Will Airway Gene Therapy for Cystic Fibrosis Improve Lung Function? New Imaging Technologies Can Help Us Find Out

Abstract

The promise of genetic therapies has turned into reality in recent years, with new first-line treatments for fatal diseases now available to patients. The development and testing of genetic therapies for respiratory diseases such as cystic fibrosis (CF) has also progressed. The addition of gene editing to the genetic agent toolbox, and its early success in other organ systems, suggests we will see rapid expansion of gene correction options for CF in the future. Although substantial progress has been made in creating techniques and genetic agents that can be highly effective for CF correction in vitro, physiologically relevant functional in vivo changes have been largely prevented by poor delivery efficiency within the lungs. Somewhat hidden from view, however, is the absence of reliable, accurate, detailed, and noninvasive outcome measures that can detect subtle disease and treatment effects in the lungs of humans or animal models. The ability to measure the fundamental function of the lung—ventilation, the effective transport of air throughout the lung—has been constrained by the available measurement technologies. Without sensitive measurement methods, it is difficult to quantify the effectiveness of genetic therapies for CF. The mainstays of lung health assessment are spirometry, which cannot provide adequate disease localization and is not sensitive enough to detect small early changes in disease; and computed tomography, which provides structural rather than functional information. Magnetic resonance imaging using hyperpolarized gases is increasingly useful for lung ventilation assessment, and it removes the radiation risk that accompanies X-ray methods. A new lung imaging technique, X-ray velocimetry, can now offer highly detailed regional lung ventilation information well suited to the diagnosis, treatment, and monitoring needs of CF lung disease, particularly after the application of genetic therapies. In this review, we discuss the options now available for imaging-based lung function measurement in the generation and use of genetic and other therapies for treating CF lung disease.

Cystic Fibrosis

There are numerous disorders that affect the lung, many of which have environmental or infectious origins. However, in the current phase of genetic therapy development and use, the monogenic inherited diseases are likely to be the most amenable to treatment in the near future. While each of the ∼5,000–8,000 monogenic diseases is by itself rare, together they will affect around 6% of the population.¹ The exact number of monogenic diseases that affect the respiratory system remains unknown, but two of these—primary ciliary dyskinesia and cystic fibrosis (CF)—are relatively well characterized.² It is expected that with success in treating the monogenic diseases, more genetically and/or physiologically complex diseases, such as asthma, allergy, lung cancer, some infections, transplant rejection, and lung injury, will all likely become targets for gene therapy in the future.³

The most common and life-shortening inherited disease for which genetic treatments are being developed is CF. CF remains an insidious disease that slowly smothers the health and potential of too many young lives. Around the world there are more than 70,000 people worldwide living with CF.⁴ It is the most common fatal genetic disease in the developed world; 1 in 25–30 people of Caucasian ancestry carry a defective copy of the CFTR gene. Although it affects almost every organ, it is the lung disease that causes most of the morbidity, and results in an early death in most cases, after a lifetime struggling to deal with progressive respiratory failure. While life expectancy has improved—people born with CF today currently have a median life expectancy of 47 years⁵—the effects on quality of life for many are significant and unrelenting. The daily demands of medications, physiotherapy, numerous prolonged hospital admissions, lost school time, restricted physical activity, and inherent social isolation mean that a CF child may never be able to fully achieve their aspirations and inherent potential. For that reason, understanding, diagnosing, and treating CF lung disease has naturally become the primary focus of much CF basic science and preclinical research, in the quest to cure or effectively treat the disease.

In the last decade a range of new CF lung disease treatment options have appeared, and many are in development.⁶ Some have reached the clinic, including the highly effective modulator therapies that began with Ivacaftor for CFTR gating mutations, such as G551D,⁷ then Orkambi⁸ and Symdeko⁹ for patients with two copies of the Phe508del trafficking mutation, and most recently Trikafta¹⁰ for patients who have at least one copy of the Phe508del mutation. These systemic oral daily treatments have boosted lung function (the volume of air that can be forcefully exhaled in one second [FEV1]) to more than 10% over standard care, and together with their beneficial effects on other organ systems have been properly recognized as a revolutionary advance in CF care. They have been transformative for many people living with CF. Nevertheless, there are limitations to these treatments. Cost is eyewatering; it is annually upward of US $200,000 per patient per year. As real-world treatment proceeds, some patients that should respond well have not, and others experience side effects that cause them to discontinue use. Perhaps most importantly, however, ∼7% of people with CF have nonsense and rare CFTR mutations that mean they do not produce any CFTR protein and will therefore never benefit from CFTR modulators. For this group of patients other therapies are required.

CF Gene Therapy

The CFTR gene was discovered in 1989,¹¹ and shortly thereafter the first retroviral,¹² adeno-associated virus (AAV),¹³ and plasmid studies¹⁴ showed CFTR could be expressed in vitro and in vivo. The first human studies occurred in 1993.¹⁵ In the early 90’s, CF lung disease was seen to be a perfect candidate disease to show the effectiveness of the new gene therapy approaches that were emerging. It is a monogenic disease for which the causative gene had been identified; the airway epithelium is in an organ with an essentially outward-facing surface, theoretically enabling easy access to the affected cells by inhalation; and since it is recessive, a gene therapy could be effective with as little as 25% of the normal CFTR gene expression level.^16,17 However, early hopes for fast delivery of new gene-addition treatments to CF patients had to be calmed as the challenges associated with barriers to gene vector delivery to the CF airway epithelium, including inflammation and immune responses, and adverse events in gene therapy trials became clearer.¹⁸ For many years gene therapy—not just for CF, but for many inherited diseases—lay in the translational “valley of death.”¹⁹ Despite more than 30 years of development since the CFTR gene was discovered, and after at least 36 gene therapy clinical trials or studies, none of these methods has yet been approved clinically for treatment of CF lung disease.²⁰

Today, the addition of a correct copy of the CFTR gene into CF airway cells (gene-addition), or correction of the CFTR gene in situ (gene-editing), remain the most biologically rational ways to prevent or treat CF airway disease for any CFTR mutation. There is a growing pipeline of gene therapy development for other diseases, with at least nine gene therapies now FDA approved for treating inherited diseases and several cancers.²¹ The first of these, Luxturna, appears curative for retinal dystrophy²² and was FDA approved in late 2017.²³ Many more are expected. For CF, the proof of principle for gene therapy has also been established. A repeat-dose Phase II CFTR nonviral (liposome) gene transfer study in the United Kingdom validated that a gene therapy can correct CF lung disease in humans.²⁴ However, the effects on the primary outcome measure (lung function, i.e., FEV1) were modest, indicating that more efficient gene transfer vehicles such as viral vectors are required.

A range of gene-based therapies are currently in development for CF.⁶ Viral vectors such as recombinant AAV (rAAV),²⁵ human bocavirus type 1,²⁶ lentivirus (LV),^20,27,28 helper-dependent Adenovirus,²⁹ and piggyBac/adenovirus³⁰ have all been tested. Nonviral approaches include minicircle DNA, lipoplexes, peptide nanoparticles, antisense oligonucleotides, and mRNA therapies.³¹ These are supplemented by gene-editing approaches using CRISPR/Cas9.³² Each of those approaches varies in the manner in which it delivers CFTR DNA or corrects the cellular DNA or protein, however, they all share the same challenges associated with delivery and measurement of success. The complexity of the lung anatomy and physiology, combined with the presence of thick sticky mucus and pathogens in the CF lung makes uniform deposition and uptake almost impossible to achieve, and hard to quantify.

