Abstract
Cystic fibrosis (CF) is a recessive disorder caused by mutations in the gene that encodes the CF transmembrane conductance regulator (CFTR) protein. CFTR protein is a chloride and bicarbonate channel that is critical for normal epithelial ion transport and hydration of epithelial surfaces. Current CF care is supportive, but recent breakthroughs have occurred with the advent of novel therapeutic strategies that assist the function of mutant CFTR proteins. The development and key clinical trial results of ivacaftor, a small molecule that targets gating defects in disease-causing CFTR mutations including G551D CFTR, are summarized in this review. The G551D mutation is reasonably common in the CF patient population and produces a CFTR protein that localizes normally to the plasma membrane, but fails to open in response to cellular cues. Ivacaftor treatment produces dramatic improvements in lung function, weight, lung disease stability, patient-reported outcomes, and CFTR biomarkers in patients with CF harboring the G551D CFTR mutation compared with placebo controls and patients with two copies of the common F508del CFTR mutation. The unprecedented success of ivacaftor treatment for the G551D CF patient population has generated excitement in the CF care community regarding the expansion of its use to other CF patient populations with primary or secondary gating defects.
Cystic fibrosis: cystic fibrosis transmembrane conductance regulator and cystic fibrosis disease
Cystic fibrosis (CF) is a progressive disease affecting over 70,000 people globally and 30,000 people in the United States [Cystic Fibrosis Foundation, 2011; Pilewski et al. 1999; Rowe et al. 2005, Rogan et al. 2011]. CF is caused by autosomal recessive mutations in the gene coding for the CF transmembrane conductance regulator (CFTR) protein, and approximately 1900 disease-causing CFTR mutations have been described since its discovery (http://www.genet.sickkids.on.ca/cftr/app). CFTR is an anion channel and traffic ATPase, and members of this protein family typically have two transmembrane domains that anchor the protein in the plasma membrane and two cytoplasmic nucleotide binding domains (NBD-1 and -2). Gating of CFTR is provided by the two NBDs, which come together in a heterodimer complex to bind and hydrolyze adenosine triphosphate (ATP) [Riordan et al. 1989, Rommens et al. 1989, Riordan, 2008]. A unique regulatory or R domain of CFTR provides protein kinase A (PKA)-dependent regulation [Cheng et al. 1991]. CFTR conducts both chloride and bicarbonate [Anderson et al. 1991; Linsdell et al. 1997; Welsh and Smith, 2001; Tang et al. 2009] and also regulates ion transporters such as other chloride channels, glutathione and thiocyanate transport, and the epithelial sodium channel (ENaC) [Stutts et al. 1995; Schwiebert et al. 1999; Hudson, 2001; Rowe et al. 2005; Moskwa et al. 2007; Lorentzen et al. 2011]. In the lungs, CFTR expression is highest in airway submucosal glands, with moderate expression in the pseudostratified epithelium of the medium and large airways and also cells lining the distal small airways [Engelhardt et al. 1992, 1994]. Expression of CFTR in the airways, intestines, pancreatic ducts, bile ducts, vas deferens, and sweat glands all have direct associations with disease manifestations.
Several lines of research support the hypothesis that CF lung disease is directly related to CFTR regulation of ion transport and hydration in airway epithelia, including the volume of the airway surface liquid (ASL) compartment and the hydration of mucus [Matsui et al. 1998a, 1998b; Joo et al. 2002; Wine and Joo, 2004; Joo et al. 2006; Boucher, 2007]. Fluid transport is largely regulated by the submucosal glands and surface airway epithelial cells, which produce a fluid compartment defined as the periciliary liquid layer (PCL, a subcompartment of the ASL). Normal hydration of the PCL supports full ciliary activity and clearance of mucus [Matsui et al. 1998a, 1998b; Boucher, 2007]. The larger gel layer (composed of hydrated mucins) functions to trap particulates and serves as a reservoir for ASL fluid. Normally, the submucosal glands and surface epithelial cells work in concert, providing hydration to both the PCL and ASL that rapidly clears secretions without the need to activate secondary host defenses. One leading hypothesis addressing how this is accomplished is that ENaC drives sodium and fluid absorption with passive chloride flow until ASL homeostasis is achieved, producing a PCL height of approximately 7 µm. If the PCL and ASL need to be expanded, chloride secretion increases. The process is primarily regulated by CFTR and balanced by the activities of local regulatory molecules [Lazarowski et al. 2004; Tarran et al. 2006], and water follows passively in response to ion flux [Donaldson et al. 2006; Boucher, 2007; Donaldson and Boucher, 2007].
