单击此处编辑母版标题样式,单击此处编辑母版文本样式,第二级,第三级,第四级,第五级,*,个体化医疗的现状与未来四,.,生物标志物研究,Outline,生物标志物的概念,如何评价生物标志物？,生物标志物的研究方法？,生物标志物的概念,什么是生物标志物,(,biomarker),？,“measurable and quantifiable biological parameters”,-,a Medical Subject Heading (MeSH) term, 1989,“ A characteristic that is objectively measured and evaluated as an indicator of normal biological processes, pathogenic processes or pharmacological responses to a therapeutic intervention.”,-Biomarker Definitions Working Group, 2001,NIH,Features of a Useful Biomarker,High sensitivity and specificity,Easy accessible sample,Correlation with histological scoring,Change in advance of clinical signs,Translational from research to clinical use,不同水平生物标志物,DNA,Primary transcript,mRNA,Transcription,protein,Translation,RNA processing,Nucleus,Biomarker Examples,Cholesterol is one of the most well-known biomarkers of cardiovascular health,Physical measurements: body temperature (fever); blood pressure (stroke risk),Other biomarkers:, blood sugar level (diabetes), antigens (hepatitis), proteins (heart attack), genetic variations (Huntingtons disease),生物标志物的临床应用,Ludwig JA,et al. Nature reviews 2005,5: 845-856,目前临床很多疾病的诊断依赖病理诊断，但不能作为常规筛查、监测手段,众多疾病缺乏早期、特异性生物标志物,治疗缺乏个体化方案,生物标志物应用现状,Clin J Am Soc Nephrol,3: 18951901, 2008.,Biomarkers for chronic kidney disease,Are we treating sub-populations?,疾病,药物,无反应率,抑郁,SSRIs, SNRIs, TCAs,40-60%,哮喘,-adrenergics, LTD4,4-75%,糖尿病,Sulfonylurea, Biguanides, Glitazones,50-75%,肿瘤,(,乳腺癌 肺癌,),Various,70-100%,From Kalow, Tyndale & Meyer, Pharmacogenomics, 2001,Novel biomarkers are needed,Early, accurate diagnosis,-Individualized therapy and improved treatment outcomes,Better defined populations will allow more specific drugs,-Better efficacy,-Fewer side effects,“The use of,biomarkers,will change medical practice from a population-based approach to an,individualized,approach”,Felix Frueh, Associate Director of Genomics at CDER, FDA,Evolution of the biomarkers research,High plasma cholesterol and cardiovascular diseases,Nearly,50,percent of all future myocardial infarction and stroke events occur in those with normal or below normal lipid levels.,EUROASPIRE Study Group, 1997,% of MI,Additional biomarkers (inflammation)Hs-CRP and cardiovascular risk,Hs-CRP is the most widely studied biomarker of inflammation in cardiovascular risk.,Since the early 1990s with the development of highly sensitive assays for its measurement, correlations of hs-CRP with both cardiovascular risk factors and future cardiovascular events has been possible.,CRP and LDLC levels and the risk of cardiovascular diseases,C-Reactive Protein (mg/L),3.0,160,LDL-cholesterol (mg/dL),3.0,2.0,1.0,0.0,Multivariable Relative Risk,Increased CRP levels are associated with increased risk of cardiovascular events independently of LDL-C levels,Ridker PM et al., 2002,27,939 women,High CRP-high LDL,High CRP-low LDL,Low CRP-low LDL,Low CRP- high LDL,Porbability of Event-free Survival,Years of Follow-up,0.99,0.98,0.97,0.96,0.00,1.00,0,2,4,6,8,Evolution of the biomarkers research :,CRP and LDL-C levels and event-free survival among women,27,939 women,The median values were as follows:,C-reactive protein,: 1.52 mg/L,LDL cholesterol,: 123.7 mg/dL or: 3.20 mmol/L,CRP and LDL-C could give better prognostic information than the two markers separately,.,Ridker,PM et al., 2002,如何评价生物标志物？