Airway and Lung Health Assessment in CF: Requirements and Limitations

Lungs are complex organs, taking in air with diaphragm movement and moving it to the alveoli, where exchange of oxygen and carbon dioxide occurs. When properly matched to perfusion, appropriate levels of oxygen are supplied to all tissues and carbon dioxide is eliminated, supporting normal body functioning. The continuous expansion and contraction during breathing both defines the operation of the lung, but also limits the types of assessments of lung and airway health that can be performed. The term “lung function” ideally encompasses the matched airflow and blood flow of the lung. In the study, we will use lung function to represent the airflow component, as this aligns with the commonly used terminology, and it is also the aspect of lung health that underlies the majority of research and clinical decision making in the treatment of CF lung disease.

The oldest, most well-established, and most readily accessible method for lung function assessment is spirometry. The measurement of FEV1—the volume produced in the first second of a maximally forceful exhalation—has long been the gold standard method for tracking changes in lung function. Spirometry can also be performed in small laboratory animals using devices such as the flexiVent (Scireq, Canada), although this is typically an invasive terminal procedure. However, spirometry is primarily a global measure, and is therefore not capable of detecting and locating airflow changes necessary to routinely show a treatment benefit is present with adequate sensitivity to detect small local changes, or any compensatory effects such as collateral ventilation. As a result, FEV1 may not be the most appropriate primary outcome measure in clinical trials when CF patients have either a normal or mild reduction in baseline FEV1.³³ There are many drugs that are efficacious in mild CF that did not produce a statistically significant improvement FEV1 during clinical trials (examples include dornase alfa,³⁴ tobramycin,³⁵ hypertonic saline,³⁶ and azithromycin³⁷), but did demonstrate other clinically beneficial endpoints (e.g., lowered exacerbation rates and improved patient-reported outcomes). It is notable that only 2 of the 36 CF lung gene therapy trials performed were sufficiently powered to detect beneficial change using spirometry³⁸ (all others were phase I/IIa designed to assess safety rather than efficacy). An analysis of the most recent of these noted that “spirometry is a variable and effort-dependent measurement and, therefore, is less than ideal” as a primary lung function endpoint, but was the only measure that was regulatory compliant.³⁹ It is also not suitable for infants or small children under ∼6 years of age and has a high day-to-day variability. Today, FEV1 remains a commonly used primary outcome measure when assessing CF therapies, and is often taken as an acceptable surrogate of survival, function, and even quality of life in the CF population.³³

The clear improvements in FEV1 produced by CFTR modulator therapies show that spirometry can play a valuable role in the detection and description of large improvements in global lung function. However, spirometry is insensitive to small airflow change, whether due to small global improvements or due to strong local improvements in function that are lost in the global picture of lung function. It is now clear that inflammation-induced lung damage occurs before any decline in FEV1, so normal spirometry does not rule out the possibility of structural lung disease.⁴⁰ The danger for the gene therapy field is that early clinical trials of promising novel and genetic therapies—particularly those that produce small initial benefit that may enhance over time due to the transduction of airway stem cells—will fail to advance because of the insensitivity of the lung function analysis, not the capability of the therapeutic.

Over the past decade, the multiple-breath washout (MBW) technique has become a more commonly used approach to quantifying lung function in CF by measuring the lung clearance index (LCI). The LCI is a marker of overall lung ventilation inhomogeneity, and is more sensitive than spirometry (FEV1) for detecting early peripheral airway damage (in airways less than 2 mm diameter) in CF.⁴¹ It is also effort independent and can be performed in young children. LCI is becoming more widely used as an outcome measure in CF clinical trials,^42

–45 but is not yet widely accepted for the management of patients with CF.⁴⁶ The MBW technique can also be performed in large and small animal models.⁴⁷

For CF lung gene therapy to become a clinical reality, the field is aware that there must be clear advances in the effectiveness and efficiency of gene delivery. What has been difficult to couple well into that research arena is the inherent requirement to be able to measure in vivo effectiveness at a sufficient spatial and temporal resolution and accuracy to ensure that small but clinically valuable treatment benefits to lung function are detectable. Accurate, detailed, and reliable analyses of improved overall function of the lung—airflow as well as its component parts and processes in CF pathogenesis—are necessary to determine how and where lung function has improved. This need for better lung health outcome measures is not restricted to CF research. The same basic impediments exist for the measurement of almost any treatment of lung disease.

Lung Genetic Therapies require new Outcome Assessment Techniques

Despite their obvious potential benefits, gene-based therapeutics—particularly the integrating viral vectors such as LV—have a range of risks that must be evaluated both preclinically and clinically. These include the theoretical risk of insertional mutagenesis (which has not been detected in animal or human studies since the use of self-inactivating vectors⁴⁸ became standard), the long-term effects of exogenous CFTR expression, and the ability to redose.³⁹ The development and testing of potential mechanisms to halt gene expression in either a worst-case adverse effect scenario, or if a more effective therapy is found in the future, may also be required.

Due to these additional safety issues compared with a pharmaceutical therapy, the method of delivery and the techniques for measurement of beneficial change will necessarily be different. We have already proposed that for integrating vectors it is unlikely that whole lung dosing will be wise, or even permitted, in the first clinical trials.²⁰ It is likely that the safest way to test genetic treatments in humans, especially for permanent gene-addition and gene-editing agents, will be through bronchoscopic delivery into segments or small regions of the lung. As a result, CF genetic therapies delivered locally and topically must necessarily produce a very large effect to be discernible using spirometry, given the averaging effect of the lung when measurements are made at the mouth. Clearly, in this situation, alternative assessment techniques will be required. If methods are available to quantify local rather than global lung function, then measurable improvements can likely be much smaller, and could allow small but beneficial changes to be detected, preventing the therapy from being falsely rejected as unsuccessful. This latter approach must first be demonstrated in small and large CF animal preclinical studies, to establish the effectiveness of promising treatment agents and methods to be translated to humans.

Careful choice of the patient population will also be essential for maximizing the ability to detect benefit. Patients must have lungs healthy enough to permit good delivery of the genetic agent, but also have disease that is not so severe as to mask any beneficial effect.⁴⁹ As already noted, people with normal baseline FEV1 are also unlikely to show benefit from treatment. An additional challenge is the design of CF gene therapy trials in the age of highly effective modulator therapies, when discontinuation of those therapies may not be possible, and simple placebo controls are unlikely to be acceptable.⁵⁰ Finally, for the first trials using integrating vectors or gene-editing agents that can target the airway stem cells, the patient population requires additional thought, as it is possible that only patients with advanced disease will be eligible, due to the unknown long-term effects of the therapy.

