Early diagnosis of lung cancer by the electronic nose : case control study on 82 subjects

Lung cancer is a major public health problem worldwide responsible of an estimated mortality that exceed the combined number of deaths from breast, prostate, colon and pancreatic. During the years, several screening programs have been developed with the aim of detecting disease at a stage when cure and control are possible. An accurate, inexpensive and non-invasive test on asymptomatic people would be a major advance to current diagnostic tools. Among the new available options for an early detection, analysis of exhaled air seems to be very promising thanks to its characteristics and stands out as a viable candidate for widespread screening. Between May 2013 and May 2014, in the Division of Thoracic Surgery of the European Institute of Oncology in Milan, we prospectively investigated the breath of 82 subjects volunteers with signed consent. This work has been funded by the Italian Association for Cancer Research (AIRC) and approved by Ethic Commission. Population consisted of 47 patients with lung cancer and 35 healthy subjects, each one older than 50 years. Exclusion criteria were previous history of cancer (within 5 years) and any treatment correlated (radio or chemotherapy). For healthy subjects was request an X-rays or CT of the chest to prove the health status. Measurements were performed immediately after the collection. We analyzed the breath of the recruited subjects with an electronic nose (TEN NOSE) to validate the ability of the instrument to distinguish between healthy subjects and lung cancer patients. The instrument is composed by eight quartz microbalance (QMB) gas sensors, coated with different metalloporphyrins. The absorption of breath by the instrument determines for each sensor a variation of the vibration frequency which is measured and stored. The data were analyzed by multivariate technical analysis. The statistical analysis found out a positive predictive value of 79% and a negative predictive value of 80%. Furthermore the sensitivity of the electronic nose was 87% and the specificity was 69%. These preliminary results are satisfactory and define the electronic nose as a promising tool for early diagnosis. The novelty of these researches concerns the realization of a robust model to transfer the instrument in the clinical practice. Further studies are necessary to extend the analysis to evaluate larger groups of patients with distinction histology subgroups of non small cell lung cancer and healthy subjects.