As drug users learned to their dismay years ago, urine can be a tattletale. But urine does not just contain remnants of illegal drugs. It also contains by-products of many chemical reactions occurring in the body.

Could urine tests be used to screen for cancer? That possibility was raised by recent work of Yinfa Ma, a chemistry professor at Truman State University in Kirksville, Mo.

It's known that some cancers produce signature chemicals. Prostate cancer, for example, makes an antigen that can be detected with a common blood test.

Ma started the study knowing that cancer cells produce an excess of pteridines, a group of chemicals that play many roles in body chemistry. Existing tests for pteridines were expensive, slow and cumbersome, so he, together with colleagues from the Chinese Academy of Sciences, set out to invent a new one.

Their research, just published in the journal Analytical Chemistry, studied urine from 19 people. If the finding is confirmed in larger studies, urine could become the gold standard for cancer screening.

How does it work?
In their search for vanishingly small concentrations of pteridines, the researchers chose a process called capillary electrophoresis. They inject a sample in one end of a very skinny tube (with an inside diameter of 50-millionths of a meter) and an electric field propels molecules in the sample toward the other end.

The molecules are separated because their speed of travel depends on their mass and charge. In general, larger molecules experience more friction and move slower than small molecules. The process is similar to gel electrophoresis, a standard method geneticists use to separate fragments of DNA by mass.

After five or ten minutes in the tube, the molecules reach the other end, where a laser zaps them, raising their electrons to higher orbits. When the electrons return to their original orbits, they release a photon, a particle of light. (If you must know, the process is called "laser-induced fluoresence.")

Counting photons tells you how many molecules were fluorescing in the detector.

To know which molecules you are counting, you simply look at your watch. Seriously -- different molecules reach the detector at different times.

Ma starts the test by running standard samples of various pteridines down the pipe and recording their arrival time at the detector. Then, when he starts testing urine samples, he already knows, for example, that pteridine "X" will reach the detector in 4.3 minutes. The resulting graph shows the concentration of each pteridine.

Revealing results
The test showed that pteridine levels do vary when the urine of cancer patients is compared to that of college students. (Was the latter, how to say this diplomatically, perhaps saturated with beer?)

If future work confirms these results, urine tests could become a cheap, non-invasive screening for cancer. But much more helpful than a generalized test for cancer would be one that identifies the type of disease. Ma's research group is trying to develop individual "pteridine signatures" for specific cancers. Unpublished work, he says, shows the existence of such signatures in urine from patients with nine types of cancer.

At best, years of work will be needed before urine tests are approved for stand-alone screening. But before then, Ma says the tests might be used to confirm a diagnosis of cancer made with other techniques. If, as he suspects, levels of pteridines reflect the tumor's growth, the levels could also be used to monitor treatment.

Beyond screening, there is a possibility that the change in pteridine levels could point the way toward some kind of drug treatment. That question may be interesting, but it won't be on the agenda unless considerably more funding arrives, Ma says. "We can see many purposes for the test, but we can only do one thing at a time."

-- David Tenenbaum

Pteridine Analysis in Urine by Capillary Electrophoresis Using Laser-Induced Fluorescence Detection, Futian Han et al, Analytical Chemistry, April, 1999.