Washington state scientists want to help revolutionize the way cancer is detected. They have begun working on simple blood tests to find cancers at their earliest stages —...

Share story

Washington state scientists want to help revolutionize the way cancer is detected. They have begun working on simple blood tests to find cancers at their earliest stages — and even identify people at high risk for the disease.

By identifying blood-borne proteins associated with certain cancers, the researchers hope physicians will someday be able to use the blood tests to find and head off cancer with early, aggressive treatment.

“We’re trying to develop a sort of blood telescope … how to look in the blood and tell what’s there,” said Dr. Martin McIntosh, a Fred Hutchinson Cancer Research Center scientist who is leading a Seattle and Washington, D.C., research team with a $9.7 million grant from the National Cancer Institute. University of Michigan scientists also are working on the two-year project, announced yesterday.

Most Read Stories

Unlimited Digital Access. $1 for 4 weeks.

McIntosh, who has worked for years on the early detection of cancer, said widely available blood tests could come as soon as five years from now. Such tests could be used to detect cancer in new patients and assure worried cancer survivors that their disease is not recurring, he said.

Collaborating in the Hutchinson-led research are the Institute for Systems Biology in Seattle, Pacific Northwest National Laboratories in Richland and the Plasma Proteome Institute in Washington, D.C. The institutions are all leaders in the advancing field of proteomics, which analyzes and catalogs the proteins made by living organisms.

The scientists will initially focus on finding proteins or groups of proteins linked to cancers of the breast, prostate, ovary, colon, skin and lung. They are among the most prevalent cancers, and protein-discovery work already has been done on some of them.

In the first year of the two-year project, the researchers will refine techniques to identify proteins extracted from blood and other tissue.

Then they will test the technology to see how well it detects cancer in laboratory mice. The animals have very humanlike cancers and complex blood similar to that of humans.

Once the proteins are identified, studies of large numbers of humans will be conducted to see how well the tests detect the different forms of cancer.

Proteins now being used to help detect cancer have a spotty record. Blood tests for proteins associated with prostate and ovarian cancer have a high rate of false positives. And Hutchinson scientists reported last May that about 15 percent of older men with normal prostate specific antigen (PSA) readings actually had prostate cancer, some of them an aggressive form.

McIntosh said the new protein studies will use state-of-the-art computer analysis to identify which proteins are most important in diagnosing cancer, including the most aggressive forms.

But it still will be a hard job, he said: So far, only a small fraction of the estimated 1 million proteins produced by the 30,000 human genes have been detected in blood. Advanced technology will be able to find plenty more, but the challenge is knowing which proteins really are important.

“The challenge is still there,” McIntosh said. “Our approach will help build understanding of how cancer develops.”

All of the research conducted by the Seattle and Michigan teams — the software, research protocols and other types of information — will be available for other scientists worldwide by way of the Internet, McIntosh said.

“It can be a resource for the early detection of cancer,” he said.

Warren King: 206-464-2247 or wking@seattletimes.com