Maastricht University Professor Frederik-Jan van Schooten and Assistant Professor Dr Agnieszka Smolinska tell HEQ about the need for non-invasive biomarker testing for colorectal cancer.
HEQ last spoke to Professor Frederik-Jan van Schooten and Assistant Professor Dr Agnieszka Smolinska, of the Department of Pharmacology and Toxicology at Maastricht University (UM) School of Nutrition and Translational Research in Metabolism, in January 2019 (see Health Europa Quarterly issue 8). They told us about their ABC-Cancer project, which aims to provide non-invasive colorectal cancer screening using biomarkers in a patient’s breath, rather than faecal blood testing, which carry a high risk of presenting false positive results.
van Schooten and Smolinska give us an update on their progress and the significance of volatile organic compounds as biomarkers for colorectal cancer.
How important is early diagnosis in the treatment of colorectal cancer? Should screening be more widely available?
van Schooten: Early diagnosis is very important. The earlier cancer is detected, the better patients’ prospects are, especially for those who are older or already in at-risk groups – for instance, a patient whose colon is already inflamed will be at higher risk for colorectal cancer. Colorectal cancers are a silent killer: patients do not notice it until it has already metastasised, so early and widely available screening programmes are important.
Smolinska: The colon screening programme in Maastricht has been on hold due to the COVID-19 pandemic, but the number of new cases has been going down, so they are now trying to gradually restart the healthcare system – because otherwise, everything will collapse. You cannot have a healthcare system spending an entire year focusing solely on COVID-19, because people will still die of other preventable diseases. Particularly in terms of colorectal cancer, the five-year survival rate for Stage 1 patients is around 90%; but by Stage 4 that drops to around 14%.
van Schooten: The stool test which is currently used in most countries’ colorectal cancer screening programmes is quite nonspecific: the test has to be quite sensitive to detect the cancer, but it does also detect a lot of non-cancerous cells as potentially cancerous. This leads to false positive results, where patients have to undergo potentially unnecessary colonoscopies; and that places a burden on endoscopist facilities, as well as on the patients themselves. That’s why we have been looking for a non-invasive test, so that more patients – particularly patients who are already at risk, such as those with ulcerative colitis or Crohn’s disease – can receive accurate screening and consequently avoid unnecessary colonoscopies.
Smolinska: Around 30% of colonoscopies conducted on suspected cancer patients are potentially unnecessary. If we can reduce that by even 10%, we will save a lot of money and time which could be put towards clinical treatments which are more urgently needed. In the future, though it is still some years away, patients may just be able to take a breath test during a regular check-up at their GP’s surgery – this would be especially beneficial for elderly and immunocompromised patients, as they would be less at risk of infection visiting their GP or even testing at home.
What are the key objectives of the ABC-Cancer project?
van Schooten: The main goal of the project is to find a screening tool which can be applied easily to high risk groups, and which can be implemented in addition to existing ongoing testing protocols. We want to use a combination of breath and faeces samples as data sources for biomarker identification, from which we will build an algorithm. Similar technologies are in place in Germany and Poland, so we are putting protocols in place to standardise our sampling methods in accordance with their metrics.
We have already conducted studies on inflammatory bowel disease, identifying the compounds in exhaled breath which are critical in the active state of the disease – it is important to know whether inflammatory bowel disease is in its active state, because then the physicians have to take action to treat it; normally they would have to perform a colonoscopy to determine whether it is in an active state or not, so this process will support clinicians in determining whether treatment is needed. We have been collaborating with the German and Polish labs, in order to obtain as many patients as possible and subsequently introduce as much variation as possible for finding robust biomarkers; it is crucial that even when using differing detection platforms and other groups of patients in different surroundings, the diagnostic algorithms should still do their job.
Smolinska: We want to be sure that if relevant volatile compounds are present in a patient’s breath or faeces in Germany or Poland or the Netherlands, it does have biological significance and won’t just be a reflection of the location of the test, or the equipment which is used, or the patient’s dietary patterns. We want to ensure that the indicative molecules we find are cancer specific and eliminate the external factors.
How has the project progressed since we last spoke to you in 2019?
Smolinska: We’ve had some ups and downs: we had some issues with PhD students beginning the project and then dropping out; and some partners were temporarily withholding funding. All this meant we only managed to begin the project in earnest around the summer of 2019, once all our partners and stakeholders were fully on board. We are currently very busy optimising our breath and faecal sampling protocols: our initial plan was to use an established method of breath sampling called the ReCIVA® breath sampler, but we are now in the process of modifying the ReCIVA to make it more sustainable.
The traditional ReCIVA uses silicone masks which are thrown away after a single use; and our plan is to sample around 3,000 to 4,000 patients overall, which translates to a lot of single use masks. Therefore, we are now trying to adjust the sampling system in order to facilitate the sterilisation and reuse of masks. Of course, this then requires a lot of checking and optimisation, because many potentially interesting compounds in the human breath are only present in very small concentrations; so if we make any adjustments to the sampling process, we have to be certain that we are mostly measuring the concentrations which are already present within the sample, rather than external contamination.
