A pilot case-control study about volatile organic compounds in exhaled breath in primary sclerosing cholangitis

Abstract

Main Outcome Measurements

Selected ion flow tube mass spectrometry was used to analyze the concentration of the subjects prevalent VOCs in bile samples. Logistic regression analysis was performed to build a predictive model for the diagnosis of CCA. Four studies examined VOCs in gastric and colorectal cancers showing a change in metabolic biomarkers of cancer patients compared to the control group. VOC analysis in the study ensured that the differences are highlighted in the lifestyle as well as the metabolism of individuals. Refinement and continuous improvement of analytical tools allowed better detection and quantifying of the core constructs, thus enabling the discovery of a broader spectrum of pathophysiology and diseases’ biomarkers.

Results

Numerous compounds (ethanol, acrylonitrile, acetonitrile, acetaldehyde, benzene, carbon disulphide, dimethyl sulphide, and 2-propanalol) were considerably different in patients with IBD complicating PSC compared with PSC and IBD. (P<.05). The model [2.3239* log (acrylonitrile) + 0.9871*log (3-methylhexane) + 0.8448*log (benzene)] < −0.12 identified the patients with IBD [area under the curve (AUC=0.89)], with 90.5% sensitivity and 72.7% specificity. (p=0.02).

Limitations

Sample size

Conclusion

The measurement of VOCs in the biliary fluid may be useful in differentiating IBD in PSC patients with PSC and IBD patients. A more extensive study with a longitudinal study design is required to confirm our pilot observations to diagnose CCA early in patients with PSC.

Keywords: Endoscopic retrograde cholangiopancreatography, volatile organic compounds, bile, cholangiocarcinoma, primary sclerosing cholangitis

Introduction

Volatile organic compounds (VOCs) can be detected in the exhaled air. “Volatile sample components diffuse into the gas phase, forming the headspace gas. Headspace analysis is the analysis of the elements present in that gas. We have previously described the utility of measuring VOCs in bile in the diagnosis of pancreas cancer and differentiating it from chronic pancreatitis. While patients with beginning pancreatic cancer may exhibit nearly similar conditions with PSC patients, PSC signs are distinct. The main challenge in PSC patients is that, once we see a dominant structure, the diagnosis of PSC in these patients is particularly challenging. In the current study, we sought to specifically study PSC patients and extend these observations to determine the discriminatory VOCs in IBD, IBD + PSC, and PSC patients. Our pilot study aimed to identify potential VOCs in the bile that could be used to distinguish PSC-IBD from PSC without IBD.

Methods

Cytology results are used as the basis through which the COVs of PSC and IBD are differentiated. Furthermore, surgical pathologies and a careful examination of clinical imagery tests could be used to identify the biomarkers that discriminately identified PSC and those that identified IBD.

Statistical Analyses

A comparative study of the two diagnosis groups of IBD and PSC concerning bile VOCs using the Wilcoxon rank-sum test for the pairwise group was done. PSC models adopted consistent regression analysis performance, considering different components. To differentiate the VOCs in the two diagnosis groups, a model was conceived. Independent variables accommodated a step by step inclusion of Bile VOCs as an independent variable premised on highly improved amongst the candidate predictors in Akaike’s information criterion

Results

There were numerous compounds which were considerably lowered with IBD in setting of PSC compared with PSC without IBD. This compounds including ethanol, acrylonitrile, acetonitrile, acetaldehyde, benzene, carbon disulphide, dimethyl sulphide, and 2-propanalol. An adjustment of VOC level for smoking and alcohol use amongst other aspects were investigated while retaining the results. Diagnosis of CCA was premised on the VOC levels of three major compounds including acrylonitrile, 3-methylhexane and benzene.  A cut point of < −0.12 optimized test performance with 90.5% sensitivity and 72.7% specificity. (p=0.02). A comparison of level of VOCs required introduction of another group of patients with common bile duct stones (N=11).

Discussion

Conventionally, it is an uphill task to try and diagnose occult CCA in PSC patients who have more than one fibrotic structures. This paper sought to examine through a research study the bio-markers through which bile could not only facilitate diagnosis of CCA but also differentiate the different bio-markers that are inherent in PSC patients from IBD patients. An observation that was made is that VOCs of bile fluid could be used to address the research questions by differentiating CCA in PSC and IBD structures. The study showed that CCA is an important peripheral as the patients with it have a specific VOC pattern which are lowered as compared to PSC without CCA.

