Wandall and his colleagues set out to design a test to detect cancer-related autoantibodies that act on a specific class of tumor antigen called mucins. In recent years, researchers have focused on mucins as possible tumor markers because they are present in abnormally high levels in multiple types of cancer. However, like other tumor markers, mucins have also been observed in noncancerous conditions.

To elicit a cancer-specific autoantibody response, the researchers theorized that they would have to engineer the mucin antigen to include a cancer-specific trait. In the past few years, scientists have discovered novel changes that occur on the surfaces of cells as they turn cancerous. Normally, the surface of healthy cells is covered with proteins attached to long chains of sugars. In cancer, these sugar chains, or glycans, are abnormally cut short--and the trait is expressed only in cancerous cells.

The Denmark team hypothesized that by combining a tumor-associated antigen like mucin with a cancer-specific glycan change, they could detect a robust autoantibody response specific to cancer.

The researchers chemically synthesized a mucin called MUC1, and attached it to short glycan chains similar to those found in cancer. They then bound the structures to the surface of a microarray. The team collected serum samples from three groups, each including 20 patients with prostate, breast, or ovarian cancer. Wandall and his team passed each serum sample through the microarray, followed by a solution containing a fluorescent agent illuminating the presence of autoantibodies.

The group found that autoantibodies were present in 20 to 30 percent of cancer patients, versus none in samples from healthy individuals. Wandall says in the future, he plans to increase the specificity of the test by combining the short glycan structures with other cancer-associated proteins.

Michael Hollingsworth, a professor of biochemistry and molecular biology at the University of Nebraska Medical Center, collaborated with Wandall in designing the screening test. Hollingsworth envisions such a test as a regular screening tool incorporated into a yearly physical. "You wouldn't want to use this as a diagnostic to say yeah, you have pancreatic cancer," says Hollingsworth. "You just want to use it to say, okay, you want to look at this more closely."

Suzanne Miyamoto, associate professor of hematology and oncology at the Cancer Center at the University of California, Davis, is developing a similar cancer screening tool by incorporating cancer-related changes in glycan structures. Miyamoto sees promise in Hollingsworth and Wandall's design. "I think it has potential," says Miyamoto. "It's only preliminary, and they will definitely need to do more validation processes to look at different cancers like benign versus malignant, and see how well that stands up to a clinical lab test. But their platform is very adaptable to the clinical lab."