Scientists Develop Breathalyzer for Disease

It may one day be possible to walk into your doctor's office and breathe into a small device that will tell you if you are in the earliest stage of a wide range of diseases from lung cancer to asthma to kidney failure.

Early detection is the key to survival when it comes to many medical problems, and the first place some diseases show up is in the breath.

Scientists have just revealed that they have produced a machine that can identify single molecules that are associated with specific diseases, and all that's required is a little breath.

"It's very noninvasive," said physicist Jun Ye, leader of the research team that is working on the technology. "There's nothing to be scared of. No blood test, just a breath test."

Ye is a research fellow at JILA, a joint institute of the National Institute of Standards and Technology and the University of Colorado at Boulder. Ye and research assistant Michael Thorpe, doctoral student Matthew Kirchner and former graduate student David Balslev-Clausen described their work in the Feb. 18 online edition of Optics Express, published by the Optical Society of America.

It has been well established that people exhale a complex mixture of gases, including oxygen, nitrogen, carbon dioxide, and others, Ye told In fact, more than 1,000 different compounds are contained in human breath. But along with those common gases and compounds, people also exhale certain molecules that are considered "biomarkers" indicating specific conditions, such as diseases.

"If you go to the medical literature you will see tons of studies that correlate certain diseases with particular molecules found in the breath," he said. "One common example is nitrous oxide, which is associated with asthma."

Recognizing that fact, scientists for some time now have been trying to develop the technology to identify those molecules in the breath, thus detecting a disease that may not show up anyplace else. Several techniques have been advanced, but Ye said they all fall short because of the old needle-in-the-haystack problem.

Normal breath consists of trillions of molecules, only a few of which are actual biomarkers. And finding just one isn't enough. There needs to be a pattern consisting of several different types of biomarkers that are all associated with a particular medical problem.

What's necessary, he said, is to create a device that will find a few molecules in a sea of background noise consisting of trillions of harmless molecules. He calls it "seeing the forest all at once, but also seeing individual trees extremely clearly."

And that, he said, is exactly what his team has accomplished.

The technology builds on the device that won a Nobel Prize in 2005, called optical frequency comb. Ye and his group applied the technology to spectroscopy, which is used to identify distinct molecules by their emission and absorption of light.

The heart of Ye's machine, which is about the size of a microwave oven, is a cavity between two curved mirrors. Laser pulses are shot into the cavity and reflect back and forth between the mirrors tens of thousands of times, bombarding any molecules in their way, before finally escaping. To test the device, the researchers recruited several students and had them breathe into the cavity.

The bouncing laser beam interacted with the billions of exhaled molecules, identifying the entire composition of the breath. The findings were very precise, Ye said. One of the participants was a smoker, and his test revealed five times the normal level of carbon monoxide.

The beauty of the system, Ye said, is the fact that it sees the entire spectrum, not just a few specific molecules. And that's important because a single molecule would mean little.

"If you have asthma, your breath will have nitrous oxide, but nitrous oxide does not necessarily mean you have asthma," he said. "But if you see several different molecules all at once, and they are associated with asthma, then you have found a real fingerprint of a certain disease."

The technology is still in its infancy, and for now it appears to be limited to diseases that somehow involve the lungs. But Ye thinks that could change, and the applications could broaden, as the technology develops further.

The device can also differentiate between different isotopes, like carbon 12 and carbon 13, and oxygen 16 and oxygen 17, and changes in those ratios could also indicate the presence of a disease. The idea, Ye said, is to build a system that will allow the earliest possible detection in the least invasive manner at a price that everyone can afford.

The technology can now identify a single molecule among billions. The next goal will be to find a single molecule among trillions. That would broaden its application even further.

The researchers built their machine at their institute, located on the University of Colorado campus, at an estimated cost of $50,000. That, of course, does not include the millions spent on earlier research, or the cost of their time.

"Once it's mass produced, the cost could be quite low," Ye said. "It could be put into every doctor's clinic so people could walk in and do painless breath tests and then walk away. They would get results in a day or so."

Of course, all of this is based on lab research, and an experiment with a few college students. The technology has not been tested yet in the medical field. But Ye's office phones were ringing constantly during the interview, and some of the calls were from companies interested in moving the technology from the laboratory to the marketplace.

Lee Dye is a former science writer for the Los Angeles Times. He now lives in Juneau, Alaska.