of
The Association of Coloproctology of Great Britain and Ireland
Advancing the cure and treatment of bowel disease /
Research Project Annual Report
Principal investigator / Dr R ArasaradnamCo-PI : Dr J Covington
Mr C Harmston / Institution / University Hospital Coventry & Warwickshire/ University of Warwick
Project Title / Characterisation of volatile organic compounds (VOCs) as early markers of colorectal cancer using an Electronic ‘e’ nose.
Start date / November 2011 / Finish date / 30th November 2013
Lay Summary
(max 500 words) / Background: Colon cancer is the largest killer amongst non-smokers with almost 40,000 new cases diagnosed annually in the UK. UHCW, the regional tertiary colon bowel disease centre, treats over 200 cases per year. If diagnosed early, the prognosis is good with ~80% five year survival. The introduction of colon cancer screening in the UK has aided somewhat in early diagnosis but uptake into the programme remains poor ~ 40%; predominantly due to the stigma of giving faecal samples. Furthermore, individuals that return positive tests have to undergo a colonoscopy which is invasive, uncomfortable and has an associated morbidity. Of these, only 10% will actually have serious pathology e.g. cancer.
Method: We believe that there may well be bio-markers of colon cancer dissolved within a patient’s urine sample. We propose to use an electronic nose (a tool that replicates the function of the human nose) to detect vapours/gas that emanate from the urine sample with the aim to create a ‘chemical fingerprint’ for those with colon cancer. We have investigated the possible detection of bio-markers in the gas/vapour emissions from urine samples for IBD (inflammatory bowel disease) with success and we plan to apply the same concept to colon cancer. We are able to distinguish patients with Crohn’s disease, diabetes and ulcerative colitis from urine samples and envisage that this or a similar method can be applied to colon cancer. To support this concept, a Japanese group have used of a dog to identify colon cancer based on this principle.
Results: We believe this methodology could become the basis for a simple, low-cost, non-invasive method for mass screening to help inform treatment pathways. Our early preliminary results of 20 patients with colon cancer have confirmed the ability of e-nose to distinguish those with colon cancer from healthy individuals as well as those with ulcerative colitis (inflammatory bowel condition with risk of developing bowel cancer). This confirms our original hypothesis on the use of electronic nose to detect colon cancer.
The change: Once validated in a large population, our method could be deployed within hospitals and GP practices, to make the maximum impact and aid in identifying those at risk. This would help make screening more socially acceptable and reduce the number of unnecessary colonoscopies. If screening uptake was improved this would hopefully translate to reduced mortality.
Background (purpose for project) / Our concept is to use an “Electronic Nose”, an instrument that artificially mimics the biological olfactory system. These instruments employ an array of different broadly tuned, gas phase, chemical sensors. This broad tuning is to a range of different gases and vapours, but importantly no two sensors are tuned to the same chemical groups. When a sample is presented to the e-nose, each sensor responds to the complex mix of chemicals within that sample. As each sensor is different their responses will be unique within the array. By taking the responses as a whole, it is possible to create a chemical ‘fingerprint’ of a smell – or as proposed here - disease. Thus, when the e-nose is presented with the same sample it is able to recognise the smell by its fingerprint. It is unique in that, like a human nose, it identifies odours through their multi-component nature. Hence, it can identify similarities and differences between samples, without the need for identifying the individual components.
The strength of the electronic nose is that it is able to identify overall changes in profile, in real-time and in a way that is easily understood. The electronic nose attempts to map a new smell profile (in our case a urine sample) onto a learnt database, thus giving a user a classification and a confidence in that classification. Such instruments have the additional advantages of small form factor, low cost, ease of use and low maintenance compared with traditional gas analysis instrumentation.
Our initial studies were based on the use of these instruments to identify the chemical fingerprint of Inflammatory Bowel Disease (IBD) from patient’s urine samples – work which is ongoing. Here we have shown the ability of the electronic nose to distinguish between Crohn’s disease, Ulcerative colitis and controls. Moreover we are also able to track VOC changes in these individuals as they move into remission (1,2). Furthermore, we have performed a pilot study (in collaboration with the Royal Marsden Hospital), on patients who have received radical pelvic radiotherapy including colo-rectal cancer. Here two groups were identified based on their VOC ‘fingerprint’ - one which developed low gastrointestinal toxicity and the other with high gastrointestinal toxicity (3). Aside from proposing that electronic nose may be able to predict those individuals who may go on to develop gastrointestinal toxicity, it also confirms the validity of using these instruments to characterise VOC ‘chemical fingerprinting’ in disease groups.
