025-11-06 修订日期：2026-05-17

DOI：

中文关键词: 胰腺癌预后炎症免疫营养列线图

英文关键词:pancreatic cancer prognosis inflammation immunity nutrition nomogram

基金项目:

作者	单位	邮编
朱仔雯	徐州医科大学附属医院肿瘤内科	221000
罗鑫婷	徐州医科大学附属医院肿瘤内科
项蓉蓉	徐州医科大学附属医院肿瘤内科
荣天明	徐州医科大学附属医院肿瘤内科
王红兵^*	徐州医科大学附属医院肿瘤内科	221000

摘要点击次数: 12

全文下载次数: 0

中文摘要:

[目的]构建基于营养－免疫－炎症相关生物标志物的新型风险评分，并建立列线图模型用于晚期胰腺癌化疗患者生存分析。[方法]回顾性收集179例晚期胰腺癌化疗患者的临床资料，随机按照7:3比例划分为训练集（n=125）和验证集（n=54），通过LASSO（Least Absolute Shrinkage and Selection Operator）-Cox回归模型筛选最具预测价值的变量，构建风险评分。采用Kaplan-Meier法进行风险评分分组的生存分析。根据多因素Cox回归分析筛选独立预后因素，并建立列线图，采用一致性指数（Concordance Index，C-index）、校准曲线（Calibration Curve）和决策曲线（Decision Curve Analysis，DCA）对列线图可信性进行评估。[结果]基于预后营养指数（Prognostic Nutritional Index，PNI）、白蛋白-碱性磷酸酶比值（Albumin-Alkaline Phosphatase Ratio，AAPR）、中性粒细胞-淋巴细胞比值（Neutrophil-to-Lymphocyte Ratio，NLR）、白蛋白-胆红素评分（Albumin-Bilirubin Score，ALBI）构建风险评分公式，Kaplan-Meier生存曲线示高风险组的总生存率显著低于低风险组（P<0.001）。多因素分析结果显示，风险评分[风险比（hazard ratio，HR）=4.113，95%置信区间（confidence interval，CI）=2.284~7.408，P<0.001）、CA19-9>37U/mL（HR=1.997，95%CI=1.099~3.629，P=0.023）和CA125>35U/mL（HR=1.979，95%CI=1.194~3.279，P=0.008）是影响总生存率的独立危险因素，基于此构建列线图。训练集与验证集的一致性指数分别为0.746和0.716，校准曲线及决策曲线均提示列线图具有良好的准确性和有效性。[结论]该风险评分可作为晚期胰腺癌化疗患者个体化预后评估有效预测工具，为临床风险分层及治疗方案选择提供依据。

英文摘要:

[Objective] To develop a new risk score based on nutritional-immune-inflammatory biomarkers and to construct a nomogram for survival analysis among patients with advanced pancreatic cancer undergoing chemotherapy. [Methods] This study retrospectively analyzed the clinical data of 179 patients with advanced pancreatic cancer undergoing chemotherapy.Patients were randomly divided into a training set (n=125) and a validation set (n=54) at a ratio of 7:3. Predictive variables were selected through LASSO-Cox regression for constructing the risk score, and a prognostic risk score was constructed based on the selected variables. Survival outcomes were evaluated using the Kaplan-Meier method based on risk score stratification. Independent prognostic factors were identified through multivariate Cox proportional hazards regression analyses. Subsequently, a nomogram was integrated, and the reliability and clinical utility of the nomogram were validated through C-index, calibration curves, and decision curve analysis (DCA). [Results] The risk score was constructed based on four biomarkers: the Neutrophil-to-Lymphocyte Ratio (NLR), the Albumin-Bilirubin (ALBI) score, the Albumin-Alkaline Phosphatase Ratio (AAPR), and the Prognostic Nutritional Index (PNI). The Kaplan-Meier survival analysis demonstrated that patients in the high-risk group had significantly poorer overall survival than those in the low-risk group (P<0.0001). Multivariate Cox regression analysis identified risk score [hazard ratio (HR) = 4.113, 95% confidence interval (CI) = 2..284~7.408, P<0.001], CA19-9 (HR=1.997, 95%CI=1.099~3.629, P=0.023) and CA125 (HR=1.979, 95%CI=1.194~3.279, P=0.008) as independent predictors of overall survival. A nomogram incorporating these variables was subsequently developed. The C-index values for the training and validation sets were 0.746 and 0.716 respectively. Both the calibration and decision curves indicated that the nomogram demonstrated good accuracy and validity. [Conclusions] The proposed risk score may serve as an effective tool for individualized survival prediction in patients with advanced pancreatic cancer undergoing chemotherapy, thereby facilitating clinical risk stratification and treatment decision-making.

在线阅读查看/发表评论下载PDF阅读器