An electronic nose enables detection of colon cancer and its precursors (adenomas, also known as polyps). The nose 'smells' traces of colon cancer in people's breath.
This is according to research by MDL doctor Peter Siersema and doctor-promotee Kelly van Keulen from Radboud UMC. In the study, they used a so-called e-Nose. Such an electronic nose is capable of automatically detecting thousands of volatile organic compounds (vocs) in the breath. Subtle differences and combinations can be found in this exhaled air. Smart algorithms distil profiles based on these. These profiles can 'indicate quite reliably whether you have something wrong with you'.
e-Nose
The e-Nose is not new. Several publications examining the potential of the electronic device have already appeared in recent months. These showed, among other things, that an e-Nose was able to detect Barret syndrome, a precursor of colon cancer. Another e-Nose has been successfully used to predict which patients with lung cancer will benefit from specific immunotherapy. A third e-Nose can smell whether a patient is carrying the bacteria that cause stomach cancer.
Siersema has been involved in many e-Nose studies. Most recently, Siersema worked with Van Keulen on detecting colon cancer and its precursors using the electronic nose. Here, the e-Nose looks for vocs generated during all kinds of metabolic processes in the body. These can be found in blood, urine and breath.
Changes in voc's
Just as tumours can be detected in blood and urine, in theory this should also be possible using breath. "With an adenoma or tumour in the gut, local changes occur, such as changes in the microbiome," says Van Keulen. "The metabolism also changes in the tumour cells and the cells on the path to tumour formation. Finally, the body as a whole also reacts to the emergence of such a tumour, think of an immune response to attack those tumour cells and other stress reactions. We think that a combination of these three factors causes the composition of voc's in the breath to change."
Currently, a stool test called the immunological faecal occult blood test (iFOBT test) is used to detect colon cancer. Everyone between the ages of 55 and 75 is sent home for an iFOBT test every two years. This test consists of a small brush that you pass through the stool and then send back packed for analysis. "Perhaps the main drawback of the test is that it screens for blood," says Siersema. "But not all intestinal tumours and their precursors bleed. And conversely, there can also be blood in the stool because of haemorrhoids or other problems unrelated to bowel cancer."
Supplement to iFOBT test
Siersema and Van Keulen therefore investigated whether searching for vocs in the breath using an e-Nose could complement the iFOBT test. Van Keulen: "An initial small trial had already been conducted by The e-Nose Company, the company that makes the device and does the analyses of the substances. Based on that initial encouraging data, we set up an initial study in over 500 patients at Radboudumc, Bernhoven and Medisch Spectrum Twente. We recently published the results, with good results. The sensitivity (sensitivity indicates the sensitivity to find what you are looking for with a test) for colon tumours is 95 per cent and for adenomas 79 per cent. The specificity (specificity indicates how specific the test is) at 64 and 59 per cent respectively. Again, these are hopeful results. But it is good to realise that we are still in the early stages of research. Meanwhile, I am working on a study involving 3,000 people. The more data we collect, the better and more reliable the analysis and test results will become. We do not yet have a system that can be used immediately in practice, we will be working on that for some time. But the results so far are very encouraging."
Siersema expects that breath analysis could eventually become very important. "If you can keep making new, well-defined combinations for different diseases based on all those voc's in the breath, you might eventually be able to combine them. Just as you can now extract a lot of data on various diseases from one tube of blood, you might eventually be able to detect multiple diseases in one breath analysis. Not a liquid biopsy, but a breath biopsy. That seems a very nice perspective to me," Siersema explains.
Author: Wouter Hoeffnagel
