A randomised, double-blind, placebo-controlled trial of repeated nebulisation of non-viral cystic fibrosis transmembrane conductance regulator (CFTR) gene therapy in patients with cystic fibrosis

Pathophysiology

Cystic fibrosis (CF), a common, genetically inherited disease, is caused by mutations in the CF transmembrane conductance regulator (CFTR) gene. Inherited in an autosomal recessive fashion, it has a prevalence of approximately 1 : 2500 and a carrier rate of around 1 : 25 in the Caucasian population. It is estimated that there are approximately 70,000–100,000 people with CF worldwide and over 10,000 on the current UK registry. It is a multisystem disease, primarily affecting the lungs, but also with significant pancreatic, liver and gastrointestinal involvement and infertility in the majority of males. Lung involvement occurs from an early age with intermittent then chronic bacterial infection, inflammation and eventual bronchiectasis, fibrosis and death from respiratory failure. 1 People affected with CF are particularly susceptible to certain bacteria, including Staphylococcus aureus Rosenbach 1884, Pseudomonas aeruginosa (Schroeter 1872) Migula 1900 and Burkholderia cepacia (Palleroni and Holmes 1981) Yabuuchi et al. 1993, although more recently the use of molecular detection techniques has revealed much more diverse bacterial populations present in the lower airway than hitherto suspected. 2

The gene responsible for CF was identified in 1985 and found to be localised to the long arm of chromosome 7, position 7q21–24,3 with the sequence being fully identified in 1989. 4 Over 2000 mutations in the CFTR gene have been described, although not all are disease-causing; some mutations still produce a functional protein and are probably more accurately termed benign polymorphisms. In its normal, wild-type, form, the CFTR gene encodes a 1480-amino-acid protein5 that sits on the cell surface and acts as a cyclic adenosine monophosphate (cAMP)-regulated ion channel. CFTR is a direct conductor of chloride and (probably) bicarbonate ions and interacts closely with the major sodium-absorbing channel, epithelial sodium channel (ENaC), inhibiting its activity. Defects in CFTR lead to reduced chloride secretion and hyperabsorption of sodium and, thus, water, the combined effect of which is dehydration of the cell surface. 6 Mucociliary clearance is impaired and the infective/inflammatory sequelae of this ensues. It is not known if the decreased volume of airway surface liquid alone causes the predisposition to infection, if the immune system is also affected causing a defect in pathogen killing,7 or whether or not other environmental factors within the CF airway prevent the innate immune system functioning as efficiently as in individuals without CF.

Clinical presentation, diagnosis and disease manifestation

The majority of today’s older children and adults, the population included in this trial, will have been diagnosed once they had developed symptoms suggestive of the disease. In many countries, the UK included, this picture has changed dramatically with the introduction of newborn screening; currently, infants are diagnosed around 3–4 weeks of age based on dried-blood-spot screening. 8 The screens differ somewhat globally, but the UK test measures immunoreactive trypsinogen (IRT) and pancreatic enzyme levels, which are increased in CF because of pancreatic duct obstruction. Samples with high levels of IRT are sent for mutation testing. This allows identification and initiation of treatment presymptomatically and before the downstream sequelae have arisen.

The gold standard diagnosis is based on measuring chloride ion levels in sweat samples. Lack of chloride ion reabsorption by CFTR in the sweat gland leads to high levels being lost in the sweat, typically > 100 mmol/l in comparison with values of < 30 mmol/l in healthy individuals. 9 Diagnosis can also be established on CFTR mutation testing, although, because of the large number of mutations, a negative test cannot rule out the diagnosis unless the entire CFTR gene has been sequenced; this is rarely employed as a first-line diagnostic tool and is reserved for difficult diagnoses. However, whereas previously it may have been sufficient to secure a diagnosis on the basis of a sweat test, with the recent development of mutation-specific small-molecule treatments,10 it is becoming increasingly necessary for a patient’s mutation(s) to be defined.

Disease presentation may be respiratory or nutritional and is often both. 1 Respiratory problems present as cough, wheeze, lower respiratory tract infections, sputum production and upper airway disease including nasal polyps and sinusitis. Lung health deteriorates over time, largely driven by infection and the host inflammatory response. Disease is thought to begin in the small airways, although this may just reflect a greater impact of obstruction at this site. Airway obstruction related to mucus accumulation and airway wall inflammation is measurable on physiological testing, the most commonly employed being forced expiratory manoeuvres with spirometry. Over time, and despite optimal treatment at specialist centres, lung function declines. A major contributor to this decline is episodes of pulmonary exacerbation, periods of increased symptoms and acute drop in lung function that is often not fully regained after treatment. 11

The most common manifestation of the pancreatic exocrine disease, present in > 95% of patients, is poor weight gain with steatorrhoea. Later, patients may develop pancreatic endocrine dysfunction with impaired glucose metabolism and CF-related diabetes mellitus; this has respiratory consequences that are poorly understood, but is clearly associated with an increased rate of progression of lung disease.

Conventional management of patients with cystic fibrosis

Cystic fibrosis is a multisystem disease and treatment can be optimally conducted only with the help of a full multidisciplinary team (CF physician, specialist nurse, physiotherapist, dietitian, clinical psychologist, pharmacist) and input from ancillary specialists with expert knowledge of CF (ear, nose and throat surgeon, obstetrician, endocrinologist). CF patients should be seen at least every 3 months by the core CF team. A large number of treatment guidelines have been published. 12–14

The main aims of respiratory care are clearance of mucus, prevention and early eradication of infection and suppression of infection once chronic. Sputum clearance is achieved with regular physiotherapy alongside adjunctive inhaled mucoactive agents: hypertonic saline has been shown to aid clearance,15 most likely be rehydrating the airway surface and stimulating cough. Recombinant human deoxyribonuclease (DNase) (dornase alfa; Pulmozyme^®, Roche Products Ltd) breaks down the high levels of deoxyribonucleic acid (DNA) in the CF airway resulting from degradation of neutrophils and decreases mucus viscosity. 16 Regular microbiological surveillance is essential for the early detection of infecting bacteria; sputum samples or cough swabs are collected at every patient contact. P. aeruginosa is an extremely common pathogen, often being acquired first in childhood and becoming chronic in adulthood in around 60–70% of patients. Its presence is associated with a more rapid decline in lung function and, therefore, aggressive eradication is employed when it is first encountered. 17 Once it becomes chronic, long-term nebulised antibiotics such as colistimethate sodium (Colomycin^®, Forest Laboratories UK Ltd) or the aminoglycoside, tobramycin for inhaled solution (Tobi^®, Novartis Pharmaceuticals UK Ltd), are administered to maintain suppression of bacterial numbers and thereby attempt to limit the host inflammatory response. 18 There are a small number of other organisms that pose particular problems, either because of high transmissibility [e.g. meticillin-resistant Staphylococcus aureus (MRSA) and B. cepacia], or because they are related to a more rapid rate of clinical decline [e.g. Mycobacterium abscessus (Moore and Frerichs 1953) Kusunoki and Ezaki 1992]. These organisms are difficult to treat and, furthermore, such patients often miss out on inclusion in clinical trials of new drugs, either because of infection control issues or concerns over baseline stability. Periods of pulmonary exacerbation are treated with increased attention to airway clearance and antibiotics, often given intravenously and requiring hospital admission.

Therefore, CF management not only poses a huge burden on the patients and their families, but also carries with it substantial health-care system costs. 19 Ultimately, once respiratory failure has ensued, lung transplantation is the only option. There is a shortage of organs in most developed countries and many patients die while on the waiting list. Even after receiving a transplant, complications frequently occur, most importantly the development of bronchiolitis obliterans, although some patients may do well for many years. 20 Currently in the UK, median age at death is around 29 years of age. 21 Thus, despite expensive and burdensome treatments, life expectancy is significantly reduced for CF patients. Progress is being made in several areas of drug development, for example with newer antibiotics, but many feel that a significant improvement in health is likely to be achievable only with novel strategies to tackle the basic defect in the CFTR gene rather than these downstream consequences. The next sections describe the progress that has been made in small-molecule and gene therapy fields to this end.

Novel small-molecule cystic fibrosis transmembrane conductance regulator modulators

An understanding of the various mechanisms by which different mutations lead to CFTR dysfunction has led to the grouping of mutations into classes. 22

Class I mutations lead to a premature ‘stop’ in the messenger ribonucleic acid (mRNA) and a short, non-functioning protein. Mutations of class II encode a structurally abnormal, misfolded protein that does not traffic to its site of action on the apical cell surface but is removed by the endoplasmic reticulum and degraded. Class III–VI mutations reach the cell surface but fail to function appropriately: class III mutations have decreased activation of the channel and remain closed; and class IV mutations cause decreased conductance of ions across the channel. Class V mutations encode proteins with splice-site mutations and result in reduced amounts of CFTR at the cell surface, so some function occurs but at a reduced level. Class VI mutations lead to a shortened half-life because of protein instability and may also impair CFTR’s regulation of neighbouring channels. Many mutations lead to defects in more than one of these classes; for example, F508del (previously ΔF508), the most common mutation worldwide, is clearly a class II misfolding defect, but also opens infrequently and has a short half-life, so also possesses class III and VI features.