There remains some uncertainty in the field regarding what ion transport abnormalities observed in CF are primarily responsible for disease pathology. In the absence of CFTR function, ENaC and sodium absorption are increased and fluid absorption has been shown to become dominant in ex vivo primary human airway epithelial cell monolayers [Boucher, 2007]. The result is dehydration of the airway luminal surface, including a decreased PCL volume and an increased percentage of solids in the ASL mucins. These processes produce mucus stasis and small airway obstruction, setting the stage for infection and inflammation. The end result is structural damage to the airway, bronchiectasis, and ultimately respiratory failure. Recent reports from the porcine model of CF question whether all aspects of this process are operative in vivo, as lower airway ENaC activity and ASL/PCL contraction were not observed in CFTR knockout piglets compared with littermate controls [Chen et al. 2010]. A critical role has also recently been described in CF lung disease for bicarbonate transport, which may be needed for effective mucin unpackaging and formation of a normal macromolecular structure [Ambort et al. 2012; Gustafsson et al. 2012]. The low pH of the ASL in CF has also been attributed to a recently described bicarbonate transport defect, reducing bacterial killing by the innate immune system [Pezzulo et al. 2012]. Also, the CF airway produces inflammation out of proportion to normal inflammatory stimuli, with high levels of cytokines that drive neutrophil influx into the airway lumen [Konstan and Berger, 1997; Sagel et al. 2007; Banner et al. 2009].
For the vast majority of patients with CF, treatment is currently supportive, and targets downstream disease manifestations (e.g. mucus obstruction, infection, inflammation) that result from lost CFTR function. Although these treatments clearly provide benefits, they are labor intensive and involve a high burden of care. Targeting early steps between lost CFTR function and disease manifestations has the potential to fundamentally change how we care for patients with CF, improving long-term outcomes and quality of life for children, young adults, and their families who live with CF [Hoffman and Ramsey, 2013].
Cystic fibrosis transmembrane conductance regulator genetics and mutation classes
Over 1900 known CF disease-causing mutations can be segregated into six specific mutation classes [Welsh and Smith, 1993; Rowe et al. 2005; Rogan et al. 2011]. While these classes are not mutually exclusive they do provide a helpful framework to consider CFTR-directed therapies (Table 1). Class I mutations result in impaired biosynthesis of CFTR at the mRNA and protein level, resulting in the absence of CFTR at the plasma membrane. Class II mutations cause defective protein folding, maturation and accelerated degradation, with failure of CFTR to traffic protein to the plasma membrane. Class III and class IV mutations localize normally to the plasma membrane, but exhibit defective gating (class III) or result in defective chloride conductance through CFTR (class IV). Class V mutations lead to diminished transcription of normal and functional CFTR, with reduced protein levels at the plasma membrane. Lastly, class VI mutations result in accelerated turnover of CFTR protein at the cell surface. Class I–III mutations are typically considered nonfunctional, while class IV–VI mutations retain some CFTR function.
CFTR mutation classes and potential treatment strategies to restore CFTR function.
Modulator strategy with studies currently in progress.
Theoretical modulator strategy based on underlying mutation class defect.
CFTR, cystic fibrosis transmembrane conductance regulator.