Introduction / The premise of our work is early detection of disease using novel, non-invasive technology that can eventually by used at point of care. Here we have employed a number of e-nose instruments (designed to mimic the human olfactory system) for the detection of colon cancer. We are also comparing these results with standard techniques such as gas chromatography/ mass spectroscopy. Our earlier results in IBD patients were promising and its potential in colon cancer has proved equally hopeful – see results and discussion section below.
Methods / Our proposed methodology is to perform a prospective observational cohort study to characterise different VOC/gas signals from both urine and faeces in patients with colorectal adenomas and cancer compared with controls. Patients will be identified from two week wait clinics as well as via the Bowel cancer screening programmes (UHCW; screening centre since 2003 - largest experience of screening in England). We aim to:
1. Collect urine and faecal samples from those with cancer and normals. Patient demographics and drug history will be collated and biological samples will be stored at -80 oC until analysis
2. Biological samples will then be thawed and sampled as per our current protocol and the air above the sample (headspace) will be analysed for VOCs/gases using a AlphaMOS Fox 4000, Owlstone Lonestar Ion Mobility Spectrometer , our in-house WOLF (Warwick OLFaction) – all types of electronic nose - and Bruker Scion SQ Gas chromatography/Mass Spectroscopy (GC/MS) instrument to aid identification of compound(s) that the differences may be attributed to.
3. Undertake analyses to determine if the e-nose can distinguish those with cancer and controls and then evaluate the GC/MS output in an attempt to find potential compound(s) resulting in the differences identified. This will be performed on urine and/or faecal samples.
Multivariate analysis will be undertaken using sophisticated software including Matlab and multisens, with different classification techniques applied to the resultant data (such as Principle Component Analysis and/or Linear discriminant Analysis).
Results and discussion / Goals Achieved:
We have already recruited 83 individuals with colon cancer. These include paired urine samples (166 urine samples) before and after treatment (surgery/chemo-radiotherapy). We also have 83 serum samples pre-therapy from these individuals.
Preliminary Results:
Our promising preliminary results of 20 patients with colorectal cancer (11 male, mean age 69, mean BMI 27), 20 with Ulcerative colitis (in remission as defined by SCAI score of <4) and 7 healthy volunteers are shown here (currently supported by the BDRF). Figure 1 shows the AlphaMOS Fox 4000 and the Owlstone lonestar (both types of e-nose) plot where the group with colon cancer are tightly grouped and are disparate compared to healthy controls and even those with Ulcerative colitis (p < 0.001). . Figure 2 shows our GCMS system, fitted with the CTC CombiPAL autosampler (purchased from the generous support of the BDRF). Figure 3 shows a chromatogram of colon cancer (green) and healthy individual (red) - superimposed. Whilst most peaks overlap, there are clear differences between these individuals.
Figure 1: Results from Electronic nose (a) and FAIMS (b) to two disease groups and healthy volunteers.
The autosampler system was only available at the end of this preliminary test, hence the reason for the limited results.
Figure 2: Bruker Scion GCMS, fitted with the CTC CombiPAL
autosampler
We are currently in the process of optimising both the electronic nose and GCMS analysis protocols. We hope that this work will improve our ability differentiate colon cancer from other disease groups and controls. .
Figure 3: GC/MS results from the headspace of urine samples from a colon cancer patient and a control
Conclusion / We have thus demonstrated the utility of the electronic nose to detect colon cancer from urine samples which are disparate from those with ulcerative colitis and healthy volunteers. GC analysis also confirms the different VOC pattern in patients with colon cancer compared to healthy controls. Further confirmation is required within a larger cohort.
Recommendations for future work / With these results we feel a further, detailed study of the potential of this new exciting method is required. We will endeavour to complete and analyse 100 patients but given our preliminary results feel that the study holds promise.
Based on our preliminary results, we wish to considerably expand our study with added support to:
1) Collect samples from far larger patient and control groups (175 patients in total compared to 20 patients studied thus far) to understand the variance/spread in both the healthy and diseased state and overlap with other IBD groups;
2) Evaluate different gas sensing technologies to identify which approach offers the highest level of successful identification (and least number of false positives);
3) Obtain a greater understanding of the chemicals (vapours and low molecular weight gases) are being modified by the disease state – for example, is this an addition/subtraction of a molecule(s) or a modulation of the existing compounds found in all urine samples.
References
(author, title, date of publication) / 1. Arasaradnam et al. Evaluation of human disease using an electronic nose. J Med Eng 2011 25:2;87-91
2. Arasaradnam et al. Novel tool to diagnose and track patients with IBD. In Press 2012
3. Covington et al. Radiation induced gut damage – identifying at risk patients using an electronic nose. In Press 2012