**3. Promising single non-invasive biomarkers of asthma**

The concept of asthma has undergone considerable changes throughout the years, from a disease mainly manifesting by variable symptoms and bronchoconstriction to airway inflam‐ mation and remodeling. More recently, heterogeneity has gained an outstanding position in asthma definition. So far, one of the most important steps in asthma history, bringing signifi‐ cant reduction in morbidity and mortality, was the recognition of airway inflammation in asthma and the introduction of efficacious and safe anti-inflammatory therapy for asthma control. Despite ongoing developments, current guidelines for both diagnosis and follow-up of patients with asthma are still grounded on clinical and lung function parameters. Thus, functional biomarkers were the first objective measures coming forward into clinical practice and, in general, the promise of delivering valuable molecular, cellular or histological biomark‐ ers to daily clinical practice has not yet been met. However, intense research in asthma has brought together scientists from academia, research institutes, the pharmaceutical industry and patient organisations, with significant progress taking place in the recent years. In this section, we discuss the currently available and more advanced non-invasive biomarkers in asthma.

Clinicians and researchers dedicated to asthma may benefit from a direct analysis of the airways, profiting the patients. In fact, non-invasive airway assessment is possible through lung function tests (LFTs) and airway sampling. Furthermore, other "more distant" to the airway biomarkers (such as blood or urinary biomarkers) can also be regarded as potentially useful, considering the systemic properties of asthma.

#### **3.1. Functional biomarkers**

LFTs are essential in routine clinical practice. They are non-invasive, well validated and reproducible. At present, LFTs provide the only generally accepted functional biomarkers to objectively aid in the diagnosis, risk assessment and monitoring of asthma. Thus, asthma definition currently implies the objective detection of variable airflow limitation, while the "best personal lung function" is a hallmark of asthma monitoring and future risk assessment.

LFTs provide relative features (phenotypes) that aid in differential diagnosis, namely in the distinction from chronic obstructive pulmonary disease (COPD), but are not diagnostic in its use. For instance, neither post-bronchodilator airway obstruction, lack of bronchodilation response or hyperinflation can be used to rule out asthma.

Presently, LFTs patterns alone are not considered to define disease subsets that respond to particular therapies. However, lung function has been shown to be predictive of clinical outcomes and provide complementary information to subphenotype asthma. For instance, variability measures of lung function can predict the loss of asthma control and response to long-term beta2-agonist treatment [15].

Airway hyperreactivity (AHR) is a basic pathophysiological hallmark of asthma, but remains a complex component of this disease. A growing number of variable airway smooth muscle (ASM) and non-muscle factors contributing to AHR has been recognised. Besides its high negative predictive value in the diagnosis of asthma, AHR has been advocated as a surrogate biomarker related to airway inflammation to guide asthma management. It has been shown that anti-inflammatory therapy directed at reducing AHR may imply higher corticosteroid doses, but leads to improved lung function and better control [16, 17]. AHR evaluation has also been suggested useful in back titration of inhaled corticosteroids. However, the reduction in AHR with higher doses appears targeted to the persistent structural component of AHR (defined as opposed to the variable inflammation component of AHR). Emerging data support that it is the structural changes of the airway that mainly contribute to AHR (i.e. reticular layer thickness and ASM hypertrophy) [16]. This effect also depends on the type of challenge used: assessing AHR to indirect bronchoconstrictor stimuli is superior in the detection of changes associated with airway inflammation, while direct stimuli, mediated through direct interaction with ASM, better reflect the structural changes. Assessment of AHR is a useful non-invasive tool providing complementary information, though its routine feasibility in general practice can be hard to settle.

Summing up, lung function measurements may not, per se, reflect the precise underlying pathological processes responsible for different phenotypes. However, in a multidimensional approach to evaluate asthma as a complex dynamic disease, functional biomarkers and their variability must definitely be part of future composite parameters in asthma.

