肿瘤血液标志物的研发及临床评价课件.pptx

1、肿瘤血液标志物的研发及临床评价肿瘤血液标志物的研发及临床评价Ten Leading Cancer Types for the Estimated New Cancer Cases and Deaths by Sex,United States,2013CA Cancer J Clin.2013,63(1):11-30Nature Reviews Cancer,2011,426-437Cell free nucleic acidsOutline of strategies for biomarker discovery through utilization of emerging techno

2、logies.Nat Clin Pract Oncol.2008,5(10):588-99Cancer biomarkers that are currently in clinical useNat Clin Pract Oncol.2008,5(10):588-99Multi-parameter panels can significantly improve the diagnostic value compared to the conventional biomarker 8MethylationprofilingofserumDNAfromhepatocellularcarcino

3、mapatientsusinganInfiniumHumanMethylation450BeadChip A A.6例早期原发性肝癌和6例健康对照的全基因组甲基化水平。B B.早期原发性肝癌组和健康对照组全基因组甲基化水平比较与健康对照组相比，早期原发性肝癌组差异甲基化位点的个数及倍数关系与健康对照组相比，早期原发性肝癌组差异甲基化位点的基因分布值大于0.5为超甲基化值小于0.2为超低甲基化我们筛选出453个健康对照组和37个原发性肝细胞癌组的超甲基化位点。超甲基化位点的聚类分析(欧氏距离)超甲基化位点的基因本体论和基因功能富集分析 原发性肝细胞癌组中超甲基化位点的GO及功能富集分析 健康对照

4、组中超甲基化位点的GO及功能富集分析 基因功能富集分析后健康对照组中超甲基化位点的基因相互作用分析DBX2和THY1甲基化位点通过亚硫酸盐测序法验证DBX2和THY1甲基化位点通过亚硫酸盐测序法进行扩大样本验证 A.A.DBX2和THY1甲基化位点在原发性肝癌组和健康对照组的甲基化水平比较。B.B.DBX2和THY1甲基化位点用于区分原发性肝癌组和健康对照组的诊断价值Multi-parametermodelfordiagnosisofcancer2129例慢性胰腺炎82例急性胰腺炎162例胰腺癌33例结肠炎62例结肠息肉101例结肠癌20例结肠癌术后40例胃增生139例胃癌31例慢性肺炎28例

5、小细胞肺癌32例肺鳞癌72例肺腺癌22例乳腺增生101例乳腺癌65例宫颈良性病变72例宫颈癌18例宫颈癌术后84例前列腺增生108例前列腺癌12例前列腺癌术后200例健康对照胰腺疾病结肠疾病胃疾病肺疾病乳腺疾病宫颈疾病前列腺疾病正常对照ALB、ALP、ALT、ApoA1、ApoB、AST、Ca、CHO、CK、CKMB、CO2、Cl、CYS、CR、DB、GGT、GLU、HDL、HCY、K、LDL、LP(a)、Mg、Na、P、SA、TP、TB、TBA、TG、UN和和UAGM-CSF、IFN-、IL-10、IL-1、IL-2、IL-4、IL-6、IL-8、MCP-1和和TNFPG1、PG2和和SCC

6、AFP、CA125、CA153、CA199、CA724、CY211、CEA、FERR和和NSE1513例For differentiating between the colorectal adenoma and colorectal cancer groups.The AUC of multivariate logistic regression was 0.945(95%CI:0.9090.981).Compared with the conventional biomarkers CEA and CA199,the AUC of multivariate logistic regress

7、ion showed significant improvements(p 0.05).AUC:Area under the curve.FutureOncology,201326 The red line was the leave-one-out cross-validation(LOOCV)accuracy curve varying with k and each asterisk represent the value of accuracy with a fixd kROCcurvesforthetopfivefeaturesandtwopanelsofbiomarkerssele

8、ctedbyourmodel.Figure 1.Flowchart of our experimental designTable 1.Clinical characteristic of the studied subjectsTable 2.List of the serum parameters in the 61-plex panelTable 3.The sensitivity of top 20 performing 4 serum parameter panels for discriminating PDAC Vs Control and PDAC Vs Benign grou

9、p identified by MMC algorithm applied to the training set at 90%specificityFigure 2.ROC curves analysis for discriminating between the PDAC versus Control and PDAC versus Benign groups in the training group.ROC curves of CA19-9,ALB,CRP and IL-8 panel(solid line)and CA19-9(dotted line)for discriminating between PDAC and Ctrl in the training group(a)and validation group(c).ROC curves of the panel consisting of CA19-9,CO2,CRP and IL-6(solid line)and CA19-9(dotted line)for discriminating between PDAC and Benign in the training group(b)and validation group(d).