The second aspect of this is that we are studying volatile organic compounds; we are surrounded by those compounds constantly, everything in our homes is made from organic compounds which have their own scent. In making adjustments to the mask, we have to be conscious that the mask itself will not release interfering compounds, so we have to select materials very carefully in order to minimise the risk of external sample contamination from the sampling device.
van Schooten: In place of the traditional masks, we now use mouthpieces in combination with barrier filters which are normally used in lung functionality tests. We produce these mouthpieces via 3D printing, using materials which have been specifically selected to release the fewest possible volatile compounds which could interfere with our tests. This is a collaborative effort with our German partner institute: they have a lot of experience both in preparing these mouthpieces and in observing the concentration of volatile compounds in different materials. We have also been collaborating closely with the manufacturer of the ReCIVA sampler, Owlstone Medical, an innovative company in Cambridge, UK, which is working on getting breath tests into clinical settings.
Smolinska: Everything depends on the temperature. If we leave a sample near a fan, it may start smelling stronger than at a temperature of 20°C. The same is true of this new mouthpiece that we are working with: we really have to check what level of contamination we can expect at different temperatures.
The device itself is very user friendly and patient friendly. It’s a mask around 30cm long which attaches to the patient’s face, then tubes filled with absorbent material are placed into the mask; the material absorbs the compounds from the patient’s breath while the patient can just sit comfortably on a couch, breathing through their mouth. The process typically takes about 10 minutes.
van Schooten: When there is a tumour in the colon, it produces volatile metabolites which are released as part of the gaseous exchange process which occurs in the lungs. The device is equipped with a CO2 sensor to ensure it is collecting air from the section of the lungs where this gas exchange occurs.
Smolinska: To further eliminate the risk of external contamination and to optimise the welfare of the patient, the device delivers clean air to the patient: the air which the patient inhales during testing is first passed through carbon and high efficiency particulate air (HEPA) filters to eliminate inhalation of ambient volatile organic compounds. If the test is conducted in a hospital setting, for instance, and half an hour beforehand a cleaner disinfects and cleans everything with using strongly scented products, the potential contamination that could cause is eliminated by the carbon and HEPA filters.
We are also in the process of assessing the instruments we use to collect faecal samples: the key challenge there is that, while breath contains a lot of water, faeces contains significantly more – and water is in effect a ‘silent killer’ for the absorption materials we use to collect compound samples, as the presence of water prevents other compounds from being absorbed.
What is the significance of identifying volatile organic compounds as potential biomarkers for disease?
Smolinska: If you know a compound which is specific to a form of cancer, the first step is to determine: can I use this knowledge to make an easily accessible diagnostic tool? Then the second step would be using detection of the significant compound in treatment monitoring, observing and comparing the presence of the compound in patients who have undergone different therapies or dosages.
van Schooten: Cancer cells have different metabolisms than normal cells; they use chemicals such as glucose differently than normal cells and they have other cancer-specific metabolic characteristics – for instance, oxidative stress is generated during cancer development – and it is still a challenge to relate the compounds to this altered metabolism. That’s why we are looking at the separated compounds using gas chromatography-mass spectrometry (GC-MS): we separate all the volatile chemicals using GC and then we identify them using MS. Once we know which of these compounds contain the information which enables us to distinguish cancer patients from non-cancer patients, we can begin to investigate where these compounds come from originally, which mechanisms produce them.
Currently, we are working on in vitro systems, whereby we try to alter mechanisms in cells and then see whether that changes the compounds they produce. We have also developed a system to capture breath from mice for further research: we have tested that instrument to ensure it is reproducible, and the mice release a similar profile of volatile compounds to humans, so we will be able to conduct more mechanistic studies which may be difficult to conduct on human patients. Research is still ongoing into the potential offered by electronic noses and other sensors – these can be based on lasers, metal oxides or other colorectal cancer detection methods.
How do you expect to be able to build on the results of your research so far?
van Schooten: I’m reluctant to put a timeframe on it as it is still very difficult to bring these technologies into a clinical setting. We need more funding and we need more staff: Owlstone is doing a great job on that front; they have a large group of people working on these kinds of technologies, doing clinical studies and accommodation studies, exploring the legal side, working on ways to protect intellectual property, determining how best to bring it into clinics, how to deal with health insurance companies, how to deal with the Food and Drug Administration (FDA).
Smolinska: It depends on the results – positive results are nice, but bringing it to the clinic will still be difficult. The compounds that we find may be difficult to build a usable and robust device around. We do hope to get our research translated into clinical protocols, but there is still a lot of research to do.
Frederik-Jan van Schooten
Dr Agnieszka Smolinska
This article is from issue 14 of Health Europa. Click here to get your free subscription today.