Choice of vector and clinical trials

Clinical trials of CFTR gene transfer to CF subjects have been undertaken using a variety of viral vector systems, including adenovirus and adeno-associated virus (AAV). Encouragingly, adenovirus-based gene transfer led to the correction of the cAMP-stimulated CFTR chloride channel defect in the nose of treated CF subjects. 39,40 However, the efficiency of adenovirus-based gene transfer to human tissues appears to be low,41,42 and is associated with lung inflammation43,44 and the development of neutralising immune responses that, to date, has precluded repeated administration. 45 AAV-based CFTR gene transfer appears less proinflammatory,46–48 although repeated administration is also inhibited by immunological reactivity. 49

Non-viral vectors have also been used in clinical trials of CFTR gene transfer to CF subjects. Ten clinical trials have been performed worldwide to date,50–58 six of which have been in the UK [Gene Therapy Advisory Committee (GTAC) numbers: 002, 007, 008, 009, 015 and 140] and conducted by current groups within the UK CF Gene Therapy Consortium (UK CFGTC). The first of these non-viral-formulation clinical trials assessed the safety and efficacy of a single nasal delivery of plasmid DNA (pDNA) containing the CFTR complementary DNA (cDNA) under the transcriptional control of the simian virus 40 (SV40) promoter complexed with 3beta-[N-(N′,N′-dimethylaminoethane)carbamoyl]cholesterol (DC-Chol)–dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE) liposomes (GTAC number: 002). 50 Evidence for pDNA transfer, vector-derived CFTR mRNA expression and partial correction of the CFTR chloride ion channel defect was obtained in a subset of subjects. Similar modest, but positive evidence of CFTR gene transfer and function were obtained after nasal delivery of a pDNA containing the CFTR cDNA under the transcriptional control of the respiratory syncytial virus 3′ long terminal repeat (RSV 3’LTR) promoter complexed with DC-Chol–DOPE liposomes (GTAC number: 007). 51 Alternative formulations in which pDNAs containing the CFTR cDNA under the transcriptional control of the human cytomegalovirus (CMV) immediate early enhancer/promoter were complexed with 1,2-dioleoyl-3-trimethylammoniumpropane liposomes (GTAC number: 008),52 p-ethyl-dimyristoylphosphaditylcholine–cholesterol liposomes,53 or a polycationic peptide consisting of a N-terminal cysteine followed by 30 lysine residues54 were broadly shown to be similarly effective. Importantly, in an additional clinical trial utilising the RSV 3′LTR promoter plasmid complexed with DC-Chol–DOPE liposomes, repeated nasal delivery of a non-viral formulation was shown to be equivalently effective after each of three successive administrations (GTAC number: 015). 55 Crucially, no important safety considerations were raised after nasal application of any of these formulations. Collectively, these clinical data provided proof of principle for CF non-viral gene therapy, but highlighted the need for formulations with enhanced efficacy. Extensive chemical optimisation of the promising DC-Chol formulation by Genzyme Inc. (Cambridge, MA, USA) generated the cationic lipid cholest-5-en-3-ol (3β)-,3-[(3-aminopropyl)[4-[(3-aminopropyl)amino]butyl]carbamate] (GL67) with improved gene transfer potency,59 well-characterised safety parameters60 and desirable stability during aerosolisation. 61 Clinical trials with GL67-based non-viral formulations have been encouraging. A single nasal administration of a pDNA containing the CFTR cDNA under the transcriptional control of the CMV promoter (plasmid pCF1-CFTR) complexed with GL67–DOPE was shown to be safe, to direct vector-derived mRNA expression and to produce an overall ≈ 20% correction of the CFTR chloride ion channel defect. 56 For efficient jet nebuliser-directed aerosol delivery, 1,2-dimyristoyl-sn-glycero-3-phosphoethanolamine-n-[methoxy(polyethylene glycol 5000)] (ammonium salt) (DMPE-PEG5000) was added to the GL67–DOPE to generate a formulation termed GL67A that consisted of a mixture of GL67–DOPE–DMPE-PEG5000 at a 1 : 2 : 0.05 molar ratio61 (for further details see GL67A cationic lipid). The GL67A formulation lacking pDNA was shown to be safe and well tolerated after a single lung administration in healthy volunteers. 62 Delivery of pCF1-CFTR complexed with GL67A to the lungs57 or nose and lungs (GTAC number: 009)58 of CF subjects led to an overall ≈ 25% correction of the CFTR chloride ion channel defect. However, CFTR expression in these subjects, and those in all other clinical trials described above, diminished quickly, such that it was essentially undetectable within 1 week of administration. Furthermore, pCF1-CFTR/GL67A lung administration was associated with a mild flu-like syndrome that resolved within 36 hours. 57,58

Further in vivo preclinical studies have identified cytosine–phosphate–guanidine (CpG) dinucleotides in the pCF1-CFTR pDNA as the likely cause of the mild flu-like symptoms following dosing and may also impact on the duration of pDNA-mediated transgene expression. The CpG motifs delivered by the pCF1-CFTR/GL67A formulation appear to be sensed by the innate immune system via the toll-like receptor 9 (TLR9) pathway. 63 Stimulation of the TLR9 pathway led to a host inflammatory reaction typified by the elevation of inflammatory cytokines within the lung milieu and recruitment of lung neutrophils. 64

Thus, proof of principle for non-viral CFTR gene therapy had been demonstrated, but improvements were needed specifically with regard to duration of expression and pro-inflammatory responses.

UK Cystic Fibrosis Gene Therapy Consortium

At this stage, in 2001, and in response to an initiative launched by the UK CF Trust, the three UK sites active in CF gene therapy (University of Edinburgh, Imperial College London/Royal Brompton Hospital and University of Oxford) came together to form the UK CF Gene Therapy Consortium (www.cfgenetherapy.org.uk/). The aim of the consortium is to develop clinically relevant gene therapy for patients with CF. The wave 1 programme was based on the recognition that repeated application would necessitate a non-viral vector and aimed to identify the best, currently available non-viral formulation coupled with a long-lasting plasmid with little or no proinflammatory potential and to take this into a clinical trial designed and powered, for the first time, to detect clinical benefit. The repeated-dose trial is the focus of this report. Work packages in the wave 1 programme leading up to this point have included a study of outcome measures on patients undergoing a pulmonary exacerbation;65 a large, longitudinal study of outcome measures in stable patients (from which power calculations were made for this trial); and a single-application safety and dose-ranging pilot study (see Choice of dose and adjunctive treatments: single-dose pilot study). The consortium is also working on a wave 2 programme developing and assessing a pseudotyped lentiviral vector, which seems to be repeatedly administrable. Although gene transfer efficiency seems greater with this approach than with our non-viral equivalents, the requirement for preclinical development and toxicity testing placed this further in the future and led to the initial focus on wave 1.

Section pGM1/GL67A for clinical trial use describes our chosen clinical trial formulation in detail and the rationale for its development.

Composition

We found GL67A (further details in GL67A cationic lipid) to be superior to others in both small and large animal models, having undertaken an extensive preclinical assessment of available non-viral formulations. GL67A has previously been administered to both healthy volunteers and CF subjects, including in a single-dose nebuliser study66 and our recently completed safety study Evaluation of safety and gene expression with a single dose of pGM169/GL67A administered to the nose and lung of individuals with cystic fibrosis (sponsor’s protocol number: cro851; EudraCT reference: 2007-004050-85). 62 For both the single-dose safety study and the trial described in this report, we made substantial changes to the CFTR plasmid, which was specifically designed with the following features:

• The promoters most commonly used are of viral origin; we had previously incorporated a CMV promoter which, although capable of high-level gene expression, is usually of short duration. The current formulation contains a human elongation factor 1 alpha (EEF1A1) gene promoter. Our preclinical work demonstrated more sustained levels of expression (up to 2 months) with this new promoter; this could translate into a less frequent dosing regimen. The gene expression profile observed in the recent single-dose study supported this hypothesis, with relatively few changes observed at early time points and several subjects demonstrating functional changes in nasal PD several months after dosing.

• Unmethylated CpG dinucleotides are present in high numbers in pDNA used in gene therapy trials. These bacterially derived components are recognised by humans as foreign and likely responsible for the flu-like responses, reported in our previous trial in the group receiving DNA/lipid but not lipid alone. Although these responses were mild and self-limiting, requiring only treatments with antipyretics, we considered them undesirable in a repeated-dose trial. In an attempt to reduce such an inflammatory response, the new plasmid has been rendered completely CpG free.

Plasmid

pGM169 is a covalently closed, circular, double-stranded pDNA molecule of 6549 base pairs purified from bacteria. It is based on a novel CpG-free plasmid backbone described in the international patent application PCT/GB2007/00110433. A diagrammatic representation of pGM169 is presented in Figure 1. pGM169 contains a CpG-free version of the CFTR coding sequence termed soCFTR2 under the transcriptional control of a novel CpG-free human CMV enhancer/elongation factor 1 alpha (hCEFI) enhancer/promoter. Other plasmid elements include a CpG-free version of the bovine growth hormone polyadenylation sequence, a CpG-free version of the R6K bacterial plasmid origin of replication and a CpG-free version of the kanamycin resistance gene under the transcriptional control of the CpG-free synthetic bacterial promoter sequence termed EM7.

FIGURE 1.

Plasmid pGM169. The basic features of pGM169 (proceeding clockwise from 0 base pairs) are the CpG-free human cytomegalovirus enhancer/elongation factor 1 alpha enhancer/promoter; a CpG-free synthetic intron sequence to enhance mRNA splicing; a CpG-free version of the CFTR coding sequence termed soCFTR2; a CpG-free version of the bovine growth hormone polyadenylation sequence; a CpG-free version of the R6K bacterial plasmid origin of replication; a CpG-free version of the kanamycin resistance gene; and a CpG-free synthetic bacterial promoter sequence termed EM7. bp, base pair; BGH, bovine growth hormone.