,常用评价指标,（一）敏感性,（二）特异性,（三）,Youden,指数,（四）阳性似然比,（五）阴性似然比,（六）阳性预报值,（七）阴性预报值,（八）,ROC,曲线,ECG,诊断试验的结果,ECG,诊断结果,心肌梗塞,合 计,出现,不出现,阳 性,阴 性,合计,416(,TP,),9,（,FP,）,425,104(,FN,),171,（,TN,）,275,520,180,700,(N),一、敏感性（,Sensitivity):,TP/(TP+FN)=TPR (true positive rate),TRP=Sen=416/(416+104)=0.8,该指标只与病例组有关，反映了诊断试验检出病例的能力,ECG,诊断试验的结果,ECG,诊断结果,心肌梗塞,合 计,出现,不出现,阳 性,阴 性,合计,416(,TP,),9,（,FP,）,425,104(,FN,),171,（,TN,）,275,520,180,700,(N),二、特异性（,Specificity),Spe=True negative rate(TNR),= TN(FP+TN)=171/(171+9)=0.95,该指标只与对照组有关，反映了诊断试验排除非病例的能力。,灵敏度与特异度的优缺点,优点：灵敏度与特异度不受患病率的影响，其取值范围均在（,0, 1,）之间，其值越接近于,1,，说明其诊断准确性越好。,缺点：当比较两个诊断试验时，单独使用灵敏度或特异度，可能出现矛盾。,解决办法：将两指标结合：,Youden,指数、阳性似然比、阴性似然比等,ECG,诊断试验的结果,ECG,诊断结果,心肌梗塞,合 计,出现,不出现,阳 性,阴 性,合计,416(,TP,),9,（,FP,）,425,104(,FN,),171,（,TN,）,275,520,180,700,(N),三、,Youden,指数，,=Sen+Spe-1=TPR-FPR,=0.8-0.05=0.75,Youden,指数取值范围在（,0,，,1,）之间，其值越接近,1,，诊断准确性越好。,ECG,诊断试验的结果,ECG,诊断结果,心肌梗塞,合 计,出现,不出现,阳 性,阴 性,合计,416(,TP,),9,（,FP,）,425,104(,FN,),171,（,TN,）,275,520,180,700,(N),ECG,诊断试验的结果,ECG,诊断结果,心肌梗塞,合 计,出现,不出现,阳 性,阴 性,合计,416(,TP,),9,（,FP,）,425,104(,FN,),171,（,TN,）,275,520,180,700,(N),医生最关心的问题：,1.,试验阳性时患病的概率多大？,2.,试验阴性时不患病的概率多大？,阳性预测值是在诊断试验阳性的受试者中，标准诊断有病的病例（真阳性）所占的比例,阴性预测值则是在诊断试验为阴性的受试者中，标准诊断证实无病的受试者（真阴性）所占的比例。,ECG,诊断结果,心肌梗塞,合 计,出现,不出现,阳 性,阴 性,合计,416(,TP,),9,（,FP,）,425,104(,FN,),171,（,TN,）,275,520,180,700,(N),阳性预报值与阴性预报值,ROC,曲线,ROC,（,receiver operating characteristic,的缩写，译为,“,接受者工作特征,”,）,ROC,曲线研究历史,1950s,雷达信号观测能力评价,1960s,中期 实验心理学、心理物理学,1970s,末与,1980s,初 诊断医学,ROC,的涵义与起源,不同诊断界值时灵敏度与特异度间的平衡,(trade off),0,20,40,60,80,100,50,60,70,80,90,100,特异度,灵敏度,百分率（）,Receiver Operating Characteristic curve Area Under Curve (AUC) - Graphed,Curve 1,= .50, Pure chance,no better than random guess,Curve 3,is better than,Curve 2,Curve 4,= 1.0,Totally Sensitive, completely accurate classification of effectively and less-effectively instructed students,完美与无用的,ROC,曲线,真阳性率,即,灵敏度,假阳性率 即,1,特异度,机率线,(chance line)(diagonal reference line),诊断准确度较低（,0.9,）,0.0,0.2,0.4,0.6,0.8,1.0,0.0,0.2,0.4,0.6,0.8,1.0,FPR,T,P,R,A,0.938,ROC,曲线下面积（,Area,）与诊断准确度高低,高,0.90-1.00 = excellent (A),中,0.80-0.90 = good (B) 0.70-0.80 = fair (C),低,0.60-0.70 = poor (D) 0.50-0.60 = fail (F),ROC,曲线小结,ROC,曲线反映了灵敏度与特异度间的平衡,(,增加灵敏度将降低特异度；增加特异度将降低灵敏度,),。,在,ROC,曲线空间，如果曲线沿着左边线，然后沿着上边线越紧密，则试验准确度越高。,在,ROC,曲线空间，如果曲线沿着机会线（,45,度对角线）越紧密，则试验准确度越低。,ROC,曲线下面积是重要的试验准确度指标。