An essential question is how can researchers and clinicians reliably measure airway and lung health in this complex and constantly moving organ? Surely there should be, or must be, better measures to base clinical decisions on than global airflow at the mouth, as assessed by spirometry. To properly describe the health state of the CF lung following airway gene therapy, multiple points along the mechanistic cascade should be assessed, ensuring: correct CFTR channel function (through assays such as potential difference measurement or Ussing chamber analysis); targeting of the airway cells required for long-term correction; airway surface health (e.g., increased [airway surface liquid (ASL)] depth); improved airway clearance through mucociliary transit; adequate gas exchange; ventilation/perfusion status; and airflow throughout the lung tree. Analytical methods that are able to visualize and quantify these processes will be advantageous.^51
–53 However, to date, imaging of the lung has been largely restricted to providing analysis of lung structure.

Structure/Function Relationships: Where do they work well, and where do they work poorly?

A time-honored and well-established biological principle, stemming from the influence of evolution on tissue, organ, and organism functioning, is that the function of an organ such as the lung can be inferred from its structure.⁵⁴ Such structure/function relationships are well demonstrated in the skeletal system where, for example, a structural change such as a bone break is highly predictive of changes in function, that is, the inability to walk. In this case, the change is clearly visible and measurable using a variety of imaging modalities, and its impact on function is clearly understood. In more complex organs this relationship still holds but is limited by our ability to resolve the structures, measure whether they have changed, and predict how any changes in structure might manifest as changes in function. This predictive ability depends on the complexity of the organ, and for lungs that complexity—as well as image capture speeds—has largely prevented the application of robust, accurate, and noninvasive technologies to measure more than simply the average lung function while the lung is operating, that is, during breathing.

At present, most tests either provide information about lung structure, or function; few provide both. As a result, they can only provide a snapshot of what is a dynamic process that is intertwined biologically and physically, in both space and time. Although MBW tests for measuring LCI can provide an indication of the relative dysfunction of small or large airways, spirometric tests essentially provide global measures of lung health because their localization data are not fine-grained enough. As a result, the information on the source of dysfunction may not be adequate for treatment targeting or understanding how and where disease is progressing. In contrast, methods that can reveal regional information about the location of disease typically rely on imaging of lung structure, and therefore cannot measure function.

Modern image-based assessments of CF lung health that can be applied in humans as well as animals now primarily rely on two imaging modalities: computed tomography (CT); and magnetic resonance imaging (MRI). These techniques have been steadily improved to gain higher temporal and spatial resolution. Lowered X-ray radiation doses have paralleled most improvements in technology, so that in children with CF a breath-hold lung CT scan can be achieved with a dose of ∼0.5 mSv.⁵⁵ Since early CF lung disease begins in the small peripheral airways, this can cause a change in ventilation distribution without registering any global change in flow parameters when measured through spirometry, since these are more influenced by the central airways. Indeed, substantial structural lung damage has been demonstrated to occur in children with normal lung function.⁵⁶ As a result, CT has been useful for visualizing structural lung disease, particularly airway wall thickening and bronchiectasis, for assessing disease progression.⁴⁰

CT scans can provide a wealth of structural information, but currently functional data are often either missing, or quite modest. Interpretation of function is typically made by highly trained radiologists through inference from visible structural data. To enhance the value from structural imaging, scoring methods have been developed to attempt to robustly quantify structural changes and use them to better predict function.^57

–61 A recent example of these is PRAGMA-CF, which overlays a grid over 10 axial slices from expiratory and inspiratory scans, to calculate the proportions of total disease, bronchiectasis, and trapped air in each region of the lung.⁶¹ The method has a high sensitivity and reproducibility for quantifying CF lung disease based on structural abnormalities. However, to our knowledge, these scoring systems have not been validated for use in CF animal models that would be used to develop new treatments that could be assessed in part using CT structural information.

The primary advantage of traditional proton MRI is that it can produce structural images without the use of ionizing radiation, which is a clear advantage for serial monitoring of CF pediatric populations through to adulthood, where radiation risk must be considered.⁶² However, the spatial and temporal resolution of MRI is substantially lower than CT, and for many years this has precluded its use for assessing lung structure in a breathing lung. However, recent improvements, such as Ultrashort Echo Time (UTE) imaging, have improved scan resolution, although at the expense of image contrast. This has made lung UTE MRI feasible for assessing CF structural lung disease, due to its ability to detect very subtle lung physiology, with scans acquired within a single breath hold.⁶³ Scoring systems can also be used for assessing lung structure through MRI, in a similar manner to CT.^64,65 As a result, UTE MRI can be used to quantify the extent and severity of lung abnormalities in very young patients with CF using a CF-specific scoring system, with a similar accuracy to CT and without ionizing radiation.^66,67 Nonetheless, despite these clear improvements in assessing lung structure through CT and MRI, methods that can provide functional information about the lung with high spatial and temporal resolution are required.

Ideal Features of an Image-Based Lung Function Test

The aphorism “if you can't measure it, you can't manage it,” from the business quality management field, applies equally well to the measurement of CF lung function. Without direct, reliable, sensitive, and localizable airflow measurements throughout the lung breath cycle, the management of lung health is very difficult. In the context of pulmonary disorders like CF, how does this lopsided information base of lung structural data affect our ability to know when and where lung function is altered by treatments using genetic therapies? One approach to answering this question is to first consider the features of an ideal imaging-based test for lung function and see where our current advantages and deficiencies may lie. We propose that an ideal imaging-based diagnostic and monitoring method would have the following features:

Noninvasive: For simplicity of measurement, as well as ensuring the measurement process does not interfere with the parameters or the wellbeing of the individual being examined.

The physical complexity of the lung is contained in the analysis: Function from the largest airway to the alveolar spaces should be available.

The regional fine detail of the lung can be examined: The redundancy and reserve in the functioning of the lung demands knowledge of the extent that disease is affecting airflow. Both the level and the maldistribution of airflow are important factors.

Assessments are possible at any or all points in the breath cycle: Lungs are dynamic organs, and disease may only be apparent at certain phases of the breath cycle, so the entire cycle of airflow must be incorporated.

Lung motion effects must be inherent in the technique: Motion is the raison-d'être of a lung and the reciprocating airflow it produces, and the measurement technique must embrace motion, not avoid it. The spatial and temporal image capture capabilities of the technique greatly influence this capability.

Assessment must be possible during normal breathing: This ensures widest applicability and patient usability. The capability for testing during special breath maneuvers—such as in exercise, maximal and minimal breath hold, different body orientation—is desirable. This ability will also be crucial for research techniques that perturb the respiratory system to enable deeper analyses of respiratory disorder causes, mechanisms, and effects.

Routine application is possible: Access should be like that of spirometry techniques, low cost, wide access, and rapid results enable clinical use for day-to-day respiratory healthcare. In animal model research these factors will be altered by the cost/benefit settings of the respiratory research questions posed.

Radiation risk is acceptable, or absent: The health cost/benefit equation controls their use, and reuse, for monitoring disease establishment and for revealing treatment benefits, particularly in children.

The patient age does not prevent or limit access: Patient age may require imaging systems of different sizes and characteristics, and testing may not be currently possible for, for example, the very young, or in special circumstances such as the very elderly and bed-bound patients.

