Multi-parameter Diagnostic Of Female Malignant Tumors Based On Peripheral Blood

Objective:Based on the serum indicators, we aimed to identify the diagnostic value of healthy controls and disease groups, which may provide a complementary diagnostic method for clinical diagnosis. We investigated the difference of genomics and serological proteomics in female cancer compared to health controls, try to explore the high sensitive and specific diagnostic model and find new biomarkers forovarian cancer and breast cancer.Methods:The serum and clinical parameter of female ovarian tumors (n=158), ovarian cancer (n=137), benign ovarian tumor (n=13), borderline ovarian tumors (n=8) breast cancer (n=60), benign breast tumor (n=20) and healthy controls (n=40) were collected for clinical data set.①The epidemiology and statistics analyzed the difference of laboratory parameters between cancer group and healthcontrol.②All the samples were analyzed using three steps based quantitative PCR method. Total RNA was isolated from serum samples using the standard Quantobio microRNA isolation procedures. All samples passed quality control. The synthesized miRNA internal control, QuantoIC1, was extrinsically supplied as normalization controls for miRNA quantification. The expression levels of miRNAs in different groups were analyzed.③All the serum samples were treated by MB-WCX beads firstly and the proteomic profilings were attained by Clinprot/Matrix-Assisted Laser Desorption/Ionization Time of Flight Mass Spectrometry (MALDI-TOF-MS).④A set of candidate cytokines and chemokines (GM-CSF,IFN-y,GRO,IL-1β,IL-2,IL-6,IL-8,MCP-1,TNF-α,VEGF,EGF,RANTES,C CL21/6Ckine,SDF-1/CXCL12) were measured by using Luminex liquid chip technique. Data were analysed by SPSS19.0and ClinPro Tools.Results:①After establishing the clinical data base, most of laboratory parameters in ovarian cancer groups were distinct with controls groups, Median level of CA125in ovarian cancer was higher than in healthcontrols (P=0.001), but only54%of the patients with CA125positive expression, positive rate of CA125increased with late stages. CA153has a sensitivity of80.3%, a specificity of72.6%in predicting ovarian cancer, while CA125has a sensitivity of64.3%, a specificity of90.4%. The combination of CA125and CA153may yield an increase sensitivity of73.5%, while still maintaining the specificity of90.5%, with the area under the curve of0.877;②Four out of the six candidate microRNAs tended to be differentially expressed in serum samples between patients with breast cancer and healthy controls: miR-451:P<0.001; miR-148a:P=0.021; miR-27a:P=0.013and miR-30b:P=0.001. Three out of the six miRNAs (miR-451, miR-27a, and miR-30b) showed different expression levels between benign breast tumor groups and healthy controls:miR-451: P=0.002; miR-27a:P=0.012and miR-30b:P=0.046. A panel of miRNAs consisting of the four down-regulated miRNAs can distinguish breast cancer from healthy controls well, with an area under the receiver operating characteristic curve (ROC) of95.3%, a sensitivity of94.7%, and a specificity of82.8%. miR-148a levels differentiated between malignant and benign breast masses, with an AUC of69.8%, a sensitivity of56.1%, and a specificity of78.9%.③The specific model Y=0.691+0.398X2487-0.638Xi867comprised of two peptides2847Da,1867Da could distinguish ovarian cancer from healthy group showed100%of sensitivity and60.9%of specificity by ROC analysis, the areaunder ROC curve attain to85.5%. Little disparity of peptidomic profiling was scantly found between homologous tumors. m/z2884.4Da showed a higher expression level in well-differentiated ovarian cancer. m/z2681.5Da,2659.8Da,2670.6Da,2952.0Da had a relatively high level in advanced ovarian cancer.④EGF,IL-6,MCP-1,6Ckine,RANTES, IL-10were significantly over-expressed in the tumor group compared with those in normal controls, while IL-2reduced. The model of binary logistic regression equationY=-3.707+0.009X EGF+0.001X MCP+0.014X Ckine, have a sensitivity of72.7%and a specificity of83.8%for predicting ovarian cancer, the receiver operating characteristic (ROC) curve analysis using five combined markers yielded an AUC of84.5%.Conclusions:Based on the clinical data base of ovarian cancer, the combination of bioinformatics and clinical data could set up simpler clinical diagnostic model; meanwhile, peptides and microRNA may be considered to be the ideal biomarker. Cytokines and chemokines have diagnostic value, but not reach the desired level.