Self-reactive CFTR T cells in humans: implications for gene therapy.

Cystic fibrosis (CF) is one of the most common autosomal recessive lethal disorders affecting white populations of northern European ancestry. To date there is no cure for CF. Life-long treatments for CF are being developed and include gene therapy and the use of small-molecule drugs designed to target specific cystic fibrosis transmembrane conductance regulator (CFTR) gene mutations. Irrespective of the type of molecular therapy for CF, which may include gene replacement, exon skipping, nonsense suppression, or molecular correctors, because all of these modulate gene expression there is an inherent risk of activation of T cells against the wild-type version of CFTR. Here we report the validation of the human interferon-γ enzyme-linked immunospot assay and its application for the analysis of CFTR-specific T cell responses in patients with CF and in non-CF subjects. We found non-CF subjects with low levels of self-reactive CFTR-specific T cells in the United States and several patients with CF with low to high levels of self-reactive CFTR-specific T cells in both the United States and the United Kingdom.

Changes in physiological, functional and structural markers of cystic fibrosis lung disease with treatment of a pulmonary exacerbation.

BACKGROUND: Clinical trials in cystic fibrosis (CF) have been hindered by the paucity of well characterised and clinically relevant outcome measures. AIM: To evaluate a range of conventional and novel biomarkers of CF lung disease in a multicentre setting as a contributing study in selecting outcome assays for a clinical trial of CFTR gene therapy. METHODS: A multicentre observational study of adult and paediatric patients with CF (>10 years) treated for a physician-defined exacerbation of CF pulmonary symptoms. Measurements were performed at commencement and immediately after a course of intravenous antibiotics. Disease activity was assessed using 46 assays across five key domains: symptoms, lung physiology, structural changes on CT, pulmonary and systemic inflammatory markers. RESULTS: Statistically significant improvements were seen in forced expiratory volume in 1 s (p<0.001, n=32), lung clearance index (p<0.01, n=32), symptoms (p<0.0001, n=37), CT scores for airway wall thickness (p<0.01, n=31), air trapping (p<0.01, n=30) and large mucus plugs (p=0.0001, n=31), serum C-reactive protein (p<0.0001, n=34), serum interleukin-6 (p<0.0001, n=33) and serum calprotectin (p<0.0001, n=31). DISCUSSION: We identify the key biomarkers of inflammation, imaging and physiology that alter alongside symptomatic improvement following treatment of an acute CF exacerbation. These data, in parallel with our study of biomarkers in patients with stable CF, provide important guidance in choosing optimal biomarkers for novel therapies. Further, they highlight that such acute therapy predominantly improves large airway parameters and systemic inflammation, but has less effect on airway inflammation.

Secreted Gaussia luciferase as a sensitive reporter gene for in vivo and ex vivo studies of airway gene transfer.

The cationic lipid GL67A is one of the more efficient non-viral gene transfer agents (GTAs) for the lungs, and is currently being evaluated in an extensive clinical trial programme for cystic fibrosis gene therapy. Despite conferring significant expression of vector-specific mRNA following transfection of differentiated human airway cells cultured on air liquid interfaces (ALI) cultures and nebulisation into sheep lung in vivo we were unable to detect robust levels of the standard reporter gene Firefly luciferase (FLuc). Recently a novel secreted luciferase isolated from Gaussia princeps (GLuc) has been described. Here, we show that (1) GLuc is a more sensitive reporter gene and offers significant advantages over the traditionally used FLuc in pre-clinical models for lung gene transfer that are difficult to transfect, (2) GL67A-mediated gene transfection leads to significant production of recombinant protein in these models, (3) promoter activity in ALI cultures mimics published in vivo data and these cultures may, therefore, be suitable to characterise promoter activity in a human ex vivo airway model and (4) detection of GLuc in large animal broncho-alveolar lavage fluid and serum facilitates assessment of duration of gene expression after gene transfer to the lungs. In summary, we have shown here that GLuc is a sensitive reporter gene and is particularly useful for monitoring gene transfer in difficult to transfect models of the airway and lung. This has allowed us to validate that GL67A, which is currently in clinical use, can generate significant amounts of recombinant protein in fully differentiated human air liquid interface cultures and the ovine lung in vivo.

Quantification of periciliary fluid height in human airway biopsies is feasible, but not suitable as a biomarker.

The "low volume hypothesis," stating that imbalanced ion movement across the cystic fibrosis (CF) airway epithelium leads to a reduction in periciliary fluid (PCL) and consequently impaired mucociliary clearance, has been the prevailing theory explaining CF pathophysiology, and has been supported by animal models and ex vivo cell culture systems. However, studies in freshly obtained human tissue have not yet been performed. Methods to quantify PCL height in freshly obtained airway biopsies may be useful to assess efficacy of new treatments aimed at restoring PCL height. Here, we established methods to quantify PCL height in freshly obtained CF and non-CF human lower airway biopsies. More than 90% of biopsies contained ciliated epithelium, and PCL height measurements were feasible in approximately 50% of these. Although the mean PCL height was reduced in CF tissue (non-CF, 5.60 ± 0.28 µm; CF, 4.52 ± 0.47 µm), this did not reach statistical significance (P = 0.06). To strengthen the data, we performed similar studies in wild-type and CF knockout mice, and confirmed the results (non-CF, 4.70 ± 0.13; CF, 4.10 ± 0.09 µm; P < 0.05). PCL height measurements in freshly obtained human airway biopsies are feasible, and PCL height appears reduced in subjects with CF, thereby further supporting the "low volume hypothesis." However, power calculations indicate that this assay can only be considered as a biomarker in large, late-phase clinical trials, because sample sizes required to achieve sufficient power are comparatively large.