Blood Test Detects Early Ovarian Cancer

The scientists who developed the test by combining proteomics with artificial intelligence computer programs say the concept may be applicable to any type of disease.

Conducted by scientists at the US Food and Drug Administration (FDA) and the US National Cancer Institute (NCI), the study was published in the February 16, 2002, issue of The Lancet. The test relies on software that can detect patterns of key proteins in blood. First, protein patterns from the blood samples of 50 women with known ovarian cancer and 50 healthy women were generated using mass spectrometry. An artificial intelligence computer program developed by Correlogic Systems, Inc., called Proteome Quest (Bethesda, MD, USA), was used to analyze the data and identify a proteomic pattern that completely discriminated cancer samples from noncancer samples.

The discovered pattern was then used to classify an independent set of 116 blinded patient blood samples, of which 50 were from cancer patients and 66 from healthy women. The researchers were able to correctly identify all 50 cases of ovarian cancer. Moreover, they identified all 18 stage 1 cases. Of the controls, 63 of the 66 were identified as noncancer. Currently, 74% of all ovarian cancer cases are not discovered until they are more advanced, when five-year survival is only 29%.

"The concept that patterns of proteins instead of single biomarkers can be used as a potential diagnostic is a brand new paradigm,” said Emanuel Petricoin, Ph.D., lead author of the study and co-director of the Clinical Proteomics Program of the FDA and NCI. "We now need to really assess the true ability of this new concept to improve and possibly even save the lives of people with cancer.”




Related Links:
Correlogic