The future of CFTR modulating therapies for cystic fibrosis

INTRODUCTION

It is estimated that 70 000– 100 000 people in the world have cystic fibrosis, a life-limiting inherited, multisystem disease. The recessive gene that causes cystic fibrosis was localized to the long arm of chromosome 7 in 1985 [1], and its sequence was identified in 1989 [2]. A normal copy of the cystic fibrosis transmembrane conductance regulator (CFTR) gene encodes a 1480 amino acid cyclic adenosine monophosphate-regulated ion channel [3], initially identified as conducting chloride and more recently, also bicarbonate. In the airway, a close inhibitory relationship with its neighbouring sodium channel, ENaC, allows regulation of sodium absorption. Thus, defective CFTR in patients with cystic fibrosis results in reduced chloride secretion and hyperabsorption of sodium and water, leading to a reduced volume of airway surface liquid and impaired mucociliary clearance [4]. Recent data from the cystic fibrosis pig have highlighted the potential importance of bicarbonate, not only in mucus composition, but also in innate defence molecule function [5]. Early and ultimately chronic, bacterial infection and inflammation are hallmarks of this disease.

Over 2000 mutations within the CFTR gene have been described to date, although not all are disease-causing [6]. Those that lead to disease can be divided into six classes according to the way in which the resulting protein is defective.

CFTR MUTATION CLASSES

Class I mutations lead, due to a premature trunca- tion codon, to a short, nonfunctioning CFTR protein. Those of class II encode a structurally abnormal, misfolded protein that fails to traffic to the cell surface. Protein of classes III–VI reach the cell surface but fail to function appropriately: class III, so-called ‘gating’ mutations fail to open nor- mally and class IV mutations lead to reduced Potentiators: increase the gating (frequency of opening) of CFTR on the cell surface.Correctors: allow trafficking of misfolded CFTR protein up to the cell membrane, where it can function, albeit not at wild-type levels.

CFTR-MODULATING THERAPIES

Until very recently, all pulmonary therapy treated the downstream consequences of abnormal CFTR: physiotherapy and mucolytics, antibiotics and anti- inflammatory agents alongside optimal nutrition. These conventional treatments have improved the prognosis for cystic fibrosis patients, but still, median age of death in the UK is under 30 years [7]. There has long been an assumption, until recently untested, that a step-change in the efficacy of therapies would require the mechanism of action to be via the basic defect itself. Three groups of small molecule drugs have been developed through to clinical trials to date:Read-through agents allow the ribosome to ‘ignore’ a premature stop codon, leading to the restoration of full length protein production.

PROGRESS TO DATE

Read-through agents

Class I, or nonsense, mutations account for approxi- mately 5– 10% of the worldwide cystic fibrosis popu- lation, with a greatly increased incidence in certain populations such as the Ashkenazi Jewish com- munity [8,9]. Following initial observations that aminoglycoside antibiotic agents could allow ribo- somal read-through, the synthetic compound, ata- luren, was developed. Despite promising proof-of- concept data from phase 2 trials, a recent large, multicentre, phase 3 trial of this oral agent failed to reach its primary outcome of change in forced expiratory volume in the 1st second (FEV ) [10&] or in a number of secondary outcomes. Interestingly, a predefined subgroup analysis did, however, show that patients who were not on inhaled aminoglyco- sides had a significant treatment-related benefit; as these drugs also act by binding to the ribosome, the hypothesis is an inhibitory interaction and a further phase 3 trial is currently underway with these drugs excluded.

