宫颈癌患者血清蛋白指纹图谱的检测及其意义

背景与目的:目前针对宫颈癌没有特异性的肿瘤标志物。表面增强激光解吸离子化飞行时间质谱(surface-enhanced laser desorption/ionization time-of-flight mass spectrometry,SELDI-TOF-MS)是最新应用的一项检测肿瘤标志物的技术。本研究应用SELDI-TOF-MS检测宫颈癌患者血清蛋白指纹图谱,筛选候选肿瘤标志物并建立诊断模型,初步探讨其在宫颈癌早期诊断中的价值。方法:取91例早期宫颈鳞癌患者和15例宫颈上皮内瘤变Ⅲ级(cervical intraepithelialneoplasia,CINⅢ)患者的血清标本进行实验,同时用55名健康人血清作为对照。用弱阳离子交换芯片(weak cation exchange,WCX2)检测各血清标本获得血清蛋白指纹图谱。用Biomarker Patterns软件分析宫颈癌差异蛋白并建立诊断模型。通过盲法分析进一步验证诊断模型的可靠性,并对结果进行统计学分析。结果:在分子量1.5～20ku范围内,共检测到122个蛋白峰,其中19个差异峰有统计学意义(P<0.001)。建立了由分子量为3977和5807的两个差异蛋白组成的宫颈癌诊断模型,其敏感性为97.29%(36/37),特异性为83.78%(31/37)。扩大样本盲法验证结果,其敏感性为94.44%(51/54),特异性为94.44%(17/18)。结论:由3977和5807两个差异蛋白组成的宫颈癌诊断模型有助于区分宫颈癌和健康人群。

Application of SELDI-TOF serum proteome profiling in cervical squamous cell carcinoma

BACKGROUND & OBJECTIVE: Up to now, there is no valid biomarker in early diagnosis of cervical cancer. Surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF-MS) is a new technique used to identify biomarkers for cancers. This study was to screen new biomarkers and build diagnostic models for early diagnosis of cervical cancer by SELDI-TOF-MS. METHODS: SELDI-TOF-MS was used to detect the serum proteomic patterns of 91 patients with early stage cervical squamous cell carcinoma, 15 patients with cervical intraepithelial neoplasia III (CIN III), and 55 healthy women (control). The serum proteomic spectra were generated on weak cation exchange (WCX2) chips. Differences in protein peaks were analyzed using Biomarker Wizard software. The diagnostic model was built by Biomarker Patterns software and further valuated by a large-scale blind test. RESULTS: A total of 122 protein peaks were detected at the molecular range of 1.5 to 20 ku, among which 19 ones were significantly different between invasive cervical squamous cell carcinomas and controls (P<0.001). A diagnostic model consisting of 2 protein peaks at 3,977 m/z and 5,807 m/z was established. Its specificity was 83.78% (31/37) and its sensitivity was 97.29% (36/37). A sensitivity of 94.44% (51/54) and a specificity of 94.44% (17/18) in a large-scale blind test were obtained. CONCLUSION: The diagnostic model consisting of 2 protein peaks at 3,977 m/z and 5,807 m/z can discriminate cervical cancer patients from healthy women.