炎症指标和肿瘤标记物联合检测对术前胃癌分期的诊断价值

Diagnostic values from combined detection of both inflammatory and tumor markers for preoperative gastric cancer staging

目的:探讨术前炎症指标中性粒细胞/淋巴细胞比值(NLR)、炎症指数(SII)及肿瘤标志物CEA、CA19-9、CA72-4单独及联合检测与胃癌(GC)临床分期的关系,并建立列线图预测模型。方法:收集2019年1月—2022年5月郑州大学第二附属医院普外科收治的192例胃癌患者临床资料。检测早期胃癌组和进展期胃癌组术前NLR、SII、白蛋白及血清CEA、CA19-9、CA72-4水平,分析两组数据之间的差异,再通过受试者工作特征(ROC)曲线确定各检测指标的截断值并绘制Logistic回归模型的列线图来评价炎症指标和肿瘤标记物单独及联合检测对术前胃癌分期的诊断价值。结果:进展期胃癌组术前NLR、SII、CEA、CA19-9、CA72-4水平均高于早期胃癌组(P<0.05)。ROC曲线分析结果显示,NLR、SII、CEA、CA19-9、CA72-4单独检测的曲线下面积(AUC)分别为0.817、0.792、0.722、0.624、0.761,最佳截断值分别为2.09、366.73、2.18、7.44、3.63。各指标联合检测的AUC为0.944,敏感度为89.3%,特异度为89.2%,构建包含NLR、SII、CEA、CA19-9、CA72-4和临床分期的列线图预测模型,模型内部验证一致性指数(C指数)为0.915,校正曲线提示该模型准确性良好。结论:术前NLR、SII、CEA、CA19-9、CA72-4与胃癌的临床分期相关,基于术前NLR、SII、CEA、CA19-9、CA72-4水平及临床分期构建的列线图模型具有良好的准确性和临床实用性。

OBJECTIVE:To investigate relationships among neutrophil/lymphocyte ratios(NLR),inflammation indices(SII),tumor markers CEA,CA19-9,CA72-4 and the clinical stages of gastric cancer(GC);and to develop a nomogram prediction model.METHODS:From January 2019 to May 2022,clinical data from 192 gastric cancer patients who were admitted to the General Surgery Department of the Second Affiliated Hospital of the Zhengzhou University were collected.The preoperative NLR,SII,albumin and serum CEA,CA19-9 and CA72-4 levels of early and advanced gastric cancer groups were determined.Differences in the collected data between the two groups of patients were analyzed.Then,the cut-off value for each test index was determined by the receiver working characteristic(ROC)curve.The nomogram of Logistic regression model was drawn to evaluate the diagnostic value of the markers alone and in combinations for preoperative gastric cancer stages.RESULTS:All preoperative NLR,SII,CEA,CA19-9,and CA72-4 levels were higher in the advanced gastric cancer group than in the early gastric cancer group(P<0.05).The ROC curve analysis showed that the area under the curve(AUC)of NLR,SII,CEA,CA19-9,CA72-4 alone were 0.817,0.792,0.722,0.724,and 0.761,respectively,and the best cut-off values were 2.09,366.73,2.18,7.44,and 3.63,respectively.The AUC for each index was 0.944,the sensitivity was 89.3%,and the specificity was 89.2%.A nomogram prediction model including NLR,SII,CEA,CA19-9,CA72-4 and clinical stage was constructed.The internal validation consistency index(C index)was 0.915,and the correction curve suggested that the accuracy of the model was good.CONCLUSION:Preoperative NLR,SII,CEA,CA19-9,and CA72-4 were associated with clinical stages of gastric cancers.In addition,the nomogram model which was constructed based on preoperative NLR,SII,CEA,CA19-9,CA72-4 levels and clinical stage had good accuracy and clinical utility.