Due to the large number of mutations that cause CF, it may be overwhelming to imagine the generation of mutation-specific CFTR drugs for clinical use. However, a select few mutations account for most CF disease, and members within a mutation class may be amenable to common strategies to restore function. For example, F508del CFTR is a class II mutation, and at least one copy is present in over 80% of patients with CF. Deletion of phenylalanine at position 508 results in abnormal protein folding and processing, with accelerated degradation and little or no F508del CFTR at the cell membrane [Riordan, 2008]. G551D CFTR is the third most common CF-causing mutation, which is present in about 4% of patients with CF. G551D is a class III gating mutation, with normal plasma membrane levels, but defective gating. The end result is a CFTR protein that cannot open and transport chloride [Rowe et al. 2005; Rogan et al. 2011]. Therefore, targeting these more common mutations or classes of mutations may be sufficient to reach the vast majority of patients with CF.
Ivacaftor: a cystic fibrosis transmembrane conductance regulator modulator restoring function to G551D CFTR
Starting in the late 1990s, the Cystic Fibrosis Foundation collaborated with pharmaceutical companies to discover and develop CFTR modulators, which are small molecules that restore function to mutant CFTR. High throughput screening of thousands of highly diverse compounds was performed and G551D CFTR was one of the original target CFTR mutations (Vertex Pharmaceuticals, Cambridge, MA, USA) [Van Goor et al. 2009]. Ivacaftor (VX-770, Kalydeco) is one result from these efforts, and the remainder of this review is focused on the evidence for ivacaftor treatment as a ‘potentiator’ of CFTR gating mutations, including G551D.
Vertex Pharmaceuticals developed the compound ivacaftor through iterative medicinal chemistry. In vitro studies describe PKA-phosphorylated CFTR as the target of ivacaftor, and recent findings indicate that it induces channel opening independent of ATP binding and hydrolysis [Eckford et al. 2012]. Ivacaftor increases chloride secretion in cultured human CF bronchial epithelia cells with the G551D mutation 10-fold, producing 50% of the total CFTR activity seen in normal (non-CF) bronchial epithelial-cell monolayers. Ivacaftor also reduces excessive amiloride-sensitive current and fluid absorption in G551D CF airway epithelia, thus increasing the ASL volume with resultant increased ciliary beat frequency [Van Goor et al. 2009]. These positive findings provided the rationale for in vivo studies of ivacaftor. More recently, in vitro studies have evaluated ivacaftor’s potentiation of CFTR gating mutations other than G551D, including G178R, G551S, G970R, G1244E, S1255P, G11349D, S549N, S549R and G1349D [Yu et al. 2012]. Ivacaftor treatment increases mutant CFTR activity by 30–118% of normal CFTR function across these gating mutations, providing support for additional studies of ivacaftor for non-G551D CFTR gating mutations.
