Feb. 13, 2008 -- Yale researchers have developed a simple blood test for
ovarian cancer that may do what no current test can -- reliably detect the
disease in its early stages while it is still highly curable.
Results from the phase II study showed the test to have an accuracy of
A phase III trial is under way and should be completed within months, Yale
School of Medicine researcher Gil Mor, MD, tells WebMD.
Ovarian cancer is the fifth leading cause of cancer deaths among women in
the U.S., even though it is much less common than many other cancers. That is
because the disease is most often diagnosed in its late stages when the cancer
has already spread beyond the ovaries.
A reliable test for detecting ovarian cancer in its early stages has been an
elusive goal, but the Yale researchers believe they may have one.
And an independent review by the National Cancer Institute's Early Detection
Research Network (EDRN) confirmed their early findings.
"We now have a test that is significantly better than anything that is
available today," Mor says.
The latest research expanded on work the Yale team first published in
The test has been modified since then and now uses six protein biomarkers
instead of four, resulting in an increase in specificity from 95% to 99.4%.
While the difference may not sound like much, from a clinical standpoint it
is a big deal.
A test that is 95% specific would result in false-positive readings in 5,000
out of every 100,000 women tested, while a 99.4% specific test would result in
a few hundred false-positives.
"A test that is 95% specific may sound good, but that means that one in 20
women who are tested will be told they may have a life-threatening malignancy,"
American Cancer Society director of cancer screening Robert A. Smith, PhD,
tells WebMD. "And these women will have to have a fairly invasive procedure to
determine if they have cancer."
The newly published trial, led by Mor, included 362 healthy women and 156
newly diagnosed ovarian cancer patients.
Using a blood test, the researchers looked for evidence of the proteins
leptin, prolactin, osteopontin, insulin-like growth factor II, macrophage
inhibitory factor, and CA-125.
While each single protein biomarker was not good in differentiating between
those with cancer and those without, the combination of the six biomarkers
together was found to be highly accurate.
The research appears in the Feb. 15 issue of the journal Clinical Cancer