Genetic mutations often get much of the credit for causing cancer, but another important factor is genes that turn on or off at the wrong times, despite being free of mutations. This abnormal gene activity often involves "epigenetic" changes to DNA and in some cases may be reversible.

Epigenetic changes alter gene activity without modifying the genetic code, and are essential to normal development. But the changes can cause problems if they occur when they shouldn't or disable genes that suppress tumors and control growth.

Studies have suggested that epigenetic changes may be as common in some tumor cells as genetic mutations, and many researchers now say that epigenetic changes are critical to understanding, detecting, and treating cancer.

"If you focus only on genetic mutations and ignore epigenetic changes in tumor cells, you may be missing half of the information you need to understand the disease," says Dr. Stephen Baylin of the Johns Hopkins School of Medicine.

A recent study led by Dr. Andrew Feinberg, also of Johns Hopkins, suggests that epigenetic changes in normal tissue can create the perfect environment for cancer to develop if genetic mutations arise later.

Dr. Feinberg's team studied mice with an epigenetic defect found in 30 percent of colon cancer patients and 10 percent of the general population. In people and in mice, the defect causes elevated levels of a cancer-related protein: insulin-like growth factor 2.

According to findings published in the March 25 Science, the flaw did not cause colon cancer in the mice, but it significantly increased the risk among mice that also carried a genetic mutation associated with the disease.

"The idea is that epigenetic and genetic factors may cooperate in causing tumors," says Dr. Feinberg. "We may therefore want to think about what sorts of epigenetic changes might occur in normal tissue that set the stage for mutations that come along later."

To identify epigenetic changes that confer cancer risk, researchers first need to know which changes are normal. This presents a challenge because unlike genetic code, epigenetic changes are dynamic and vary by cell type, age, and sex, among other factors.

The most commonly observed epigenetic change is DNA methylation, in which chemicals called methyl groups attach to DNA, often silencing a nearby gene. DNA methylation can be detected in body fluids such as blood and urine, and new technologies can rapidly scan genomes for epigenetic changes.

Perhaps the most exciting prospect for the field is the potential for detecting epigenetic changes linked to cancer and then doing something about them.

"We cannot reverse genetic mutations, but there is the hope and the potential for reversing epigenetic changes," says Dr. Mukesh Verma of NCI's Division of Cancer Control and Population Sciences. "That's why interest in epigenetics is so high right now."

At least a dozen drugs that target epigenetic changes such as methylation are in clinical trials and more are in development. Last May, the Food and Drug Administration (FDA) approved a methylation inhibitor to treat the rare bone marrow disorder myelodysplastic syndrome (MDS), which can lead to leukemia.

The drug, azacitidine (Vidaza), had nearly been abandoned two decades ago but found new life as researchers showed that it helps MDS patients when given at low doses.

Diet and certain nutrients such as folic acid may also have an effect on epigenetic changes. An NCI-sponsored clinical trial is evaluating folic acid as a tool for preventing colon cancer, and the researchers will be documenting certain epigenetic changes in participants.

"Our diets may alter epigenetic information over the course of many years, but right now we know very little about the effects of diet on epigenetic changes in the long term," comments Dr. Jean-Pierre Issa of M.D. Anderson Cancer Center.

Similarly, Dr. Issa adds, the links between aging and epigenetic changes are important but poorly understood: "We need to develop a better understanding of the epigenome and its interactions with environmental exposures if we hope to understand age-related diseases."

Two collaborative pilot projects on epigenetics are underway in Europe, and there have been discussions about a U.S.-led effort analogous to the Human Genome Project. A meeting on the subject, sponsored by the American Association for Cancer Research, will take place near Washington, D.C., in June.

By Edward R. Winstead