#### **3.2. Exhaled air biomarkers**

characteristics (biomarkers) indicative of a specific asthma subphenotype/endotype. Subse‐ quent studies are being undertaken to test if one's "handprint" can predict the disease course and can indicate a response to (targeted) asthma treatments. This approach will provide a key

Generally, an ideal biomarker should possess the following key characteristics: clinical relevance, adequate sensitivity and specificity for (targeted) treatment effects, repeatability,

The concept of asthma has undergone considerable changes throughout the years, from a disease mainly manifesting by variable symptoms and bronchoconstriction to airway inflam‐ mation and remodeling. More recently, heterogeneity has gained an outstanding position in asthma definition. So far, one of the most important steps in asthma history, bringing signifi‐ cant reduction in morbidity and mortality, was the recognition of airway inflammation in asthma and the introduction of efficacious and safe anti-inflammatory therapy for asthma control. Despite ongoing developments, current guidelines for both diagnosis and follow-up of patients with asthma are still grounded on clinical and lung function parameters. Thus, functional biomarkers were the first objective measures coming forward into clinical practice and, in general, the promise of delivering valuable molecular, cellular or histological biomark‐ ers to daily clinical practice has not yet been met. However, intense research in asthma has brought together scientists from academia, research institutes, the pharmaceutical industry and patient organisations, with significant progress taking place in the recent years. In this section, we discuss the currently available and more advanced non-invasive biomarkers in

Clinicians and researchers dedicated to asthma may benefit from a direct analysis of the airways, profiting the patients. In fact, non-invasive airway assessment is possible through lung function tests (LFTs) and airway sampling. Furthermore, other "more distant" to the airway biomarkers (such as blood or urinary biomarkers) can also be regarded as potentially

LFTs are essential in routine clinical practice. They are non-invasive, well validated and reproducible. At present, LFTs provide the only generally accepted functional biomarkers to objectively aid in the diagnosis, risk assessment and monitoring of asthma. Thus, asthma definition currently implies the objective detection of variable airflow limitation, while the "best personal lung function" is a hallmark of asthma monitoring and future risk assessment.

LFTs provide relative features (phenotypes) that aid in differential diagnosis, namely in the distinction from chronic obstructive pulmonary disease (COPD), but are not diagnostic in its use. For instance, neither post-bronchodilator airway obstruction, lack of bronchodilation

**3. Promising single non-invasive biomarkers of asthma**

useful, considering the systemic properties of asthma.

response or hyperinflation can be used to rule out asthma.

step to personalised medicine [12–14].

18 Asthma - From Childhood Asthma to ACOS Phenotypes

simplicity and cost-effectiveness [10].

asthma.

**3.1. Functional biomarkers**

Exhaled breath can be sampled in a fully non-invasive manner across all age groups. However, exhaled breath analysis is not useful for analysing cellular or histological biomarkers and, in general, the search for useful molecular biomarkers has been hampered by methodologic difficulties mainly dealing with very low molecular concentrations, variability and lack of sampling and analysing methods standardisation [10].

*FeNO* is so far the most commonly used molecular biomarker in exhaled air. Nitric oxide (NO) is a gaseous chemical compound, which can be measured in exhaled breath either by chemi‐ luminescence and electrochemical analysers. The American Thoracic Society and the European Respiratory Society recommendations for standardised procedures for the FeNO measure‐ ment have been published [18]. Accordingly, FeNO is measured at a flow rate of 50 mL/s, thus reflecting NO production from the central airways. Currently available devices allow accurate and highly reproducible measurements, through simple, fast and non-invasive methodology. Hand-held devices are now widely available in clinical practice and used in both adults and children (since preschool age, usually above the age of 4 years) [10].