Equipment, training, and expertise is readily available: While highly specialized technologies can provide unique insights to lung health, they are often restricted to use in well-funded hospitals, institutes, and academic-research settings. Ready access that can reach out into community health care settings is required. Indeed, this is a major advantage of the current spirometric methods for assessing global lung function.

Scalability from preclinical to clinical models: The technology is lung-size independent and can be applied from mice up to large animal models and humans.

This list of features is by no means exhaustive, nor are all items essential for a novel technique to provide value. However, we propose that they should be kept in mind while developing or evaluating lung airflow assessment tools.

Functional Lung and Airway Imaging Methods to Measure Lung Health

A range of new tools have been developed to attempt to observe lung function throughout the breath to provide a more sensitive and accurate assessment of disease. These include functional MRI, dynamic CT, and X-ray velocimetry (XV).

Hyperpolarized gas MRI

Originally using ³He, and more recently with ¹²⁹Xe this method is becoming more common for assessing changes in ventilation heterogeneity resulting from progressive CF lung disease. Most previous work has utilized ³He, as it is easier to polarize than ¹²⁹Xe, and produces a higher signal-to-noise ratio. However ¹²⁹Xe is cheaper and more readily available,⁶³ although it can be an anesthetic at high concentrations. The gases are administered immediately before MRI, using a breath hold to acquire images without respiratory motion. Both provide a greater airspace T2 MRI signal than air, with regions of low signal intensity indicating poor ventilation.⁶⁸ The ventilation distribution images allow a measure of ventilation defect percentage (VDP)—the percentage of nonventilated lung volume—to be reported. Hyperpolarized MRI methods are robust and repeatable.⁶³ They have also been shown to have a higher diagnostic sensitivity than spirometry, LCI, and CT for detecting early stage CF lung disease.⁶⁹ Other studies have demonstrated that these techniques can be applied to unsedated infants and children,⁷⁰ suggesting that they should have high clinical utility in tracking CF disease progression. However, at present, it is suggested that hyperpolarized gas MRI for assessing function should be performed in combination with structural CT or MRI to achieve a complete clinical picture.⁶³ Using complementary structure/function imaging methods enable the location of flow defects within the lung to be identified, and potentially improve clinical management. However, one of the primary drawbacks is that the hyperpolarized gas must be produced on site, typically by spin exchange optical pumping,⁷¹ which increases the cost and decreases the accessibility of the technique. Hyperpolarized MRI methods can also be applied in animals as small as mice.^72,73

Dynamic CT

The challenges associated with CT and traditional proton MRI being a snapshot of lung structure at a particular time have been clearly described above. Dynamic CT attempts to overcome this issue, by capturing multiple CT snapshots of the lung at different phases of the breath in an attempt to understand dynamic lung motion. This technique has been used in humans, with CT scans at end-inspiration and end-expiration used to measure lung tissue deformation to identify and track tumors.^74
–76 However, this type of technology has not been widely utilized outside of evaluating tumor motion and is not fine-grained enough to quantify lung motion as a measure of lung function. To better capture dynamic lung motion CT scans can be performed during multiple inspiratory breath-hold maneuvers at increasing pressures to simulate the lung structure during normal motion.⁷⁷ With this approach, the motion of the lung parenchyma could not be tracked, so a registration algorithm was used to quantify the motion of manually identified landmarks at the edge of the lung, and within five transverse slices. By doing this, a regional lung compliance map could be created. However, we suggest that this technique does not satisfy the criteria outlined above: it does not capture images during normal breathing; is not noninvasive because it probably cannot be successfully performed in conscious patients; and the radiation dose would be very large due to the multiple CT acquisitions that are required.

X-ray velocimetry

XV is a new technique to measure lung ventilation that overcomes the limitations of CT and MRI structural imaging. XV uses X-ray imaging to track the motion of the distinctive speckle pattern that is created by the overlapping alveoli within the lung,⁷⁸ and from the lung motion can calculate functional information about airflow. Regardless of whether data are acquired in two dimensional (2D) or three dimensional (3D), the principle of XV remains the same.

The preclinical validation utilized β-ENac mice⁷⁹ that possess CF-like airway disease, and was performed by our team using CT scans acquired at the SPring-8 Synchrotron in Japan.⁸⁰ In those scans, the image acquisitions were gated to ventilation to capture ∼15 images throughout every breath, at multiple angles over 180°. The CT projections were separated into phases of the breath and reconstructed to produce 15 separate CT volumes, representing different points throughout the respiratory cycle. XV analysis was then applied to that complete dataset to track lung tissue displacement throughout the breath.⁸¹ From that data, the regional air volume in the lungs was quantified at every point within the breath. When that information was associated with the airway tree, the airflow at every point within the tree could be calculated throughout the breath.⁸² XV ventilation maps showed that β-ENac mice exhibited patchy lung disease, with the spatial distribution of airflow disease quantified using a parameter called the expiratory time constant. The regions of reduced airflows corresponded to mucus obstructions that were identified histologically and correlated with gold-standard flexiVent spirometric analyses. We have since translated 3D XV techniques to the Australian Synchrotron Imaging and Medical Beamline,⁸³ and to a laboratory-based liquid metal jet X-ray source setup.^84,85 A commercial product called the Permetium (4DMedical, Melbourne, Australia) is now also available. This is a turnkey system that can perform XV scans that identify the regions and levels of lung disease in small laboratory animals, with XV data analysis performed in the cloud by 4DMedical.

An example of the capability of small animal XV imaging with the Permetium is shown by altered ventilation produced by a single intratracheal infusion of a bolus of saline into a mouse lung (Fig. 1). A range of respiratory research questions are now being investigated by employing the features of XV lung ventilation to provide unusual insight into related areas in small animal models, such as analysis of airflow movement that occurs in the lung due to the beating of the heart,⁸⁶ and the effect of mechanical ventilation on lung injury.^87
–89

Figure 1.

The mean breath XV ventilation scans from an anesthetized mouse secured vertically, before and after delivery of 40 μL of normal saline, acquired using a Permetium small animal XV scanner. Substantial reductions in airflow (red regions) in the bases of the right and left lungs, extending some distance up the lateral edges are evident. Airflow compensation appears to be present in the remainder of the right lung, as shown by the increased specific ventilation there (more widely distributed and deeper blue). Images acquired by the authors at the Preclinical Imaging and Research Laboratory (South Australian Health and Medical Research Institute Animal Ethics approval number SAM 424.19) in Adelaide. XV, X-ray velocimetry. Color images are available online.

XV provides the ability to visualize the heterogeneity of CF disease, and to do so throughout the breath cycle with high spatial and temporal detail. While the radiation doses in the formative synchrotron small-animal 3D studies were very high, subsequent refinement of the technology and protocols means that repeated 2D measures over long periods are now feasible. Such detailed measurement and localization of the patchiness of CF disease has not previously been possible in small animals, and it should enable unique structure/function analyses that can better understand this hallmark of CF lung disease known from structural imaging studies.