Good manufacturing practice (GMP) manufacture of pGM169 was conducted by VGXi Inc. (The Woodlands, TX, USA). Bacteria containing the plasmid were fermented to a high density and harvested. The bacteria were then lysed to release their contents, including the plasmid, into solution. The lysate was subjected to three significant purification steps: (1) solid/liquid separation, (2) ion-exchange chromatography and (3) hydrophobic interaction chromatography. Subsequently, the purified plasmid was concentrated and desalted by ultrafiltration/diafiltration into a sterile 8 mM sodium chloride (NaCl) solution and finally subjected to aseptic filtration to provide the bulk drug substance. This bulk was aseptically filled into single-unit vials and stored at ≤ –70 °C. To prepare the final drug substance, single or multiple pooled lots of bulk drug substance were, if necessary, diluted to 5.3 ± 0.3 mg/ml with sterile 8 mM NaCl and then filled into 10-ml clear glass vials at a fill level of 5.2 ± 0.2 ml. Vials were stored at –80 °C. The material is stable for at least 3 years.

GL67A cationic lipid

The cationic lipid mixture GL67A is an excipient, consisting of a mixture of three components (the structure of which is shown in Figures 2–4; GL67, DOPE, and DMPE-PEG5000) formulated at a 1 : 2 : 0.05 molar ratio.

FIGURE 2.

Structure of GL67. GL67 is included in the GL67A cationic lipid mixture for its DNA-binding properties (i.e. binding of pGM169).

FIGURE 3.

Structure of DOPE. DOPE is included in the GL67A cationic lipid mixture for its endocytosis-inducing properties.

FIGURE 4.

Structure of DMPE-PEG5000. DMPE-PEG5000 is included in the GL67A cationic lipid mixture for its charge-shielding properties that facilitate final preparation of the pGM169/GL67A drug product and allow efficient nebulisation.

The final formulation containing GL67, DOPE and DMPE-PEG5000 at a 1 : 2 : 0.05 molar ratio and a GL67-to-DNA ratio of 0.75 : 1 was termed GL67A.

Good manufacturing practice-grade GL67 was manufactured by Sanofi-Genzyme (Haverhill, UK). GMP-grade DOPE and DMPE-PEG5000 were purchased from Avanti Polar Lipids (Alabaster, AL, USA). GL67A was formulated from GMP-grade GL67, DOPE and DMPE-PEG5000 by OctoPlus N.V. (Leiden, Netherlands). Briefly, the individual constituents were first dissolved in 2-methylpropan-2-ol (9 : 1 weight-to-weight ratio of 2-methylpropan-2-ol to water) and then mixed in appropriate quantities to obtain a GL67–DOPE–DMPE-PEG5000 molar ratio of 1 : 2 : 0.05. After sterile filtration the lipid mixture was dispensed into individual 10-ml glass lyophilisation vials such that each vial contained 38.8 mg of GL67, 93.7 mg of DOPE and 17.8 g of DMPE-PEG5000. The vials were freeze-dried under nitrogen gas for approximately 94 hours at temperatures ranging from –50 °C to +10 °C. The vials were capped with aluminium crimp caps, coded and stored at –80 °C. The material is stable for at least 3 years.

Preclinical toxicology of the selected formulation: murine

To comply with the Medicines and Healthcare products Regulatory Agency (MHRA) regulations, a multidose toxicology study in rodents (mice) was performed under good laboratory practice and the study was outsourced to a clinical research organisation (CRO).

Additional research questions

1. Will markers of molecular efficacy (mRNA and PD) measured in the lung correlate with changes in clinical end points?

2. Will markers of molecular efficacy measured in the nose correlate with these markers measured in the lungs in the same patients?

3. Will molecular efficacy increase with repeated administration when measured serially in the same patient?

4. Will it be possible to identify the phenotype of responders or non-responders to allow stratification of patients for future gene therapy trials?

5. Will it be possible to identify stratifiable biomarkers from the secondary outcomes, based on disease severity?

6. Will one (or more) of the secondary outcome measures provide different or better information than FEV₁, allowing it to be proposed as a novel registrable biomarker for future trials?

7. Will the extensive preclinical molecular efficacy data produced in support of this study correlate with similar data in humans?

8. Will the extensive two species toxicology package produced in support of this study correlate with the findings in humans?

Major secondary outcomes

These include:

• change in other spirometric values [forced vital capacity (FVC), mid-expiratory flow 25–75% (MEF_25–75%)]

• lung clearance index (LCI)

• structural parameters on computed tomography (CT) scans of the chest

• a validated quality-of-life score (as measured by the Cystic Fibrosis Questionnaire – Revised; CFQ-R). 71

Other secondary outcomes

Other measures of lung physiology including:

• exercise capacity [maximal oxygen uptake (VO_2max)] and activity monitoring

• gas transfer

• Inflammatory markers:

  • blood – white blood cell count and differential count, C-reactive protein (CRP), serum calprotectin

  • sputum – soluble inflammatory markers, cellular inflammation, 24-hour weight, solid content, DNA content, microbiology.

Adverse events (AEs) and other safety measures including:

• blood biochemistry

• haematology

• liver and renal function

• urinary markers

• histology

• fat-laden airway cells.

Mechanistic outcomes

• Transgene-specific DNA and mRNA on bronchial and nasal brushings.

• Nasal and lower airway PD measurements (CFTR protein function).

Proposed sample and effect sizes

The sample size derivation was based on outcomes of previous clinical trials in CF and available data on what is widely considered the minimum clinically meaningful difference. Over the last two decades, a relatively small number of new drugs have been licensed for the treatment of patients with CF and adopted internationally as standards of care. Those of particular note include the mucolytic agent, recombinant human DNase [rhDNase; dornase alpha (Pulmozyme^®, Roche)]; tobramycin for the inhaled solution; and the macrolide, azithromycin.

1. Recombinant DNase acts by breaking down neutrophil-derived DNA, which contributes significantly to mucus viscosity. In the most widely cited double-blind, placebo-controlled trial, the drug led to a relative improvement in FEV₁ of 5.8%; this was accompanied by a significant reduction in the frequency of infective exacerbations. 72 This agent is now a licensed product in routine clinical use worldwide.

2. Tobramycin for inhalation solution (TOBI) was designed as an antipseudomonal antibiotic to be administered on alternate months, in contrast to other agents that are administered continuously. At the end of a 6-month trial period, the relative improvement in FEV₁ was around 7%. 73 This agent is now considered the gold standard by the regulatory agencies both in Europe and in the USA, to the extent that all other inhaled antibiotics are required to undergo head-to-head testing with it.

3. Azithromycin is an orally bioavailable macrolide that was first considered as a useful therapy in CF after significant success in Japanese panbronchiolitis sufferers. Its mechanism of action is unknown, but is thought likely to relate to its known anti-inflammatory actions. Several trials have been conducted in CF, the largest of which reported a 6.2% TE on FEV₁ and a significant reduction in the frequency of exacerbations. 74 Azithromycin is now in widespread clinical use for CF.

Our decision to power for a relative change of 6% in FEV₁ was made, in part based on these data, and in part on pragmatic considerations of patient recruitment feasibility and cost of materials. We suggest that the current consensus supports this change as being a clinically meaningful improvement. Based on run-in study data [longitudinal assessment of clinical measurements in patients with cystic fibrosis in preparation for a clinical trial of CFTR gene therapy (unpublished data, Gene Therapy Consortium)], and using the mean of two measurements at baseline and the mean of two measurements at 12 months, we estimate the standard deviation (SD) of the percentage change in percentage predicted FEV₁ over 12 months to be 10.0%. The use of duplicate measurements reduces variability substantially, with resultant increases in power. The corresponding SD using only single measurements is 12.2%. We also demonstrated, using data from the run-in study, that analysis of percentage predicted FEV₁ is more sensitive than analysis of absolute FEV₁; the corresponding SD for percentage change from baseline based on duplicate measurements was 11.6% for absolute FEV₁. With the SD of 10.0%, a total sample of 120 evaluable patients would provide 90% power at the 5% significance level (2-sided) to detect a difference of 6% between the randomised groups in the mean change from baseline. With a plan to recruit 130 patients, we would allow a safety margin for subjects leaving the study prior to completion. This number also allows us to be powered at ≥ 80% to detect changes in the secondary outcomes LCI and CT scan parameters, which were smaller than we had previously observed in a study of intravenous antibiotics. 65 The above power calculation was conservative in that covariate adjustment can be anticipated to increase the statistical power.

Statistical analysis

We reviewed the approach taken to the analysis of FEV₁ in over 40 published randomised trials of interventions in CF. There is no consistent approach, with roughly half of the trials using absolute changes from baseline (for either absolute FEV₁, measured in litres, or percentage predicted FEV₁) and the other half using relative changes (i.e. percentage change from baseline in terms of either absolute FEV₁ or percentage predicted FEV₁). Analysis of absolute changes is typically based on an analysis of covariance (ANCOVA) with baseline FEV₁ as a covariate, and in some instances relative changes were also analysed with an ANCOVA to adjust for baseline FEV₁. 75 The clinical investigators were of the opinion that in the context of this trial and, in particular, given the age range of the recruits, the clinically relevant measure of TE is the relative (percentage) change from baseline in the percentage predicted FEV₁ and not the absolute change. The primary analysis was designed to compare the two randomised groups in terms of the mean percentage change in percentage predicted FEV₁ from baseline to end of treatment.

