Wang Hsin-Yao, Hsieh Chia-Hsun, Wen Chiao-Ni, Wen Ying-Hao, Chen Chun-Hsien, Lu Jang-Jih
Department of Laboratory Medicine, Chang Gung Memorial Hospital at Linkou, Taoyuan City, Taiwan.
Division of Hematology-Oncology, Department of Internal Medicine, Chang Gung Memorial Hospital at Linkou and Chang Gung University, Taoyuan City, Taiwan.
PLoS One. 2016 Jun 29;11(6):e0158285. doi: 10.1371/journal.pone.0158285. eCollection 2016.
Analytic measurement of serum tumour markers is one of commonly used methods for cancer risk management in certain areas of the world (e.g. Taiwan). Recently, cancer screening based on multiple serum tumour markers has been frequently discussed. However, the risk-benefit outcomes appear to be unfavourable for patients because of the low sensitivity and specificity. In this study, cancer screening models based on multiple serum tumour markers were designed using machine learning methods, namely support vector machine (SVM), k-nearest neighbour (KNN), and logistic regression, to improve the screening performance for multiple cancers in a large asymptomatic population.
AFP, CEA, CA19-9, CYFRA21-1, and SCC were determined for 20 696 eligible individuals. PSA was measured in men and CA15-3 and CA125 in women. A variable selection process was applied to select robust variables from these serum tumour markers to design cancer detection models. The sensitivity, specificity, positive predictive value (PPV), negative predictive value, area under the curve, and Youden index of the models based on single tumour markers, combined test, and machine learning methods were compared. Moreover, relative risk reduction, absolute risk reduction (ARR), and absolute risk increase (ARI) were evaluated.
To design cancer detection models using machine learning methods, CYFRA21-1 and SCC were selected for women, and all tumour markers were selected for men. SVM and KNN models significantly outperformed the single tumour markers and the combined test for men. All 3 studied machine learning methods outperformed single tumour markers and the combined test for women. For either men or women, the ARRs were between 0.003-0.008; the ARIs were between 0.119-0.306.
Machine learning methods outperformed the combined test in analysing multiple tumour markers for cancer detection. However, cancer screening based solely on the application of multiple tumour markers remains unfavourable because of the inadequate PPV, ARR, and ARI, even when machine learning methods were incorporated into the analysis.
血清肿瘤标志物的分析测量是世界某些地区（如台湾）癌症风险管理常用的方法之一。最近，基于多种血清肿瘤标志物的癌症筛查经常被讨论。然而，由于敏感性和特异性较低，风险效益结果对患者似乎不利。在本研究中，使用机器学习方法，即支持向量机（SVM）、k近邻（KNN）和逻辑回归，设计了基于多种血清肿瘤标志物的癌症筛查模型，以提高对大量无症状人群中多种癌症的筛查性能。
对20696名符合条件的个体测定甲胎蛋白（AFP）、癌胚抗原（CEA）、糖类抗原19-9（CA19-9）、细胞角蛋白19片段（CYFRA21-1）和鳞状细胞癌抗原（SCC）。对男性测量前列腺特异性抗原（PSA），对女性测量糖类抗原15-3（CA15-3）和糖类抗原125（CA125）。应用变量选择过程从这些血清肿瘤标志物中选择稳健的变量来设计癌症检测模型。比较了基于单一肿瘤标志物、联合检测和机器学习方法的模型的敏感性、特异性、阳性预测值（PPV）、阴性预测值、曲线下面积和尤登指数。此外，评估了相对风险降低、绝对风险降低（ARR）和绝对风险增加（ARI）。
为了使用机器学习方法设计癌症检测模型，女性选择了CYFRA21-1和SCC，男性选择了所有肿瘤标志物。SVM和KNN模型在男性中显著优于单一肿瘤标志物和联合检测。所有3种研究的机器学习方法在女性中均优于单一肿瘤标志物和联合检测。对于男性或女性，ARR在0.003 - 0.008之间；ARI在0.119 - 0.306之间。
在分析多种肿瘤标志物进行癌症检测方面，机器学习方法优于联合检测。然而，即使将机器学习方法纳入分析，仅基于多种肿瘤标志物的应用进行癌症筛查仍然不利，因为PPV、ARR和ARI不足。
