Cystic fibrosis transmembrane conductance regulator dysfunction and its treatment

Introduction

Cystic fibrosis is a genetic condition caused by mutations in the cystic fibrosis transmembrane conductance regulator gene (CFTR). Loss or altered function of the product of this gene, the CFTR protein, affects the transport of ions across cell membranes in several tissues in the body which in turn, through mechanisms which are still not fully understood, results in the disease.

Normal function

CFTR is complex protein found on the surface membrane of cells in a wide variety of tissues where it functions as a regulated chloride ion channel. In the lung CFTR is found on the apical membrane of the cells lining the airways. To open, the channel must undergo a conformational change induced by cAMP-dependent phosphorylation and by binding and hydroysing ATP.¹ When the channel is open, chloride ions move from inside the cell to the airway lumen. Electrophysiological studies on the airways of subjects with CF have shown that in addition to a defect in chloride secretion, there is a baseline state of sodium hyper-absorption,² which turns out to result from an interaction between CFTR and the epithelia sodium channel, eNaC. Normally functioning CFTR inhibits the opening of eNaC and when CFTR is dysfunctional or absent, eNaC is open more of the time, allowing greater sodium influx. The mechanism by which CFTR regulates eNaC is not understood.³

Pathophysiology of CF lung disease

There have been several hypotheses to try and explain how CFTR dysfunction results in recurrent airway infection and lung damage. The hypothesis which is now most widely accepted is the low volume hypothesis first proposed by Boucher in 1994.^4,5 Effective ciliary function is essential in preventing airway infection. The cilia move within a watery fluid layer called airway surface liquid (ASL), the depth of which is tightly regulated to be equal to the length of the cilia. Mucous ‘rafts’ float on the surface of the ASL and are propelled towards the larynx by the underlying cilia. CFTR appears to be critical in regulating ASL height (Figure 1). The combination of loss of chloride secretion and sodium hyper-absorption via ENaC results in loss of ASL, which in turn results in the overlying mucous becoming adherent to the airway cells and prevents effective ciliary activity. Inhaled bacteria are no longer cleared but instead set up low grade persistent infection; the associated inflammatory and immune responses lead to airway damage and airways obstruction. CF lung disease is usually significantly more severe than that seen in primary ciliary dyskinesia. This suggests that it is not just loss of ciliary activity that is important; loss of ASL volume and adherence of mucous to airway cells seems to be critical in the development of CF lung disease, and particularly to biofilm formation which leads to chronic infection.

Figure 1.

Low volume hypothesis for CF lung disease. Sodium hyperabsorption and defective chloride secretion drives increased absorption of water by osmosis and reduces the height of the airway surface liquid (ASL), leading to impaired muco-ciliary clearance and adherence of mucus to airway epithelial cells

CFTR mutations

Over 1800 CFTR mutations have been described; most are very rare, and occur in only a few families, and only 20 or so occur at a frequency above 0.1% worldwide. In the UK CF population, 32 mutations account for 85% of all CFTR defects. CFTR mutations have been classified in two different ways; either by their effects on CFTR function (See Figure 2) or by the underlying molecular defect (See Box 1).

Figure 2.

Classification of CFTR protein dysfunction into mutations that result in either: no CFTR protein at the apical membrane; reduced CFTR at the apical membrane; normal levels of CFTR at the apical membrane but with decreased chloride conductance; normal levels of CFTR at the apical membrane but with abnormal regulation

Box 1. Classification of CFTR mutations by underlying molecular defect, based on Welsh and Smith 1993⁶.

Class I and II mutations are common; Class III and IV are uncommon (1–5% of CF mutations) and Class V and VI mutations all very rare (<1% of CF mutations)

Class I mutations completely or nearly completely abolish production of full length CFTR, either because the mutation itself forms a stop codon, or results in a shift of the reading frame resulting in a downstream stop codon. Examples: G542X, W1282X, 621 + 1GtoT

Class II mutations result in changes that affect how the CFTR protein is processed within the endoplasmic reticulum and Golgi systems, such that nearly all the protein is degraded and recycled. A small amount may make it to the cell surface. The amino acid changes may also affect CFTR function (class III, IV or VI effects) of any protein that makes it to the cell surface. Examples: dF508, N1303K

Class III mutations result in amino acid substitutions which affect how the channel is regulated, usually resulting decreased channel opening. Examples: G551D, R560T

Class IV mutations result in amino acid substitutions which affect how well the channel conducts chloride ions. Examples: R117H, R347P

Class V mutations result in the production of reduced amounts of normal CFTR protein. The commonest are mutations which reduce the efficiency of CFTR messenger RNA splicing but still allow a variable proportion to be spliced normally. Example: 3849 + 10kb C to T; 2789 + 5GtoA

Class VI mutations are thought to affect the stability of CFTR at the cell surface. Examples: Q1412X, N287Y, and possibly dF508

Therapy

Osmotic agents

Osmotic therapies, such as nebulized hypertonic saline and inhaled mannitol have been designed to draw water into the airway lumen and to reconstitute the ASL, although whether this is actually how they work is debatable.⁷ Short term use of hypertonic saline is associated with improvements in FEV1 (by 2–10%)^8,9 and in lung clearance index.¹⁰ There is only one longer term study of hypertonic saline and this showed relatively small improvements in absolute percent predicted FEV1 of around 3% over a 48 week period, with the additional benefit of a 50% reduction in number of respiratory exacerbations.⁸ Inhaled dry powder mannitol is a more recently developed drug and results in similar benefits – small but sustained improvements of FEV1 of around 5% and a 35% reduction in exacerbation rate over a 12 month period.¹¹ A UK license for the use of inhaled mannitol is CF is being sought, but has not yet been approved.

