Han Zongqiang, Wen Lina, Feng Linlin
Department of Laboratory Medicine, Beijing Xiaotangshan Hospital, Beijing, China.
Clinical Nutrition, Beijing Shijitan Hospital, Capital Medical University, Beijing, China.
Ann Transl Med. 2020 Sep;8(18):1186. doi: 10.21037/atm-20-6071.
Serum samples of patients with hemorrhagic cerebral infarction (HCI), cerebral infarction (CI), and healthy controls (HCs) were used to screen statistically different protein peaks as potential biomarkers and to establish a decision tree classification model.
The serum samples from clinically confirmed patients with HCI and CI from November 2018 to October 2019 were collected, along with those of HCs who visited our hospital during the same period. Surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF-MS) with CM10 ProteinChip was used to analyze the differences in serum protein expression profiles of 30 patients with HCI, 32 patients with CI, and 31 HCs in the training group, and a decision tree classification model was established. At the same time, the blind test group (18 patients with HCI, 21 patients with CI, and 17 HCs) was tested by a blind method.
Model 1 was successfully established by software analysis with a mass-to-charge ratio of 3,495.2, 8,941.0, and 15,890.4 as a differential protein peak. The sensitivity, specificity, and accuracy of model 1 in distinguishing HCI from HCs were 86.8%, 87.1%, and 86.9%, respectively. After verification of model 1 by the blind test group, the results showed that the sensitivity, specificity, and accuracy were 88.9%, 94.1%, and 91.4%, respectively. The sensitivity, specificity, and accuracy of model 2 with a mass-to-charge ratio of 2,941.3 as a differential protein peak were 86.7%, 75.0%, and 80.6%, respectively. After verification of model 2 by the blind test group, the results showed that the sensitivity, specificity, and accuracy were 83.3%, 90.4%, and 87.2%, respectively.
Surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF-MS) and CM10 ProteinChip can be used to screen serum protein markers in patients with HCI. Mass-to-charge ratio of 3,495.2, 8,941.0, 15,890.4, and 2,941.3 may be potential protein biomarkers of HCI and used to distinguish HCI patients from HCs and CI.
采用出血性脑梗死（HCI）患者、脑梗死（CI）患者及健康对照者（HCs）的血清样本，筛选具有统计学差异的蛋白峰作为潜在生物标志物，并建立决策树分类模型。
收集2018年11月至2019年10月临床确诊的HCI和CI患者的血清样本，以及同期来我院就诊的HCs的血清样本。采用CM10蛋白芯片的表面增强激光解吸/电离飞行时间质谱（SELDI-TOF-MS）分析训练组中30例HCI患者、32例CI患者和31例HCs的血清蛋白表达谱差异，并建立决策树分类模型。同时，对盲测组（18例HCI患者、21例CI患者和17例HCs）进行盲法检测。
通过软件分析成功建立模型1，以质荷比3495.2、8941.0和15890.4作为差异蛋白峰。模型1区分HCI与HCs的灵敏度、特异度和准确度分别为86.8%、87.1%和86.9%。经盲测组验证模型1，结果显示灵敏度、特异度和准确度分别为88.9%、94.1%和91.4%。以质荷比2941.3作为差异蛋白峰的模型2的灵敏度、特异度和准确度分别为86.7%、75.0%和80.6%。经盲测组验证模型2，结果显示灵敏度、特异度和准确度分别为83.3%、90.4%和87.2%。
表面增强激光解吸/电离飞行时间质谱（SELDI-TOF-MS）和CM10蛋白芯片可用于筛选HCI患者的血清蛋白标志物。质荷比3495.2、8941.0、15890.4和2941.3可能是HCI的潜在蛋白生物标志物，可用于区分HCI患者与HCs及CI患者。