An ANCOVA model was designed to include baseline percentage predicted FEV₁, together with the other variables used in the randomisation algorithm, as covariates. ‘Baseline’ will be defined as the mean of the FEV₁ values from the two pretreatment assessments. ‘End of treatment’ will be taken as the mean of the values obtained at 14 days and 28 days following the final treatment. The TE will be presented as an adjusted difference in mean percentage change along with its corresponding 95% confidence interval (CI). No interim efficacy analyses were planned. As a sensitivity analysis the above analysis would be repeated, but with the logarithm of the end of treatment percentage predicted FEV₁ taken as the response variable, and the logarithm of the baseline percentage predicted FEV₁ included as a covariate in place of its raw value. An exploratory analysis compared the two randomised groups in terms of the evolution of FEV₁ over the 12 months of treatment. The study was not adequately powered to explore subgroup effects for the primary outcome measure, although we did plan to look at the stability of the TE over subgroups defined by the covariates included in the ANCOVA model. The formal analyses were performed by including interaction terms in the model. A similar approach could be used with certain secondary outcome measures that are closer to the direct mechanism of action of the study intervention, as there is likely to be more statistical power with such variables to explore subgroup effects, which could support a ‘stratified medicine’ approach to the use of gene therapy.

The intention-to-treat (ITT) group was defined as patients randomised and having any subsequent follow-up data. The per-protocol (PP) population was defined as patients receiving ≥ 9 doses. Demographic and safety data are presented for both populations. All efficacy data are presented for the PP population.

Inclusion criteria

1. Cystic fibrosis confirmed by sweat testing or genetic analysis.

2. Males and females aged ≥ 12 years.

3. FEV₁ between 50% and 90% predicted inclusive (Global Lung Function Initiative reference ranges; www.ers-education.org/guidelines/global-lungfunction-initiative.aspx).

4. Clinical stability at screening defined by:

   1. not on any additional antibiotics (excluding routine, long-term treatments) for the previous 2 weeks

   2. no increase in symptoms such as change in sputum production/colour, increased wheeze or breathlessness over the previous 2 weeks

   3. no change in regular respiratory treatments over the previous 4 weeks

   4. if any of these apply, entry into the study can be deferred.

5. Prepared to take effective contraceptive precautions for the duration of their participation in the study and for 3 months thereafter (as stated in GTAC guidelines)

6. If taking regular rhDNase and is willing, and considered able by independent medical carers, to withhold treatment for 24 hours before and 24 hours after the gene therapy dose

7. Written informed consent obtained.

8. Permission to inform general practitioner of participation in study.

Exclusion criteria

1. Infection with Burkholderia cepacia complex organisms, MRSA or M. abscessus.

2. Significant nasal pathology including polyps, clinically significant rhinosinusitis, or recurrent severe epistaxis (nose bleeds; nasal cohort only).

3. Acute upper respiratory tract infection within the last 2 weeks (entry can be deferred).

4. Previous spontaneous pneumothorax without pleurodesis (bronchoscopic subgroup only).

5. Recurrent severe haemoptysis (bronchoscopic subgroup only).

6. Current smoker.

7. Significant comorbidity including:

   1. moderate/severe CF liver disease (varices or significant, sustained elevation of transaminases: alanine transaminase/aspartate transaminase levels of > 100 IU/l)

   2. significant renal impairment (serum creatinine levels of > 150 µmol/l)

   3. significant coagulopathy (bronchoscopic group only)

8. Receiving second-line immunosuppressant drugs, such as methotrexate, ciclosporin and intravenous immunoglobulin preparations.

9. Pregnant or breastfeeding.

Patient recruitment

We had originally anticipated that the majority of trial participants would be recruited from an earlier study (the Gene Therapy Run-in) assessing outcome measures at periods of stability (unpublised data, Gene Therapy Consortuim). However, delays because of funding issues and the resultant drop-off in patient interest meant that recruitment had to be spread more widely. Initial patients were recruited by study staff from the adult and paediatric clinics at the Royal Brompton Hospital and from across Scottish CF centres. Once it was apparent that the pace of recruitment was slower than required, we activated participant identification centres. Local CF centres were activated in November 2012 and nationwide CF centres followed in January 2013; in total, 16 outside centres referred patients. These patients attended one of the three sites for trial-related visits, but continued their clinical care at their referring hospital. All patients had read a detailed, age-specific information sheet and underwent a thorough briefing before signing consent. Parents provided consent for participation of paediatric subjects, who were also asked for their assent to take part. Patient information sheets are available in Appendix 1.

Randomisation

Randomisation was performed by trial staff following a successful screening visit using the electronic InForm™, version 4.6 (Phase Forward Incorporated, Oracle, CA, USA) system. Patients consenting to participate in one or more of the mechanistic subgroups were randomised 2 : 1 to enrich for subjects receiving active treatment; all others were randomised on a 1 : 1 basis. The data required for randomisation and stratification [centre (London or Edinburgh), age (< 18/≥ 18 years), FEV₁ (< 70/≥ 70%) and inclusion in the mechanistic subgroups] were entered by a member of the trial team, which led to the generation of a unique patient number corresponding to a blinded randomised arm of the trial. In the event of computer system failure, a prearranged manual randomisation method was available from the InForm team. The unique patient number was entered onto the subject’s prescription sheet and submitted to the trial pharmacists who had access to the unblinding code and prepared the active or placebo product as appropriate.

Assembly of pGM169/GL67A complexes

The pGM169 drug substance and GL67A cationic lipid mixture are imported into the UK by The Clinical Biotechnology Centre of the University of Bristol, Bristol, UK, by a qualified person acting on behalf of the UK National Blood Service, a division of the NHS Blood and Transplant Special Health Authority. The dose preparation is described in full in Appendix 2. Briefly, immediately prior to use, pGM169 and GL67A are thawed to room temperature and GL67A is hydrated in water for injection. pGM169/GL67A complexes are prepared by rapidly mixing 5 ml of pGM169 and 5 ml of GL67A using a disposable syringe-based static mixer device [Plas Pak Industries, Norwich, CT, USA (Figure 5)]. The final dosage form of pGM169/GL67A contains nominally 2.6 mg/ml of pGM169, 3.7 mg/ml of GL67, 8.9 mg/ml of DOPE and 1.7 mg/ml of DMPE-PEG5000. The shelf life of the final formulation was 6 hours.

FIGURE 5.

Mixing of the trial formulation. The pGM169 and GL67A were mixed in a dual-barrel mixing device by loading the components into the syringe (a) and using a fixed-rate plunge mechanism (b) to deliver this into the receiving container.

Blinding

pGM169/GL67A or placebo (0.9% saline) were then filled into AeroEclipse II breath-actuated nebulisers by unblinded trial pharmacists. To avoid any inadvertent unblinding by clinical staff, nebulisers were taped and a tamper-proof seal was attached (Figure 6 and see Appendix 3 for details). Clinical trial staff, participants and trial analysts were blind to allocation until database lock. The 10-ml volumes were placed in opaque nasal spray devices (GlaxoSmithKline parts number AR5989 30 ml bottle/AR9488 30 ml actuator) and the device was primed.

FIGURE 6.

Masking of the nebuliser pots to ensure blinding. AeroEclipse II nebuliser pots are transparent (a) as the gene therapy formulation has a milky appearance that contrasts clearly with that of the 0.9% saline (clear, colourless liquid), it was necessary to mask the pots, which was achieved with tape and (b) a tamper-proof seal was applied to the lid, which also prevented any unintentional spillages.

Early patient safety cohort adaptive design

An adaptive early recruitment phase was designed to permit the early identification of any cumulative side effects: 20 subjects (10 active treatment and 10 placebo) would receive three doses at 4-weekly intervals before any further subjects were dosed. In addition to the visits described below undertaken by the entire cohort, they were also seen on day 2 post each dose. Clinical examination findings, lung function, gas transfer and systemic inflammatory markers would then be reviewed in an unblinded fashion by the Data Monitoring and Ethics Committee (DMEC).

• Should data prove satisfactory, these subjects would continue with subsequent visits and the remainder of the cohort will begin dosing.

• Should the data be of sufficient concern to the DMEC that progression was unacceptable, a second cohort of 20 patients would receive three doses of the formulation at a 2.5-ml dose, followed by DMEC review in a similar fashion.

  • If these data were acceptable, the trial would continue with all subjects receiving a 2.5-ml dose of either gene therapy or placebo. In this instance, the initial cohort would be discontinued and an additional 20 ‘naive’ subjects recruited.

  • Should the DMEC have considered these data unacceptable, the trial would have been halted.

The adaptive design was not in fact required as there were no concerning safety signals. Safety data from this cohort are presented in Chapter 5.

Schedule of trial visits

Scheduled study visits and interventions are shown in Table 1.

Clinical examination

Clinical examination included recording pulse rate, blood pressure (BP), respiratory rate, temperature, pulse oximetry, height, weight and lung auscultation. Patients were assessed by a team comprising medical, nursing and respiratory physiology personnel.