Correcting sodium hyper-absorption

The observation that sodium hyper-absorption seen in human nasal epithelium could be blocked by infusing amiloride on the epithelial surface, led to several studies of nebulized amiloride as a treatment for CF. Pilot studies were optimistic about the benefits of amiloride.^12,13 Three subsequent randomized control trials^14–16 failed to show any benefit of amiloride on respiratory function tests over a 6 to 12 month period (see also Cochrane review 2006).¹⁷ It has been suggested that the lack of clinical benefit of amiloride is due to its short half-life on the airway surface because of rapid absorption. Higher potency sodium channel blockers with longer airway surface half-lives have been developed, most recently a compound called PS552. This has been shown to increase mucocilary clearance in sheep for a longer duration than amiloride.¹⁸ No clinical studies on PS552 have been published.

An alternative method of inhibiting the activity of ENaC is to reduce the number of functioning ENaC channels on the apical surface. To become active, ENaC must interact with specific cell-surface proteases. A number of cell-surface protease inhibitors have been indentified and have the potential to reduce ENaC-mediated Na+ transport.¹⁹ Clinical studies on these compounds are awaited.

Alternative chloride channels

CFTR is not the only chloride channel on the apical surface of airway epithelial cells. Increasing the activity of other chloride channels may replace lost chloride secretion when CFTR is dysfunctional. Whether correction of chloride transport without correction of sodium hyper-absorption will be enough to prevent the development of CF lung disease is not known. The main target for this approach has been the calcium-activated chloride channel (CaCC).

Increases in intracellular calcium resulting in increased CaCC activity have been achieved via the P2Y2 apical surface receptor. Denufosol is a P2Y2 receptor agonist, and has been shown to increase chloride and fluid secretion in cell culture systems. Denufosol has also been shown to have a modest ability to inhibit ENaC, as well as increasing ciliary beat frequency.²⁰ In a randomized controlled trial of 352 patients (80% were children) with cystic fibrosis who had mild lung disease (mean FEV1 92%, mean FEF25–75 84%) inhaled denufosol, 60 mg three times daily given over a 6 month period, was associated with a mean increase in FEV1 of 48 ml (about 2% of baseline) compared to 3 ml for placebo, with a P value of 0.047.²¹ There were no differences in exacerbation rate or other measures of lung function between the 2 groups. Denufosol was well tolerated. In the subsequent 6-month open-label phase of the study, the improvement in FEV1 increased to 115 ml in the treatment group and to 78 ml in the placebo group switched to active treatment. It is possible that drugs like denufosol will be more effective at preventing deterioration of lung function, rather than working as rescue therapies, and these trial results should be interpreted in this light. The improvement in absolute volume of FEV1 is that expected over time for a growing population, and this was protected more effectively in the denofosul group than the placebo group. A longer term phase III trial, also in children over 5 years of age, is in progress to confirm these results. Ideally denofosol, like any to treatment designed to correct the underlying CFTR defect, would be best assessed by starting treatment in asymptomatic infants and following them for several years.

Duramycin, also known as Moli-1901, is a polypeptide of 19 amino acids that appears to be able to increase intracellular calcium and activate the CaCC, although its precise mechanism of action is unclear.^22–24 A phase 1 study of intranasal duramycin showed changes in chloride transport in the nasal mucosa of healthy subjects and those with CF consistent with the activation of a chloride channel.²⁵ It is reported to be undergoing phase II clinical trials, but there are as yet no published data.

