| 陆丽红,杨小毛,吴辰,魏云海.胃癌相关肿瘤特征预测淋巴结转移及死亡风险与列线图的构建[J].浙江中西医结合杂志,2021,31(1): |
| 胃癌相关肿瘤特征预测淋巴结转移及死亡风险与列线图的构建 |
| Characteristics of Gastric Cancer-Related Tumors to Predict Lymph Node Metastasis and Death Risk and the Construction of a Line Map |
| 投稿时间：2020-04-20 修订日期：2020-12-25 |
| DOI： |
| 中文关键词: 肿瘤特征 淋巴结转移 风险评估模型 |
| 英文关键词:tumor characteristics lymph node metastasis risk assessment model |
| 基金项目:浙江省医药卫生平台项目(2019ZD050) |
|
| 摘要点击次数: 898 |
| 全文下载次数: 16 |
| 中文摘要: |
| 目的：胃癌相关肿瘤特征预测淋巴结转移及死亡风险与列线图模型的构建。方法：回顾性分析我院肿瘤科胃癌患者的临床病理资料,85例患者被纳入研究。通过单因素和多因素分析胃癌患者的年龄、性别、生存时间、肿瘤位置,大小,浸润深度,区域淋巴结转移,脉管侵犯,神经侵犯,病例类型,分化程度,是否溃疡型等肿瘤特征预测淋巴结的转移以及死亡风险和生存率的评估,并通过R软件的rms包构建淋巴结转移及死亡风险和生存率评估模型的列线图,并得到各项指标的评分。将每个患者的各项指标对应的分数相加,得到模型的总分；总分越高,对应的风险就越高。并且绘制ROC曲线计算列线图建模型预测风险的准确性。结果：在85例胃癌患者中,首先通过多因素分析分析出显著性小于0.05的因素,之后通过单因素分析显示,性别、脉管是否侵犯、病理类型、是否是溃疡性(P<0.05)是淋巴结转移的高危险因素。根据列线图分析建立的淋巴结转移风险评估模型总分为260分,并根据这四项因素建立了生存风险的模型,观测其能生存多久。风险模型包括性别,脉管侵犯,病理类型,溃疡型,病理类型是最危险的因素,随着病理类型逐渐向结节形转变,其淋巴结转移的风险就越大,分值就越高。最佳阈值为126分,在ROC曲线中可知敏感性为0.83,特异性为0.59,AUC为0.785。在预测生存率模型中,病理类型同样也是一个影响生存时间的决定性因素,是否溃疡所带来的影响要大于风险评估中的数值,随着分值的增加,存活率逐渐变小。结论：由性别、脉管侵犯、病理类型、溃疡型构建的列线图风险和存活率模型可以较好地预测胃癌患者淋巴结转移风险和生存率,为胃癌患者提供更精确的临床治疗方案,可以及时的选择如何治疗以及治疗的生存几率。 |
| 英文摘要: |
| Objective: To predict lymph node metastasis and death risk and build a nomogram model for gastric cancer-related tumor characteristics. Methods: The clinical and pathological data of patients with gastric cancer in our oncology department were retrospectively analyzed, and 85 patients were included in the study. Analysis of the age, sex, survival time, tumor location, size, depth of invasion, regional lymph node metastasis, vascular invasion, nerve invasion, case type, degree of differentiation, and whether ulcer type predicts lymph nodes by univariate and multivariate analysis Evaluation of the risk of metastasis and death and survival rate, and a nomogram of lymph node metastasis and death risk and survival rate evaluation model was constructed by the rms package of R software, and scores of various indicators were obtained. Add the scores corresponding to each indicator of each patient to get the total score of the model; the higher the total score, the higher the corresponding risk. And draw the ROC curve to calculate the accuracy of the collinear map construction model to predict the risk.Results: Among 85 patients with gastric cancer, the factors with significance less than 0.05 were first analyzed by multivariate analysis, and then univariate analysis showed that gender, vascular invasion, pathological type, and ulcerative (P <0.05) were lymph node metastases High risk factors. The lymph node metastasis risk assessment model established based on the nomogram analysis has a total score of 260 points, and a survival risk model is established based on these four factors to observe how long it can survive. The risk model includes gender, vascular invasion, pathological type, ulcer type, and pathological type are the most dangerous factors. As the pathological type gradually changes to nodular shape, the greater the risk of lymph node metastasis, the higher the score. The optimal threshold is 126 points. From the ROC curve, the sensitivity is 0.83, the specificity is 0.59, and the AUC is 0.785. In the predictive survival model, the type of pathology is also a decisive factor influencing survival time. Whether the impact of ulcers is greater than the value in the risk assessment. As the score increases, the survival rate gradually decreases.Conclusion: The collinear risk and survival model constructed by gender, vascular invasion, pathological type, and ulcer type can better predict the risk and survival rate of lymph node metastasis in gastric cancer patients, and provide more accurate clinical treatment options for gastric cancer patients. Choose how to treat in a timely manner and the survival chance of the treatment. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|
