Diagnosing one of the many types of cancer can be a complex process involving time-consuming imaging and invasive procedures such as diagnostic surgery. Now, a research team at the Technion−Israel Institute of Technology in Haifa, Israel has developed sophisticated sensors that can “sniff” out ovarian cancer in the volatile organic compounds (VOCs) of patient breath samples. Their study, under the supervision of Hossam Haick, professor of chemical engineering at the Technion, was published in American Chemical Society’s journal Nano Letters. 

While breath analysis is not a new process, according to the study’s lead author, Nicole Kahn, “The flexible sensors allow the collection of a significantly larger amount of data, which allows us to build a physically smaller sensing system, as well as a cheaper one.”

The sensors are composed of a flexible polymer substrate, overlaid with gold nanoparticles to which the VOCs attach. The VOC nanoparticle films are placed over the electrodes and a voltage bias is applied. Thousands of VOCs are exhaled in every breath at very low concentrations—in the parts per billion (ppb). Kahn tells mental_floss, “Changes in metabolism that accompany a specific illness cause changes in the composition and/or concentration of VOCs in the breath and form a distinct pattern.”

This pattern can then be matched to individual diseases and used to “train” a sensory array to discriminate between healthy and sick individuals. Kahn makes clear that the test isn’t detecting a single biomarker for a disease. “A training set of breath samples collected from sick and control individuals is used to teach the array of what a sick person’s breath looks like,” she says.

The team used their sensitive array to test the breath of 43 volunteers that included 17 ovarian cancer patients and achieved an 82 percent rate of accuracy in detection. This success rate is promising for a type of cancer for which the current diagnostic technologies “provide limited sensitivity and selectivity,” says Kahn. “And because the next step after diagnosis would be a high-risk surgical procedure to confirm diagnosis and choose a treatment, only high-risk individuals are screened, to prevent numbers of false positives.”

Cancer may go by one name, but Kahn makes clear that it’s not realistic to speak about cancer in general terms. “It’s a group of many diseases; it isn’t possible to relate to all as a single unit.”

Though the team’s study focused only on ovarian cancer, they have seen some success with diagnosing other cancers by breath, such as lung, breast, colon, prostrate, gastric, and liver cancers. She also says the same technology can be applied to detect non-cancerous diseases such as Parkinson’s, Alzheimer’s, hypertension, and tuberculosis, though much more research is needed.

Their success with these sensors is a promising extension of current breath-based diagnostics, though it will be some time before these arrays are made clinically available. Kahn says that improvements can be made to future generations of sensors “in terms of nanoparticle ligands and film morphology," which will make them smaller and more sensitive. And, Kahn adds, "strain sensitivity of the sensors can be tailored to find the most effective sensors for diagnostic applications.”