Photo by Todd McNaught
Researchers in the Public Health Sciences Division have developed a new test that could improve diagnosis of ovarian cancer. The test, which detects the amounts of two proteins in a woman's blood serum, accurately identified more cases of cancer than did tests for either of the two proteins alone. One of the two proteins, CA125, is believed to be the best single blood-serum marker for the disease.
Dr. Martin McIntosh and colleagues examined the performance of the combined-marker test using serum samples from women who already had been diagnosed with symptomatic advanced-stage and localized ovarian cancer. In the next phase of the work, the scientists plan to determine whether the two proteins can be used to detect tumors at much earlier times, before symptoms develop and when the disease is more likely to be curable.
The findings appear in the October issue of Gynecologic Oncology. Co-authors included PHS investigators Drs. Charles Drescher, Nathalie Scholler and Nicole Urban; Dr. Beth Karlan of Cedar Sinai Medical Center in Los Angeles; and Drs. Karl and Ingegerd Hellstrom at Pacific Northwest Research Institute (PNRI). The research was funded by a Specialized Program of Research Excellence grant for ovarian cancer from the National Cancer Institute, for which Urban serves as principal investigator. The Listwin Family Foundation supports Scholler's ongoing investigations on blood-serum markers for ovarian cancer.
Scientists have been eager to develop a diagnostic tool to identify women who have a high risk of having or soon developing ovarian cancer. Previous studies have shown that CA125, a protein produced in abundance by the majority of advanced ovarian tumors, rises to high levels in only about half of women with early stage disease. That has prompted Fred Hutchinson investigators to search for additional markers which, when measured in conjunction with CA125 levels, could constitute a test that detects the majority of early-stage tumors.
"It's unlikely that a single marker can be used to detect all cases of pre-symptomatic early stage ovarian cancer," McIntosh said. "We'd like to find a panel of markers that when used in combination can pick up the vast majority of cases and that has a low false-positive rate. What's encouraging about these new results is that we've found a second marker that increases the sensitivity of detection and does not increase the number of false-positives."
The second marker is a protein called mesothelin, which is overproduced in the tissue of about 95 percent of ovarian cancers and is made at low levels in normal tissue. Scholler, a senior staff scientist in PHS, developed a test to detect a breakdown product of mesothelin in blood serum when she was a researcher at PNRI. The current study found that this protein is at elevated levels in the blood of women with ovarian cancer.
The researchers compared the sensitivity of the combined-marker test with CA125 using serum samples obtained from 52 women with ovarian cancer, 43 controls with benign ovarian disease and 220 normal-risk controls from women who participated in an ovarian-cancer screening program.
The combined-marker test correctly identified 45 out of 52 (86.5 percent) of the cases compared to 41 of 52 (78.8 percent) for CA125 alone. The mesothelin test alone detects fewer than half of the cancer cases. The researchers also compared the false-positive rate — also known as marker specificity — of the combined-marker test to CA125 alone and found the rates to be similar.
"This is the first time we've seen that two markers combined can pick up significantly more cancers than a single marker," McIntosh said. "We are searching for additional markers that are not completely redundant with CA125 and mesothelin, which could allow us to detect even more cancer cases."
Last year, the research team developed an ovarian-cancer test based on a marker known as HE4. The test detects the same number of cases as CA125 but is far less likely than CA 125 to pick up benign ovarian conditions that can lead to unneeded surgeries. McIntosh's group is now looking at whether HE4 also can be used to improve the use of mesothelin and CA 125 for detecting cancers.
McIntosh said that the definitive phase of the work requires evaluating the performance of the markers using blood serum obtained from women months or years before cancer develops. That project, led by Dr. Garnet Anderson, will reveal the ultimate usefulness of the tests as screening tools for early detection of ovarian cancer.