Test shown to find ovarian cancer early

Three blood proteins increase with disease, Hopkins researchers say

August 15, 2004|By Frank D. Roylance | Frank D. Roylance,SUN STAFF

For most women with ovarian cancer, a diagnosis comes too late. The malignancy has usually spread from their ovaries, and their chances of survival are slim.

But researchers at the Johns Hopkins Medical Institutions say they've fashioned a test that can identify the disease in its early stages by measuring three blood proteins that increase markedly in ovarian cancer patients.

Combined with an existing blood test, they say, the new "biomarkers" could help doctors detect as many as 83 percent of early ovarian cancers, vastly improving their patients' chances of beating the disease.

Daniel W. Chan, director of Hopkins' Biomarker Discovery Center in Baltimore, said more work is needed to improve the testing technology and to verify its accuracy in large, independent clinical tests. But so far, he said, "it looks promising."

"A lot of people are dedicating their careers to blood tests, or assays, that can identify ovarian cancer," said Dr. Dennis Chi, co-director of pelvic reconstructive surgery at Memorial Sloan-Kettering Cancer Center in New York.

Others have devised tests based on similar "profiling" of blood proteins, with assertions of even greater accuracy than the Hopkins test, Chi said. But none of the proposed tests has been proved or readied for widespread use.

A report on Chan's work appears today in the journal Cancer Research.

In announcing the findings, Hopkins disclosed that the university and Chan have a financial stake in the outcome. The research was funded by the National Cancer Institute and Ciphergen Biosystems Inc., a private company in which Chan is a stockholder, consultant and scientific advisory board member.

Under a licensing agreement with Ciphergen, Chan may also share with Hopkins any royalties that result from the sale of products developed using the research.

Chan said it's likely to take several years, more public and private funding, and additional development before a routine test will be ready for clinical use.

The National Cancer Society expects more than 25,500 new diagnoses of ovarian cancer this year in the United States. More than 16,000 women will die from the disease. Although it ranks behind cancers of the breast, lungs, colon and uterus as a killer of women, ovarian cancer is particularly deadly because it's so likely to be missed in its early stages.

According to the National Cancer Society, almost 70 percent of women with the common epithelial form of ovarian cancer get the diagnosis after their cancer has spread to other parts of their body. Only 15 to 20 percent of those patients will be alive five years later. Early detection boosts survival rates to 70 percent to 90 percent.

"There's no question that we need a test to diagnose ovarian cancer, either in its earliest stage or even before it becomes cancer," said Chi, who was not involved in the Hopkins study. Without more accurate tests, doctors frequently wind up performing diagnostic surgery that finds nothing. "It happens probably 20 to 40 times for every early-stage cancer that we find," Chi said.

Chan said the current blood test for ovarian cancer looks for a blood protein called CA-125. But because it is not very sensitive or specific to ovarian cancer, the FDA has approved the CA-125 test for detecting only the return of cancer in women who have been treated for it. As an overall screening test, he said, it detects a third of ovarian cancers.

That leaves doctors with less precise diagnostic tools, such as patients' descriptions of their symptoms and basic vaginal or rectal exams, sonograms or X-rays, or invasive diagnostic procedures such as laparoscopy - the insertion of a slender instrument into the abdomen and the extraction of ovarian cells for study.

Chan's team decided to search for cancer biomarkers that could be measured from a simple blood test. Because blood samples from patients with early ovarian cancer are scarce, they teamed up with colleagues at other cancer centers, including the Duke University Medical Center, the MD Anderson Cancer Center in Houston, and institutions in Australia and the Netherlands.

Together, they assembled blood samples from 195 women with ovarian cancers. Sixty-five were early cases, some discovered by chance during unrelated surgery. With mass spectroscopy, the researchers then identified a list of blood proteins in the samples and measured their concentrations. The findings were compared with those from more than 300 healthy women and women with benign pelvic tumors.

Using a computer program designed by Zhen Zhang, an associate professor of pathology at Hopkins, the group sorted through the results and found three proteins that increased in the ovarian cancer patients but not among healthy women or those with other diseases.

Baltimore Sun Articles
Please note the green-lined linked article text has been applied commercially without any involvement from our newsroom editors, reporters or any other editorial staff.