Teams of scientists around the world are working to develop ways to detect early signs of ovarian cancer in blood. Considerable progress has been made in understanding the disease, but as yet there are no proven effective biological markers or panels of biomarkers for early detection.
For the most promising markers to date, a critical issue is the question of lead time. That is, how long before the disease is diagnosed can the biomarkers distinguish individuals with and without the disease? A new report provides some answers and illustrates just how difficult it can be to find markers for this relatively rare and deadly disease.
In one of the first such studies, Dr. Garnet Anderson of the Fred Hutchinson Cancer Research Center and her colleagues tested the lead times for six promising markers, using prediagnostic blood samples from 34 women with cancer and a matched comparison group. All were participants in a cancer prevention study called the Carotene and Retinol Efficacy Trial.
The analysis revealed changes in three markers several years before diagnosis, but the markers typically did not give a strong signal until less than a year before diagnosis. By the time they were diagnosed just a few months later, most women had advanced tumors, the researchers reported online in the Journal of the National Cancer Institute on December 30.
“These three markers have the potential to move diagnoses of ovarian cancer a few months earlier, but we also found evidence of changes as early as 3 years before diagnosis,” said Dr. Anderson, who co-led the study with Dr. Nicole Urban.
The study demonstrates one reason that screening regimens based on markers—or panels of markers—can fail: because the blood levels of markers do not increase early enough, said Dr. Patricia Hartge of NCI’s Division of Cancer Epidemiology and Genetics in an accompanying editorial. This information could inform the design of future early detection programs of ovarian cancer, she wrote.
“The good news is that this study shows you can find informative markers within an interval of about a year of diagnosis,” Dr. Hartge said in an interview. “The bad news is we don’t have a foolproof marker that will be put into clinical use and save women’s lives.”
The three informative markers were CA-125, HE4, and mesothelin. (The others tested were B7-H4, decoy receptor 3, and spondin-2.) While CA-125 performed the best of the three, the combination of the three markers was more informative than CA-125 alone. Doctors routinely track blood levels of CA-125 in women with ovarian cancer to monitor a treatment or see whether the cancer has come back. Levels of the marker can rise for reasons other than cancer, however, and they are not always elevated in the disease.
CA-125 and HE4 were the top performers in a validation study reported last spring. “The new findings reinforce the idea that these two markers are currently the best and should be evaluated in future studies,” said lead investigator Dr. Daniel W. Cramer of Brigham and Women’s Hospital, who is supported by NCI’s Early Detection Research Network. The study was a joint effort and was cosponsored by NCI's Specialized Programs of Research Excellence (SPOREs). The lead investigator among ovarian SPOREs was Dr. Urban.
Dr. Cramer cautioned against being too pessimistic about the current results, noting that the study had limitations that could be avoided in future studies. For instance, blood from the women was not drawn annually on a regular schedule as would happen in formal screening studies. The study was also small and the women were all past or current heavy smokers, so it is not clear how the findings would apply to other groups.
“When it comes to early detection for ovarian cancer, we’re not there yet,” said Dr. Anderson. “But we’re still working hard, and we still hold out hope that we’re going to find something with a significant impact for women. We think the three markers are a reasonable core we can build on and add other markers to.”
Some creative new approaches might be needed for finding informative biomarkers. Most potential biomarkers to date have come from studying the blood of women with advanced disease, but this may not be the best approach for finding markers that are elevated early in the disease, the study authors said.
Many experts agree that ovarian cancer screening will eventually involve testing biomarkers as well as an imaging component. Two ongoing trials are evaluating combinations of CA-125 and transvaginal ultrasound. The UK Collaborative Trial of Ovarian Cancer Screening requires both CA-125 and transvaginal ultrasound to be positive before referring women to surgery; the NCI-sponsored Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial requires that one or the other is positive.
“Most of us believe that multimodal screening will be needed and that means combining biomarkers with some sort of imaging technology,” said Dr. Anderson. “But we will need improvements in both imaging technologies and in biomarkers to meet our goals of developing effective screening strategies.”
––Edward R. Winstead