Abstract 769P
Background
Ovarian cancer (OC) is one of the gynecological cancers with the worst prognosis, causing 185,000 deaths and 300,000 women diagnosed worldwide annually. A major challenge to improve treatment outcomes in ovarian cancer is the lack of effective screening methods for early detection. microRNAs modulate gene expression in various cancer types and are recognized as promising biomarkers that could lead to clinical applications, since their profiles differ in healthy and cancer groups of people. Identification of tumor-specific microRNAs presents a powerful opportunity to potentially reduce cancer mortality through early detection. In order to achieve early detection through population screening, noninvasive approaches are more favorable. Urine-based liquid biopsy has some advantages compared to blood-based liquid biopsies, such as being non-invasive and easy handling and sampling. Here, we successfully identified OC-related microRNA ensembles in urine through a combination of microfluidic microRNA extraction device and machine learning-based analysis.
Methods
We analyzed 80 urine samples from 56 OC patients and 24 healthy individuals (controls). An initial set of 22 microRNAs was selected by microarray microRNA profiling assay as biomarker candidates. Quantitative reverse-transcription qPCR (RT-qPCR) was used for further analysis to validate the expression of microRNAs.
Results
We selected a group of 13 microRNAs for validation; (miR-339-5p, miR-4525, miR-7704, miR-148a-3p, miR-143-3p, miR-4488, miR-6090, miR-26a-5p, miR-9-5p, miR-4505, miR-6089, miR-1207-5p, miR-4784). Five of them (miR-339-5p, hsa-miR-143-3p, miR-6090, miR-9-5p, miR-1207-5p) were confirmed to be statistically significantly different in OC and controls. We adopted a logistic regression model and successfully developed a classifier based on these 13 microRNAs, which showed remarkably high sensitivity (98.21%) and specificity (91.67%).
Conclusions
These microRNAs could be potentially developed as biomarkers for OC, and this approach may be useful as a test for early detection of ovarian cancer.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
Craif. Inc.
Disclosure
All authors have declared no conflicts of interest.