Spirometry

Forced expiratory manoeuvres from which the FEV₁ and the total volume exhaled (FVC) are conventional measures of airway patency and lung health, and are widely used both clinically and in trial settings for CF. The MEF (flow at varying percentages of vital capacity) are thought to more closely reflect small, distal airway disease, but are recognised as being more inherently variable. In CF, FEV₁ is closely linked to survival and is the most widely used primary outcome in clinical trials of respiratory interventions. It is recognised by regulatory agencies as a surrogate outcome and for this reason, alongside our longitudinal data demonstrating good power, it was chosen as the primary outcome for this trial. Spirometry was performed on the EasyOne spirometer (New Diagnostic Design Technologies, Zurich, Switzerland) with disposable mouthpiece and filter according to the American Thoracic Society (ATS)/European Respiratory Society (ERS) criteria. 76 A minimum of three measurements and the best FEV₁, FVC and MEF_25–75% were recorded in absolute values. These values were converted to percentage predicted values based on the Stanojevic references. 77

Lung clearance index

The LCI is a measure derived from a multiple breath washout (MBW), which provides a global measurement of ventilation inhomogeneity. It can be performed either with inhalation of an inert tracer gas, such as sulphur hexafluoride (SF6), or by using 100% oxygen to wash out resident nitrogen. In the case of an exogenous tracer, the gas is inspired until equilibrium is reached between the inhaled and exhaled air at which point the gas source is removed. The number of lung volume turnovers required until the expired tracer gas concentration falls to an arbitrary 1/40th of its starting value is used to calculate the LCI. Individuals with lung disease and greater ventilation inhomogeneity require longer to clear the tracer gas and, therefore, will have a higher (more abnormal) LCI.

Many different systems have been used to measure MBW in clinical trials in CF and these have been summarised in a recent ATS/ERS consensus document. 78 Although the mass spectrometer is considered the gold standard gas analyser, it is very expensive, custom built for MBW and, therefore, not suitable for widespread use. 78 The method adopted in this trial used the tracer gas, SF6, detected by the Innocor™ photoacoustic gas analyser (Innovision, Glamsbjerg, Denmark). Methodology was as described in a recent trial of the small molecule, ivacaftor. 79 In brief, the procedure is performed with a mouthpiece during tidal breathing while patients wear a nose clip (Figure 8). Triplicate measures are undertaken, with each test taking approximately 15 minutes. Data are subsequently analysed off-line according to strict standard operating procedures (SOPs) for quality control using software developed within the consortium. Every seventh trace was analysed in duplicate by a corresponding team member from the other site.

FIGURE 8.

Lung clearance index. (a) A child performs a LCI on the Innocor gas analyser; (b) wash-in of the tracer gas, SF6 (in green) is undertaken until equilibrium; and (c) the gas supply is disconnected and concentration falls with each subsequent breath as washout occurs (bottom). The black trace denotes flow.

Cycle ergometry

This test of exercise capacity was performed on a stationary exercise bike with breath-by-breath analysis on the Innocor photoacoustic gas analyser and was used to calculate VO_2max. Subjects were asked to pedal at a set speed of 70 rpm and maintain this speed throughout the test while wearing a nose clip and breathing via a mouthpiece attached to a gas analyser. The resistance increased automatically each minute for as long as pedalling could be sustained. The starting workload and increase in workload was dependent on the patient’s height, and calculated using the Godfrey protocol.

• Patients < 120 cm use 10 W starting resistance and 10-W increments.

• Patients 120–150 cm use 15 W starting resistance and 15-W increments.

• Patients > 150 cm use 20 W starting resistance and 25-W increments.

The subject’s heart rate, oxygen saturations and Borg scale were measured at rest and at each minute during exercise. The test was stopped by the operator if the patient’s oxygen saturation fell below 80%, if they were unable to maintain the required pedalling rate or if there was any concern about the patient’s condition. Once the test was complete, patients had a 2-minute cool-down period and were monitored until their oxygen saturation returned to baseline. The Innocor system was validated against standard calibrated VO₂ equipment using paired exercise tests in 12 CF patients. For VO_2max, the mean difference (reference equipment – Innocor) was –0.026 l/minute and the 95% CI was –0.27 to 0.22 l/minute.

Body-worn activity monitoring

Patients were asked to wear the SenseWear^® body monitoring system^™ (BodyMedia, Pittsburgh, PA, USA) for at least 7 whole days at time points throughout the trial. The device uses accelerometry, heat flux, skin temperature and galvanic skin response sensors to gather physiological data on movement and daily physical activity patterns. These armbands have been validated in normal subjects and a number of patient groups. 80 Data from activity monitors were downloaded and analysed using proprietary software. Final data were analysed when there were at least 4 days of data, at least 1 day was on a weekend and there were at least 10 hours of data for each day.

This analysis yields data for the proportion of time that wearers spend at various levels of energy expenditure. For the purposes of summarising these data in the present trial, a comparison was made (between active and placebo groups) of the percentage of time spent in vigorous exercise (defined as > 3 metabolic equivalents; METs) compared with the percentage of time spent at ≤ 3 METs of activity, as well as the mean number of steps taken per day.

Gas transfer

The transfer factor of the lung for carbon monoxide (TLCO) was measured using a single-breath technique (London site: Jaeger Masterscreen, CareFusion, Germany; Edinburgh site: CPL, nSpire Health, England). Subjects, who were seated and wearing a nose clip, were asked to exhale to residual volume, then take a maximal inspiration (required to be > 90% vital capacity) of the test gas (medical air with 0.28% carbon monoxide and 9% helium) prior to a breath hold of 10 seconds’ duration and a smooth exhalation. This procedure was repeated at least twice on each test day (minimal interval between tests was 4 minutes). Duplicate estimates of TLCO were required to be within 10% or 1.0 mmol/minute/kPa (whichever was greater) to meet requirements for inclusion. At least two technically acceptable tests from up to 10 attempts were obtained and the mean values were used for analysis. The following values were recorded for each test: single-breath TLCO, alveolar volume (VA), transfer coefficient (KCO; derived from TLCO/VA) and inspiratory vital capacity. TLCO and KCO were corrected for most recent haemoglobin (usually obtained on the same day) and expressed as corrected TLCO (TLCOc) and corrected KCO (KCOc) using standard formulae.

Quality-of-life questionnaire

The validated CFQ-R for either adults or children was completed by subjects at time points throughout the trial, always at the start of a visit before any other study investigations. The questionnaire is available in Appendix 4.

Lung computed tomography scanning

High-resolution CT scans were obtained on 64-channel multidetector scanners (SOMATOM Sensation 16 or 64; Siemens Medical Solutions, Erlangen, Germany). On two occasions (pre-dosing and at 28 ± 5 days after dose 12) patients underwent a full scan comprising contiguous thin sections (1.25 mm) through the entire volume of the lungs obtained during inspiration and also interspaced (1-mm sections at 10-mm increments) at end-expiration. A high-spatial-resolution algorithm was used for image reconstruction. All CT scans were scored independently by the same two radiologists specialising in thoracic imaging; scans had been anonymised and scores were assigned with the radiologists blinded to both subject identification and whether pre- or post-dosing. All scoring was performed directly from workstations with access to image manipulation including window settings (default: width 1500 Hounsfield units, centre 500 Hounsfield units). The presence and severity of specific CT scan features were recorded for each lobe as previously described. 81 The features studied were extent of bronchiectasis, severity of bronchiectasis, bronchial wall thickness, small and large airway mucus plugs (all scored per lobe on a range of 0–4 and the results meaned) and gas trapping (scored on a percentage basis) (features illustrated on examples in Figure 9). In addition, as a safety outcome, all subjects underwent a limited-cut inspiratory CT scan pre-dosing at the dose 4 visit; these scans were not formally scored, but were reviewed in real time by the clinical radiology team for any acute changes and the reports were provided to the DMEC.

FIGURE 9.

Computed tomography parameters. (a) Examples of bronchiectasis; (b) gas trapping (areas of mosaic attenuation highlighted in yellow where gas has failed to empty from areas of lung on expiration); and (c) an area of dense small mucus plugs in the left-lower lobe highlighted in yellow.

Venepuncture for blood sampling

This was performed by an experienced practitioner after the application of topical anaesthetic cream, if desired by the patient, via a peripheral vein. The use of central indwelling intravenous catheters was discouraged but not prohibited.

Urine sampling

Urine was collected into a sterile container and tested immediately by dipstick.

Sputum sampling

Sputum was collected after spontaneous expectoration whenever possible. In patients unable to expectorate, induction was considered at screening and follow-up visits only; all other visits included administration of study drug and it was considered undesirable to expose the patients to a procedure known to induce a degree of bronchospasm prior to dosing. The induction protocol followed clinical guidelines and included pretreatment with an inhaled bronchodilator (200 µg of salbutamol or equivalent, 15 minutes before test). Hypertonic saline (7%) was administered via a deVilbiss 2000 (DeVilbiss Healthcare Ltd, Tipton, UK) or equivalent ultrasonic nebuliser. Attempts to induce sputum were made during and after nebulisation and FEV₁ was monitored carefully throughout. Prior to dose 1, sputum was expectorated in 72 out of 116 (62%) of patients [active group, 37/62 (60%); placebo group, 35/54 (65%)]. After the last dose, sputum was expectorated in 57 out 116 (49%) of patients [active group: 27/62 (44%); placebo group: 30/54 (56%)].

At certain visits, patients were asked to collect all sputum expectorated over the previous 24 hours into 50-ml Falcon tubes (VWR International Ltd, Leicestershire, UK) and to bring these with them to the study visit.

Nasal potential difference

The characteristic CF ion transport defects, involving the charged ions chloride and sodium, result in an altered PD across the respiratory epithelium, measurable in millivolts. Techniques to measure this for both clinical diagnostic purposes and in the context of trials have been developed. 82,83 When compared with healthy cells expressing CFTR, in CF the absent chloride secretion and sodium hyperabsorption result in an apical cell surface that is electrically more negative under basal or resting (non-stimulated) conditions. Amiloride, as a sodium channel blocker, results in a greater reduction (depolarisation) in nasal PD for CF than non-CF subjects and subsequent ionic and pharmacological challenges to stimulate chloride secretion (via chemical gradients and cAMP stimulation) are largely ineffective in CF but lead to pronounced hyperpolarisation in the presence of functional CFTR (Figure 10).