XV has also been validated through human clinical studies, which supported approval of XV Lung Ventilation Analysis Software (LVAS) by the FDA in May 2020 for all respiratory indications in adults. XV analysis in humans uses a low-dose imaging protocol to capture cinefluorographs with existing hospital fluoroscopy equipment. Scans at five different angles are acquired during tidal breathing, with a single breath captured at each angle. Altogether, these five scans are used to construct a four-dimensional map of regional lung tissue displacement during the breath. Tissue displacement calculations then provide regional ventilation at over 10⁴ locations throughout the lung, over the course of the breath. In addition to a visual map of regional ventilation, XV analysis also provides quantitative metrics of lung health such as ventilation heterogeneity and VDP. XV-derived VDP is a similar quantity to that measured through hyperpolarized MRI; in this case VDP is defined as the percentage of the lung with ventilation below 60% of the mean.

XV analysis overcomes many of the limitations of other functional imaging modalities. It does not require inhaled contrast agents, breath-holds, or specific breathing maneuvers, which makes it suitable for a wide range of patients regardless of age or disease severity. Additionally, it is based on direct measurement of lung tissue movement throughout normal breathing, instead of relying on snapshots. Most hospitals already have XV-compatible fluoroscopes, making XV analysis widely available to clinicians and researchers without any additional capital expenses. Since the imaging period is so short (five acceptable resting breaths) the use of fluoroscopic X-ray allows the acquisition of moving images while minimizing patient radiation exposure. The effective radiation dose for the entire protocol is low, ∼0.2 mSv, equivalent to the dose of two chest X-rays (assuming an average effective dose for posteroanterior and lateral projections is 0.1 mSv⁹⁰) or less than 4% of a standard adult high-resolution chest CT (assuming a typical dose is 5.5 mSv.⁹⁰) Fig. 2 highlights the advantage of XV analysis in detecting regional differences in function across the lung, where XV analysis has provided both quantitative metrics of lung health as well as identification of the regions of localized dysfunction.

Figure 2.

Example XV LVAS output from two patients whose spirometry indicated normal lung function (FEV1 and FVC >90% expected, FEV1/FVC >80% for both cases). The regional ventilation visualization shows specific ventilation (normalized to the mean) throughout the lung. Red = low ventilation, Green = average ventilation, Blue = high ventilation. The graphs show frequency distribution of the specific ventilation values throughout the lung. The VDP data region is block shaded in pink. (A) Illustrates mostly homogeneous, average ventilation throughout the lungs, supported by the tight peak centered at 1.0 in the histogram. In contrast, the lungs in (B) are more heterogeneous in their ventilation, as shown by the additional areas of red, and the greater area and color depth of the blue region. Clear differences between the left and right lung are visible. The histogram displays a much wider distribution and a greater VDP, and there is an increased proportion of the lungs with higher-than-average (blue) ventilation. Increased ventilation in the right lung compared with the left suggests compensation for declining function in the left lung. Clinical Trial Registration Number NCT02735746. FEV1, the volume of air that can be forcefully exhaled in one second; FVC, forced vital capacity; VDP, ventilation defect percentage. Color images are available online.

X-ray-based contrast-free pulmonary angiography (CFPA) analysis that quantifies the pulmonary vasculature using a single chest CT is also possible. While CT and MR angiography methods are available, they often require injection of contrast agents and very specific imaging protocols.^91,92 Unlike these methods, CFPA analysis can be done with prior CT scans and does not require contrast agents. CFPA has been validated through preclinical studies,^93,94 and is currently in the process of clinical validation. CFPA can be combined with XV analysis to further enhance the functional data, providing a contrast-free ventilation/perfusion (VQ) analysis.

At present, the limitations of XV appear to be largely centered on the need to use ionizing radiation. Although the dose per scan is much lower than the CT scans currently used for monitoring structural lung disease progression in children with CF,⁹⁵ radiation effects are cumulative. XV analysis also currently requires a chest CT to determine the boundaries of the lung, however, a prior CT can be used to minimize the additional radiation exposure to the patient. Subsequent XV analyses can also be done using the same CT, allowing repeated analysis over time with minimal added radiation dose. For safety, especially where XV imaging is applied to measuring lung disease in young CF children—or where repeated measurements are needed to monitor disease progression and the effectiveness of generic or other lung treatments over a lifetime—cumulative doses are likely to become important. As with CT, technological advances that continue to reduce the radiation dose during XV scans will help to establish a CF lung health diagnostic environment, where radiation dose becomes a minor and perhaps unnecessary question in the ordering of lung ventilation scans in children and adults with CF.

In summary, XV imaging can reveal ventilation defects in a similar manner to hyperpolarized gas imaging, as well as provide quantitative detail of airflow within each airway. Data identifying the level and the location of changes in local airflow throughout the lung that are produced by the disease or treatments will greatly augment the functional clinical information currently provided primarily by spirometry. A range of clinical studies should now be performed to evaluate the efficacy of XV for a wide range of respiratory indications, including CF, chronic obstructive pulmonary disease, asthma, pulmonary hypertension, lung transplant, and acute respiratory distress syndrome. The potential is high for XV imaging to more effectively and efficiently diagnose and monitor COVID-19-induced lung injury with sensitivity and an ability to detect disease localization like never before. We expect that CF clinicians, researchers, and patients will benefit rapidly from XV technology, since it is not limited in its application by patient age, provides identification and quantification of ventilation in any part of the lung and breath cycle, and uses existing hospital X-ray capabilities with a cloud-based analysis and reporting system cycle. Until CF airway gene therapy is ready for clinical evaluation, the potential of XV for the longitudinal tracking of the benefits produced by CFTR modulator therapies is clear.

How do the Choices for Lung Imaging in CF Genetic Therapies Align to the Ideal Test?

As potential new treatments for CF lung disease appear—regardless of whether they are genetic therapies or new pharmaceutical options—reliable and sensitive lung function measurement techniques that can provide accurate localization information are required. These will be essential for researchers working with preclinical small and large animal models, for human clinical trials, and for clinical assessments. They are also vital to avoid the potential loss of promising agents and methods that may show only mild benefit in the early phases of research, particularly for gene therapies applied to small regions of the lung in a targeted manner. Ideally, CF genetic therapies would be a single dose (or a short series of repeated early doses) that produce very long-lasting benefit, and so there will be a need for repeated and safe detailed assessments of disease reduction, progression, or stasis. A capability to continually gather evidence of lung health value with low procedure risk will properly support evidence-based treatment decisions, as well as continued insurer/national health system coverage of the treatment regime.

CT and conventional MRI will continue to bring important aspects of the analysis of structural lung disease to researchers and clinicians. Both techniques continue to advance, and the ability to resolve the fine detail of the lung structure will likely remain essential to building a complete picture of lung health. In their review of functional imaging options for CF lung diagnosis and treatment, Kolodziej et al. noted that “none of the currently available methodologies is simple and detailed enough.”⁶⁸ In the last 3 years, the field has advanced quickly, and gained the capability to visualize airflow in human lungs with the maturation of hyperpolarized gas MRI and the availability of XV lung ventilation analysis. Both tests fulfill many of the features of the ideal imaging-based diagnostic and monitoring method we detailed above, and so should be of special benefit to respiratory researchers. We feel that these two technologies show the greatest future potential to provide CF patients the best opportunity to measure and treat their earliest disease.

