Article Text

Download PDFPDF

Give me a sign, any sign
  1. Peter J Mazzone
  1. Correspondence to Dr P J Mazzone, The Cleveland Clinic, 9500 Euclid Avenue, Cleveland, OH 44195, USA; mazzonp{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

For clinical progress to occur in lung cancer, advances must be made in many inter-related areas. Advances in chemoprevention will be most useful if testing is able to identify those at greatest risk of developing lung cancer. Advances in surgical and ablative therapies will be most useful if testing is able to identify lung cancer at the earliest possible stage. Advances in systemic, targeted and individualised therapies will be most useful if testing is able to predict the nature of a patient’s lung cancer and the response to specific treatment choices.

A new test can improve on currently used tests by being more accurate, less invasive, less expensive and/or novel in its intent. To have a clinical impact, the result of the test must affect a decision to the benefit of the patient. The most recently developed tests that have had this sort of impact in lung cancer are positron emission tomographic imaging,1 advances in diagnostic bronchoscopy (electromagnetic navigation,2 endobronchial ultrasound3) and perhaps epidermal growth factor receptor analysis.4

Progress is occurring on many fronts in lung cancer testing. The clinical impact of imaging advances such as dual energy imaging,5 temporal subtraction,6 computer-aided detection/diagnosis7 and volumetric analysis8 of lung nodules is being assessed. Bronchoscopic advances in the use of ultrathin bronchoscopes,9 navigation programmes,2 ultrasound-guided sampling procedures,3 narrow band imaging10 and optical coherence tomography11 are occurring. We have seen advances in molecular testing through the analysis of tumour and airway tissue genomes12 13 and proteomes,14 as well as blood proteomes,15 DNA methylation,16 circulating tumour cells,17 antibodies to tumour-associated antigens18 and microRNA profiles.19 Finally, there is hope that analysis of the volatile and non-volatile20 components of exhaled breath will provide useful information about our patients.

In this issue of Thorax, progress on the testing of exhaled breath volatile compounds as a diagnostic in lung cancer is presented. Volatile organic compounds (VOCs) are present in the breath in very low concentrations (parts per billion to parts per trillion volume). They can be inhaled into the lungs and absorbed through the skin from exogenous sources, or can be generated directly from the cellular biochemical processes of the body (eg, lipid peroxidation of fatty acid components of cell membranes). The origin of many endogenous VOCs is not known. Several lines of evidence suggest that the biochemical processes of lung cancer cells differ from those of normal cells. Thus, it is reasonable to expect that the pattern of exhaled volatile compounds in patients with lung cancer would be different from that of individuals without lung cancer.

The concept that a unique pattern of exhaled VOCs exists in those with lung cancer has been studied previously. Some researchers have evaluated gas chromatography/mass spectrometry (GC/MS) systems, while others have investigated the use of non-specific chemical sensing matrix devices for this purpose. There are benefits and downsides to the use of either of these technologies. GC/MS devices are able to identify the specific components of a gas mixture at low concentrations, but they are cumbersome to use and expensive. Chemical sensing matrix devices are easier to use as a point of care test and relatively inexpensive, but they do not identify the actual compounds and may lack enough sensitivity to various volatile compounds to be accurate. Despite these concerns, both techniques have shown promise with accuracies in the 70–85% range for the identification of lung cancer being reported. The studies have differed in the breath collection methods used, the populations tested and the statistical methods applied to identify the unique patterns as well as to validate the model developed.21

Technological advances should lead to progress in breath testing. Chemical sensing devices have been developed that can detect VOCs at lower concentrations than traditional GC/MS in near real-time. In this issue of Thorax, Westhoff et al22 describe the use of one such device for the analysis of exhaled breath (see page 744). This device, called an ion mobility spectrometer, is able to detect volatile compounds at a tenfold lower concentration than standard GC/MS devices in approximately one-fifth of the time. It does not, however, identify the specific compounds in the mixture. The authors enrolled subjects with lung cancer and healthy controls. They were able to demonstrate a complete separation of the breath signals between these groups. These results hold promise for this or similar devices to be developed into an accurate diagnostic tool, but should be viewed as pilot information only. The authors did not include control subjects with other medical problems, their control group was younger than the cancer group, most of the cancer subjects had relatively advanced disease and they did not use a separate validation cohort. All of these factors may influence the accuracy of the tested device in future studies.

The field of lung cancer test development is on the verge of some major advances. In the near future we may have testing available that will help us to assess the risk of developing lung cancer, diagnose lung cancer at an earlier stage, improve our ability to predict the course of lung cancer once diagnosed and intelligently individualise treatment decisions. These tests will foster advances in other areas of lung cancer management to the benefit of our patients.


View Abstract


  • Competing interests None.

Linked Articles