Evidence-based guidelines for adequate interpretation of FeNO measurement have been developed [19]. This biomarker can be affected by several perturbing factors, mainly age, height and recent active or passive smoking. Other variables that have been reported to affect FeNO levels include weight, gender, race, atopic status, diet or alcohol intake [20]. Large variation of normal FeNO values exists, with wide inter-individual differences and significant overlaps between healthy/non-asthmatic and asthmatic populations. Intriguingly, the afore‐ mentioned confounding factors explain few of the substantial variations within the general population [20]. For these reasons, guideline-recommended cut-points are supported for routine interpretation of FeNO levels [19].

Presently, there is evidence to support the use of FeNO thresholds essentially for assessing the likelihood of Th2-mediated airway inflammation and responsiveness to corticosteroids [19]. Low FeNO levels do not rule out asthma [19].

Persistently high FeNO levels may be attributed to poor adherence to corticosteroid therapy, poor inhaled drug delivery or persistent/high allergen exposure [19]. This has also been suggested to reflect a highly reactive asthma phenotype [21]. Although FeNO may be indica‐ tive of loss of disease control or exacerbation, some patients remain with high FeNO despite good clinical asthma control, and clinical trials of FeNO-guided management have yielded conflicting results [22–24]. Increased knowledge on asthma pathophysiology and the source and biochemistry of FeNO may help to further understand these findings. Traditionally, FeNO is known to originate in the airway epithelium as a result of inducible nitric oxide synthase (iNOS) upregulation, which occurs with inflammation [19]. Recent data give further support to this view by showing iNOS overexpression in the airway epithelium of patients with asthma [25]. However, it is interesting to note that despite the strong association between FeNO and Th2-mediated/eosinophilic inflammation and atopy, eosinophils are not the principal cells in the airways that express iNOS and this enzyme is upregulated by Th1 cytokines [26]. Anti-IL-5 and anti-IgE therapy for asthma reduced sputum eosinophilia without affecting FeNO, contrary to IL-13 inhibition that significantly decreased FeNO [27]. Studies have shown that FeNO levels are not elevated in many patients with severe asthma, compared to mild and moderate asthma, despite evidence of airway inflammation [13, 28]. Other sources of FeNO need also to be considered. For instance, as NO is a highly reactive molecule, it can be trapped and directly regenerated by abundant free thiol-containing biomolecules [26]. One of these thiols is S-nitrosoglutathione, which has been shown to be depleted in severe asthma, possibly contributing to comparative lower FeNO levels in these patients. Another important reservoir of nitrogen species is nitrite/nitrous acid. These agents are physiologically recycled in blood and tissues to form NO and other bioactive nitrogen oxides. When airway pH increases, more nitrite is formed and FeNO levels fall. On the other hand, FeNO may be high with acidification [26]. Still, many questions regarding the source of FeNO and its specific role need to be explored. Another area of research that may bring additional knowledge and clinical useful‐ ness is dedicated to partitioning of FeNO. In particular, alveolar FeNO can be obtained by measuring FeNO at multiple flow rates and has been shown to be an independent parameter that is putatively associated with increased distal lung inflammation and more severe disease [29].

In summary, the clinical importance of FeNO as a marker of Th2-mediated airway inflamma‐ tion that is likely to respond to corticosteroid treatment may be "indirect," but is well estab‐ lished. Further analysis is needed to address the possible need to define FeNO levels cut-points in different situations, according to the presence or absence of pertinent confounders. The application of FeNO measurement to identify particular asthma phenotypes or as part of a more comprehensive panel of biomarkers including also other "Th2 type" biomarkers may allow taking better profit of this readily available biomarker [30]. Partitioning of FeNO is a promising area of research, whose clinical usefulness is yet to be established.