In our research, analyses of exhaled breath has been done in gastrointestinal and liver illnesses[1]. However, of importance is that bile is aspirated spontaneously and readily during ERCP. Also, it was noted that the VOC were very unlikely to be influenced and affected by environmental conditions or factors inherent in diet rather it was more likely to be influenced by local perturbations along the biliary tract. Related literature has shown that VOCs may aid in the diagnoses of cancer. The exploration of VOCs contained in faces and breath were found to effectively identify colorectal cancers with a sensitivity of around 85%[2]. A further examination also showed that a particular VOC pattern could be identified in pancreatic cancer.

It is an interesting phenomenon to note the lower rates of VOC in CCA among patients with PSC which is in sharp contrast among patients with PSC but lacking CCA. One of the components that can be used to expand on this phenomenon is benzene. Benzene is actually a pollutant whose production could be attributed to tobacco and automobile exhausts. This compound has been heavily association with alterations in the genetic makeup. The study found out that the concentration of benzene was lower in CCA. A deeper analyses of breath released from smokers showed that Nitrogen having VOCs such as acetonitrile and acrylonitrile could be found.

The study found out that there were very minimal levels of benzene as well as acetonitrile and acrylonitrile amongst patients with CCA. Consistency was observed even after adjusting the smoking pattern. The formation of Carbon disulphide, dimethyl sulphide and mercaptopurines are due to partial liver metabolism of sulphur carrying amino acids. However, of importance is that an enhanced level of sulphur containing VOCs had their role in fetor hepaticus particularly in patients with liver cirrhosis. PSC patients were found with a high level of carbon disulphide and dimethyl disulphide as compared to patients with CCA.

Acetonitrile found in acetic acid commonly referred to as ethanoic acid has been in the past been used widely in IBD models to reproduce or recreate IBD conditions. While acetaldehyde is an important marker for IBD, it was identified to occasion antimicrobial activity. Studies have also shown that it could be a potential marker for differentiation between CD and UC diagnosis. One of the research questions in the study was to differentiate the VOCs among IBD and PSC patients and the study has done that through examination of acetic acid where acetonitrile levels amongst patients with PSC and CCA while in IBD patients it was found out that acetaldehyde was inherent among IBD patients making it an important marker.

Acetonitrile was found to be an important marker for CCA patients, however, there was not data whatsoever linking acetonitrile with selective diagnosis of IBDs but as discussed above there are other components that are elated to acetic acid that are important markers for IBD. This calls for further investigations into whether a verification can be done to ascertain whether acetonitrile can be used suitably as a marker for IBD. Furthermore, acetonitrile was included within one of the innovative studies that was aimed at evaluating the VOCs profile amongst patients with regards to their body positions in order to pinpoint VOCs or group of VOCs that could be considered as bio- markers.

Endogenous alkanes including ethane and pentane formation is due to the peroxidation of polyunsaturated fatty acids which are inherent within the cell membrane. Inflammatory conditions lead to oxidative stress and it was found that the role of oxidative stress in the pathogenesis of liver diseases that could lead to infectious and autoimmune etiologies have been shown from studies. This study showed that the levels of the investigated pentane and ethane were much higher in PSC relative to CCA. This suggests that they could be playing a role in inflammatory conditions although they could not be sufficiently backed up by data analysis. Furthermore, our study, methyl hexane which denotes a branched chain alkane was demonstrated to be one of the optimal linear predictors of CCA as showed from logic regression analyses.

As compared to the potential roles of alkanes in prediction of CCA, their role in IBD is largely unsatisfying. Propene which is a hydrocarbon is known to be as a result of the metabolism of gram-positive as well as negative bacteria. The particular origin of propene and subsequently its role in IBD is unknown. However, it could only be attributed to its degradation as a result of β–oxidation. Methane was indirectly quantified in the examination but with no particular calibration.  Butanone which has never been identified as a specific ketone in VOC studies just like the other methyl ketones are produced from decarboxylation of fatty acids. However, their production from non-farmenting sources can be traced to different origins. The production of methane within the distal colon has been long been attributed to endogenous as well as exogenous compounds. Methane has been found by the study as a crucial biomarker of bacterial overgrowth which is synonymous with IBD condition.