Treatments directed at CFTR dysfunction

Premature termination codons

A premature termination codon (PTC) arises when a point mutation results in a premature stop (nonsense) codon, resulting in a truncated and usually functionally useless protein. Nonsense mutations, such as G542X, are present in 5–10% of the UK CF population. Some aminoglycosides and several synthetic ribosomal-binding molecules have the potential to allow the protein machinery to ignore the premature stop signal and continue with full-length protein production. The best studied investigational drug is ataluren (PTC124). A phase II study of ataluren, taken orally for 14 days by 23 subjects with CF, who had at least one nonsense mutation, has shown that it is well tolerated and that it was associated with significant improvements in transepithelial nasal ion transport.²⁶ Nineteen of the same 23 patients took part in a longer study over 12 weeks.²⁷ There was a significant improvement in nasal chloride transport whilst taking ataluren; the improvement was lost in all subjects within 4 weeks of stopping the drug. There were small non-significant improvements in lung function, and an apparent 20% reduction in cough frequency, but these data are hard to interpret without a control group. Similar results have been reported in a 14-day study of 30 children (6 to 18 years) with CF, who had at least one CFTR nonsense mutation. Subsequently 15 of the 30 children had an improvement in nasal chloride transport of at least 5.0 mV whilst taking ataluren and nasal biopsies showed an increased proportion of cells with full-length CFTR in the apical cell membrane.²⁸ Variation in response to ataluren between individuals may reflect mutation-specific differences (it may work better for some nonsense mutations than others), or may reflect differences in the amount of CFTR messenger RNA present in different individuals.²⁹ A phase III study of over 200 children and adults with CFTR nonsense mutations is underway, with a primary endpoint of improved lung function, and is expected to report its findings in 2012.

Correctors

Correctors are small molecules which assist trafficking of mutant CFTR to the cell surface membrane. Though dF508 CFTR can function as a chloride channel, albeit at lower efficiency than normal CFTR, if it is assisted in reaching the cell surface. High throughput drug discovery systems using cell culture models testing thousands of diverse molecules, have identified several corrector candidates. Of these, VX-809, produced by a joint venture of the Vertex Company and the American Cystic Fibrosis Foundation, has undergone limited clinical studies in CF patients. In cell culture systems, VX-809 is associated with restoration of chloride transport to 15% of that seen in normal cells;³⁰ this level of activity is thought to be sufficient to restore effective muco-ciliary transport and to protect against the development of severe CF lung disease.³¹ A multi-dose phase IIa study of 89 adult subjects homozygous for dF508, taking a range of doses of VX-809 orally for 28 days, showed that the drug was well tolerated and was associated with a small (5–15 mmol/L) but statistically significant correction of sweat chloride, which was dose dependent and sustained during the treatment period. There were no changes in nasal chloride transport, lung function or detectable mature CFTR in rectal biopsy samples.³² These results suggest that VX-809 can improve CFTR function, but that its effects on the airway were below the detection level of the assays used. Higher doses, use for a longer period, or possibly use combined with a potentiator drug (see below) may be necessary to demonstrate benefit in the airway. Safety concerns persist about the specificity of correctors for CFTR – assisting trafficking of other mutated proteins through the cell's quality control systems may have unforeseen adverse effects.

Potentiators

Potentiators are small molecules which increase the channel activity of mutant CFTR in the cell membrane and, as with correctors, several active potentiator molecules have been identified by high throughput screening. Ivacaftor (previously known as VX-770) is an orally bio-available molecule which can increase chloride transport to 50% of normal levels in cells in culture expressing one copy of dF508 CFTR and one copy of G551D CFTR.³³ In a 28-day randomized double-blind placebo-controlled phase IIa trial of 19 patients with CF who had at least one G551D mutation, there was a significant improvement in nasal chloride transport (9 out of 15 had a 5 mV or more improvement compared to 0 out of 4 on placebo) and an astonishing reduction in sweat chloride (all 15 subjects on active drug had lower sweat chloride; median change in sweat chloride −45 mmol/L).³⁴ Two phase III studies, one in adults and one in children, have recently been completed. The adult study (161 subjects, mean age 25.5 years, mean FEV1 64% predicted) has just been published, and shows that compared to placebo, patients taking ivacaftor had an impressive absolute improvement in FEV1 of 11%, a 55% reduction in pulmonary exacerbations and a 2.7 kg weight gain over a 48 week period.³⁵ Publication of the study in children is awaited. FDA and EMA approval is being sought to bring the drug to market soon. The benefits of this drug have only been shown in patients carrying the G551D mutation – whether it will work for other class III mutations, or for dF508 CFTR brought to the cell surface by a corrector drug, is not yet known.

Conclusions

The lack of new therapies that correct the underlying defect in CF has been frustrating, particularly so after the excitement that surrounded the identification of the CFTR gene in 1984. Real progress is now being made, and clinically effective drugs may soon be available. Critically, for the first time with ivacaftor, there is convincing evidence that correcting CFTR function does lead to rapid improvements in simple measures of lung function and wellbeing. This finding will be relevant when evaluating other therapies, including gene therapy. Introduction of mutation-specific therapies to the clinic will need to be done sensitively since, by their very nature, they will not be suitable for all patients. The molecular mechanisms of many rare mutations are not known, and there will be a need to test individual patients' mutations to see if they might benefit from new therapies. These investigations are likely to be expensive and combined with the likely high cost of any new therapies, will bring added pressures on already very tight budgets. It seems likely however, that therapies which correct the basic defect, and which can be started in asymptomatic infants, will be cost-effective over the life-time of the individual.

DECLARATIONS

Competing interests

None declared

Funding

None

Ethical approval

Not applicable

Guarantor

Not applicable

Contributorship

Jeremy Hull was the sole contributor

Acknowledgements

None