Potentiators

Ivacaftor (Kalydeco) is the first CFTR-modulating agent to have received regulatory approval, regarded by many to herald the beginning of a new era in the treatment of cystic fibrosis. The commonest class 3 mutation, Gly551Asp (previously known as G551D), leads to a protein which is correctly situ- ated in the apical cell membrane, but which fails to open. ‘Open probability’ was significantly improved in cells in vitro and the drug has since shown impres- sive clinical benefit in patients with at least one copy of Gly551Asp [11,12] and, more recently, with one of a larger number of rarer gating mutations [13]. Alongside a reduction in the CFTR biomarker sweat chloride, clinical benefits were demonstrated including improvements in percentage predicted FEV1% (approximately 10% absolute in the gating mutations), weight gain (thought likely related to improved gut bicarbonate secretion) and in some studies a reduction in exacerbation frequency and improved quality of life; even patients with well preserved [14] or end-stage [15&] lung disease have been shown to be able to benefit. A trial in children 2–5 years of age, the first to include stool assays,appeared to demonstrate at least partial restoration of pancreatic exocrine function, indicating that there may be a window in early life where presumed ‘irreversible’ organ damage may be amenable to treatment. Longer term follow-up of these children is currently ongoing, particularly with regard to rises in liver function tests which appeared more often than in older patients. As the drug has been licensed, many patients are being followed-up in an observa- tional study of biomarkers [16&]; one interesting observation is an apparent reduction in positive cultures for the damaging Gram-negative pathogen, Pseudomonas aeruginosa [17&]. These findings require further research, but could relate to improved bicarbonate secretion and thus innate defence at the airway surface.

Correctors

Class II defects are the commonest worldwide; F508del (previously DF508) is carried by approxi- mately 90% of the cystic fibrosis population, 50% being homozygous and 40% heterozygous [18]. Fol- lowing an early observation that low temperatures or the compound glycerol [19,20] stabilized mis- folded CFTR protein, so-called corrector agents were actively sought via high-throughput screening pro- grammes. The first of these to enter clinical trials is the Vertex compound, lumacaftor, which had led in vitro to chloride transport of around 14% of wild- type [21]. However, the drug was shown to have very modest effects on sweat chloride and not to improve any clinical outcomes in patients homozygous for the F508del mutation [22].

In contrast, in combination with ivacaftor (which is thought to potentiate CFTR which has been successfully trafficked to the cell surface by the lumacaftor), the drug has recently been found safe and effective, with 3– 4% improvement in FEV1 and a significant reduction in exacerbations [23&]. It has not demonstrated efficacy in patients com- pound heterozygous for F508del and possessing a second, non-ivacaftor sensitive mutation [24&]. The combination preparation (Orkambi) has been granted approval by the Food & Drug Adminis- tration (FDA) for F508del homozygous patients aged 12 years and above and at the time of writing is under review in Europe.

Vertex has also developed an alternative correc- tor, VX-661, which possesses a possible advantage over VX-809 in that its metabolic pathway does not interact with that of ivacaftor and achieving thera- peutic levels may be easier. Phase 3, global, clinical trials of this agent in combination with ivacaftor are currently underway in patients with F508del and a variety of other mutations [25–27].

In summary, significant progress has been made in the last few years; cystic fibrosis is now seen as a disease in which the basic defect can be targeted, which can lead to very significant clinical improve- ments. Several other pharmaceutical agencies are beginning to launch small molecules into clinical trials [28–31] and others are at earlier stages of preclinical development. Thus, there is much room for optimism. However, a number of challenges remain; these will be discussed in the remainder of this article.

OUTSTANDING CHALLENGES

Increasing coverage to other mutations

Patients with gating mutations responsive to single- agent ivacaftor are in a minority and, to date, only patients with two copies, not one, of F508del have shown clinical benefit with Orkambi. This leaves a significant proportion of the global cystic fibrosis population without a mutation-specific therapy. Theoretically, an agent which corrects and poten- tiates two copies of F508del should also be active when only one is present, so this lack of effect likely represents the threshold of a drug with limited efficacy; this may in part relate to the interaction of the two components of Orkambi. The investi- gation of combinations without such inhibitory interactions or with greater inherent potency could allow those patients with only one copy of F508del to benefit also. Similarly, the misfolding of class 2-mutated CFTR is complex and multistep [32]; some have suggested that a triple combination including two correctors may ultimately be required. Neither correctors nor potentiators would be predicted to work either alone or in combination for patients with class I mutations, and so continu- ing efforts to understand and overcome the current limitations of readthrough agents is to be encour- aged. In addition to cystic fibrosis, a large number of other inherited diseases are caused by premature stop mutations, so a successful drug could have much broader indications; indeed the drug has recently received conditional approval for certain types of Duchenne muscular dystrophy [33&]. Finally, the vast majority of CFTR mutations are extremely rare, and many are incompletely under- stood; undertaking sufficient research to fully characterize all of these and allow access to those patients with potential to benefit will require crea- tive alternatives to conventional preclinical and clinical trial design; ex-vivo drug testing strategies, for example using organoids, has been successful [34&&] as may so-called n-of-1 trials, whereby a strict protocol allows a patient to act as their own control in establishing efficacy [35]. An alternative to mutation-specific small molecules is gene-based approaches: the UK CF Gene Therapy Consortium has recently reported modest but statistically signifi- cant clinical benefits in patients receiving lipid- mediated CFTR gene transfer compared with placebo, an effect which was mutation-independent [36&] and others are investigating the potential of stem cell therapy [37&] and mRNA repair [38]. Although challenging in their own right, these approaches do offer wider coverage, in addition to potentially less frequent therapeutic interventions.

