VX-561

An update on new and emerging therapies for cystic fibrosis

1. Background
1.1. Clinical manifestations and CFTR biology

Cystic fibrosis (CF) is a fatal genetic disease that can cause multiorgan dysfunction including recurrent sinopulmonary infections, male infertility, pancreatic insufficiency, bowel obstruction, and liver impairment [1,2]. It is the most common autosomal recessive disorder leading to premature death in Caucasians [3]. It is typically diagnosed in infancy, often before the onset of symptoms, through newborn screening programs that have been adopted across the United States, Australia, United Kingdom, and much of Europe [4]. CF is caused by mutations in both alleles of the cystic fibrosis transmembrane conductance regulator (CFTR) gene, which encodes the CFTR protein. Disease severity is partially dictated by patient-specific CFTR mutations, which influence the level of functional CFTR, as well as other factors including genetic modifiers, environ- mental exposures, nutrition, and adherence to therapies [3,5]. CFTR is an anion channel that transports chloride and bicarbonate across the plasma membrane of several epithelia [6,7]. CFTR is also a regulator of epithelial sodium channels (ENaC) and other chloride channels [8]. Dysfunctional CFTR in airway epithelia causes increased sodium resorption via ENaC, dehydration of the air surface liquid (ASL), and impaired mucociliary clearance [9]. There are some data suggesting that humans with CF have decreased ASL pH in vivo [10]. Results from CF pigs indicate that this impairs pH-dependent bacterial killing and interrupts normal mucus production and release that contributes to mucus stasis [11,12]. CFTR is expressed in many cell types including epithelial cells that line the respiratory tract and submucosal glands, the GI tract, the pancreatic duct, biliary tree, sweat gland duct, and immune cells, resulting in clinical manifestations across numerous affected organs [13–15].

Most morbidity and mortality in CF is due to respiratory disease, which occurs as a result of mucus accumulation that leads to infection and excessive inflammation, culminating in airway remodeling and bronchiectasis [1,16]. Many of these pulmonary manifestations are targets of established and emerging therapies and will be the focus of this review. Other critical contributors to CF morbidity and mortality include pancreatic insufficiency (90%+ of patients), CF-related diabetes (−35% of CF adults), CF-related liver disease (~5% of patients), infertility (98% of CF males), and pulmonary infec- tions with difficult to treat pathogens [17–25]. These are cur- rently managed through a variety of traditional symptom- based therapies, and while they are important contributors to improvement in overall CF morbidity and mortality, they are not addressed in this review, which will focus on new and emerging pulmonary and systemic modulator therapies [26–29].

1.2. Mutations in CFTR leading to disease

To date, >1900 mutations in CFTR have been identified and >280 variants are known to cause CF disease. The extent and severity of CF disease is in part attributed to the nature and severity of the CFTR defect and can be the result of insertions, deletions, and duplications [30,31]. These mutations are grouped into genotypic classes with similar characteristics (Classes I–VI) [31,32]. Class I mutations, e.g. R553X, produce biosynthetic defects (such as premature termination codons [PTCs] or frameshift mutations) that result in little if any func- tional CFTR protein expression (Table 1), often causing severe disease [31,33]. Class II mutations cause protein folding and trafficking defects; F508del CFTR is a member of this class and is found in ≥85% of CF patients. F508del produces an in-frame deletion of a ‘CTT’ codon in exon 10, which results in loss of a phenylalanine at position 508. The resultant protein has abnormal folding, fails to mature and reach the plasma mem- brane, and is subsequently degraded in the proteosome [32,34,35]. Class III mutations have defective channel gating; G551D CFTR is the most common representative of this class and is found in ~3–5% of CF patients. These mutations have normal levels in the plasma membrane, but there is a severely reduced open channel probability (impaired gating – described in Table 1) [32,36]. Class IV mutations have reduced single channel conductance, with normal protein levels at the plasma membrane, but anion permeation through the chan- nel pore is restricted. The most commonly described CFTR mutation with decreased conductance is R117H CFTR [37]. Class V mutations typically have splice abnormalities that reduce the amount of functional CFTR at the plasma mem- brane. An example of this is the 2789+5G-A mutation [38,39]. Class VI mutations have increased turnover and reduced time at the plasma membrane [40]. It is important to note that these defects are not mutually exclusive, and that a given mutation can demonstrate abnormalities across >1 mutation class. The newest drugs on the market, CFTR modulators, target-specific CFTR protein defects such as protein folding/ trafficking or gating [31]. To date, FDA approval for use of these modulators is specific to CFTR mutations shown to be effective in clinical trials, and recently for patients with extre- mely rare mutations based primarily on in vitro data [41].

