Novel predictive model harnesses technology to diagnose ovarian cancer more accurately than ultrasound.
Researchers recently developed a blood test that may detect ovarian cancer earlier and more accurately than current methods. The new technique detects circulating microRNAs that are linked to ovarian cancer, according to a study published by eLife.
A majority of women with ovarian cancer are diagnosed at later stages of the disease, which severely impacts survival rates. There are currently no FDA-approved screening approaches for ovarian cancer, which makes it difficult to diagnose it early.
Early detection tests for ovarian cancer—including ultrasound or the CA125 protein—are not very accurate, according to the study authors. Previous clinical trials have shown that the utilization of these tests may not have an impact on survival rates.
In the new study, the authors analyzed whether microRNAs could detect ovarian cancer earlier and more accurately than current approaches.
“MicroRNAs are the copywrite editors of the genome: Before a gene gets transcribed into a protein, they modify the message, adding proofreading notes to the genome,” said lead author Kevin Elias, MD, in a press release.
The authors discovered that ovarian cancer cells and healthy cells have different microRNAs. Since microRNAs circulate in the blood, it is possible to measure their levels, according to the authors.
The investigators sequenced the blood levels of microRNAs for 135 women prior to treatment to train a computer program to differentiate between ovarian cancer, benign tumors, and healthy tissue.
Using machine-learning, the researchers can leverage large amounts of data and develop different predictive models, according to the study. The model was observed to accurately differentiate ovarian cancer from benign tissue.
“When we train a computer to find the best microRNA model, it’s a bit like identifying constellations in the night sky,” Dr Elias said. “At first, there are just lots of bright dots, but once you find a pattern, wherever you are in the world, you can pick it out.”
Once the accuracy of the test was confirmed in a small patient population, the authors evaluated the model in 859 patient samples.
The authors said that the new model was significantly more accurate at detecting ovarian cancer than an ultrasound.
For ultrasounds, less than 5% of abnormal test results are confirmed cases of ovarian cancer, but nearly 100% of abnormal results with the microRNA test were confirmed to be cancer, according to the study.
To further validate their findings, the authors used the microRNA test to determine whether 51 patients receiving surgical care had ovarian cancer. They discovered that 91.3% of the abnormal test results were ovarian cancer cases, while negative test results predicted absence of disease 80% of the time, according to the study.
Additionally, the authors found that microRNAs changed after surgery, which suggests that their signal decreases once the tumor is removed, according to the study.
To move the test into the clinical setting, the authors said they must confirm how the microRNA signatures change over time and whether it could be useful for women at high risk of ovarian cancer.
“The key is that this test is very unlikely to misdiagnose ovarian cancer and give a positive signal when there is no malignant tumor,” said senior author Dipanjan Chowdhury, PhD. “This is the hallmark of an effective diagnostic test.”