To maximize the potential of these new functional imaging methods in clinical trials and clinical assessments, and to improve the treatment of CF disease, efforts must now be directed toward conducting longitudinal randomized controlled trials that utilize and compare each of these modalities, develop publicly accessible data repositories, and promote technique standardization.

Conclusions

One of the clear failures in the history of lung function testing for CF has been the inability to bring a lung function evidence base to the treatment of infants and young children, as they are unable to perform spirometry tests. CT imaging and other analyses have greatly advanced our understanding of early disease and shown that CF lung disease is producing structural damage from soon after birth, much of it irreversible. New functional lung imaging technologies will bring sufficiently sensitive and detailed quantification of lung disease to researchers and clinicians as a routine capability. Knowledge of the location of disease within the CF lung will enable the application of the next generation of highly targeted effective genetic treatments, to halt disease progression. The opportunity now with hyperpolarized gas MRI and XV ventilation imaging is for lung function testing to move from one where diagnostics and outcome measurements have been technology limited to a time where a patient focus can come fully to the fore. Altogether these functional imaging technologies will allow us to determine whether airway gene therapy for CF works.

Footnotes

Acknowledgments

The authors thank Dr Juliette Delhove for expert article editing and 4DMedical for providing the XV LVAS output shown in . The authors acknowledge the facilities and scientific and technical assistance of the National Imaging Faculty, a National Collaboration Research Infrastructure Strategy (NCRIS) capability, at the Preclinical, Imaging and Research Laboratories of the South Australian Health and Medical Research Institute.

Author Contributions

D.P. and M.D. wrote the article.

Author Disclosure

DP has received renumeration for research advice from 4DMedical. DP and MD have both purchased shares in 4DMedical Limited.

Funding Information

Authors supported in part by the NHMRC (GNT1160011), the Cystic Fibrosis Foundation (PARSON18GO), and an MRFF Frontier Health and Medical Research Grant.

References

Prakash

, Moore

, Yanez-Munoz

. Current progress in therapeutic gene editing for monogenic diseases. Mol Ther, 2016; 24:465–474.

Yao

, Shen

. Monogenic diseases in respiratory medicine: clinical perspectives. Pediatr Invest, 2017; 1:27–31.

Albelda

, Wiewrodt

, Zuckerman

. Gene therapy for lung disease: hype or hope?. Ann Intern Med, 2000; 132:649–660.

About Cystic

Fibrosis

. Bethesda

: Cystic Fibrosis

Foundation

, 2020. https://www.cff.org/What-is-CF/About-Cystic-Fibrosis/ (last accessed August 17, 2020 ).

Understanding Changes in Life

Expectancy

. Bethesda

: Cystic Fibrosis

Foundation

, 2020. https://www.cff.org/Research/Researcher-Resources/Patient-Registry/Understanding-Changes-in-Life-Expectancy/ (last accessed August 17, 2020 ).

Pranke

, Golec

, Hinzpeter

, et al. Emerging therapeutic approaches for cystic fibrosis. From gene editing to personalized medicine. Front Pharmacol, 2019; 10:121.

McPhail

, Clancy

. Ivacaftor: the first therapy acting on the primary cause of cystic fibrosis. Drugs Today (Barc), 2013; 49:253–260.

Kuk

, Taylor-Cousar

. Lumacaftor and ivacaftor in the management of patients with cystic fibrosis: current evidence and future prospects. Ther Adv Respir Dis, 2015; 9:313–326.

Taylor-Cousar

, Munck

, McKone

, et al. Tezacaftor–Ivacaftor in patients with cystic fibrosis homozygous for Phe508del. N Engl J Med, 2017; 377:2013–2023.

10.

Middleton

, Mall

, Dřevínek

, et al. Elexacaftor–Tezacaftor–Ivacaftor for cystic fibrosis with a single Phe508del Allele. N Engl J Med, 2019; 381:1809–1819.

11.

Riordan

, Rommens

, Kerem

, et al. Identification of the cystic fibrosis gene: cloning and characterization of complementary DNA. Science, 1989; 245:1066–1073.

12.

Drumm

, Pope

, Cliff

, et al. Correction of the cystic fibrosis defect in vitro by retrovirus-mediated gene transfer. Cell, 1990; 62:1227–1233.

13.

Flotte

, Solow

, Owens

, et al. Gene expression from adeno-associated virus vectors in airway epithelial cells. Am J Respir Cell Mol Biol, 1992; 7:349–356.

14.

Yoshimura

, Rosenfeld

, Nakamura

, et al. Expression of the human cystic fibrosis transmembrane conductance regulator gene in the mouse lung after in vivo intratracheal plasmid-mediated gene transfer. Nucleic Acids Res, 1992; 20:3233–3240.

15.

Zabner

, Couture

, Gregory

, et al. Adenovirus-mediated gene transfer transiently corrects the chloride transport defect in nasal epithelia of patients with cystic fibrosis. Cell, 1993; 75:207–216.

16.

Farmen

, Karp

, Ng

, et al. Gene transfer of CFTR to airway epithelia: low levels of expression are sufficient to correct Cl- transport and overexpression can generate basolateral CFTR. Am J Physiol Lung Cell Mol Physiol, 2005; 289:L1123–L1130.

17.

Zhang

, Button

, Gabriel

, et al. CFTR delivery to 25% of surface epithelial cells restores normal rates of mucus transport to human cystic fibrosis airway epithelium. PLoS Biol, 2009; 7:e1000155.

18.

Cooney

, McCray

Jr , Sinn

. Cystic fibrosis gene therapy: looking back, looking forward. Genes (Basel), 2018; 9:538.

19.

Gupta

, Gaston

. From genetics to precision therapy: finding a path through the scientific valley of death. Am J Respir Cell Mol Biol, 2019; 61:671–672.

20.

Donnelley

, Parsons

. Gene therapy for cystic fibrosis lung disease: overcoming the barriers to translation to the clinic. Front Pharmacol, 2018; 9:1381.

21.

Daley

. Gene therapy arrives. Sci Am, 2020; 322:1.

22.

Russell

, Bennett

, Wellman

, et al. Efficacy and safety of voretigene neparvovec (AAV2-hRPE65v2) in patients with RPE65-mediated inherited retinal dystrophy: a randomised, controlled, open-label, phase 3 trial. Lancet, 2017; 390:849–860.

23.

Morrison

Fresh from the biotech pipeline-2017. Nat Biotechnol 2018. [Epub ahead of print]. DOI: 10.1038/nbt.4068

24.

Alton

, Armstrong

, Ashby

, et al. Repeated nebulisation of non-viral CFTR gene therapy in patients with cystic fibrosis: a randomised, double-blind, placebo-controlled, phase 2b trial. Lancet Respir Med, 2015; 3:684–691.

25.

Guggino

, Cebotaru

. Gene therapy for cystic fibrosis paved the way for the use of adeno-associated virus in gene therapy. Hum Gene Ther, 2020; 31:538–541.

26.

Yan

, Feng

, Sun

, et al. Human Bocavirus Type-1 capsid facilitates the transduction of Ferret airways by adeno-associated virus genomes. Hum Gene Ther, 2017; 28:612–625.

27.

