Biomarkers: 'I'm Pretty Optimistic'

The best way to attack cancers and other diseases is to be watchful for the earliest signs. That's why medical researchers are excited about the potential for diagnostic techniques involving proteins--workhorse molecules that carry out myriad tasks required to keep the human body functioning. Each disease may trigger its own unique set of protein "biomarkers," which doctors might someday be able to detect in simple blood tests. Geneticist Lee Hartwell, president and director of the Fred Hutchinson Cancer Research Center in Seattle and co-winner of the 2001 Nobel Prize in Physiology or Medicine, spoke with NEWSWEEK's Fred Guterl about the future promise of protein diagnostics. Excerpts:

HARTWELL: You are born with a certain set of genes, so the most you can get out of DNA is inherited risk. The exception is cancer, where there are some DNA changes during the course of the disease. But you're pretty much limited to biopsying the tumor to find the DNA, and if you're interested in early detection or even signs of increased risks, you don't have a tumor to biopsy. Proteins are much more general, they apply to all diseases and they are in easily accessible bodily fluids like blood, urine and saliva. Proteins are also much more diverse than DNA and much closer to the function of the cell. They tell us what is going on in real time.

There are precursor stages for all the cancers we can examine easily--melanoma, cervical cancer, breast cancer or even colon cancer. If you look for certain proteins, you can find them present in these precursor stages. It's just going to take time before we are able to identify all the right biomarkers [proteins] for each type of cancer. The technology for reading that information is at a transition point. Over the next decade we could develop biomarkers for every disease that allow us to intervene much earlier.

We do, but they're not as good as they could be. With mammography, we've reduced the mortality of breast cancer by something like 15 or 20 percent--not a huge amount--and that is fairly expensive to apply to the whole population. With colon cancer, [colonoscopies are] effective because you can clearly see the polyps and remove them while you are doing it, but it is expensive. It is also so invasive that many people will not do it. If a blood test showed that you had a protein that indicates there's a high likelihood that you have colon cancer, you'd go in and get a colonoscopy. If you had a test that said you were having an early-onset disease, like diabetes, then you could modify your diet and other approaches to keep it from getting worse.

That's what we are working on. It's a huge project. We've got blood samples, and we've identified a few thousand proteins. Now we've got to figure out which proteins are associated with the onset of disease. Within three to five years we will know how well that is working. I'm pretty optimistic.

The whole game in cancer is early detection. Currently five-year survival of cancer is 64 percent, which means 36 percent of people die within five years. Decades of increasingly expensive therapeutics have only slightly changed that picture.