Evidence-based guidelines for adequate interpretation of FeNO measurement have been developed [19]. This biomarker can be affected by several perturbing factors, mainly age, height and recent active or passive smoking. Other variables that have been reported to affect FeNO levels include weight, gender, race, atopic status, diet or alcohol intake [20]. Large variation of normal FeNO values exists, with wide inter-individual differences and significant overlaps between healthy/non-asthmatic and asthmatic populations. Intriguingly, the afore‐ mentioned confounding factors explain few of the substantial variations within the general population [20]. For these reasons, guideline-recommended cut-points are supported for

Presently, there is evidence to support the use of FeNO thresholds essentially for assessing the likelihood of Th2-mediated airway inflammation and responsiveness to corticosteroids [19].

Persistently high FeNO levels may be attributed to poor adherence to corticosteroid therapy, poor inhaled drug delivery or persistent/high allergen exposure [19]. This has also been suggested to reflect a highly reactive asthma phenotype [21]. Although FeNO may be indica‐ tive of loss of disease control or exacerbation, some patients remain with high FeNO despite good clinical asthma control, and clinical trials of FeNO-guided management have yielded conflicting results [22–24]. Increased knowledge on asthma pathophysiology and the source and biochemistry of FeNO may help to further understand these findings. Traditionally, FeNO is known to originate in the airway epithelium as a result of inducible nitric oxide synthase (iNOS) upregulation, which occurs with inflammation [19]. Recent data give further support to this view by showing iNOS overexpression in the airway epithelium of patients with asthma [25]. However, it is interesting to note that despite the strong association between FeNO and Th2-mediated/eosinophilic inflammation and atopy, eosinophils are not the principal cells in the airways that express iNOS and this enzyme is upregulated by Th1 cytokines [26]. Anti-IL-5 and anti-IgE therapy for asthma reduced sputum eosinophilia without affecting FeNO, contrary to IL-13 inhibition that significantly decreased FeNO [27]. Studies have shown that FeNO levels are not elevated in many patients with severe asthma, compared to mild and moderate asthma, despite evidence of airway inflammation [13, 28]. Other sources of FeNO need also to be considered. For instance, as NO is a highly reactive molecule, it can be trapped and directly regenerated by abundant free thiol-containing biomolecules [26]. One of these thiols is S-nitrosoglutathione, which has been shown to be depleted in severe asthma, possibly contributing to comparative lower FeNO levels in these patients. Another important reservoir of nitrogen species is nitrite/nitrous acid. These agents are physiologically recycled in blood and tissues to form NO and other bioactive nitrogen oxides. When airway pH increases, more nitrite is formed and FeNO levels fall. On the other hand, FeNO may be high with acidification [26]. Still, many questions regarding the source of FeNO and its specific role need to be explored. Another area of research that may bring additional knowledge and clinical useful‐ ness is dedicated to partitioning of FeNO. In particular, alveolar FeNO can be obtained by measuring FeNO at multiple flow rates and has been shown to be an independent parameter that is putatively associated with increased distal lung inflammation and more severe disease

routine interpretation of FeNO levels [19].

20 Asthma - From Childhood Asthma to ACOS Phenotypes

Low FeNO levels do not rule out asthma [19].

[29].

Other biomarkers have been studied in exhaled breath vapor namely *volatile organic compounds* (VOCs). In general, reactive oxygen species result from inflammation and promote polyunsaturated fatty acids degradation, originating volatile hydrocarbons. These VOCs are subsequently excreted in exhaled breath. Thus, exhaled VOCs may originate from systemic metabolism or from local airway inflammation. It is important to consider also that VOCs in exhaled breath may also be originated from pathogenic bacteria or from exogenous sources such as ambient air pollution [31]. Some studies have suggested that single VOCs such as pentane or ethane could be significantly higher in patients with asthma. However, VOCs profiles analyses bring significant additional value [31].

Another potential single biomarker in exhaled air is *exhaled breath temperature* (EBT), which reflects heat, a cardinal sign of inflammation. EBT has been shown to correlate with bronchial blood flow [32], which is advocated as the main mechanism to explain EBT changes in disease status.

