Diagnosis of lung cancer by canine olfactory detection in urine and breath samples

• Published in: Journal of clinical oncology Vol. 37; no. 15_suppl; p. e13067
• Main Authors: Feil, Charlotte, Stein, Thorsten, Forster, Andreas, Schmidtmann, Irene, Riemann-Seibert, Thomas, Berger, Martin, Owen, Robert, Schimanski, Carl Christoph
• Format: Journal Article
• Language: English
• Published: 20.05.2019
• ISSN: 0732-183X; 1527-7755
• DOI: 10.1200/JCO.2019.37.15_suppl.e13067

Abstract only e13067 Background: Lung cancer is the leading oncological cause of death in western countries. The WHO estimated 2.09 million newly diagnosed lung cancer patients in 2018 worldwide. Although early detection is crucial for patients outcome, no surveillance tools exist. Dogs have a highly sensitive olfactory system which is already used in several ways, such as drug and ketone detection. The aim of the study was to evaluate the capability of a classically conditioned domestic dog to accurately distinguish samples of lung cancer patients of all tumor stages in urine and breath from healthy controls. Methods: This monocentric clinical trial was an original study and approved by the local ethics committee. After conditioning a domestic dog with samples of 186 patients (36 cancer patients and 150 control patients), further 246 patients aged between 45 and 80 entered into the study: 41 patients with a histologically proven lung cancer comprising all different stages and 205 control samples of healthy individuals with no cancer history. Two urine and two breath samples were collected of each patient and immediately shock frozen at -80°C. Urine and breath samples were separately exposed to the dog in a randomized, double-blinded manner, resulting in a specific conditioned reaction indicating the cancer sample. Results: The dog correctly predicted cancer in 36 of 41 urine samples, corresponding to a sensitivity of 87.8% and a specificity of 97.6%. Concerning the breath samples, the dog correctly predicted cancer in 32 of 41 samples, corresponding to a sensitivity of 78% and a specificity of 95.6%. Combining both techniques, the dog correctly identified 40 of 41 cancer samples, leading to an overall sensitivity of 97.6%. The specificity is not evaluable. Conclusions: Urine and breath carry volatile organic compounds indicating cancer growth, as previously reported. Canine olfactory detection of lung cancer is a simple tool to detect lung cancer non-invasively. Further identification of target compounds for technical translation of this approach is under way, with the potential for the development of a bionic electronic nose.