Long-term surveillance to seek evidence of ‘disease modification’ and confirm safety

Recent long-term follow-up data suggest that, in addition to the relatively acute and sustained improvements in FEV1 in patients with Gly551Asp on ivacaftor, there is a decrease in the chronic rate of decline of lung function, providing the first evi- dence that CFTR modulators may be genuinely disease modifying. Seeking such confirmation is important for a chronic disease for which such therapies are designed to be life long, but come at a high financial cost. Similarly, in a rare disease with only relatively small numbers of patients receiving drugs and the current exposure period being short, rare or late side-effects may be difficult to identify. In both of these scenarios, the use of existing national and multinational registries to aid the real-world collection of such data is an elegant solution; much has been invested into these resources by healthcare teams,national/continental organizations and by patients themselves. Building on this investment by employing registries as tools for phase 4 trials and long-term surveillance can provide a win-win if undertaken with sufficient care and planning.

Equality of access

Personalized or precision medicines, particularly for rare diseases, are expensive to develop, so ensuring equality of access once licensed drugs are available is a major challenge. Although insurance companies may be prepared to cover the cost of one or two patients, nationalized healthcare systems are likely to have different concerns; indeed, reimbursement for the first of these drugs, ivacaftor, took signifi- cantly variable time periods across the globe. As the eligible population expands with the licensing of Orkambi, these concerns will likely increase and may significantly impact patients in developing and less well-resourced countries. Across Europe, for example, there are already significant inequal- ities in access to high-standard healthcare. One proposal arising from a European Commission Workshop addressing this is the development of European Union-wide best-practice initiatives [39], but although there is much reference in this and similar documents to equality of access as para- mount, the funding frameworks within which this will be achievable are not currently clear.

Treatment early in the disease course: prevention better than cure?

There is clearly a window early in the life of a cystic fibrosis patient, where lungs are completely healthy, any impairment in mucus clearance or host defence having not yet led to detectable disease. The advent of widespread, although not yet universal, newborn screening presents an opportunity to utilize that window, and now that highly effective basic defect therapies are coming of age, that option seems, in many ways, tantalizing. However, there are two main concerns. The first is that it is likely unwise to extrapolate safety data from older patients down to small babies with rapidly growing and adapting, and thus biologically different, organs. Clearly then, new drugs need testing in these younger age groups. It may be considered unethical to conduct placebo- controlled trials in this age group, for example exposing young children to interventions such as repeated blood tests, and this may be particularly so once efficacy has been established in older patients. But, in small populations, undertaking open-label safety trials may make interpretation of possible adverse events difficult. There is also a second con- cern relating to determining efficacy in a relatively healthy baby; if there are no or few abnormalities to measure, demonstrating that things have improved becomes increasingly difficult. In addition, many of the more conventional outcome measures, for example spirometry, are complex to perform in this age group and yet these are currently favoured by regulatory agencies. There is currently much collab- orative focus on establishing the utility of alterna- tive outcome measures, for example lung clearance index and standardized imaging, which could be of particular value in patients with early stage disease.

CONCLUSION

Increased understanding of the underlying causes of cystic fibrosis has led to a significant expansion of therapeutic targets and the first licensed therapies targeting the basic defect in cystic fibrosis. This is not only of potential benefit to those individuals with responsive mutations, but also proves the prin- ciple that restoration of CFTR protein function can lead to significant clinical benefits and has thus energized the translational research community, for whom improving the outlook for this,Vanzacaftor as still life- limiting, disease is the major goal.