Several studies have shown that EBT is higher in patients with asthma [32–34]. Conflicting data have been reported regarding a possible association between EBT and asthma control, with several studies supporting [34, 35], and others rejecting this relation [36, 37]. Correlation between EBT and other biomarkers, such as sputum eosinophils and FeNO, has resulted in inconsistent reports [32, 37]. Furthermore, EBT has been shown to increase after eucapnic voluntary hyperventilation, methacholine challenge test or exercise, but no difference was found between asthmatics and healthy individuals [38], suggesting this increase in EBT to be physiologic.

However, it is important to stress that different methods have been used to measure EBT. Some studies used a flow and pressure-controlled maximal slow continuous exhalation to residual volume to measure EBT, while others measured EBT in tidal volume until a temperature plateau was reached. Different variables have been analysed: plateau EBT, rate of temperature increase, time to achieve plateau EBT. These different methods preclude results comparison and, to our knowledge, no study has analysed both methods simultaneously. The recent development of improved, easier-to-use, portable devices has improved feasibility, including in children and in the elderly [34, 36, 39].

Moreover, further studies are needed when it concerns interpretation of the results. Variables such as room temperature and relative-ambient humidity may influence the results [39]. Some studies point a correlation between gender [37, 39], age [36, 39] and lung volume [36], which needs to be addressed. No significant correlation has been documented between EBT and auricular temperature, suggesting EBT to be a distinct variable and not just another measurement of body temperature [33, 34].

Conclusively, EBT assessment may be an appealing method enabling completely non-invasive and patient-friendly evaluation and deserves further standardisation and validation as a potentially useful biomarker in asthma.

#### **3.3. Exhaled breath condensate biomarkers**

Exhaled breath has been a source for intense research in the latest years and many other biomarkers have been studied. *Exhaled breath condensate* (EBC) has the advantage of being a more stable matrix than exhaled breath vapor, including volatile and also non-volatile compounds. It is obtained by cooling exhaled air and is thought to reflect the composition of the airway lining fluid. Many molecules have been analysed in EBC, including metabolites and also proteins. Although methodological recommendations for exhaled breath sampling and analysis have been published [40], the procedures for EBC collection and biomarker detection are not fully standardised and there is significant heterogeneity between different working groups yielding (highly) variable data.

Many biomarkers analysed in EBC reflect oxidative stress. Among these, the most extensively studied include H2O2 and isoprostanes.

*H2O2* is a reactive oxygen species that contributes to oxidative stress within the airways. A meta-analysis has reported that EBC H2O2 concentrations were significantly higher in adults with asthma, and associated with disease severity and control [41]. This has also been reported in children. Of importance, smoking increases H2O2 levels. EBC H2O2 levels were inversely correlated with lung function parameters and improved with inhaled corticosteroids [41]. Thus, EBC H2O2 has been suggested a promising biomarker for asthma control monitoring.

Oxidative stress can also be assessed through the determination of lipid peroxidation-derived products. *8-isoprostane* derives from arachidonic acid peroxidation. Increased levels of 8 isoprostane have been found in EBC in patients with asthma, correlating with disease severity [42]. EBC 8-isoprostane levels have been shown to be particularly useful to indicate asthma control and severity in childhood when combined with different markers [30]. Increased 8 isoprostane levels in EBC of children with exercise-induced bronchoconstriction (EIB) have been described, suggesting a role for oxidative stress in EIB [43].

Markers of inflammation have also been addressed. *Leukotrienes (LT)* are important mediators of airway inflammation in asthma, and the most extensively studied molecular biomarkers of inflammation in EBC. Increased levels of LTs have been detected in EBC of patients with asthma, correlated with disease severity and were effectively reduced by oral corticosteroids or LT receptor antagonist [44, 45]. However, the reported effect of inhaled corticosteroids on LTB4 EBC levels is controversial [46]. LTs have been suggested as markers of asthma severity [42]. Likewise, LTs have been associated with EIB severity [47].