We believe that the preliminary observations that have been made in this paper offer an attractive alternative for treating patients with cancer. VOCs are present in breath and in gaseous state and rightly so because of high vapour pressure. It would be intriguing to see whether the biomarkers from the analysis in this study would have a corresponding exhaled air print that can non- invasively diagnose CCA[3]. As for IBD, the investigated VOCs showed positive prospects as biomarkers for non-invasive detection of IBD therefore calling out for further investigations to ascertain whether the technical aspects regarding our research could be improved for a better conclusion.

The study has shown that bile VOCs may aid in the diagnosis of CCA in the setting of PSC. This was done on the premise of three compounds including acrylonitrile, methyl hexane and benzene[4]. The specificity and the sensitivity of the given results were at 72.7% and 90.5% respectively. Some uncertainties remain over some other VOCs because more components inherent in the biomarkers may be containing the same molecular weights. Following the hypothesis which identified these molecular weights to be fundamentally the primary molecules rather than being fragments after ionization, a comparison of our results with other literature encompassing alveolar air was done. Some correlations could be found s with MW 114 which corresponded with 2-3-3-trimethylpentane which was found by Filipiak et al. as well as MW 76 that could be identified with carbon disulphide as referred to by Navaneethan et al.

In our opinion, this study should be considered as a strong starting point. The coming up with predictive models premised on VOC profiles with highly precise instruments and more sophisticated statistical tools could help come up with new non-invasive, quicker and relatively cheaper diagnostic tools that symbolizes  crucial steps towards etiology of PSC as well as IBD illnesses. The bio-markers and VOCs in PSC patients could also aid in the diagnosis of CAA and further provide a distinct diagnosis of CAA and PSC. The results gotten from observation of bile headspace could be translated into breath analysis for true diagnosis of CCA.

References

Altomare, D. F., Di Lena, M., Porcelli, F., Trizio, L., Travaglio, E., Tutino, M., & De Gennaro, G. (2013). Exhaled volatile organic compounds identify patients with colorectal cancer. British journal of surgery, 100(1), 144-150.

Buszewski, B., Ulanowska, A., Ligor, T., Denderz, N., & Amann, A. (2009). Analysis of exhaled breath from smokers, passive smokers and non‐smokers by solid‐phase micro extraction gas chromatography/mass spectrometry. Biomedical Chromatography, 23(5), 551-556.

Dadamio, J., Van den Velde, S., Laleman, W., Van Hee, P., Coucke, W., Nevens, F., & Quirynen, M. (2012). Breath biomarkers of liver cirrhosis. Journal of Chromatography B, 905, 17-22.

de Meij, T. G., Larbi, I. B., van der Schee, M. P., Lentferink, Y. E., Paff, T., Terhaar sive Droste, J. S., … & de Boer, N. K. (2014). Electronic nose can discriminate colorectal carcinoma and advanced adenomas by faecal volatile biomarker analysis: proof of principle study. International journal of cancer, 134(5), 1132-1138.

[1] Altomare, D. F., Di Lena, M., Porcelli, F., Trizio, L., Travaglio, E., Tutino, M., & De Gennaro, G. (2013). Exhaled volatile organic compounds identify patients with colorectal cancer. British journal of surgery, 100(1), 144-150.

[2] de Meij, T. G., Larbi, I. B., van der Schee, M. P., Lentferink, Y. E., Paff, T., Terhaar sive Droste, J. S., … & de Boer, N. K. (2014). Electronic nose can discriminate colorectal carcinoma and advanced adenomas by faecal volatile biomarker analysis: proof of principle study. International journal of cancer, 134(5), 1132-1138.

[3] Dadamio, J., Van den Velde, S., Laleman, W., Van Hee, P., Coucke, W., Nevens, F., & Quirynen, M. (2012). Breath biomarkers of liver cirrhosis. Journal of Chromatography B, 905, 17-22.

[4] Buszewski, B., Ulanowska, A., Ligor, T., Denderz, N., & Amann, A. (2009). Analysis of exhaled breath from smokers, passive smokers and non‐smokers by solid‐phase microextraction gas chromatography/mass spectrometry. Biomedical Chromatography, 23(5), 551-556.