Cmielewski

, Donnelley

, Parsons

. Long-term therapeutic and reporter gene expression in lentiviral vector treated cystic fibrosis mice. J Gene Med, 2014; 16:291–299.

28.

Cooney

, Abou Alaiwa

, Shah

, et al. Lentiviral-mediated phenotypic correction of cystic fibrosis pigs. JCI Insight, 2016; 1:e88730.

29.

Cao

, Ouyang

, Grasemann

, et al. Transducing airway basal cells with a helper-dependent adenoviral vector for lung gene therapy. Hum Gene Ther, 2018; 29:643–652.

30.

Cooney

, Singh

, Loza

, et al. Widespread airway distribution and short-term phenotypic correction of cystic fibrosis pigs following aerosol delivery of piggyBac/adenovirus. Nucleic Acids Res, 2018; 46:9591–9600.

31.

Velino

, Carella

, Adamiano

, et al. Nanomedicine approaches for the pulmonary treatment of cystic fibrosis. Front Bioeng Biotechnol, 2019; 7:406.

32.

Zhou

, Yang

, Cao

, et al. In Vitro Validation of a CRISPR-Mediated CFTR correction strategy for preclinical translation in pigs. Hum Gene Ther, 2019; 30:1101–1116.

33.

Moss

. Pitfalls of drug development: lessons learned from trials of denufosol in cystic fibrosis. J Pediatr, 2013; 162:676–680.

34.

Quan

, Tiddens

, Sy

, et al. A two-year randomized, placebo-controlled trial of dornase alfa in young patients with cystic fibrosis with mild lung function abnormalities. J Pediatr, 2001; 139:813–820.

35.

Murphy

, Anbar

, Lester

, et al. Treatment with tobramycin solution for inhalation reduces hospitalizations in young CF subjects with mild lung disease. Pediatr Pulmonol, 2004; 38:314–320.

36.

Elkins

, Robinson

, Rose

, et al. A controlled trial of long-term inhaled hypertonic saline in patients with cystic fibrosis. N Engl J Med, 2006; 354:229–240.

37.

Saiman

, Anstead

, Mayer-Hamblett

, et al. Effect of azithromycin on pulmonary function in patients with cystic fibrosis uninfected with Pseudomonas aeruginosa: a randomized controlled trial. JAMA, 2010; 303:1707–1715.

38.

Tang

, Yan

, Engelhardt

. Viral vectors, animal models, and cellular targets for gene therapy of cystic fibrosis lung disease. Hum Gene Ther, 2020; 31:524–537.

39.

Alton

EWFW

, Boyd

, Davies

, et al. Genetic medicines for CF: hype versus reality. Pediatr Pulmonol, 2016; 51:S5–S17.

40.

Breuer

, Caudri

, Stick

, et al. Predicting disease progression in cystic fibrosis. Expert Rev Respir Med, 2018; 12:905–917.

41.

Fretzayas

, Douros

, Moustaki

, et al. Applications of lung clearance index in monitoring children with cystic fibrosis. World J Clin Pediatr, 2019; 8:15–22.

42.

Amin

, Subbarao

, Jabar

, et al. Hypertonic saline improves the LCI in paediatric patients with CF with normal lung function. Thorax, 2010; 65:379–383.

43.

Amin

, Subbarao

, Lou

, et al. The effect of dornase alfa on ventilation inhomogeneity in patients with cystic fibrosis. Eur Respir J, 2011; 37:806–812.

44.

Milla

, Ratjen

, Marigowda

, et al. Lumacaftor/ivacaftor in patients aged 6–11 years with cystic fibrosis and homozygous for F508del-CFTR. Am J Respir Crit Care Med, 2017; 195:912–920.

45.

Ratjen

, Hug

, Marigowda

, et al. Efficacy and safety of lumacaftor and ivacaftor in patients aged 6–11 years with cystic fibrosis homozygous for F508del-CFTR: a randomised, placebo-controlled phase 3 trial. Lancet Respir Med, 2017; 5:557–567.

46.

Poncin

, Lebecque

. [Lung clearance index in cystic fibrosis]. Rev Mal Respir, 2019; 36:377–395.

47.

Dharmakumara

, Prisk

, Royce

, et al. The effect of gas exchange on multiple-breath nitrogen washout measures of ventilation inhomogeneity in the mouse. J Appl Physiol (1985), 2014; 117:1049–1054.

48.

Zufferey

, Dull

, Mandel

, et al. Self-inactivating lentivirus vector for safe and efficient in vivo gene delivery. J Virol, 1998; 72:9873–9880.

49.

Griesenbach

, Pytel

, Alton

EWFW

. Cystic fibrosis gene therapy in the UK and elsewhere. Hum Gene Ther, 2015; 26:266–275.

50.

VanDevanter

, Mayer-Hamblett

. Innovating cystic fibrosis clinical trial designs in an era of successful standard of care therapies. Curr Opin Pulmon Med, 2017; 23:530–535.

51.

Donnelley

, Morgan

, Awadalla

, et al. High-resolution mucociliary transport measurement in live excised large animal trachea using synchrotron X-ray imaging. Respir Res, 2017; 18:95.

52.

Donnelley

, Morgan

, Siu

, et al. Tracking extended mucociliary transport activity of individual deposited particles: longitudinal synchrotron X-ray imaging in live mice. J Synchrotron Radiat, 2014; 21:768–773.

53.

Morgan

, Donnelley

, Farrow

, et al. In vivo X-ray imaging reveals improved airway surface hydration after a therapy designed for cystic fibrosis. Am J Respir Crit Care Med, 2014; 190:469–471.

54.

Horsfield

. The relation between structure and function in the airways of the lung. Br J Dis Chest, 1974; 68:145–160.

55.

Ward

, Carroll

, Cunningham

, et al. Radiation dose from common radiological investigations and cumulative exposure in children with cystic fibrosis: an observational study from a single UK centre. BMJ Open, 2017; 7:e017548.

56.

de Jong

, Nakano

, Lequin

, et al. Progressive damage on high resolution computed tomography despite stable lung function in cystic fibrosis. Eur Respir J, 2004; 23:93–97.

57.

Bhalla

, Turcios

, Aponte

, et al. Cystic fibrosis: scoring system with thin-section CT. Radiology, 1991; 179:783–788.

58.

Brody

, Klein

, Molina

, et al. High-resolution computed tomography in young patients with cystic fibrosis: distribution of abnormalities and correlation with pulmonary function tests. J Pediatr, 2004; 145:32–38.

59.

Calder

, Bush

, Brody

, et al. Scoring of chest CT in children with cystic fibrosis: state of the art. Pediatr Radiol, 2014; 44:1496–1506.

60.

Loeve

, van Hal

, Robinson

, et al. The spectrum of structural abnormalities on CT scans from patients with CF with severe advanced lung disease. Thorax, 2009; 64:876–882.

61.

Rosenow

, Oudraad

, Murray

, et al. PRAGMA-CF. a quantitative structural lung disease computed tomography outcome in young children with cystic fibrosis. Am J Respir Crit Care Med, 2015; 191:1158–1165.

62.