Various *cytokines* and other molecules have been analysed in EBC. In particular, IL-4 has been found to be higher in EBC of patients with asthma, especially in asthma associated with atopy [30, 42]. Cytokine ratios and biomarker panels in EBC including cytokines have been suggested to be useful to assess asthma control (including IL-4 and interferon-gamma) and to predict asthma exacerbations (e.g. IL-5) [30, 48].

Last but definitely not least, the measurement of *pH* is one of the simplest and most technically validated biomarkers in EBC. EBC pH reflects airway acidification [49]. Several research groups have found higher pH levels in healthy subjects, compared to patients with asthma [10]. Significant decline in EBC pH occurred during asthma exacerbations. EBC pH shows good reproducibility, having low running costs and normal data sets have been published in selfreported healthy subjects [50].

Although some biomarkers may be useful to measure in EBC, samples are highly diluted, biomarker concentrations are difficult to measure, require specialised equipment, laboratory techniques and normalisation standards are lacking. Unfortunately, EBC has been hampered by serious drawbacks in the methodology, detection techniques and result interpretation, all consistent with large intra and intersubject variability, precluding validation for most single biomarkers.

#### **3.4. Biomarkers in non-respiratory specimens**

needs to be addressed. No significant correlation has been documented between EBT and auricular temperature, suggesting EBT to be a distinct variable and not just another

Conclusively, EBT assessment may be an appealing method enabling completely non-invasive and patient-friendly evaluation and deserves further standardisation and validation as a

Exhaled breath has been a source for intense research in the latest years and many other biomarkers have been studied. *Exhaled breath condensate* (EBC) has the advantage of being a more stable matrix than exhaled breath vapor, including volatile and also non-volatile compounds. It is obtained by cooling exhaled air and is thought to reflect the composition of the airway lining fluid. Many molecules have been analysed in EBC, including metabolites and also proteins. Although methodological recommendations for exhaled breath sampling and analysis have been published [40], the procedures for EBC collection and biomarker detection are not fully standardised and there is significant heterogeneity between different

Many biomarkers analysed in EBC reflect oxidative stress. Among these, the most extensively

*H2O2* is a reactive oxygen species that contributes to oxidative stress within the airways. A meta-analysis has reported that EBC H2O2 concentrations were significantly higher in adults with asthma, and associated with disease severity and control [41]. This has also been reported in children. Of importance, smoking increases H2O2 levels. EBC H2O2 levels were inversely correlated with lung function parameters and improved with inhaled corticosteroids [41]. Thus, EBC H2O2 has been suggested a promising biomarker for asthma control monitoring.

Oxidative stress can also be assessed through the determination of lipid peroxidation-derived products. *8-isoprostane* derives from arachidonic acid peroxidation. Increased levels of 8 isoprostane have been found in EBC in patients with asthma, correlating with disease severity [42]. EBC 8-isoprostane levels have been shown to be particularly useful to indicate asthma control and severity in childhood when combined with different markers [30]. Increased 8 isoprostane levels in EBC of children with exercise-induced bronchoconstriction (EIB) have

Markers of inflammation have also been addressed. *Leukotrienes (LT)* are important mediators of airway inflammation in asthma, and the most extensively studied molecular biomarkers of inflammation in EBC. Increased levels of LTs have been detected in EBC of patients with asthma, correlated with disease severity and were effectively reduced by oral corticosteroids or LT receptor antagonist [44, 45]. However, the reported effect of inhaled corticosteroids on LTB4 EBC levels is controversial [46]. LTs have been suggested as markers of asthma severity

been described, suggesting a role for oxidative stress in EIB [43].

[42]. Likewise, LTs have been associated with EIB severity [47].

measurement of body temperature [33, 34].

22 Asthma - From Childhood Asthma to ACOS Phenotypes

potentially useful biomarker in asthma.

**3.3. Exhaled breath condensate biomarkers**

working groups yielding (highly) variable data.

studied include H2O2 and isoprostanes.