FIGURE 10.

Representative nasal PD traces from (a) a CF patient; and (b) a healthy volunteer. The CF basal (Ringer’s, R) measures are higher (more negative) than non-CF largely related to sodium hyperabsorption. For the same reason, there is a larger drop (depolarisation) to the sodium channel blocker, amiloride (RA). There is little chloride secretion in response to either a zero chloride solution (ZC, passive gradient) or the cAMP stimulator, isoproterenol (ZCAI), in contrast to the hyperpolarisation observed in the healthy volunteer.

There have been over 20 previous clinical trials of CF gene therapy, which have involved dosing to the nose, lung or both. 84 In trials involving nasal dosing, nasal PD has been used as a main outcome for efficacy. 39,42,50–52,55,58,84–86 However, these tests are inherently variable and gene therapy-related results have been similarly so; correlation of any change in nasal PD with molecular evidence of CFTR has also often been lacking. 50,52,58 Where changes in PD have been reported following gene therapy, these have almost uniquely been restricted to the chloride secretion phase rather than correcting sodium hyperabsorption phases. This likely reflects the higher proportion of cells required to be corrected before changes in sodium hyperabsorption can be corrected (> 90% vs. < 10% for restoration of chloride transport). 87–89 Use is not limited to gene therapy evaluation, and recent studies have included trials of the small-molecule drugs ataluren and ivacaftor. Interestingly, while the latter small-molecule potentiator demonstrated significant and large-magnitude clinical changes, changes in nasal PD were much more modest.

For reasons of variability, patients electing to participate in the nasal dosing substudy underwent up to three nasal PDs on different days prior to dosing and two at follow-up appointments. A minority of subjects also underwent on-treatment recordings at other time points, but these are not included in the analysis owing to the small number.

At the time of the first measurement in any subject, a single-use, disposable, dual-lumen catheter (Marquat, France) was fixed in place at the site of maximal unstimulated PD in one nostril; for all subsequent measurements, the same nostril and the same distance into the nasal cavity was used. Solutions comprised chemicals listed in the Cystic Fibrosis Foundation Therapeutic Development Network (CFF TDN)’s SOP90 and were perfused at room temperature at a rate of 4 ml/minute in the following sequence: Ringer’s solution, Ringer’s plus amiloride (0.1 mM), Ringer’s plus amiloride plus zero chloride and Ringer’s plus amiloride plus zero chloride containing isoproterenol (0.01 mM) (Figure 11). Post-dosing measurements were obtained after 14 ± 2 days and 28 ± 5 days.

FIGURE 11.

Nasal PD. The subject has a double-lumen catheter inserted in one nostril. Solutions are pumped through one lumen while the other is connected to an electrode via conducting cream. There is a reference electrode placed subcutaneously on the forearm.

Traces were obtained using a high-impedance, low-resistance voltmeter (LR-4; Logan Research UK Ltd, Rochester, UK) onto a laptop computer and were scored by four investigators experienced in the technique (EWFWA, JCD, SNS and MDW). The four components (baseline, amiloride, zero chloride and isoproterenol responses) were scored individually using standard criteria.

Nasal brushings

Nasal brushings were performed in subjects in the nasal subgroup both pre-dosing and at 28 ± 5 days post final dose from under the middle or inferior turbinate of both nostrils using interdental brushes (Dent-o-care, London, UK). Cells were suspended in 800 µl of phosphate-buffered saline (PBS) and processed for transgene DNA/mRNA analysis (see Nasal and bronchial brushings).

Bronchoscopy and associated procedures

Bronchoscopy was undertaken in a subgroup of patients for both efficacy (gene transfer assessed by mRNA on brushings and PD) and safety assays (remodelling or lipid deposition in endobronchial biopsies).

Potential difference

Cystic fibrosis transmembrane conductance regulator-mediated ion transport is clearly of more relevance in the lower airway than the nose, but is more complex and invasive to measure. No standardised technique has been developed and only a few research groups perform this procedure. General anaesthesia is required to overcome the challenges of interference from patient movement or coughing, and to avoid local anaesthetic agents that affect ion channel function. The single operating channel within most fibreoptic bronchoscopes limits the size of the PD catheters used. Although some systems allow for a double-lumen PD catheter to pass through this port, our group has designed a single-lumen technique. 91 The technique permits basal and stimulated measurements, although pooling of solutions which then dilute subsequent perfusate needs to be carefully avoided; for this reason, perfusion rate is slower than for nasal measurements and an amiloride phase is omitted. We have previously shown clear separation between CF and non-CF basal PD in the proximal airways, although in both groups PD falls with distal progression and a disease-specific difference is not seen in the most peripheral sites measured. As expected, chloride secretory responses are significantly different between CF and non-CF (Figure 12). This technique was used in our previous trial of GL67-mediated CFTR (with a previous version of the CFTR plasmid incorporating a CMV promoter) and showed significant improvement in chloride secretion on day 2 post dosing. 58

FIGURE 12.

Chloride secretory responses on lower airway PD. Summary data demonstrating the increase in millivolts in lower airway PD in non-CF patients following perfusion with a zero-chloride solution containing isoprenaline (indicating chloride secretion). The effect is absent in patients with CF. The pre–post design of measurements in this trial is in an attempt to detect a signal over and above pre-dosing traces in the actively treated group. 91

In the current trial, all bronchoscopies and associated procedures were performed by a single operator (JCD), with the same senior anaesthetist (BK). Solutions (Ringer’s and a zero-chloride solution) were manufactured by the pharmacy at Eastbourne Hospital, UK and constituted the chemicals in the CFF TDN SOP,90 with the exception that amiloride was omitted from the latter. They were warmed prior to use to reduce the formation of microbubbles, but were used at room temperature and perfused at 100 ml/hour (Figure 13). Basal (non-stimulated) PD was measured on each wall of the distal trachea with sterile Ringer’s solution. Where possible, a stable period (< 1 mV change over 30 seconds) was recorded. Subsequently, at three sites more distally (approximately fifth-generation airways; all in one lung), following a second period with Ringer’s solution, the solution was switched to a zero-chloride equivalent containing isoproterenol (10 µM) for 5 minutes. The catheter was removed and reprimed between measurements. Hardware and software were as previously described. 91 Measurements were completed before the samples described below were obtained.

FIGURE 13.

Bronchial PD measurement. Under video control, the PD catheter is advanced via the bronchoscope onto the lower airway epithelial surface and can be observed on the screen towards the top of the picture. Solutions are perfused via pumps to the right and recordings made via a millivoltmeter connected to a laptop computer.

For analysis, recordings (1–3) from the same patient at the same time point were pre-grouped and scored blinded both for acceptability and response. Recordings were only accepted for scoring if (a) the time of any inflexion owing to the zero chloride/isoproterenol solution was approximately 50–80 seconds after onset of perfusion, this duration having previously been shown to represent clearing of the dead space within the catheter (≈50 seconds) and allowing ≈30 seconds for the initiation of any isoproterenol response; and (b) the response fell into one of three categories namely no response, continuous upwards (hyperpolarisation) or continuous downwards (depolarisation) deflection. Recordings were excluded if (a) the response at the 50- to 80-second time point was unstable (rising or falling); or (b) the deflection was characterised by both a depolarisation and a hyperpolarisation. Traces were scored by four investigators with experience in the technique (EWFWA, JCD, KH and SNS).

Bronchial brushing

Ten bronchial brushing samples were obtained from throughout the airways of the same lung using disposable cytology brushes (BC-202D-5010; Olympus UK, Southend-on-Sea, UK); to prevent cell loss that occurs while resheathing the brush, the entire bronchoscope was removed each time and the brush cut while still exposed at the distal end of the scope. Brushed cells were placed into two aliquots of PBS (five brushings in each) and processed for DNA/mRNA analysis (see Nasal and bronchial brushings).

Endobronchial biopsy

Two endobronchial biopsies were obtained from approximately fifth-generation subcarinae with disposable forceps (BC-202D-5010, Olympus UK, Southend-on-Sea, UK; Figure 14) and processed as described in Endobronchial biopsies. Bronchoalveolar lavage was not undertaken, but, where possible, samples of secretions were sent for microbiological culture.

FIGURE 14.

Endobronchial biopsy. (a) The biopsy forceps were advanced under direct vision to a subcarina of an approximate third- to fifth-generation airway where two biopsies were obtained; and (b) representative morphology on a haematoxylin and eosin (H&E)-stained section of an endobronchial biopsy demonstrating pseudostratified columnar epithelium and a submucosa containing smooth muscle, glandular tissue and an inflammatory cell infiltrate.

Post procedure, patients were observed for 4–6 hours prior to discharge.

Blood

Sputum

Samples collected into Falcon tubes for 24 hours prior to a visit were weighed. Freshly expectorated sputum obtained during the trial visit was stored on ice for a maximum of 2 hours and processed as previously described. 65 If patients did not produce sufficient sputum to perform all sputum assays, the following priority for assays was assigned: (1) clinical microbiology (generated from cough swabs if patients were completely non-productive); (2) soluble sputum that generated samples for quantification of interleukin 8 (IL-8), calprotectin and extracellular DNA as well as sputum cells; and (3) sputum solid content. Samples to prepare soluble sputum were available from 130 out of 232 samples (56%).

