整合生物标志物与临床表型以精确预测子痫前期发病风险

Pre-eclampsia remains an important cause of maternal and perinatal morbidity and mortality. The multifactorial pathogenesis of pre-eclampsia (PE) has not been fully elucidated. No single test met the clinical standards for a predictive test. Improving the prediction of pre-eclampsia has been the focus of a significant amount of research. Recent studies suggested that PE prediction performance based on clinical phenotypes may be improved by incorporating biomarkers. The purpose of this research is to develop the pre-eclampsia classification system based on that system’s ability to predict risk of PE in routine practice in China. Case-control study nested in a cohort will be incorporated to prospective collection of information on clinical risk factors for PE in routine clinical practice. Based on our pilot experiment on possible DNA copy number variations of PE, further identification and verification of reliable and accurate biomarkers will be conducted with larger sample size. Integrated multivariable predictive models for PE will then be constructed on specific key clinical risk factors present in early pregnancy alone or in combination with biomarkers. The predictive performance of different models will be compared and determined, so as to provide an evidence-based tool to identify pregnant women at increased risk of PE in the settings of current models of antenatal care, particularly in the public system.

子痫前期（PE）严重威胁母婴健康，由于病因不明，临床缺乏有效治疗措施。建立PE早期预测方法一直是研究热点与难点。国际新近研究提示：根据PE相关危险因素建立统计模型可较好预测PE发病风险。PE多因素发病又存在种族差异，目前的挑战是如何建立稳健的适用于我国产前保健体系的预测模型并加入何种可靠生物标志物以改进早期预警能力。响应精确医学理念，并基于PE遗传易感性学说和流行病学证据，我们前期开展了高通量基因芯片预试验，结果已提示PE相关联的拷贝数变异区域。本课题拟在前期工作基础上扩大样本量，进一步筛选可靠生物标志物；结合公立医院产前保健常规建立孕妇队列开展高效的巢式病例对照研究收集相关资料。通过集成新颖统计算法整合生物标志物与临床表型以精确预测子痫前期发病风险。PE预测模型研究将对PE的预防和个体化分类诊治提供循证基础，从而服务于个性化分类诊疗的医疗保健服务新模式。

子痫前期（Pre-eclampsia, PE）病因未明。对于已经发病的重度PE患者，终止妊娠仍是目前最为直接的手段。迫切需要早期识别PE发生风险以早期获得临床干预的时机。本课题批准为2年期，首先侧重于筛选并验证可靠的PE生物标志物。在此前全基因组Affymetrix Ctyo HD芯片检测工作基础上扩大样本量，基于10对孕产妇血样芯片检测结果，根据在两分组中事件发生的差异达到了30%以上的条件筛选出了44个全基因组拷贝数差异（CNV）区域，其中差异CNV区域与DGV数据库中的正常人高频拷贝数变异区域(在正常人群中CNV达到10%及以上)没有重叠的有27个分别是1q23.1、3p14.2、5q35.3、6q26、7q36.2等等，涉及到27个差异基因，其中LPA在关于子痫前期的文献中曾被提到与之有关。SNP位点经过分析也得到了30个差异突变位点，涉及到26个突变基因，其中KIR3DL2和MUC16也在文献中被报道与子痫前期有关。采用基质辅助激光解吸电离飞行时间质谱(MALDI-TOF-MS)方法进行SNP的验证结果显示只有OR5B3和EFCAB5两个基因对应的SNP位点即：rs11229411和rs9897794在100对样本中存在统计学意义上的差异。本研究提示孕妇外周血DNA分析所获得的差异CNV和SNP可能成为临床子痫前期预测的潜在生物标志物。在此基础上对接当地产检常规中的检查指标和临床相关资料，即可能与子痫前期预测有关的临床信息及检查指标，包括：孕产妇一般情况、孕产妇疾病史、家族史、本次怀孕情况、孕期实验室指标等，综合这些临床危险因素、多维度整合分析将有助于建立本地化可靠的预测模型。