Clinical trials of ivacaftor monotherapy
There are four publications that summarize the clinical experience of ivacaftor in patients with CF [Accurso et al. 2010; Ramsey et al. 2011; Flume et al. 2012; Davies et al. 2013]. These are summarized in Table 2. The first study was a phase II, randomized, double-blind placebo-controlled trial of ivacaftor performed in 39 adults with CF with at least one copy of the G551D CFTR mutation and a baseline measure of lung function [forced expiratory volume in 1 s (FEV1)] of at least 40% of predicted [Accurso et al. 2010]. The study was performed in two parts. In the first, 20 patients were randomly assigned to receive increasing doses of ivacaftor (25, 75, or 150 mg) every 12 h or placebo for 14 days in a modified crossover design. In the second part, 19 patients were randomly assigned to receive ivacaftor (150 or 250 mg) every 12 h or placebo for 28 days in a parallel design. The gender distribution, median age, body mass index (BMI), and baseline FEV1 were comparable across all groups, and adherence to the study drug was over 90%. The primary endpoints included safety and adverse events (AEs) relative to placebo, with key secondary endpoints including biomarkers of CFTR function [nasal potential difference (NPD), sweat chloride], pulmonary function tests (PFTs) and a validated patient reported outcome [the Revised CF Health-related Quality of Life Questionnaire (CFQ-R)] [Quittner et al. 2009]. Overall, the most frequent AEs included fever, cough, nausea, pain, and rhinorrhea, and severe adverse events included rash and abnormal blood glucose. Significant within-subject changes in the NPD (a measure of nasal epithelial CFTR activity) from baseline to day 14 were observed in both the 75 mg (mean −4.7 mV, p = 0.003) and 150 mg ivacaftor groups (mean −5.4 mV, p = 0.01). No significant NPD improvements were noted in either the placebo or 25 mg ivacaftor group from baseline. The mean change in sweat chloride concentration from baseline varied from −32.9 mMol in the 25 mg ivacaftor group to −42.3 mMol in the 150 mg ivacaftor group (p < 0.001 for all dose groups, within-group changes from baseline and compared with placebo). Both CFTR biomarkers rapidly returned to baseline after washout. The mean change in FEV1 % predicted compared with baseline ranged from 4.9% in the 25 mg ivacaftor group to 10.5% in the 150 mg ivacaftor group. The within-subject improvements in the FEV1 % predicted were significant in both the 75 mg (p = 0.002) and 150 mg (p = 0.008) ivacaftor groups.
Summation of completed ivacaftor monotherapy clinical trials.
Adapted from Accurso et al. [2010], Ramsey et al. [2011], Flume et al. [2012], Davies et al. [2013].
Sweat chloride values = mMol (negative value indicates increased CFTR function).
FEV1, sweat chloride, CFQ-R, and NPD values are median for Accurso and colleagues and the rest are presented as mean values.
Statistically significant change reported in original study (p < 0.05).
CFQ-R focused on the respiratory domain. Disease-specific health-related quality of life questions are answered by the subject on a four-point Likert scale and then rescaled within the respiratory domain to a score range from 0 to 100 points, with higher points indicating better health.
NPD in millivolts (negative value indicates increased CFTR function).
CFQ-R, Cystic Fibrosis Questionnaire – Revised; CFTR, cystic fibrosis transmembrane conductance regulator; FEV1, forced expiratory volume in 1 s; NPD, nasal potential difference; w/pulm. exac., number of patients with pulmonary exacerbations at the end of treatment time point.
In the second part of the study (28-day treatment, 150 mg or 250 mg every 12 h compared with placebo in a randomized, double-blind parallel design), the 150 mg ivacaftor group had significant within-subject improvements in the NPD, sweat chloride concentrations, and FEV1 at day 28 that were not seen in the placebo group. The median within-subject change in CFTR activity by NPD was −3.5 mV at day 28 (range −8.3 to +0.5 mV, p = 0.02), −59.5 mMol for sweat chloride (range −66 to −19 mMol, p = 0.008), and +8.7% predicted for FEV1 (range +2.3 to +31.3%, p = 0.008). NPD, sweat chloride and FEV1 % predicted improvements were also seen in the 250 mg dose group, while CFQ-R changes did not meet statistical significance in either dose group.