Other non-invasive matrices have also been analysed in search for biomarkers in asthma. *Saliva* is a readily available specimen and allows metabolites, proteins and also deoxyribonucleic acid (DNA)/ribonucleic acid (RNA) extracting (although buccal swabs perform better), including also oral microbiota assessment. *Cotinine* in saliva has been one of the most extensively studied biomarkers, with interest in asthma as a measure of tobacco exposure. Salivary *cortisol* has also been used for the evaluation of adrenal function. Morning salivary cortisol was significantly lower in patients with asthma than in healthy individuals, and poor asthma control has recently been associated with lower salivary cortisol levels [51]. Preliminary data have suggested that *inflammatory salivary markers* may also be associated with asthma control, including eosinophil-related (such as eosinophil cationic protein) and myeloid/innate media‐ tors [52]. Additionally, a significant decrease in salivary antioxidant enzyme-peroxidase activity was observed in children during asthma exacerbations [53]. A salivary pH decline has also been associated with asthma and AHR [54]. Another area of research includes the analysis of oral microbiota, which may change in asthma, either through disease status or its pharma‐ cotherapy. The interest in saliva studies in relation to asthma is still preliminary and the role of many possible confounders needs to be considered.

Although *urine* does not directly reflect the airways, samples are easily obtained across the full age spectrum. Several urine molecular biomarkers have been described to be associated with asthma. Here, we focus on four molecules which have been studied in more detail.

Of the potent lipid inflammatory mediators comprising the cysteinyl *LTs*, only LTE4 is stable, making this molecule the dominant LT detectable in biological fluids. Urinary levels of this end product of LT metabolism have been shown to be elevated in asthma, both in children and adults, and in patients with aspirin-exacerbated respiratory disease [55, 56]. It has been associated with the degree of airflow limitation and acute exacerbations [55, 57]. Although inhaled corticosteroids are the most effective treatment for asthma, they do not alter LTE4 excretion. Urinary LTE4 levels have been suggested as potential predictors of better response to anti-LT therapy compared to other therapeutic approaches, though further studies are needed, including other biomarkers, to predict individual responses.

As LTs, *prostaglandins* (PG) are the end products of arachidonic metabolism. PGD2 results from cyclooxygenase pathway and is excreted in urine after being metabolised to 9α,11β-PGF2. Increased urinary excretion occurs in patients with asthma and after challenge tests, and a negative association has been found with lung function [58].

*Bromotyrosine* is another molecule with possible interest in asthma. It is generated from protein oxidation by eosinophils. The oxidised amino acid is stable and excreted in urine. Urinary bromotyrosine levels are higher in patients with asthma and have been associated with asthma control and lung function, predicting exacerbations [59]. Its levels have been shown to reduce during inhaled corticosteroid therapy. High urinary bromotyrosine levels could predict a favorable clinical response to inhaled corticosteroid therapy, especially in combination with high FeNO values [59]. These results warrant further developments.

Though urinary biomarkers may become useful tools, many require specialised equipment and their measurement is not fully validated or standardised. There is a current need for normalisation standards and assessment of intra and inter-individual variation to select the potentially useful biomarkers. It is also important to address urine dilution when reporting quantitative absolute results.

#### **3.5. Airway imaging biomarkers**

Airway imaging biomarkers are also emerging, offering the potential of adding complemen‐ tary information, namely on small airways function and remodeling. High-resolution com‐ puted tomography (HRCT) images are used to measure airway narrowing, wall thickening, air trapping and ventilation inhomogeneity [27]. The first two measures have been correlated with lung function and asthma severity. Increased parenchymal lucency has also been associated with severe asthma exacerbations, lung function and neutrophilic inflammation. HRCT is easily performed though it requires that lungs are scanned at a standard volume for validity and reproducibility. The risk of exposure to significant ionising radiation needs to be considered and normal ranges have not been established.
