Serum MicroRNAs Profile From Genom-wide Serves As A Fingerprint For Diagnosis Of Cervical Cancer And Endometrial Cancer

Background:Cervical cancer and endometrial cancer are the most commom malignancies of the female genital tract. Cervical cancer is the fourth leading cause of cancer death in females worldwide; endometrial cancer is the third most common gynecologic malignancies (behind ovarian and cervical cancer). Serum microRNAs (miRNAs), the remarktable stability and their unique concentration profiles in patients with various diseases, make them promising noninvasive biomarkers for tumor detection. Our group has systematically discovered that circulating miRNAs are correlated with certain human cancer. We investigated the altered profile of serum microRNAs in cervical cancer and endometrial cancer patients in order to predict the two commom gynecologic oncologies at a relative early stage.Methods:Serum samples were taken from 213 cervical cancer patients,158 age-and ethnicity-matched controls; 33 endometrial cancer patients,42 age-and ethnicity-matched controls. In the initial screening phase, we perform Solexa sequencing to identify miRNAs that showed significant differences in the cervical cancer cases; we also used TaqMan low density array to find miRNAs that showed great upregulation differences in the endometrial cancer cases. Differential expression was then validated using hydrolysis probe-based stem-loop quantitative reverse transcription polymerase chain reaction (qRT-PCR) in the validation phase. We constructed the ROC curve and Risk score analysis to evaluate the specificity and sensitivity of cervical cancer and endometrial cancer diagnosis.Results:(1)The Solexa sequencing results demonstrated 12 marked upregulation serum miRNAs in cervical cancer patients compared with controls. RT-qPCR analysis identified a profile of five serum miRNAs (miR-21, miR-29a, miR-25, miR-200a and miR-486-5p) as cervical cancer biomarkers. The ROC curves indicated a panel of five miRNAs has a great potential to offer more sensitive and specific diagnostic tests than single miRNA-based assay, squamous cell carcinoma (SCC) and carbohydrate antigen 125 (CA125). More importantly, miR-29a, miR-200a may indicate tumor histology grade and progression stages. (2)The TaqMan low density array results demonstrated that 22 serum miRNAs were markedly upregulated in the endometrial cancer patients compared to the controls. The qRT-PCR analysis further identified a profile of four serum miRNAs (miR-222, miR-223, miR-186 and miR-204) as a fingerprint for endometrial cancer detection. The areas under the receiver operating characteristic (ROC) curve of this four-serum miRNA signature were 0.927, markedly higher than carbohydrate antigen 125 (CA125) (0.673).Conclusions:miR-21, miR-29a, miR-25, miR-200a and miR-486-5p were significantly up-regulated in cervical cancer patients. A five-miRNAs signature identified from genome-wide serum miRNAs expression profiling may serve as a fingerprint for cervical cancer diagnosis. MiR-222, miR-223, miR-186 and miR-204 showed great upregulation in endometrial cancer patients.The four-miRNA signature identified by genome-wide serum miRNA expression profiling provides a novel, noninvasive approach for endometrial cancer diagnosis.