The scientists are not sure what biochemical pathway might be generating those two compounds or what process is causing the change in concentration. "It's not surprising that there might be a physical or chemical basis to this distinctive pattern of odors," says dermatologist Paul Nghiem of the University of Washington and the Fred Hutchinson Cancer Research Center, in Seattle, who was not involved in the study. Skin-cancer cells act differently on many levels, from gene expression to protein synthesis. "If some of those proteins are volatile, it's not a stretch" that differences could be detected, Nghiem says.

Gallagher and Preti are continuing to look for other biomarkers for skin cancer. When they're identified, sensor technology like the electronic noses under development could be programmed with the levels of chemicals indicative of a tumor. "Instead of a visual exam and a biopsy, you could have a sensor that you could wave across the body," Gallagher says.

The current study focused on patients with basal cell carcinoma because there are many more cases of this cancer than there are of the more serious melanoma, making it easier to recruit volunteers. The researchers are actively looking for patients with melanoma so that they can repeat the study. "We have to have a window where it's safe for them to be diagnosed and not come back to the doctor for a week or two," Gallagher says, giving the patients time to do the wash-out and participate in the odor sampling.

It is with melanoma that a rapid detection method would be most useful, says Nghiem. "It's a life-threatening disease and can be clinically difficult to diagnose," he says. Basal cell carcinoma, on the other hand, is fairly easy to diagnose, rarely fatal, and readily curable by surgery.