Reproducibility of exhaled biomarkers in COPD—the road less traveled

In recent years, interest in the measurement of exhaled biomarkers has increased, mainly driven by the unmet clinical need to monitor airway inflammation and the response to antiinflammatory therapy. Beside exhaled nitric oxide measurement that entered clinical practice, measurement of biomarkers in exhaled breath condensate (EBC) is a rapidly expanding area of this field. EBC is easy to collect as it only requires the noninvasive collection of exhaled breath in a cold trap. The obtained fluid is a complex diluted solution of diverse biomarkers with various chemical stabilities. Due to the complexity of EBC and the fact that it is a much diluted sample, there are still several unsolved issues about standard protocols for measurements and there is a lack of reproducibility data for different biomarkers, particularly in disease states. The lack of knowledge and evidence on certain areas limited the ability of the European Respiratory Society/American Thoracic Society Task Force to make firm guidelines for each aspect of this sampling technique and prompted the authors to highlight areas for further research (Horvath et al 2005).

The current issue of the International Journal of COPD contains an important methodological contribution to this rapidly growing field (Borrill et al 2007). The authors performed a carefully designed study to assess within assay, within and between day reproducibility of EBC leukotriene B₄ (LTB₄) and 8-isoprostane concentration measured by enzyme immunoassays (ELISA). They demonstrated that within assay variability was small, but there was a considerable within and between day variability for these biomarkers. Their two main conclusions derived from the study are: 1) the cause of relatively low reproducibility of EBC LTB₄ and 8-isoprostane is multi-factorial including both biological and methodological variability; 2) the high level of variability they observed casts doubt on the current EBC methodology used to assess LTB₄ and 8-isoprostane.

The very important contribution of this article to the field is the careful analysis of reproducibility for two biomarkers. It has been shown for some EBC biomarkers that oral contamination during sampling (Gaber et al 2006; Marteus et al 2005), differences in collecting surfaces (Rosias et al 2006), different assay techniques (Huszar et al 2005), and mode of standardization (Kullmann et al 2007) all can contribute to limited reproducibility of results. It is extremely difficult to address all potential methodological biases of ELISA when assessing EBC. Intra-assay variability is relatively easy to assess: one can compare readings of duplicate or triplicate aliquots of the same samples. Determination of inter-assay differences again is best done from the same samples adding aliquots of it to two different plates of the assay. If results of inter-assay reproducibility are good, then different batches of the assay can be used to determine intra-day or intra-week reproducibility. This potential difference needs to be taken into account when using EBC as a source of airway biomarkers (Huszar et al 2005). Determination of intra-week reproducibility then can be performed by 1) keeping the first sample for a week and measuring it together with the second (in this case however the potential effect of storing on mediator concentration needs to be taken into account); or 2) measuring the two samples by different batches of ELISA from the same manufacturer (in this case inter-assay variability complicates interpretation). Although it cannot be established from the study by Borrill and colleagues (2007) what kind of arrangement they used for the determination of intra-day and intra-week variability, their conclusion about the complexity of reproducibility holds very well.

Another issue their study touches upon is the limitations of the usefulness of biomarkers with limited reproducibility. Precise standardizations of traditional lung function tests and exhaled nitric oxide measurements were essential for these variables to provide accurate, well reproducible readings and gain clinical acceptance. EBC biomarkers have to travel through the same path before they can become clinically meaningful tools. The study by Borrill and colleagues (2007) adds important insight into the interpretation of observed differences of EBC LTB₄ and 8-isoprostane and it also emphasizes the need for reproducibility data, not only in healthy subjects, but also in patients. COPD is a disease with fluctuation of airway inflammation and oxidative stress that only partially reflected by clinical symptoms therefore it is reasonable to assume a lower than normal degree of reproducibility of biomarkers representing pathophysiological events of the disease even under stable conditions. Therefore, long-term follow-up studies assessing biomarker reproducibility can help to better understand the dynamics of EBC biomarker profile in this multiplex disease.

I hope you will find the article by Borrill and colleagues (2007) stimulating from both clinical and research point of views.