Urine

Samples were dipsticked routinely for glucose, protein and blood. At the start of the trial and on visits including CT scans, post-menarche female patients underwent urinary pregnancy testing.

Nasal and bronchial brushings

Deoxyribonucleic acid and ribonucleic acid (RNA) were simultaneously isolated from nasal and bronchial brushing samples using the AllPrep DNA/RNA Mini Kit (QIAGEN, Manchester, UK). Levels of pGM169-specific and endogenous CFTR DNA and mRNA were quantified using TaqMan^® (Life Technologies, Paisley, UK) qRT-PCR instruments and supplies, following reverse transcription, when appropriate, as described by Rose et al. 95 Absolute DNA and in vitro transcribed RNA calibration standards allowed precise copy number quantification. pGM169 DNA levels were determined using a qPCR primer set specific for the soCFTR2 cDNA:

• 50 nM forward primer: 5′-GGAACAGCTCCAAGTGCAAGA-3′

• 900 nM reverse primer: 5′-CCTGGTGTCCTGCACTTCCT-3′

• 100 nM probe: 5′-FAM-CAAGCCCCAGATTGCTGCCCTG-TAMRA-3′.

Levels were normalised to total genomic DNA, as determined using a qPCR primer set specific for the endogenous CFTR gene:

• 300 nM forward primer: 5′-CTTCCCCCATCTTGGTTGTTC-3′

• 300 nM reverse primer: 5′-TGACAGTTGACAATGAAGATAAAGATGA-3′

• 100 nM probe: 5′-VIC-TGTCCCCATTCCAGCCATTTGTATCCT-TAMRA-3′.

pGM169-derived mRNA was determined using qRT-PCR. Reverse transcription reactions (20 µl) contained ≤ 5 µl total RNA, 1.25 units of MultiScribe™ Reverse Transcriptase (Life Technologies) and 0.4 units RNAse inhibitor (Life Technologies) and 400 nM of the appropriate primer:

• soCFTR2 mRNA: 5′-CCAGCTGAAGAACAGCTTGCT-3′

• human CFTR (hCFTR) mRNA: 5′-CCAGGCGCTGTCTGTATCCT-3′.

pGM169-derived cDNA levels were determined using a qPCR primer set specific for the correctly spliced exon 1–2 soCFTR2 mRNA:

• 300 nM forward primer: 5’-TCTCCCTCCTGTGAGTTTGGTT-3′

• 300 nM reverse primer: 5’-GCTCACCACAGAGGCCTTCT-3′

• 100 nM probe: 5’-FAM-CTAGCCACCATGCAGAGAAGCCCTCTG-TAMRA-3′.

Levels were normalised to endogenous hCFTR mRNA as determined using a qPCR primer set specific for the correctly spliced exon 1–2 endogenous CFTR mRNA:

• 300 nM forward primer: 5’-GGAAAAGGCCAGCGTTGTC-3′

• 300 nM reverse primer: 5’-CCAGGCGCTGTCTGTATCCT-3′

• 100 nM probe: 5’-VIC-CCAAACTTTTTTTCAGCTGGACCAGACCAA-TAMRA-3′.

The qPCR reactions (10 µl) contained 0.6–6 ng of total DNA or 2 µl of cDNA and 1 × TaqMan^® Universal MasterMix (Life Technologies Manchester, UK). Thermocycling was performed in 384-well plates, on a 7900HT cycler (Applied Biosystems, Thermofisher, Foster City, CA, USA), with SDS 2.2 software, and the following cycling conditions: 50 °C for 2 minutes; 95 °C for 10 minutes; then 40 cycles of 95 °C for 15 seconds and 60 °C for 1 minute. Data were calculated as the percentage of pGM169 specific to endogenous CFTR copy number. When samples were positive for pGM169-specific signal, but either quantification fell below the linear range of the standards (limit of quantification; LOQ) or quantification of endogenous CFTR signal was negative, they were scored as positive but not quantifiable (PBNQ). Remaining samples that were negative for endogenous CFTR signal were scored as not determined. Samples negative for the pGM169-specific signal but positive for endogenous CFTR signal were scored as zero. When data are presented as post–pretreatment difference, a conservative approach was adopted for samples scored with the nominal value of PBNQ by substituting a value of either zero or the LOQ as appropriate to maximise the post–pre difference for placebo samples and minimise the post–pre difference for active samples. For bronchial brushings, two nominally identical samples were typically available (each pools of five independent bronchial brushings) and the maximum score was reported.

Endobronchial biopsies

Two biopsies were collected from each patient. One was frozen in 2-methylbutane-cooled liquid nitrogen at –196 °C and the other was immediately fixed in 10% formal saline. The formalin-fixed samples were processed using a Tissue Tek Sakura VIP processor (Sakura Finetek UK Ltd, Thatcham, UK) and embedded in paraffin wax blocks. Three micron sections were then cut from these blocks for haematoxylin and eosin (H&E) staining. For frozen samples, 6- to 8-µm sections were cut for H&E and ORO staining. For H&E staining, slides were thawed at room temperature for 30 minutes, and then placed in Harris’s haematoxylin for 1 minute. Subsequently, they were differentiated in acid alcohol for 5 seconds and left to blue in tap water for 5 minutes. They were then counterstained with eosin for 30 seconds and cover-slipped using an aqueous mountant.

For ORO staining, frozen sections were thawed and air dried at room temperature for 30 minutes. Slides were rinsed in 60% 2-propanol then placed in filtered ORO working solution for 15 minutes. Subsequently, the slides were taken out of the ORO solution and excess dye was washed off with 60% 2-propanol, before rinsing in three changes of distilled water. The sections were counterstained using Harris’s haematoxylin for 30 seconds in order to visualise the nuclei. The sections were left to blue in running tap water for 5 minutes and mounted with a coverslip using Aquatex (an aqueous mountant). H&E slides were scored independently by two blinded pathologists using a semiquantitative scoring system for goblet cell hyperplasia, basement membrane thickening, presence of chronic inflammatory cells (lymphocytes and plasma cells), neutrophils, eosinophils and seromucinous gland hyperplasia. ORO slides were scored for the presence of lipid-laden macrophages using a semiquantitative scoring system.

Trial database, monitoring and audit

The Imperial College Trials Unit (ICTU) built an InForm database specifically for the trial and the staff working on the trial entered the data from source. The InForm database was monitored by both ICTU on a regular basis and the Imperial College Research Office undertook on-site source data verification. Two patients had 100% source data verification and the remainder had 10%. Any queries were flagged on the database as well as in the case report forms and dealt with by the study team.

Data analysis

A detailed statistical analysis plan (SAP) was drafted by the trial statistician (GDM) and was further refined with input from the Trial Management Group (see Appendix 5). The SAP was approved by the Trial Steering Committee (TSC) and finalised ahead of the database being locked and the trial unblinded. Data management was undertaken using Microsoft Excel^® (version 14.4.6 for Mac OSX; Microsoft Corporation, Redmond, WA, USA). Descriptive statistics and standard analyses were performed using Prism (Version 5.0c for Mac OSX, Graph Pad Software Inc., San Diego, CA, USA) and or IBM SPSS (version 22.0, IBM Corporation, Armonk, NY, USA). There was no exploration of the impact of missing values in the primary analysis as data were available for 114 out of the 116 PP patients.

Physiology

There was also a significant TE in FVC (p = 0.031; Figure 18; see also Figures 25 and 26). No difference was observed in the LCI at the follow-up period, although the serial time point graph (Figure 19) demonstrates that the active-treatment group appear to have a better-preserved LCI in the earlier stages of the trial.

FIGURE 18.

Forced vital capacity. Time course of the response of FVC to either placebo or active treatment. ‘Pre’ and ‘post’ indicate the mean of two measurements at the respective time points. Error bars indicate SEM. There was a significant (p = 0.031) TE, with the active group having an ANCOVA-adjusted improvement of 3.0% (95% CI 0.3% to 5.8%) compared with placebo at 12 months’ follow-up. SEM, standard error of mean.

FIGURE 19.

Change in LCI. There was no significant TE in LCI, both groups increasing (worsening) slightly over the year of the trial.

Computed tomography scans

Paired pre and post follow-up high-resolution CT scans were available for 115 patients; the one PP patient missing is the same patient also missing from the analysis of the primary end point as she had been lost to follow-up. The other patient missing from the primary outcome is represented here, as CT scanning did not pose the same risks following her pneumothorax as a forced expiratory manoeuvre would have done.

Bronchiectasis is defined as dilatation and thickening of the airways. It is considered irreversible. An intervention applied for this period of time would not therefore be expected to have any impact on the extent of bronchiectasis and this was the case in this cohort (Figure 20). Although it was not statistically significant, it was interesting to note a trend for less worsening in the severity of bronchiectasis in the active-treatment group than in the placebo.

FIGURE 20.

Change in CT scores for bronchiectasis (a) extent; and (b) severity.

Gas trapping is a feature visible on expiratory films when areas of lung fail to empty properly because of airway obstruction. There were increases in the percentage of gas trapping in the placebo group that were not demonstrated in the active-treatment group and which led to a significant TE (p = 0.048; Figure 21).

FIGURE 21.

Gas trapping on CT scan. Over the course of the study, gas trapping indicative of small airways disease, increased in the placebo group (worsened) but did not in the active-treatment group leading to a statistically significant TE (p = 0.048).

Scores of mucus plugging (large or small; Figure 22) and airway wall thickness (Figure 23) did not differ significantly between the two groups, although, for every parameter, the change was smaller in and, therefore, favoured the active-treatment group.