Kuo

, Ciet

, Tiddens

, et al. Monitoring cystic fibrosis lung disease by computed tomography. Radiation risk in perspective. Am J Respir Crit Care Med, 2014; 189:1328–1336.

63.

Woods

, Wild

, Wielputz

, et al. Current state of the art MRI for the longitudinal assessment of cystic fibrosis. J Magn Reson Imaging 2019. [Epub ahead of print]; DOI:10.1002/jmri.27030

64.

Eichinger

, Optazaite

, Kopp-Schneider

, et al. Morphologic and functional scoring of cystic fibrosis lung disease using MRI. Eur J Radiol, 2012; 81:1321–1329.

65.

Schaefer

, Hector

, Schmidt

, et al. A semiquantitative MRI-Score can predict loss of lung function in patients with cystic fibrosis: preliminary results. Eur Radiol, 2018; 28:74–84.

66.

Dournes

, Menut

, Macey

, et al. Lung morphology assessment of cystic fibrosis using MRI with ultra-short echo time at submillimeter spatial resolution. Eur Radiol, 2016; 26:3811–3820.

67.

Roach

, Cremillieux

, Fleck

, et al. Ultrashort echo-time magnetic resonance imaging is a sensitive method for the evaluation of early cystic fibrosis lung disease. Ann Am Thorac Soc, 2016; 13:1923–1931.

68.

Kolodziej

, de Veer

, Cholewa

, et al. Lung function imaging methods in cystic fibrosis pulmonary disease. Respir Res, 2017; 18:96.

69.

Marshall

, Horsley

, Taylor

, et al. Detection of early subclinical lung disease in children with cystic fibrosis by lung ventilation imaging with hyperpolarised gas MRI. Thorax, 2017; 72:760–762.

70.

Altes

, Meyer

, Mata

, et al. Hyperpolarized helium-3 magnetic resonance lung imaging of non-sedated infants and young children: a proof-of-concept study. Clin Imaging, 2017; 45:105–110.

71.

Walker

, Happer

. Spin-exchange optical pumping of noble-gas nuclei. Rev Modern Phys, 1997; 69:629–642.

72.

Imai

, Matsumoto

, Miyakoshi

, et al. Regional fractional ventilation mapping in spontaneously breathing mice using hyperpolarized (1)(2)(9)Xe MRI. NMR Biomed, 2015; 28:24–29.

73.

Loza

, Kadlecek

, Pourfathi

, et al. Quantification of ventilation and gas uptake in free-breathing mice with hyperpolarized (129)Xe MRI. IEEE Trans Med Imaging, 2019; 38:2081–2091.

74.

Chen

, Xie

, Zhang

, et al. Lung respiration motion modeling: a sparse motion field presentation method using biplane x-ray images. Phys Med Biol, 2017; 62:7855–7873.

75.

Guerrero

, Zhang

, Huang

, et al. Intrathoracic tumour motion estimation from CT imaging using the 3D optical flow method. Phys Med Biol, 2004; 49:4147–4161.

76.

Zhang

, Huang

T-C

, Guerrero

, et al. Use of three-dimensional (3D) optical flow method in mapping 3D anatomic structure and tumor contours across four-dimensional computed tomography data. J Appl Clin Med Phys, 2008; 9:59–69.

77.

Perchiazzi

, Rylander

, Derosa

, et al. Regional distribution of lung compliance by image analysis of computed tomograms. Respir Physiol Neurobiol, 2014; 201:60–70.

78.

Kitchen

, Paganin

, Lewis

, et al. On the origin of speckle in x-ray phase contrast images of lung tissue. Phys Med Biol, 2004; 49:4335–4348.

79.

Mall

, Grubb

, Harkema

, et al. Increased airway epithelial Na+ absorption produces cystic fibrosis-like lung disease in mice. Nat Med, 2004; 10:487–493.

80.

Stahr

, Samarage

, Donnelley

, et al. Quantification of heterogeneity in lung disease with image-based pulmonary function testing. Sci Rep, 2016; 6:29438.

81.

Dubsky

, Jamison

, Irvine

, et al. Computed tomographic x-ray velocimetry. Appl Phys Lett 2010;96:2,023702.

82.

Dubsky

, Hooper

, Siu

, et al. Synchrotron-based dynamic computed tomography of tissue motion for regional lung function measurement. J R Soc Interface, 2012; 9:2213–2224.

83.

Murrie

, Morgan

, Maksimenko

, et al. Live small-animal X-ray lung velocimetry and lung micro-tomography at the Australian Synchrotron Imaging and Medical Beamline. J Synchrotron Radiat, 2015; 22:1049–1055.

84.

Murrie

, Werdiger

, Donnelley

, et al. Real-time in vivo imaging of regional lung function in a mouse model of cystic fibrosis on a laboratory X-ray source. Sci Rep, 2020; 10:447.

85.

Werdiger

, Morgan

, Donnelley

, et al. Laboratory-based regional lung function analysis via x-ray imaging, for the delineation of Cystic Fibrosis lung disease presentations in Beta-ENaC-Tg mice. Sci Rep 2020.

86.

Dubsky

, Thurgood

, Fouras

, et al. Cardiogenic airflow in the lung revealed using synchrotron-based dynamic lung imaging. Sci Rep, 2018; 8:4930.

87.

Preissner

, Murrie

, Pinar

, et al. High resolution propagation-based imaging system for in vivo dynamic computed tomography of lungs in small animals. Phys Med Biol, 2018; 63:08NT03.

88.

Yen

, Preissner

, Bennett

, et al. The link between regional tidal stretch and lung injury during mechanical ventilation. Am J Respir Cell Mol Biol, 2019; 60:569–577.

89.

Yen

, Preissner

, Bennett

, et al. Interaction between regional lung volumes and ventilator-induced lung injury in the normal and endotoxemic lung. Am J Physiol Lung Cell Mol Physiol, 2020; 318:L494–L499.

90.

Kroft

LJM

, van der Velden

, Giron

, et al. Added value of ultra-low-dose computed tomography, dose equivalent to chest X-Ray radiography, for diagnosing chest pathology. J Thorac Imaging, 2019; 34:179–186.

91.

Kaya

, Ufuk

, Karabulut

. Diagnostic performance of contrast-enhanced and unenhanced combined pulmonary artery MRI and magnetic resonance venography techniques in the diagnosis of venous thromboembolism. Br J Radiol, 2019; 92:20180695.

92.

Ohno

, Koyama

, Lee

, et al. Contrast-enhanced CT- and MRI-based perfusion assessment for pulmonary diseases: basics and clinical applications. Diagn Interv Radiol, 2016; 22:407–421.

93.

Preissner

, Murrie

, Bresee

, et al. Application of a novel in vivo imaging approach to measure pulmonary vascular responses in mice. Physiol Rep, 2018; 6:e13875.

94.

Samarage

, Carnibella

, Preissner

, et al. Technical Note: contrast free angiography of the pulmonary vasculature in live mice using a laboratory x-ray source. Med Phys, 2016; 43(11):6017–6023.

95.

Newbegin

, Pilkington

, Shanthikumar

, et al. Clinical utility of surveillance computed tomography scans in infants with cystic fibrosis. Pediatr Pulmonol, 2018; 53:1387–1390.