The second publication was a phase II trial examining ivacaftor (150 mg every 12 h) in patients with CF possessing two copies of the F508del CFTR mutation [Flume et al. 2012]. As noted previously, the F508del CFTR mutation differs from G551D in that it primarily interrupts trafficking of CFTR to the plasma membrane. However, some studies suggested that small amounts of F508del CFTR may be at the plasma membrane of some patients with F508del CF and thus are available for potentiation by ivacaftor [Van Barneveld et al. 2010]. Adults with CF were enrolled in a 16-week double-blind trial with 4:1 randomization to ivacaftor versus placebo (part A). A total of 140 total patients were enrolled and 104 completed ivacaftor treatment compared with 26 placebo controls. Overall, the safety and AE frequency were comparable between the ivacaftor and placebo treatment groups. The difference in the change of FEV1 % predicted from baseline through week 16 between the ivacaftor and placebo groups was +1.7% (p = 0.15). Sweat chloride levels were slightly reduced in the ivacaftor group compared with the placebo group by −2.9 mMol (p = 0.04) from baseline through week 16. The results indicate that ivacaftor monotherapy was insufficient to produce clinically meaningful benefits to patients with CF who were homozygous for the F508del CFTR mutation. In part B of this study, patients who had a 10% increase in FEV1 % predicted or a decrease in sweat chloride level of 15 mMol during part A were enrolled in an open-label 96-week extension to assess whether any lung function or sweat chloride changes from part A were sustained. There were no meaningful changes noted in efficacy or biomarker outcomes during the extension. The results further support the conclusion that ivacaftor monotherapy is insufficient to improve biomarker or clinical outcome measures in patients homozygous for the F508del mutation. Ivacaftor is under investigation as a cotherapy with F508del CFTR correctors in patients with CF with the F508del CFTR mutation, but a detailed review of these studies in progress is beyond the scope of this article (see http://www.clinicaltrials.gov/).
Phase III trials have examined ivacaftor monotherapy in patients with CF possessing the G551D CFTR mutation [Ramsey et al. 2011; Davies et al. 2013]. Both trials demonstrated rapid and sustained improvement in clinical outcome measures and CFTR biomarkers with an acceptable safety profile. The first study was a randomized, double-blind placebo-controlled trial that evaluated ivacaftor (150 mg every 12 h) in patients with G551D CF 12 years of age and older with an FEV1 of 40–90% predicted. Subjects were treated with ivacaftor (n = 84) or placebo (n = 83) for 48 weeks, and adherence was reported to be 91%. The primary endpoint was change in FEV1 % predicted from baseline through 24 weeks. Secondary endpoints included change in FEV1 through 48 weeks, time to first pulmonary exacerbations, changes in the respiratory domain of the CFQ-R, weight, and sweat chloride concentration. The ivacaftor-treated group demonstrated significant improvements in FEV1, pulmonary exacerbation frequency, CFQ-R (respiratory domain), weight, and sweat chloride concentrations at weeks 24 and 48 compared with the placebo group. In the ivacaftor group compared with the placebo group, the change in mean FEV1 was +10.5% predicted at week 48 (p < 0.001), with a 55% relative reduction in the risk of pulmonary exacerbation (p < 0.001), a change in mean weight of +2.7 kg at week 48 (p < 0.001), and change in mean sweat chloride values of −48.1 mMol (p < 0.001). The CFQ-R respiratory domain scores also improved in the ivacaftor group above the Minimal Clinically Important Difference (p < 0.001). Ivacaftor was generally well tolerated, with a similar frequency of AEs in the treatment and placebo groups [Ramsey et al. 2011].
Results have recently been published for the second study, a similar phase III study evaluating the efficacy and safety of ivacaftor monotherapy in patients with CF patients aged 6–11 years possessing the G551D CFTR mutation [Davies et al. 2013]. In this study, a total of 52 patients were randomized 1:1 to either ivacaftor treatment (150 mg every 12 h) or placebo. Interestingly, despite ‘normal’ baseline values in FEV1 % predicted (mean FEV1 = 84.7% and 83.7% in the ivacaftor and placebo groups respectively), patients receiving ivacaftor had a significant increase in FEV1 % predicted through 24 weeks versus placebo (treatment effect of +12.5%; p < 0.001). The effect on pulmonary function was evident by 2 weeks and was maintained through 48 weeks. Patients treated with ivacaftor gained an average 2.8 kg above the placebo group at week 48 (p < 0.001). The change in sweat chloride concentration through week 48 was −53.5 mMol (p < 0.001) for the ivacaftor group compared with placebo, while the incidence of AEs was similar between the two groups. The results of these studies led to US Food and Drug Administration approval of ivacaftor monotherapy (150 mg every 12 h) for the treatment of patients with CF aged over 6 years with the G551D CFTR mutation. This approval is independent of lung function or other measures of disease state and has been followed by similar approvals in other countries.
