Breath biomarkers for lung cancer detection and assessment of smoking related effects — confounding variables, influence of normalization and statistical algorithms Sabine Kischkel a, ⁎, Wolfram Miekisch a , Annika Sawacki a , Eva M. Straker a , Phillip Trefz a , Anton Amann b,c , Jochen K. Schubert a a Department of Anaesthesiology and Intensive Care Medicine, University Rostock, Schillingallee 35, D-18057 Rostock, Germany b Department of Operative Medicine, Innsbruck Medical University, Anichstraße 35, A-6020 Innsbruck, Austria c Breath Research Institute, Austrian Academy of Science, Dammstraße 22, A-6850 Dornbirn, Austria abstract article info Article history: Received 16 April 2010 Received in revised form 4 June 2010 Accepted 4 June 2010 Available online 10 June 2010 Keywords: Breath gas analysis Lung cancer SPME-GC-MS Data processing Inspired concentrations Confounding variables Background: Up to now, none of the breath biomarkers or marker sets proposed for cancer recognition has reached clinical relevance. Possible reasons are the lack of standardized methods of sampling, analysis and data processing and effects of environmental contaminants. Methods: Concentration profiles of endogenous and exogenous breath markers were determined in exhaled breath of 31 lung cancer patients, 31 smokers and 31 healthy controls by means of SPME-GC-MS. Different correcting and normalization algorithms and a principal component analysis were applied to the data. Results: Differences of exhalation profiles in cancer and non-cancer patients did not persist if physiology and confounding variables were taken into account. Smoking history, inspired substance concentrations, age and gender were recognized as the most important confounding variables. Normalization onto PCO 2 or BSA or correction for inspired concentrations only partially solved the problem. In contrast, previous smoking behaviour could be recognized unequivocally. Conclusion: Exhaled substance concentrations may depend on a variety of parameters other than the disease under investigation. Normalization and correcting parameters have to be chosen with care as compensating effects may be different from one substance to the other. Only well-founded biomarker identification, normalization and data processing will provide clinically relevant information from breath analysis. © 2010 Elsevier B.V. All rights reserved. 1. Introduction More than 3000 different substances can be determined from human breath [1–3] by means of hyphenated analytical techniques [4,5]. Some of these substances have been described as being linked to lung disease, inflammatory and malignant processes in the body [6–9]. In addition, previous exposure to various chemical substances may be recognized in this way. As breath analysis is completely non-invasive it holds promise for screening purposes. Prognosis of malignant diseases such as bronchial carcinoma [10,11] could be significantly improved if early diagnosis was possible by means of non-invasive screening tests. Correlations between lung cancer and different exhaled breath biomarkers have been reported [12–18]. But up to now, none of the markers or marker sets proposed for cancer recognition reached clinical relevance in terms of reliable disease recognition and sufficient sensitivity and specificity. Crucial and still unsolved issues in breath analysis are ambient concentrations of potential biomarkers, prior intake and actual excretion of environmental contaminants and the lack of standardized and generally accepted methods of sampling, analysis and data processing. Clear distinction of endogenous disease related biomarkers from contaminants originating from the actual environment or from prior uptake is indispensable for clinically relevant breath testing. Hence reliable methods for breath sampling, separation and identification of volatile substances have to be set up, and, finally, physiologically sound and smart algorithms for data processing have to be applied. Within a clinical study in lung cancer patients, smokers and healthy non-smoking controls, we looked upon endogenous volatile substances, compounds occurring in cigarette smoke and contami- nants from the clinical and laboratory environment. Finally, different algorithms were applied to the data in order to account for inspired concentrations and physiological variables such as body surface area (BSA) or minute ventilation. Clinica Chimica Acta 411 (2010) 1637–1644 ⁎ Corresponding author. Department of Anaesthesiology and Intensive Care Medicine, University Rostock, Schillingallee 70, D-18057 Rostock, Germany. Tel.: + 49 381 494 5955; fax: +49 381 494 5942. E-mail addresses: sabine.kischkel@uni-rostock.de (S. Kischkel), wolfram.miekisch@uni-rostock.de (W. Miekisch), a.sawacki@web.de (A. Sawacki), e.str@gmx.net (E.M. Straker), phillip.trefz@uni-rostock.de (P. Trefz), anton.amann@i-med.ac.at (A. Amann), jochen.schubert@uni-rostock.de (J.K. Schubert). 0009-8981/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.cca.2010.06.005 Contents lists available at ScienceDirect Clinica Chimica Acta journal homepage: www.elsevier.com/locate/clinchim