Page 93 - ADULT-ONSET ASTHMA PREDICTORS OF CLINICAL COURSE AND SEVERITY

BIOMARKERS FOR DIAGNOSING ASTHMA: NOT QUITE A SMOKING GUN?
these results add important data to the growing evidence of FeNO and blood eosinophils as biomarkers for diagnosing allergic asthma.
The ndings of Giovannelli and colleagues have important clinical and scien c implica ons. Firstly, they extend the results from previous studies, indica ng that FeNO and blood eosinophils may act di erently in smokers and in non-smokers 4, 5. Physicians should take this into account when using FeNO and blood eosinophils in the clinical work-up of asthma pa ents.
Secondly, the results highlight the urgent need for more research in asthma pa ents with a smoking history. In clinical asthma trials, smokers and ex-smokers are usually excluded, which implies that about one third of the asthma popula on is not taken into account. Therefore, results of these trials may not necessarily be extrapolated to smoking pa ents. Accumula ng evidence suggests that smoking asthma pa ents represent a separate asthma phenotype, in par cular with respect to the clinical value of biomarkers of airway in amma on in predic ng response to treatment 10. The ndings by Giovannelli and colleagues further amplify this message.
Thirdly, the results con rm that FeNO and blood eosinophils have only moderate accuracy for diagnosing asthma, both in smokers and in non-smokers, as has been reported in previously published smaller evalua ons 11. Physicians should realize that the use of these biomarkers as single diagnos c tests in clinical prac ce can easily lead to false posi ve or false nega ve test results.
This does not necessarily mean that these biomarkers have no value at all. Single markers that are insu ciently accurate to serve as stand-alone diagnos c tests may have important value as rule-in or rule-out tests 12. A test that is very sensi ve but less speci c may be useful to rule-out a disease, as the number of false nega ves is low. Vice versa, a test that is very speci c but less sensi ve may be useful to rule-in a disease. Markers can also be combined to develop a predic on model with improved diagnos c accuracy 13. Unfortunately, there are currently no well-de ned and reproducible thresholds for FeNO and blood eosinophils for ruling-in or ruling-out allergic asthma, and e orts to build such predic on models have so far been limited 11. Future studies should focus on the iden ca on and valida on of such thresholds and models, because if there is a role for FeNO and blood eosinophils in the diagnosis of asthma, it will probably lie there. Such studies should take into account that thresholds may be di erent between smokers and non-smokers, as the current study once again illustrates 11.
Giovannelli and colleagues conclude: “These ndings raise ques ons about the clinical value of FeNO and blood eosinophils in smokers”. Although this is certainly true for diagnosing asthma, it should be noted that these biomarkers may have other clinical applica ons as well. For example, they may be used for iden fying speci c in ammatory asthma phenotypes 14, 15, for disease monitoring and assessing the risk of asthma exacerba ons 16-18, or for selec ng pa ents that are likely to respond to treatment 11, 19.
91