Currently, there are at least four studies evaluating ivacaftor monotherapy in new disease populations. One is examining ivacaftor in patients with CF between the ages of 2 and 5 years (‘A Phase 3, 2-Part, Open-Label Study to Evaluate the Safety, Pharmacokinetics and Pharmacodynamics of Ivacaftor in Subjects With Cystic Fibrosis Who Are 2 Through 5 Years of Age and Have a CFTR Gating Mutation’) [ClinicalTrials.gov identifier: NCT01705145]. The second is examining ivacaftor monotherapy in patients with CF with non-G551D gating mutations (KONNECTION trial: ‘A Phase 3, Two-Part, Randomized, Double-Blind, Placebo-Controlled, Crossover Study With an Open-Label Period to Evaluate the Efficacy and Safety of Ivacaftor in Subjects With Cystic Fibrosis Who Have a Non-G551D CFTR Gating Mutation’) [ClinicalTrials.gov identifier: NCT01614470]. The rationale for this study is provided by in vitro studies demonstrating that ivacaftor increases the activity of several CFTR gating mutations in addition to G551D [Yu et al. 2012]. The third study is testing the efficacy of ivacaftor monotherapy in patients with CF possessing the R117H CFTR mutation (KONDUCT trial: ‘A Phase 3, Randomized, Double-Blind, Placebo-Controlled, Parallel-Group Study to Evaluate the Efficacy and Safety of Ivacaftor in Subjects With Cystic Fibrosis Who Have the R117H-CFTR Mutation’) [ClinicalTrials.gov identifier: NCT01614457]. R117H is typically considered primarily a class IV mutation, characterized by reduced chloride channel conductance. This study represents an extension of ivacaftor monotherapy outside of mutations characterized primarily by gating abnormalities. Finally, ivacaftor monotherapy is being studied in patients with CF with residual-function CFTR mutations (‘A Pilot Study Testing the Effect of Ivacaftor on Lung Function in Subjects With Cystic Fibrosis, Residual CFTR Function, and FEV1 ≥40% Predicted’) [ClinicalTrials.gov identifier: NCT01685801]. This study is enrolling patients with a wide variety of surface-localized CFTR mutations (including patients with splicing mutations) and is utilizing a novel design that includes repeated periods of active drug and placebo treatment.
Conclusion
Preclinical studies and several clinical trials have shown that ivacaftor activates G551D CFTR. Ivacaftor treatment leads to statistically significant improvements in lung function (FEV1 % predicted), decreased pulmonary exacerbation rates, increased weight, and reduced sweat chloride values in patients with CF patients harboring the G551D CFTR mutation. The effects on clinically relevant outcomes were sustained up to 48 weeks in patients with minimal or advanced disease, and these effects were not seen in patients with the F508del homozygous mutation. Side effects of ivacaftor have been consistent with placebo-treated patients and with underlying CF disease manifestations. Work in progress is attempting to apply this novel therapeutic to new CF populations, including younger patients, those with non-G551D gating mutations or the R117H CFTR mutation, and patients with residual CFTR function. Ivacaftor has the potential to fundamentally change the way in which CF patients with drug-responsive mutations are treated, and leads the way in bringing genotype and patient-specific CFTR modulators to the general CF population.
Footnotes
Acknowledgements
The authors are grateful to J. Denise Wetzel, CCHMC Medical Writer, for critical review of the manuscript.
Funding
This review article received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflict of interest statement
Dr Clancy’s institution has received research funding from Vertex Pharmaceuticals for the conduct of CF clinical trials. Dr Clancy has received funding from Vertex Pharmaceuticals for participation on the Vertex Global CF Scientific Advisory Board and to provide educational talks regarding the use of ivacaftor in CF clinical care.