2. Medical need
2.1. Life limitations due to CF

Despite improvements in CF care, patients have limitations in their day-to-day activities and a limited lifespan (survival in the United States is approximately half of that seen in the non-CF population) [17]. In a recent study in France, more than half of the 200 CF participants reported that they were limited in their ability to do their job, and two-thirds felt that CF limited them from having their desired career [42]. In the US Cystic Fibrosis Foundation (CFF) patient registry, 11% also required use of supplemental oxygen to function in their daily lives [17]. Thus, the impact of CF on longevity and quality of life (QOL) is significant. Lung transplant is an option for many CF patients with end-stage lung disease, which improves QOL, but 5-year survival is 50–60%, with many recipients develop- ing transplant-related complications (e.g. bronchiolitis obliter- ans) [43–45]. There are large differences in life expectancy between different countries with a recent analysis estimating the median age of survival in Canada was 50.9 years and in the United States was 40.6 years, which may be attributable to ‘differences in health-care systems.’ [46] Comparisons of coun- try-specific lifespans are challenging because of numerous external factors and the variety of data collection methods chosen in reports [47].

3. Emerging CFTR modulator treatment
3.1. Ivacaftor

A breakthrough in CF therapeutics occurred in 2011 with the results of the trial of ivacaftor (VX-770) in people with CF who possessed a G551D CFTR mutation (impaired channel gating). This subsequently led to FDA approval of ivacaftor in 2012. Ivacaftor is a potentiator of the CFTR channel, which means that it enhances the time that the channel spends in the open configuration. In vitro preclinical studies showed that ivacaftor increased the amount of chloride transport (described in Table 1) in G551D airway epithelial cells to 50% of that seen in healthy controls [48]. Ivacaftor requires CFTR to be present in the plasma membrane for its activity. Consistent with this, ivacaftor produced small and variable improvements in chlor- ide transport in airway cells from F508del/F508del CFTR donors, and a phase II clinical trial in F508del homozygous CF adults demonstrated minimal improvement in a pulmonary function test (PFT), the forced expiratory volume in 1 s (FEV1), or sweat chloride [48,49]. Table 2 describes the FEV1 (percent predicted) and sweat chloride tests, both commonly assessed in clinical care and studies of CF patients [50,51].

Trials supporting regulatory approval of ivacaftor for CF patients with a G551D mutation came from a phase II safety and dosing trial followed by a phase III randomized, double- blind, placebo-controlled study of 162 people ≥12 years old with stable disease dosed with oral ivacaftor (150 mg every 12 h) or placebo for 48 weeks [60,61]. The study demonstrated improvement in important clinical end points including a >10% absolute mean change in FEV1 (or 0.367 L) from baseline. Ivacaftor reduced the risk of pulmonary exacerbations by 55%,
improved QOL scores (key clinical end points – Table 2), and increased weight, with an average of 2.7 kg gained relative to placebo. The mean sweat chloride concentration of the ivacaf- tor-treated cohort dropped below the diagnostic threshold for CF, while values in the placebo group remained unchanged [60,61]. There was no concerning adverse event (AE) profile identified, but monitoring of liver function tests is recom- mended [61]. A subsequent randomized, double-blind, pla- cebo-controlled trial of 52 children aged 6–11 years old with a G551D mutation also demonstrated dramatic improvements with ivacaftor [62]. The starting mean FEV1 was in the normal range (83–84%) across the two groups, which was higher than that seen in the aforementioned study of older patients. Ivacaftor treatment produced improvements in FEV1 >12%, weight increase, and sweat chloride reductions over the study period [62,63]. Encouraging in vitro data prompted placebo- controlled trials that have demonstrated efficacy and helped to extend ivacaftor monotherapy to new populations, including patients with rare gating mutations (G178R, S549N, S549R, G551S, G1244E, S1251N, S1255P, or G1349D) and the R117H mutation (with both conductance and gating defects) [64,65]. A subsequent observational study has provided evidence that prolonged ivacaftor treatment (3 years) slows the rate of lung function decline compared with observational controls and also decreases the risk of pulmonary exacerbation [66].
In May of 2017, the FDA expanded approval of ivacaftor monotherapy to CF patients >age 2 years possessing 1 of an additional 23 residual function CFTR mutations likely to respond to ivacaftor [41]. This approval is unique within CFTR modulator development, including both in vitro and in vivo data in the decision process. This approval provides an additional 900 CF patients with extremely rare CF-causing mutations in the United States with potential access to CFTR modulator therapy and in many ways is groundbreaking as developing a pivotal clinical trial for this heterogeneous patient group is extremely challenging [67]. Subsequently, FDA approval for ivacaftor extended to CF patients ≥2 years old with a residual function mutation due to one of five specific splicing defects based on clinical data [68]. A phase III open-label trial is evaluating ivacaftor in ~35 children <24 months of age with gating mutations (NCT02725567). 3.2. Lumacaftor–ivacaftor The biology of the F508del CFTR mutation includes protein misfolding that interrupts its maturation, leading to endoplas- mic reticulum-associated degradation in the 26S proteosome. The result is little (if any) mature protein detectable in the plasma membrane. Prior work also indicated that F508del CFTR in the plasma membrane has additional deficits, includ- ing reduced gating and plasma membrane residence time [48,69,70]. