“Using Imaging as a Biomarker for Asthma”

Abstract

There have been significant advancements in the various imaging techniques that are being used in the evaluation of patients with asthma, both from a clinical and research perspective. Imaging characteristics can be used to identify specific asthmatic phenotypes and provide a more detailed understanding of endotypes contributing to the pathophysiology of the disease. Computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography (PET) can be used to assess pulmonary structure and function. It has been shown that specific airway and lung density measurements by CT correlate with clinical parameters including severity of disease and pathology but also provide unique phenotypes. Hyperpolarized ¹²⁹Xe and ³He are gases used as contrast media for MRI that provide measurement of distal lung ventilation reflecting small airway disease. PET scanning can be useful to identify and target lung inflammation in asthma. Furthermore, imaging techniques can serve as a potential biomarker and be used to assess response to therapies, including newer biological treatments and bronchial thermoplasty.

Introduction

Although chest radiography and computed tomography (CT) remain the primary imaging methods used in the clinical and research evaluation of asthmatic patients, there have been parallel advancements in a growing range of imaging techniques that are now available in both research and clinical arena (Table 1) [1]. Modalities such as magnetic resonance imaging (MRI), endobronchial ultrasound (EBUS), optical coherence tomography (OCT) and positron emission tomography (PET) are among the techniques that can be used in pulmonary imaging of asthma patients to assess both structure and function, so that these parameters can be related back to more traditional clinical parameters for a broadened, personalized understanding of the individual patient [2]. Imaging characteristics are now providing an understanding of endotypes and further define asthmatic phenotypes [3-5] and can potentially serve as predictive and response biomarkers [6]. Lung imaging can be used to assess response to standard treatment such as inhaled corticosteroids, newer pharmacologic therapies including biologic agents, and non-pharmacologic therapy such as bronchial thermoplasty [7,8]. This review will discuss the imaging modalities that are being used or have the potential for use to evaluate patients with asthma and assist the clinician in understanding the clinical benefits of their use. We will discuss the implications of imaging techniques as a biomarker and the utility of imaging in assessing and predicting response to treatment.

Table 1.

Asthma Imaging Summary

Modality	Structural Assessment	Functional Assessment	Clinical Utility	Disadvantages
CT	Detailed assessment -Airway tree -Vascular tree -Lung parenchyma	- Regional ventilation - Parenchymal perfusion	-Non-invasive measure of airway remodeling - Biomarker to assess response to therapy	- Radiation exposure prohibits serial examinations
MR	- Lung microstructure using ADC - Combined with CT for detailed structural evaluation	- High spatial resolution evaluation of regional ventilation - Gas exchange	-Biomarker to assess response to therapy - Assessment of V/Q ratio	- Less structural detail than CT - Limited to specialized MR centers
EBUS	- Access airways as small as 4mm with visualization of multiple layers of airway wall	- None	- Monitor serial airways changes	- Requires bronchoscopy - No functional assessment - Standards not yet established
OCT	- 2-D images of airway wall with spatial resolution of 1-15μm and penetration of 2-4mm	- None	- Microscopic view of wall thickness and subepithelial matrix - Monitor serial airway changes	- Requires bronchoscopy - Subject to respiratory cycle movement - Standards not yet established
PET	- Combine with CT for detailed structural evaluation	- Pulmonary inflammation - Ventilation/perfusion	- Response to anti-inflammatory therapies - Evaluate inhaled drug delivery	- Limited spatial resolution -Radiation exposure

Abbreviations: CT = computed tomography, MR = magnetic resonance, ADC = apparent diffusion coefficient, EBUS = endobronchial ultrasound, OCT = optical coherence tomography, PET = positron emission tomography

Use of imaging in the clinical evaluation of asthma

Imaging of the lungs in patients with asthma has evolved dramatically over the last decade; however, the clinical diagnosis of asthma is still based on a compatible history, exam findings and evidence of variable airflow obstruction. Chest imaging is most helpful in evaluating complications from asthma and ruling out alternative diagnoses. The chest radiograph findings are non-specific and often may be normal. The most common abnormal finding is bronchial wall thickening, present in 48-71% of radiographs [9,10], followed by hyperinflation found in 24% of cases in one series [9]. Marked hyperinflation is most often seen in the setting of emphysema. Previous studies evaluating the need for chest radiographs in acute asthma exacerbations revealed that patients presenting to the emergency department with uncomplicated asthma have abnormal chest x-rays only 1-2.2% of the time [11,12]. However, this number increases to nearly 34% in patients who are unresponsive to initial bronchodilator therapy and require admission to the hospital [13]. Abnormalities that may change management include pneumothorax, pulmonary vascular congestion, focal parenchymal opacities, and enlarged cardiac silhouette.

As with chest radiography, CT of the chest is not indicated in the routine management of asthma. However, it is helpful if the clinician suspects an alternative diagnosis or complication based on an unusual history (e.g. large amounts of expectorated sputum suggestive of bronchiectasis) or physical exam finding (e.g. inspiratory crackles suggestive of interstitial lung disease). Common CT findings include bronchial wall thickening, air trapping and bronchiectasis [14]. Automated techniques are now being utilized to evaluate the extent of airway wall thickening in a highly objective and reproducible manner. Multiple studies have demonstrated that disease severity correlates with the degree of bronchial thickening and air trapping [15-17]. Importantly, CT imaging may be necessary to rule out diseases that masquerade as asthma such as intra- or extra-thoracic airway obstruction, obliterative bronchiolitis, chronic obstructive pulmonary disease, congestive heart failure, hypersensitivity pneumonitis, hyper-eosinophilic syndromes, allergic bronchopulmonary aspergillosis and eosinophilic granulamatosis with polyangiitis (Churg-Strauss syndrome).

In the evaluation of asthma, MRI is an appealing complimentary or, in some cases, alternative modality to CT imaging due to the lack of ionizing radiation. However, at this time MRI is limited to asthma research due to the paucity of anatomic detail using conventional proton MRI. Hyperpolarized helium and xenon gas have emerged as a method for evaluating the functional changes of the distal small airways [18]. Using this technique ventilation defects can be quantified by evaluating the percentage of voxels whose signal intensity is below a threshold of 60% of the total lung mean signal intensity. Studies have demonstrated that patients with severe asthma have larger ventilation defects than those without severe asthma [19].

The comparative contributions of CT and MRI to an improved understanding of the lungs has been recently reviewed [20]. Typically imaging is performed at full inspiration (total lung capacity (TLC)) (Figure 1A) and at end expiration either at residual volume (RV) or functional residual capacity (FRC) (Figure 1B). Protocols for specific scanners [21] and automated image analysis software is used to identify the lungs, lobes (Figure 1A), airway tree (Figure 2), and vascular tree. Empirically derived thresholds have been established whereby the percent of the imaged voxels within the lung field at TLC falling below −950 Hounsfield Units (HU) is considered emphysema-like or hyperinflated. Voxels falling below −856HU on the expiratory image is considered air trapped (Figure 3). Image matching methods have been used (Parametric response mapping (PRM) (76) and disease probability mapping (DPM) [77-78]) (Figure 4A and B) whereby inspiratory and expiratory scans are warped together such that voxels can be assigned to categories of air trapped, normal and hyperinflated lung can be assigned. Airway metrics include luminal area, minimum and maximum diameters, hydraulic diameter, tapering, wall area (WA) and thickness (WT), wall area percent (WA%), and branch angles. Previous applications of these various metrics to asthma have been reviewed by Castro et al [2].

Figure 1. CT chest lung density.

Panel A demonstrates a 3D volume rendition of the lung, lobes and bronchial tree detected from a CT image of the fully inflated (total lung capacity) lung of a normal subject. Panel B. CT chest shows a similar volume rendition using the expiratory image (in this case, functional residual capacity) of a subject with severe asthma. Note areas of air trapping and pruning of the airways. Image processing derived from Apollo software (VIDA Diagnostics, Coralville, IA).

Figure 2. CT chest 3D bronchial tree.

Figure demonstrates labeling of the bronchial tree out to the segmental bronchi of a subject with severe asthma enabling each segmental bronchial wall thickness to be measured quantitatively. Image processing derived from Apollo software (VIDA Diagnostics, Coralville, IA).

Figure 3. CT chest air trapping distribution.

Figure demonstrates the concentration of regions determined to represent air trapping (voxels <−856) on the expiratory CT image of the same subject with severe asthma in Figure 1B. Trapped air, defined as voxels within the lung field falling below −856 HU, are demonstrated by sphericals proportional to area of air trapping (volume rendered view). Each lobe is color-coded.

Figure 4. MDCT chest image matching.

Parametric response mapping (PRM) (76) and disease probability mapping (DPM) [77-78]) methods are demonstrated whereby inspiratory and expiratory scans are warped together such that voxels can be assigned to categories of air trapped, normal and emphysema/hyper-inflated can be assigned. Panel A maps voxels from TLC (y axis) and FRC (x axis) in terms of their probability of a being hyper-inflated vs. ventilated in a plot from a severe asthmatic; a normal subject is shown in the insert (upper left). The green represents the normal end of the scale, yellow represents the probability of being air trapped (poorly ventilated) and red represents hyper-inflation. Because the image is a probability map, the colors are shown blended. Panel B serves to quantitate clusters of air trapped vs. normal lung tissue as a function of lung location.

CT examinations have increased from approximately 3 million exams in 1980 to 80 million in 2011 [22,23]. The expansion of CT use in medicine and in the novel phenotyping of the the lung in both asthma as well as chronic obstructive lung diseases (COPD) has led to a growing public concern about the potential risk of excess cancers. As radiation risks are greatest in women and younger patients, the dose delivered should remain a special concern in these populations [24, 25]. In this context, physicians and researchers must consider the risk/benefit ratio when seeking to employ this imaging modality. There have been considerable improvments in the sensitivity and spatial resolution of detector technologies, improvements in iterative reconstruction methodologies for image noise reduction associated with low dose imaging [26], use of more powerful x-ray tubes with smaller focal spots allowing for beam shaping with tin filters [27], and use of variable doses [28] as the scanner spirals about the thorax based upon local path lengths and densities. Thus, with the appropriate technologies CT imaging can be carried out at 1-3% of previous doses [29,30]. Therefore, CT chest examinations should be programmed with techniques that conform to the as-low-as-reasonably-achievable (ALARA) principal yet provide adequate image quality.

Assessing lung structure and function with via CT, PET, and MR imaging

With the drive to more broadly utilize CT for its newfound roles in clinical assessment of the lung, screening, phenotyping, as well as for drug and device discovery and outcomes assessment, there have been rapid advances, bringing highly evolved CT technologies into the clinical environment. CT imaging can provide comprehensive evaluation of the lung by allowing for detailed descriptions of not only the airway tree and lung parenchyma, but also regional ventilation [2]. Multidetector row CT allows for faster acquisition of multiple cross-sectional slices of images with a high spatial resolution.

Optical coherence tomography [OCT] is a tool that produces a two-dimensional image of the airway wall using near-infrared light through a fiberoptic catheter [31]. Distinct imaging patterns are produced as a result of the varying optical refractive properties of the different tissue layers (Figure 5). A previous study of COPD patients showed a strong correlation between OCT and CT measurements of AWA and LA [32]. Patients with asthma when evaluated by OCT had greater distension of airways at a given pressure and had decreased LA compared to controls [33]. Therefore, OCT can be used to monitor serial airway changes after a therapeutic intervention, such as bronchial thermoplasty, while avoiding cumulative radiation exposure (Figure 5).

Figure 5. Optical Coherence Tomography.

Optical coherence tomography (OCT) images (A) and mean ± standard deviation airway measurements (B) prior to bronchial thermoplasty (BT), 6 months post-BT and 2 years post-BT, with C) the corresponding bronchial biopsy at 6 months post-BT. Epi: epithelium; BM: basement membrane; SM: smooth muscle: WA: airway wall. Scale bars=1 mm. Modified with permission from [82].

Another method for assessing airway structure is through the use of endobronchial ultrasound [EBUS]. It is performed with an ultrasound probe through the working channel of a fiberoptic bronchoscope. EBUS can access airways as small as 4mm of internal diameter and it can visualize multiple layers of the airway wall (Figure 6). Studies using EBUS demonstrated an increase in AWT in patients with asthma compared to healthy controls comparable to that measured by CT [34, 35]. Like OCT, EBUS offers the ability to monitor serial changes without exposure to ionizing radiation.

Figure 6. Endobronchial Ultrasound.

Endobronchial ultrasound (EBUS) of bronchial wall from equine asthma model (A) and corresponding histological (B) images. Only a portion of the second layer (L2) area and corresponding smooth muscle area have been encircled in yellow, to allow the reader appreciate the rest of the image. L1-5: ultrasound layers 1-5; D1 and D2: perpendicular diameters (blue dotted lines); LA: lumen area (filled light green area); Pi: airway perimeter (continuous green line). Modified from [83]. © 2015 Bullone et al.

While OCT and EBUS can be used to identify structural changes, CT in combination with MRI and PET imaging can provide an objective quantitative assessment of the interactions between structure and function. Several imaging modalities, including CT, MRI, PET, or any combination of the three, have been employed to assess regional lung function. Hyperpolarized ³He and ¹²⁹Xe gases are utilized as MRI contrast media for measuring pulmonary functional biomarkers which include lung ventilation, quantification of airway microstructures, and gas exchange[7,18,36-41] (Figure 7A). Ventilation defects observed in hyperpolarized gas MR images of asthmatic patients (Figure 7B) have been shown to correspond spatially to regions of air trapping as identified on CT [42]. One study determined that the number of ventilation defects per image slice correlated to asthma severity and degree of airflow limitation [19]. Furthermore, the ventilation defect percentage as measured via hyperpolarized ³He MRI has been shown to correlate with the clinical features of asthma patients and including medication requirement, airway pathology, severity, symptom score, and atopic markers [43]. It was also shown that many areas of regional obstruction persisted over time, with 67% of ventilation defects persisting over an interval of 31 days and 38% persisting over 85 days [39]. It has also been shown that ³He MRI can be used to measure treatment effects after bronchial thermoplasty [7].

Figure 7. Magnetic Resonance Imaging.

MRI lung demonstrate ventilation maps based upon the distribution of hyperpolarized Xe gas assessed via MRI of a normal (A) and an asthmatic subject (B) respectively. Note the patchy regions of poor to no ventilation in subject with severe asthma.

Apparent diffusion coefficient (ADC), a method that takes advantage of the diffusive nature of gases in diffusion weighted MRI, can be utilized to infer the structure of alveoli and terminal bronchioles [79]. Regions where the diffusive motion of gas atoms are restricted by normal alveoli walls have lower ADC values, whereas areas of increased alveoli size or alveoli destruction allow for increased diffusion and are characterized by higher ADC values (Figure 8). Higher values of ADC have been shown to correlate with areas of emphysema-like lung by CT and lower diffusion capacity on pulmonary function testing [80]. Asthma patients have been found to have small focal areas of increased ADC values that may represent air trapping [81]. Although hyperpolarized gas MRI provides a direct measure of lung ventilation, the high cost and shortage of ³He, the need for polarizers and specialized hardware, limit this technique to certain research institutions [44].

Figure 8. Apparent Diffusion Coefficient.

Apparent Diffusion Coefficient (ADC) map of healthy non-smoker (A), Chronic Obstructive Pulmonary Disease (COPD) gold stage 2 (B), and severe asthmatic (C). The color scale on the right represents diffusion coefficients in cm2/sec with blue representing low ADC values and yellow representing higher ADC values. Notice the regions of higher ADC values in the COPD patient corresponding to areas of alveoli destruction. Image courtesy of James Quirk, PhD, Washington University in St. Louis

Multi-volume CT and ¹H MR imaging are also employed as another means for evaluating lung function by extracting the per voxel signal/density change between inspiration and expiration [45-48]. Previous multivolume CT studies have demonstrated the ability of this technique to assess regional changes in lung density (interpreted as ventilation) and regional variations in ventilation in both healthy subjects and diseased patients [45,47,48]. While this technique using CT has been shown to be exquisitely sensitive to regional variation, concern over exposure to ionizing radiation makes serial imaging with CT impractical, particularly for vulnerable populations [49,50]. Multivolume MR imaging has recently been developed as a potential surrogate for assessing ventilation inhomogeneity due to gravitational dependence and regional abnormalities caused by lung disease [46]. One advantage of gated MR imaging is that patients can freely breathe during a scan; through a technique of either prospective or retrospective gating, images at inspiration and expiration can be utilized for ventilation mapping [51].

Positron emission tomography (PET) is able to measure pulmonary perfusion and ventilation when used in combination with a positron emitter isotope of nitrogen, ¹³NN. These PET ventilation studies have been undertaken to assess asthma pathophysiology, and have allowed for the developments of models which help explain the heterogeneity of peripheral airway involvement [52,53]. Although ¹³NN-PET imaging can provide a direct measure of lung ventilation it is limited by low resolution, low signal-to-noise, nontrivial radiation dose, and time averaged acquisitions which might not reflect tidal breathing [54].

In addition to measuring lung ventilation, positron emission tomography (PET) imaging, most often with fluorine-18-fluorodeoxyglucose (¹⁸F-FDG), a commonly used radiolabeled molecule, is able to visualize metabolically active cells. FDG-PET shows promise as an imaging biomarker of lung inflammation in asthma patients as previous studies suggest that neutrophils are the primary source of increased FDG uptake in the lungs [55]. Furthermore, in studies involving human subjects, increased regional FDG uptake was shown to correlate with conditions characterized by inflammation, sarcoidosis, cystic fibrosis, and chronic obstructive pulmonary disease [56-59]. The ability of FDG-PET to assess an inflammatory response points to its potential as a tool to better understand asthma pathogenesis, phenotype differences, and responses to anti-inflammatory therapies [2].

Conventional lung function measuring techniques such as pulmonary function tests and exercise tests, which measure aerobic capacity and dynamic hyperinflation, lack the ability to assess regional distribution of changes in local airway resistances and airway volume growth [60]. Computational fluid dynamics is a technique whereby airflow patterns of the respiratory system are simulated in three-dimensional computer models, and provides a quantitative basis for predicting airflow and transport of inhaled material [61]. Computational fluid dynamics combined with HRCT was utilized to study the asthmatic response to bronchodilation [62]. This study determined that changes in CFD airway model parameters, after administration of a bronchodilator, correlated to the observed changes in clinical outcomes via spirometry measurements [62]. A separate study utilized hyperpolarized ³He phase-contrast (PC) MRI in conjunction with CFD to compare measured and predicted airflow patterns in rat pulmonary airways. Their findings demonstrated that integration of these two techniques can be used to develop and assess predicted airflow in-vivo and test mass-transfer models, which are fundamental to gas mixing in respiratory physiology [63]. Additionally, CFD models have been used in an attempt to assess the bronchiolar airway deposition of inhaled aerosolized medication, in order to better understand delivery of anti-inflammatory drugs in asthma patients [64].

CT and proton MRI tend to overlap somewhat in the implementation of visualizing lung structure and function, however, each imaging modality has its own set of drawbacks and advantages. Specifically, CT is typically the gold standard in pulmonary imaging and offers greater image resolution and easily-interpretable x-ray attenuation in hounsfeld units, but it exposes patients to ionizing radiation. Although the resolution of MRI is often slightly lower, it can serve as a longitudinal imaging surrogate for CT to limit a patient's exposure to ionizing radiation, with signal that is less readily quantifiable as macroscopic density.

Relating structure and function to clinical parameters

Airway remodeling is a term used to describe increased airway wall thickness in patients with asthma. This condition encompasses a range of processes including mucous gland hyperplasia, smooth muscle hypertrophy, inflammatory cell infiltration, and collagen deposition [65]. CT has been used to evaluate the extent of airway wall thickening [66,67]. Wall area percentage (WA%) and wall thickness percentage (WT%) measured by CT were increased in patients with severe asthma and correlated with airway epithelial thickness on endobronchial biopsies. Not only did airway measurements correlate with pathology, clinical correlations were apparent as well. Patients with increased WA% and WT% had a lower FEV₁ and greater bronchodilator response [17]. A separate study found a significant correlation between WA and asthma control score for all bronchi and WA/BSA in the subsegmental bronchi only [68]. These imaging characteristics can help characterize asthmatic phenotypes and differentiate between severe and non-severe disease as those with severe disease exhibited increased airway remodeling.

Further evaluation of differences in airway remodeling among severe asthma cluster subphenotypes was undertaken by Gupta et al [3]. Unbiased phenotyping of patients with severe asthma was completed using a cluster analysis previously described [69] and showed four distinct phenotypes. When compared to the healthy control group, all four phenotypes exhibited a decrease in lumen area and an increase in WA%. Although there was no significant difference between the subphenotypes, those severe asthma patients with persistent airflow obstruction had a significantly increased WA% when compared to those without fixed obstruction. Another key finding was that WA% was increased in patients with persistent neutrophilic inflammation measured by repeat sputum evaluation. Neutrophilic inflammation has also been associated with the air-trapping phenotype. Patients with air trapping were significantly more likely to have asthma-related hospitalizations, ICU admissions, and greater airflow limitations compared to those without air trapping [4]. This suggests that air trapping quantified by CT may identify a unique, more severe phenotype of asthma.

A novel cluster method was recently implemented to identify CT determined phenotypes [5]. Three unique clusters were described using CT measures of air trapping and proximal airway remodeling. All clusters demonstrated air trapping, but Clusters 1 and 3 had more significant air trapping and worse lung function than Cluster 2. Cluster 1 patients had increased lumen and wall measurements, Cluster 2 did not have proximal airway remodeling, and Cluster 3 patients demonstrated luminal narrowing. Imaging characteristics can be incorporated in future cluster analyses to better understand specific asthmatic phenotypes. Choi et al. in a series of papers [70-72] have demonstrated the use of a combination of metrics including lung shape, hydraulic diameter of the airway segments, airway cross sectional shape, airway branch angles, lung densities, and more to improve the ability of quantitative CT to differentiate between non-asthmatics, non-severe asthmatics and severe asthmatics. These growing set of metrics are being used to advance the concept of identifying clusters serving to differentiate the astmatic subject into sub-groups with the hopes of improving the discovery of more targeted interventions.

Currently, serum total IgE and blood eosinophil counts are examples of biomarkers being utilized to characterize specific asthmatic phenotypes to determine the appropriate patient for specific treatments. These markers can also be used to monitor response to treatment with immunomodulatory and biological therapies. Airway imaging is an emerging biomarker, but standardization is required [6]. While awaiting further validation studies, it is important to note that there is some current evidence to support the use of imaging for assessing and predicting response to specific treatments.

Imaging as a biomarker

The effect of inhaled corticosteroid use on air trapping in mild to moderate asthma patients with uncontrolled symptoms has been assessed using CT. After completing 3 months of therapy with an inhaled corticosteroid, patients exhibited a decrease in air trapping as measured by CT [73]. Thus, air trapping can serve as a potential outcome related to disease control. Recently, biologic therapy with anti-IL5 monoclonal antibody (mAb) has shown promise to reverse airway remodeling process. Haldar et al demonstrated in 26 patients with severe refractory asthma with sputum eosinophilia that treatment with mepolizumab (an anti-IL5 mAb) significantly decreased average wall area over one year compared with placebo [74]. Current analysis tools allow for measurement of the ventilation defect percentage (VDP) from ¹²⁹Xe and ³He MRI, which is the volume of lung not involved in ventilation. Texture features can also be generated from MRI ventilation images and can be used to quantify differences in lung ventilation post bronchodilator therapy [75]. Further studies are needed to evaluate the optimal imaging biomarker to assess response to biologic therapy in patients with asthma.

A combination of two imaging modalities can be used to assess response and guide therapy. Regional lung ventilation has been quantified in severe asthma patients using multi detector CT and ³He MRI [7]. A majority of the patients with severe asthma underwent bronchial thermoplasty and had repeat ³He MRI. Comparisons were made between the pre- and post-treatment images in relation to segmental defect percentages and whole-lung defect percentages. Although ventilation defects increased immediately after bronchial thermoplasty, the ventilation defects decreased after a longer period of observation. This information can be used to potentially guide bronchial thermoplasty and can be used to target specific segments, potentially decreasing the number of treatment sessions or the number of segments needing treatment.

Conclusion

Current use of CT of the chest in asthma has been to identify alternative diagnosis or complicating conditions that may be contributing to uncontrolled disease. Recent studies have now demonstrated that quantitative CT of the chest and hyperpolarized gas MRI can be used as a biomarker of airway remodeling. Prospective longitudinal trials of targeted biologics (anti-IgE, IL-5, IL-4a and IL-13) and non-pharmacologic (bronchial thermoplasty) treatments using these quantitative imaging biomarkers are needed to assess if these treatments are modifying the course of this disease. If these disease modifying effects of treatment can be demonstrated then perhaps they can be introduced earlier in the disease process.

What Do We Know?

• Several imaging modalities are clinically available for use in the evaluation of patients with asthma.

• CT, MRI, and PET are among the modalities that can provide detailed assessment of lung structure and function.

• Measurements from imaging, such as wall thickness, air trapping, ventilation defects, can serve as a biomarker and be used to assess response to new therapies.

What Is Still Unknown?

• Prospective trials of biologic agents and bronchial thermoplasty using imaging endpoints are necessary to determine if these therapies are truly disease modifying agents.

• Further studies are needed to determine the optimal biomarker to assess response to therapies.

Abbreviations Used

AWA

Average wall area

CT

Computed tomography

DPM

Disease probability mapping

EBUS

Endobronchial ultrasound

FRC

Functional residual capacity

HU

Hounsfield units

MRI

Magnetic resonance imaging

OCT

Optical coherence tomography

PRM

Parametric response mapping

PET

Positron emission tomography

RV

Residual volume

TLC

Total lung capacity

WA

Wall area

WT

Wall thickness