A breather for lung cancer suspects

Researchers in Israel have used cheminformatics methods to "train" an array of gold-nanoparticle sensors to rapidly distinguish between the out breath of lung cancer patients and that of healthy individuals.

Lung cancer accounts for almost a third of cancer-related deaths, with approximately 1.3 million people dying of the condition worldwide every year. Conventional diagnostic methods for lung cancer are expensive, however, and can sometimes miss tumours.

In recent years, researchers have discovered that one's breath can harbour all kinds of markers and signals indicative of health status. Most recently, gas chromatography and mass spectrometry have been used to reveal volatile organic compounds in the breath of patients with lung cancer that are at much lower concentrations in healthy breath.

Now, Gang Peng, Ulrike Tisch, Orna Adams, Meggie Hakim, Nisrean Shehada, Yoav Broza, Salem Billan, Roxolyana Abdah-Bortnyak, Abraham Kuten, and Hossam Haick of the Technion-Israel Institute of Technology, in Haifa, Israel, and colleagues are developing a device that could be relatively inexpensive, portable and amenable to widespread use in hospitals. Haick's system requires no pre-treatment of dehumidification of the breath prior to analysis.

The team have built a nine-sensor array consisting of gold nanoparticles that have been functionalized with different organic groups to respond to various volatile organic compounds present in the breath of lung cancer patients. The system can detect compounds that normally appear at levels of 1 to 20 parts per billion in healthy individuals but are elevated to levels between 10 and 100 ppb in lung cancer patients.

Initially, the team used solid-phase microextraction and gas chromatography/mass spectrometry to identify 42 volatile organic compounds that represent lung cancer biomarkers. They selected four of these to train and optimize the sensor array and demonstrated good agreement between patient and simulated breath samples using principal component analysis (PCA).

The team simulated healthy and cancerous breath made from artificial mixtures of volatile organics and tested them using their sensor array. The two breath types were clearly distinguishable using pattern recognition technology, and the results correspond well with those gathered in experiments using actual human breath. The team collected breath samples from 56 healthy volunteers and 40 lung cancer patients prior to chemotherapy, aged 28 to 60 years to validate their sensor.

The compounds detected are mostly straight chain and methylated alkanes containing between 4 and 11 carbon atoms, such as 4-methyl-octane and 2,3-dimethylhexane, as well as certain derivatives of benzene, including butylated hydroxytoluene. "The compounds that were observed in both healthy breath and lung cancer breath were presented not only at different concentrations but also in distinctively different mixture compositions," the team points out, "The higher concentration and lower concentration in lung cancer breath compared with healthy breath can be understood, respectively, in terms of the release of certain VOCs and the consumption of other VOCs by the lung cancer cells," they explain.

The team also points out that they observed 33 lung cancer biomarker VOCs compared with the 22 cited in earlier reports using GC-MS, which they explain may be due to the pre-concentration method used. However, further stepwise discriminant analysis also revealed nine uncommon VOCs that were present in the breath of at least 83% of the patients but not among most healthy subjects.

The researchers also emphasise that the PCA of the healthy control group showed that the results from the gold nanoparticle sensors were not affected by subject gender, age or smoking habits, which is important if such a diagnostic system is to be able to discriminate between healthy and cancerous breath quickly and easily. They add that experiments with a wider population of volunteers is underway that will take into account diet, alcohol consumption, metabolic state and genetics.

"Our results show that sensors based on gold nanoparticles could form the basis of an inexpensive and non-invasive diagnostic tool for lung cancer," the researchers conclude.