A new, computer-based diagnostic test to detect ovarian cancer from blood samples has begun to receive attention in recent weeks. The results of a preliminary study, released in The Lancet on February 16, suggest that the test may be a great leap forward in the early detection of this particularly deadly form of cancer. If it is caught early, ovarian cancer is treatable, but there is currently no reliable screening test available.

The new test, which is still in the research phase of development, combines artificial intelligence technology and proteomics (the study of proteins’ structure and function) to look for patterns of proteins in the blood. In the study conducted by the National Cancer Institute, the FDA, and Correlogic Systems Inc., which developed the computer program, the test was able to detect ovarian cancer correctly in all of 50 cancerous samples, 18 of which were stage I (early stage). The test reported a false-positive result in only 5% of the 66 non-cancerous samples.

Although the test population was small, the test’s initial success in detecting all of the stage I ovarian cancer cases inspires optimism that this method will allow for earlier diagnosis of the disease and will greatly improve patients’ chances of survival. The test also has the benefit of taking just 30 minutes to complete and requires only a drop of blood from a fingerstick.

Researchers say that more studies of the test’s sensitivity and accuracy are needed before it will be available and, then, it may at first be limited to use among women at high risk for the disease. However, they are hopeful that this test will help clinicians to diagnose ovarian cancer earlier than other methods currently available and that this technique could be applied to other cancers for similar early detection screening tests.