消化道肿瘤患者接种新型冠状病毒疫苗后免疫原性的预测模型构建

A Predictive Model to Estimate Immunogenicity after Receipt of SARS-CoV-2 Vaccines in People with Gastrointestinal Cancer

目的:本研究旨在寻找影响消化道肿瘤患者血清抗体滴度降低的主要因素,并应用诺曼图构建预测模型。方法:我们采用单因素和多因素Logistic回归分析筛选消化道肿瘤患者血清抗体滴度降低的独立危险因素。基于回归分析的结果,以回归系数为基础,建立了相应的诺曼图预测模型,我们通过一致性指数(C-index)值、曲线下面积(AUC)、校准曲线和决策曲线分析(DCA)来评估诺曼图的性能和区分能力。结果:多因素Logistic回归分析显示,性别、ASA评分、进行积极治疗和化疗是新冠病毒中和抗体低的独立危险因素,ASA评分和化疗是RBD抗体低的独立危险因素。两种模型之间无统计学差异。我们使用Bootstrap方法进行了内部验证,NABS抗体模型的C指数为0.803,RBD抗体模型的C指数为0.799。校准曲线显示,患者的实际血清学阳性率与预测的血清学阳性率一致。临床决策曲线的结果表明,这两种预测模型在临床上是有用的。结论:这项研究提供了两个诺模图,可以预测胃肠道癌症患者的血清学反应。每个诺模图代表新冠肺炎疫苗的一种抗体,其中包含一些危险因素。它们方便为后续接种新冠疫苗的消化道肿瘤患者并提供参考信息。

Objective: This study aimed to find the main factors that determine the lower serum anti-body titer and to establish a predictive model by using nomogram. Methods: We used univariate and multi-variate logistic regression analyses to screen the independent risk factors of lower serum anti-body titer associated with the patients with gastrointestinal cancer. Based on the regression coefficients, the corresponding nomogram prediction model was drawn. We established two models, nomogram Nabs, (neutralizing antibodies) and nomogram RBD (anti-receptor binding domain (RBD)-IgG) based on two antibodies. We evaluated nomogram performance and discriminative power by con-cordance index (C-index) values, area under the curve (AUC), calibration curve and decision curve analyses (DCA). Results: Multivariate logistic regression analysis indicated that the four risk factors including gender, ASA, active treatment and chemotherapy were independent risk factors of lower neutralizing antibodies (Nabs), while ASA and chemotherapy were independent risk factors of low-er anti-receptor binding domain (RBD)-IgG. There existed no statistical difference between the two models. We used the bootstrap method to perform an internal verification, and the C index of nom-ogram Nabs was 0.803, while the C index of nomogram RBD was 0.799. The calibration curves showed that the actual serological positive rate of the patients was consistent with the predicted serological positive rate. The DCA demonstrated that the nomograms were clinically useful. Conclu-sions: This study presents two nomograms, which can predict seronegativity in patients with gas-trointestinal cancer. Each nomogram represents a type of antibody induced by the COVID-19 vac-cine and incorporates some risk factors. They are convenient and provide customizing information.