Johns Hopkins Medicine
Office of Corporate Communications
Media Contact: Audrey Huang;
410-614-5105; [email protected]

September 9, 2005


An interdisciplinary team of scientists from Johns Hopkins and elsewhere has been selected to receive a $2.25 million, five-year grant from the Doris Duke Charitable Foundation to develop a practical test to predict a person's risk of colon cancer by looking for a particular biological marker in the blood.

If successful, the test could be broadly applied to identify those who most need to be monitored with invasive but standard colonoscopy or who should make changes in their diet or in other factors to help lower their risk of developing colon cancer.

The researchers' test is based on the fact that five percent to 10 percent of people have improper control of a growth-promoting gene called insulin-like growth factor 2 or IGF-2. The Johns Hopkins scientists reported in February that, in mice, the double dose of IGF-2 protein sets the stage for cancer development -- by increasing the number of primitive cells found in the colon -- and that early evidence suggests the same could be true in people.

"If everything works out -- if IGF-2 status is tied to colon cancer risk in people and the blood test is workable -- then IGF-2 status could be the colon cancer equivalent of cholesterol levels as a risk factor for heart disease," says Andrew Feinberg, M.D., Ph.D., professor of medicine at Johns Hopkins and one of the leaders of the project. "We've never had a broad molecular screening tool like that for any cancer."

But there's a lot of work to do first. With the new Doris Duke Charitable Foundation grant, part of the foundation's Clinical Interfaces Award Program, Feinberg and his colleagues will develop and validate the blood test, study how IGF-2 status changes over time and evaluate whether IGF-2 status is a predictor of colon cancer. Patients will be enrolled at Johns Hopkins Hospital and at the Cleveland Clinic Foundation in Ft. Lauderdale, Fla.

"We've long suspected that genomic changes underlie cancer risk, but that knowledge hasn't translated yet into a broadly applicable way to identify individuals most at risk," says Feinberg. "The BRCA1 and BRCA2 genes implicated in breast and ovarian cancers are great examples, but they are applicable only to a small percentage -- less than one percent -- of women."

But while the troublesome BRCA1 and BRCA2 genes stem from changes in the sequence of the DNA itself, the problems with IGF-2 are epigenetic -- changes in what is attached to the DNA.

The IGF-2 gene normally is controlled by an epigenetic process called "imprinting," in which the copy that used to make proteins depends on which parent the copy came from, rather than the Mendelian notion of dominant or recessive. For IGF-2, only the gene copy inherited from the father should be turned on, and that from the mother should be off. In about five percent to 10 percent of people, however, both copies of the gene are turned on -- a problem called loss of imprinting.

To check for loss of imprinting in people, the researchers will be looking for small methyl groups that should be attached to the DNA near the IGF-2 gene in blood cells from study participants. If neither gene copy has these methyl groups, both will be turned on, resulting in loss of imprinting.

The new project will build on the team's earlier work, funded by a planning grant from the Foundation, which shows that altered IGF-2 status can be detected in blood samples from Caucasians, African-Americans and Hispanics, and that the extra IGF-2 also seems to increase the number of primitive cells that allow the colon to regenerate its lining.

In addition to the practical science, the team will also evaluate the effectiveness of the informed consent process.

"All research investigators are required to obtain informed consent from human subjects. But there is a lot evidence that most consent forms are difficult to read and understand and that relying on the form alone is inadequate to inform a potential research subject," says Holly Taylor, M.P.H., Ph.D., an assistant professor of health policy and management and the Johns Hopkins Bloomberg School of Public Health and research director at Hopkins' Phoebe R. Berman Bioethics Institute. "We're going to try something new and see if we can facilitate understanding."

For this part of their project, the researchers will measure participants' understanding of the research, the blood test under study and the test's results to see whether an enhanced consent process is better than standard practice. The researchers also will explore how a positive test results affects participants' preventive health behavior.

Heading the project is Andrew Feinberg of Johns Hopkins. Other key investigators are Marcia Cruz-Correa of the University of Puerto Rico; Frank Giardiello, Elizabeth Platz, Christine Iacobuzio-Donahue and Hengmi Cui of Johns Hopkins; and Ben Wilfond of the National Human Genome Research Institute. Also collaborating on the project are Edward Giovannucci and Walter Willett of the Harvard School of Public Health.

(Data note: According to the National Cancer Institute, approximately 2.3 percent of women of Ashkenazi Jewish descent have troublesome forms of the BRCA1 and/or BRCA2 genes, a rate five times higher than the general population.)


 Related Stories:
Andrew Feinberg of the Center for Epigenetics on charm school, London and complex diseases

Gene Dose Affects Tumor Growth