,生物标志物研究方法,phase,Phase1,Preclinical Exploratory,Phase2,Clinical Assay and Validation,Phase3,Retrospective,Longitudinal,Phase4,Prospective Screening,Phase5,Disease control,Objective,Target biomarker identification, feasibility,Study assay in people with and without disease,Case-control studies using specimens,Longitudinal studies to predict disease,Clinical use,Site,Biomarker development lab,Biomarker validation lab,Clinical epidemiologic centers,Cohort studies,Community,Design,Cross-sectional,Cross-sectional,Case-control,prospective,RCT,Sample size,small,small,modest,medium,large,Vasan RS. Circulation. 2006;113:2335-2362,.,The Agendia MammaPrint Test,首个,FDA,批准的基因组检测试验,- Feb. 2007,How they got there,？,2002 Discovery of 70 gene signature (117 patients),2002 Duplication of results (in another sample set:295 patients),2006 Assay performance,2006 Optimized array format: reproducibility; back to,original sample set,2006 External confirmation (307 patients, 5 hospitals),2007 Approval by FDA,生物标志物研究技术,传统研究方法：,PCR, Western blotting, ELISA, et al,新型研究方法：,基因组学技术,蛋白质组学技术：,2-DIGE/MS,蛋白质芯片,生物标志物研究方法,Question1.,What human samples should be collected, and how should they be used?,Does this vary between discovery, validation and implementation?,Answer1.,All biological samples are eligible for collection,Collected biological material depends on analyte and tissue source,Examples,Biological fluids,Serum, plasma, urine, csf,Secretions,Saliva, seminal fluid,Body cavity fluids,Pleural fluid, peritoneal fluid, etc,Specific tissue material,Specialized cells reproductive cells,Non-cellular,Ideal, Biomarker discovery samples should be identical to the projected testing situation, (e.g. Do not study plasma for discovery, and then validate or implement assay using serum),Practical, set up study with samples that are as close to the testing situation as possible,Question 2.,What is the role of routinely accessible biofluids such as plasma, serum, and urine?,What is the role of “proximal” fluids like CSF, synovial fluid, ascites, pancreatic ductal fluid, etc?,What is the role of solid tissues?,Role of routinely accessible biofluids,Very important in the discovery of biomarkers of diseases (systemic vs. organ specific/local),Important for:,early detection, disease severity, tumor burden, prognosis, monitoring of response to therapy,“Proximal fluids”, Can reflect disease perturbations in the organs or tissues from which they are secreted,Solid tissues, Very important for the development of novel insitu biomarkers, Immunofluorescence, immunocytochemistry, Imaging mass spectrometry,Question 3.,Must human sample collection be prospective, or can existing repositories be used?,What considerations are important in determining the adequacy of repository samples?,Answer3:,Ultimate confirmation of the validity of a biomarker has to be proven in a prospective study,Nevertheless, well designed retrospective studies using well characterized samples in repositories can be performed and frequently yield viable candidates,生物标志物研究面临的挑战,The multidisciplinary nature of biomarker discovery & development,Complex,Multiple disciplines,Heterogeneous populations,Standards not established,Expensive,Human resources,Multiple technologies,