MADISON, WIS.–"Targeted medicine" has been a tantalizing yet elusive goal. The lung cancer drug Iressa, tailored to a particular kind of cancer DNA, has in a few patients worked spectacularly well but in the majority has not. Now Paul Bertics, a biomolecular chemist, may have figured out why. Not only that, he may have designed an easy test to determine who will do well on the drug and who will not–a test using liquid crystals like the ones in your digital wristwatch. Recently Bertics sat down in his office at the University of Wisconsin–Madison to explain how this works:

What's the problem with Iressa?

It's a great drug for a few people. In maybe 5 percent of patients, it stops tumor growth or shrinks cancers for months or years. That's great in lung cancer, which is a deadly disease. But we haven't been able to tell which 5 percent will respond to the drug, so most of the people who've received it have not benefited. If we could figure this out, we'd have a wonderful treatment.

Does it work differently in people with different genes?

Yes, that's what we think.

So why not just look at their genes?

It's really hard to get big enough samples of lung cancer cells so you can do genetic analysis. The cells are hidden in the chest, in the lungs, and you can't get enough material. If we could figure out a way to amplify the signal from this tiny material, however, we could figure out who should get the drug. And that's when an engineering colleague of mine here, Nick Abbott, thought we could do this with liquid crystals.

Liquid crystals? Like the ones in your watch?

Well, not my watch, because I have an old-fashioned analog one with hands. But like the ones in your watch, I see.

How can crystals test drugs?

Because they form tiny, nanoscale walls that block light. We found that if a drug binds strongly to parts of a cancer cell, and crystals stack on top of that, it changes the light, and we can see that change.

This actually works?

Yes, amazingly. You have to understand a little bit about these crystals to see how. We can blow a stream of them at a microscope slide, and the crystals will stack on top of one another in the same orientation, like bricks in a wall. Actually, they will form a series of parallel walls. So if we shine a light at that slide from one side, the walls will block the light. But if you mess up their formation, more light will get through. Like a badly built brick wall.

And does Iressa mess up that formation so you see a different amount of light?

Not Iressa by itself, but Iressa if it binds to parts of cancer cells. Iressa binds tightest–and works best–to cells with a particular string of genes. So here's how our test works: We can lay down Iressa on the slide, and then lay down a cancer sample from a lung patient. Remember, this is a tiny amount of material. But if Iressa binds to the cancer material, that means extra molecules are on the slide surface. When we next try and build our crystals walls, those molecules disrupt the alignment. On the other hand, if Iressa does not bind to the cancer, the extra molecules aren't there, and the crystal walls are not disrupted. In either case, when we shine a light at the slide, we will see a different signal.

How long does this test take?

A matter of minutes. And it's a lot easier to detect light than it is to detect genes, so it's actually fairly simple.

How would this help a patient?

If the test reveals Iressa binds tightly, that patient should benefit from the drug. If not, the drug wouldn't work.

So now what's the holdup? Why isn't this helping patients right now?

We need to do more tests. We need to figure out what amount of change in the light we detect correlates with a medical change in a patient. Once we have reliable benchmarks for that, labs can easily use the test and report back to a doctor that the drug is or is not appropriate.

Is this just a test for Iressa?

It should work on a variety of compounds, drugs, or cell material. It doesn't require a lot of complex instruments, so we're pretty hopeful the technique can reveal a lot of biological activity that's previously been hidden from us.

Find out more: Go to the U.S. News cancer center to get in-depth information about symptoms, tests, treatment, and prevention.