Wednesday April 20 2011
It could be years before the test is available in a clinical setting
“A breath test that can detect cancer with an ‘electronic nose’ has been developed by scientists,” reported the Daily Express. The newspaper said the device detects microscopic chemical changes that are emitted in the breath of people with head and neck tumours.
The news report is based on a small study that looked at breath samples from 62 people using the Na-Nose, 16 of whom had head and neck cancer, 20 had lung cancer and 26 were healthy. The test was able to distinguish between these individuals by looking at the presence and levels of certain chemicals in their breath samples. Only two healthy individuals were falsely suggested to have head and neck cancer.
This small study had promising results, but the findings will need to be confirmed in much larger samples, representative of the different types and stages of head and neck cancer. If the results are favourable, then the Na-Nose would need to be trialled against existing standards of care to assess its potential benefits and risks.
A non-invasive breath test for various cancers is an appealing option, particularly for those cancers that are difficult to detect in other ways. Undoubtedly, this is an area that will receive much more research in the future. As the Daily Express reports, it would probably be many more years before a breath test could be available in a clinical environment.
Where did the story come from?
The study was carried out by researchers from Technion – The Israel Institute of Technology – and Rambam Health Care Campus in Israel. Funding was provided by the European Commission. The study was published in the peer-reviewed British Journal of Cancer.
The BBC and the Daily Express both cover this study well, indicating that the study is preliminary, and that it could take years to see whether the test could be used in a clinical setting.
What kind of research was this?
This was a cross-sectional study, looking at the ability of a breath test to distinguish between people with head and neck cancer (HNC), healthy individuals and those with lung cancer. The breath test in question uses a device called the Nanoscale Artificial Nose (or Na-Nose). The Na-Nose uses gold nanoparticle gas sensors that can detect and separate different odours, even at very low concentrations. The device has already been tested for its ability to detect other cancers (breast, lung, colon and prostate cancer).
Head and neck cancer can affect various tissues in the head and neck, including bone, soft tissues, salivary glands, skin and mucous membranes. The researchers report that HNC is the eighth most common malignancy worldwide, that it is often diagnosed late as it lacks specific symptoms and that there are no screening methods available. They say that less than half of individuals achieve an overall cure from HNC, and that lifelong follow-up is needed, as patients often develop a second primary cancer (a new cancer), most commonly HNC or lung cancer.
This sort of study is usually used in the earlier stages of assessing the performance of a new diagnostic or screening test. Tests that show promise in these early stages need then to be tested in larger samples that are more representative of the population in which the test might be used. Finally, if the test performs well enough it might go on to be tested in randomised controlled trials. These trials usually compare the test with standard current practice, to assess whether it offers any benefits (such as reducing the number of deaths) and what the associated harms might be (for example, false negatives, psychological distress and unnecessary investigations due to false positives). If the test was to be used in screening the entire population, or a subgroup of the population, then the cost of providing the test would need to be weighed up against the potential benefits and harms.
What did the research involve?
The researchers collected breath samples from 40 healthy individuals, 22 people with head and neck cancer (HNC) and 25 people with lung cancer (who were matched for age and gender to those with HNC). These breath samples were then tested using the Na-Nose device, to see whether it could detect differences between the three groups.
The participants were all adults (age 24-78 years), some were smokers and some non-smokers. Participants with HNC had been diagnosed using standard methods, including biopsies. Among the participants with HNC, four had earlier stage cancers (stages I and II), and 18 had later stage cancers (stages III and IV). Participants with lung cancer all had stage III or IV cancers. The average age of the healthy controls was younger than that of the participants with cancer (45 years versus 60 years for HNC participants, and 66 years for lung cancer participants). There was a greater proportion of women in the healthy sample (57% female, while less than 15% of the cancer groups were female). The researchers report that the Na-Nose test has been designed not to be sensitive to differences in age, gender and smoking habits.
The researchers collected breath samples under the same conditions for all participants. The inhaled air was filtered to remove any particles or volatile organic chemicals using a mouthpiece. The exhaled samples were collected in a way that allowed collection of air that had been in the balloon-like structures in the lungs called the alveoli, where gases diffuse into and out of the blood. The breath samples from the HNC patients were collected before they received any treatment.
Sixty-two breath samples were tested using the Na-Nose (16 HNC, 20 lung cancer, 26 healthy). The test used five sensors to detect five different volatile organic chemicals. Statistical methods were used to group (or “cluster”) the breath samples based on the similarity of the chemical profiles seen.
To support and validate the findings of the Na-Nose analysis, the chemical composition of 40 of the breath samples was also assessed using standard techniques called gas chromatography and mass spectrometry.
What were the basic results?
The researchers found that the Na-Nose device could distinguish between breath samples from the three different groups of individuals: those with head and neck cancer, those with lung cancer and healthy individuals.
The Na-Nose correctly identified 24 out of 26 healthy individuals (two were falsely identified as having HNC). All 16 people with HNC were correctly identified, as were all 20 individuals with lung cancer.
Gas chromatography and mass spectrometry showed that there were differences in the chemical composition of breath samples from these three groups. The researchers identified groups of six or seven volatile organic compounds that could be used to distinguish between the three groups.
How did the researchers interpret the results?
The researchers concluded that their findings “could lead to the development of a cost-effective, fast, and reliable” breath-based test for identifying individuals with head and neck cancer. They say that the Na-Nose test has potential for use as a screening tool.
This study tested whether the Na-Nose device could distinguish between individuals with head and neck cancer, healthy individuals and those with lung cancer. Although the results from this study appear promising, they are very preliminary. Only 16 patients with head and neck cancer were tested, and much larger samples would be needed to confirm the results. Ideally, such a study would include a broader spread of individuals who are representative of the different stages and types of head and neck cancer.
One limitation of the study is that it is not clear whether the people carrying out the analyses knew which individuals had which diagnosis, which could lead to bias. The authors themselves acknowledge that a larger, blinded trial is needed.
If the test continues to show promise in larger studies, it might go on to be tested in randomised controlled trials, to assess whether it offers any benefits (such as reducing the number of deaths from head and neck cancer) and what the associated harms might be (for example false negatives, psychological distress and unnecessary investigations due to false positives). The tool is most likely to be considered as a screening device for head and neck cancer, which is not currently screened for.
Head and neck cancers are currently diagnosed using methods such as endoscopy and biopsy when suspicion is raised due to symptoms of cancer (which will be variable depending on the type of cancer). However, symptoms of some cancers may be non-specific or only reveal themselves when the cancer is already advanced. If the new breath test were approved for screening, one of the most important considerations would be who to screen. If the test was to be used in screening the entire population, or a subgroup of the population, then the cost would also need to be weighed up against the potential benefits and harms.
A non-invasive breath test for various cancers is an appealing option, particularly if it is difficult to detect the cancer in other ways. Undoubtedly, this is an area that will receive much more research in the future. As the Daily Express reports, it would probably be many more years before a breath test could be available in a clinical environment.