FIGURE 22.

Mucus plugging on CT scan. (a) Large mucus plugging; and (b) small mucus plugging (sometimes termed ‘tree in bud’).

FIGURE 23.

Airway wall thickness on CT scan.

Quality-of-life scores

Baseline quality-of-life scores for the domains of major interest in a trial of respiratory treatment (respiratory and physical) were high for the group as a whole (median 87.5%). There were no statistically significant TEs at the end of the trial, although, again, the TE appeared to favour the active-treatment group (Figure 24).

FIGURE 24.

Quality-of-life (QoL) scores. Change in the (a) respiratory and (b) physical domains on the validated CF quality-of-life questionnaire CFQ-R.

Other secondary outcomes

Figures 25 and 26 show that for all the assays, even those which did not reach significance, the standardised TE favoured the active-treatment group.

FIGURE 25.

Secondary outcome measures. Forest plot showing the responses of secondary outcome measures to placebo or active treatment. To allow results from different end points to be plotted on a common scale, the estimated TEs were standardised to be presented as multiples of the underlying SD (standardised TE). The size of the circle is proportional to the number of subjects represented and the bars indicate 95% CI. *, statistically significant; ESR, erythrocyte sedimentation rate; WBC, white blood cell.

FIGURE 26.

Responses of active and placebo groups separately for secondary outcomes. Forest plot showing the responses of the placebo or active-treatment arms when assessed by the predefined subgroup values shown. To allow results from different endpoints to be plotted on a common scale, the estimated TEs were standardised to be presented as multiples of the underlying SD. The size of the circle is proportional to the number of subjects represented and the bars indicate 95% CI. *, statistically significant; Con meds, concurrent medication; MP, mucus plugs; Pa, P. aeruginosa; TA AE, treatment-associated AE.

Nasal

In the nasal arm of the substudy, the assay could quantify a dosing-dependent increase in vector-specific DNA in 15 out of 15 active-dose subjects post treatment, although in four of these modest levels of DNA were detected prior to dosing. In the placebo patients, DNA could also be detected in samples from three out of four subjects post dosing (none prior to dosing); no vector-specific mRNA was quantifiable in either group.

Lower airway

In the bronchial arm of the substudy, the assay quantified vector-specific DNA in 10 out of 10 active-treatment patients and zero out of seven placebo subjects post dosing (Figure 30); no vector-specific mRNA was quantifiable in either group.

FIGURE 30.

Assessment of DNA from bronchial brushings in the placebo (n = 7) and active-treatment (n = 10) subgroups. Each circle represents an individual patient.

Nasal

The nasal arm comprised 24 patients in the ITT group, 20 of whom were in the PP. One patient had a mean total chloride secretory response ≥ 5 mV pre-dosing and was, therefore, excluded as prespecified in the protocol. Day 28 post-treatment recordings for two active-treatment patients had to be delayed beyond a window of > 7 days of the prespecified interval and were therefore excluded; both patients had usable values at the day 14 post treatment time point. Overall, 75 out of 106 (70.8%) of zero chloride, and 70 out of 106 (66.0%) of isoproterenol recordings were interpretable.

There were no significant changes in baseline values or amiloride responses in either the placebo or active-treatment groups. No significant changes in either the zero chloride or isoproterenol responses were seen. For the zero chloride component 10 out of 14 active-treatment patients and three out of six placebo patients showed net secretion (i.e. a more negative value post treatment than pretreatment); 4 out of 14 active-treatment patients showed mean pre–post-treatment responses (ranging from –3.4 mV to –7.0 mV), which were more negative than the largest placebo response. For the isoproterenol component, 11 out of 12 active-treatment patients and three out of six placebo patients showed net secretion (Figure 31).

FIGURE 31.

The response of the nasal epithelium to perfusion with (a) a zero chloride solution; and (b) a zero chloride solution containing isoproterenol (10 µM). Each symbol indicates the change in this response from trial start to finish for the relevant treatment in an individual patient. The plotted value is the most negative value obtained from all interpretable recordings (range 1–3) at each time point for that patient. A more negative value is in the non-CF direction. The placebo group had a median pre–post trial change of + 0.1 mV (range +1.1 to –2.3 mV) and the active-treatment group –0.6 mV (range +3.5 to –7.0 mV) (p = 0.509). The response of the nasal epithelium to perfusion with a zero chloride solution containing isoproterenol (10 µM). The placebo group showed a median pre–post trial change of +0.3 mV (range +2.9 to –2.6 mV) and the active-treatment group –1.0 mV (range +0.1 to –3.8 mV) (p = 0.424).

Bronchial

In the bronchial subgroup, one active-treatment patient withdrew after six doses, but consented to post-dosing bronchoscopy within the prespecified timing (28 ± 5 days) after the last dose; values for this patient are included in the analysis. Post-dosing bronchoscopy was performed on two placebo patients 58 and 62 days after their last doses, and on two active-treatment patients 49 and 111 days after their last dose. The data for the two placebo patients are included in the final analysis, on the basis that the natural history of bronchial electrophysiology is unlikely to be influenced by a dose of saline approximately 60 days previously. Values for the two active-treatment patients were omitted from the final analysis given the monthly dosing schedule in the trial based on previously generated gene expression data, although their samples were included in the safety analysis. Overall, 66 out of 102 (64.7%) recordings fulfilled the acceptability criteria.

There were no significant changes in basal values in either the placebo or active-treatment groups. Figure 32 shows bronchial chloride responses using the mean of all available interpretable tracings for each patient; a negative value indicates a non-CF direction. The placebo group (n = 7) had a median pre–post trial change of +3.1 mV (range +9.3 to –1.2 mV) and the active-treatment group (n = 10) –1.3 mV (range +4.0 to –5.8 mV) (p = 0.032). Five out of 10 active-treatment patients had values more negative than the largest placebo response. Figure 32b shows the same analysis with only the most negative value recorded for each patient at any time point. The placebo group showed a median pre–post trial change of +2.6 mV (range +9.3 to –1.2 mV) and the active group –2.8 mV (range + 4.0 to –16.8 mV) (p = 0.087). Six out of 10 active-treatment patients had values more negative than the largest placebo response.

FIGURE 32.

Change from pretreatment in the response of the bronchial epithelium to perfusion with a zero chloride solution containing isoproterenol (10 µM). (a) The response of the bronchial epithelium to perfusion with a zero chloride solution containing isoproterenol (10 µM). Each symbol indicates the change in this response from trial start to finish for the relevant treatment in an individual patient. The plotted value is the mean of all interpretable recordings (range 1–3) at each time point for that patient. A more negative value is in the non-CF direction. The placebo group had a median pre–post trial change of +3.1 mV (range +9.3 to –1.2 mV) and the active group –1.0 mV (range +4.0 to –5.8 mV) (p = 0.032). (b) The response of the bronchial epithelium to perfusion with a zero chloride solution containing isoproterenol (10 µM). Each symbol indicates the change in this response from trial start to finish for the relevant treatment in an individual patient. The plotted value is the most negative value obtained from all interpretable recordings (range 1–3) at each time point for that patient. A more negative value is in the non-CF direction. The placebo group showed a median pre–post trial change of +2.6 mV (range +9.3 to –1.2 mV) and the active group –2.8 mV (range + 4.0 to –15.6 mV) (p = 0.087).

Adverse events

The AEs are presented as an aggregate number of events and also are listed by site and treatment group. The total numbers of AEs are presented in Tables 3 and 4.

Table 3 presents the total number of AEs that were observed initially in the trial, we observed that that there was only one severe AE.

Table 4 presents the number of AEs that were observed 2 days after dosing the subjects. No SAEs were observed 2 days after dosing the subjects. Table 5 presents changes in spirometric indices.

We observe that, overall, there were no SAEs occurring during the trial.

Spirometry data

Clinical examination

Table 6 presents the change in clinical examination parameters from screening day to dosing day. The changes in the clinical examination are presented as a number of cases that change from a positive result to a negative result; change from a negative result to a positive result; and do not change.

Current symptoms

Table 7 presents the symptoms expressed as change from the dosing day to 2 days after dosing.

Vital signs

Table 8 presents the change from screening day to dosing day. A negative change indicates a decrease in vital sign parameters.

Biochemistry parameters

Table 9 presents the change from screening day to dosing day. A negative change indicates a decrease in the biochemistry parameters.

Haematology parameters

Table 10 presents the change from screening day to dosing day. A negative change indicates a decrease in the haematology parameters.

Gas transfer

Table 11 presents the change from the screening day to 2 days after dosing. A negative change indicates a decrease in the gas transfer.

In summary, the DMEC did not consider there was a safety signal on reviewing these results and wrote approving continuation of full recruitment into the trial at the 5-ml dose.

General adverse events

All recorded AEs were categorised as in Table 12, and the frequency of their occurrence expressed proportional to the number of patients in each group. All patients in both groups reported AEs, as would be expected in a population of CF patients monitored for 12 months. However, the majority of AEs were of a nature expected in this clinical context and frequencies were similar between groups; there were no statistically significant differences in number overall or once they were broken down into the categories as tabulated.

One patient in the placebo group (fatigue and increased respiratory symptoms) and one in the active-treatment group (flu-like symptoms) discontinued study treatment because of AEs.

Serious adverse events

There were no deaths during the study. Six SAEs were documented, all in the active-treatment group (Table 13). Neither the DMEC nor the TSC considered any SAE was related to the trial medication, but that one (case 2) was probably related to a trial procedure (bronchoscopy).