New Colon Cancer Marker Discovered in Stool

Researchers ID Colon Cancer Markers in Stool


Findings could lead to diagnostic tests beyond colonoscopy

September 2009

In a series of studies, researchers at Johns Hopkins Kimmel Cancer Center and other institutions around the world have helped identify four potential biomarkers for colorectal cancer that appear in affected patients’ stool, paving the way for new, noninvasive methods of screening for colorectal cancers.

The studies indicate that methylation – a chemical process that alters a gene’s expression – in the genes TFPI2, GATA4, GATA5, and NDRG4 occurs early and frequently in colorectal cancers. The altered genes were identified in stool samples from colorectal cancer patients but were rarely seen in samples from people who did not have cancer, indicating that stool tests to check for these changes could be effective in early identification of colorectal cancers.

The research also suggests that the four genes could help suppress cancerous activity when they function correctly.

Though colonoscopy is considered the gold standard for colorectal cancer detection, recommended for all adults starting at age 50, only half of adults ever go for screening because it is invasive, says Nita Ahuja, M.D., a study co-author and an assistant professor of surgery and oncology at Johns Hopkins. Other testing methods that look for blood in the stool are not accurate enough to detect all cancers, she says.

Identifying methylation markers that are sensitive and specific for colorectal cancer detection may improve early detection of the disease, Ahuja says: “We’re hoping if we can’t get patients to get a colonoscopy, then this could be another means of testing for what is a preventable disease.”

In the first study, published in the June 1 issue of the journal Cancer Research, researchers examined tissue factor pathway inhibitor (TFPI2), a gene that protects the substance surrounding cancer cells from degradation and inhibits cell colony formation and proliferation. The scientists studied colorectal cancer cell lines, tumors from patients who underwent surgery for colorectal cancers, and stool samples from healthy, noncancerous patients and from patients whose colonoscopies confirmed the presence of colorectal tumors.

In lab tests, researchers detected methylation of TFPI2 in 97 percent of colorectal cancerous tumors, and in 99 percent of cancers ranging in severity from Stage I to IV. Then, they found DNA tests of the stool samples to correctly detect aberrant methylation of TFPI2 up to 89 percent of the time in cancerous patients, and to correctly rule out cancer up to 93 percent of the time.

A second study, reported in the June 15 issue of the journal Clinical Cancer Research, looked at the proteins GATA4 and GATA5, which play an essential role in the development and differentiation of the gastrointestinal tract. By analyzing large series of stool samples from colorectal cancer patients and healthy, noncancerous patients, the researchers found that methylation of GATA4 occurs in about 70 percent of colorectal cancers and methylation of GATA5 occurs in about 79 percent of colorectal cancers, but happens at low levels (up to 13 percent) in noncancerous patients.

In an additional experiment looking at GATA4 and GATA5’s effects on colorectal cancer cells, the scientists found the proteins normally function as tumor suppressors, curbing the cancerous cells’ ability to form colonies, proliferate, migrate and invade. Finally, by examining fecal DNA in two series of samples from colorectal cancer patients and healthy, noncancerous patients, the researchers found that testing for GATA4 methylation correctly identified patients with cancer up to 71 percent of the time, and correctly ruled out cancer up to 93 percent of the time.

In a third paper, published in the July 1 issue of the Journal of the National Cancer Institute, the researchers studied N-Myc downstream-regulated gene 4 (NDRG4), which encodes a protein that may be involved in the regulation of chemical signaling in cells forming blood vessel walls. They found the prevalence of NDRG4 methylation in two series of colorectal cancer cell lines to be up to 86 percent, compared with only 4 percent in noncancerous colon tissue. And, they observed that NDRG4’s messenger RNA (a compound that transmits genetic information from DNA to the protein-forming system of the cell) and protein expression were decreased in colorectal cancer tissue compared with noncancerous tissue.

Overexpression of NDRG4 in colorectal cancer cell lines was found to suppress cancerous activity such as cell colony formation, cell proliferation and invasion. In studies of fecal DNA from colorectal cancer patients and healthy, noncancerous patients, testing for NDRG4 methylation correctly identified cancer in up to 61 percent of cases and correctly ruled out cancer up to 100 percent of the time.

Ongoing research efforts will examine if searching for more than one marker at a time could identify additional cancer patients; look for additional genes that may serve as potential biomarkers of disease; and see if testing methods can be made more accurate, says James G. Herman, M.D., a study co-author and professor of oncology at Johns Hopkins.

The papers’ coauthors were from Belgium, New York, Germany, the Netherlands, and Japan.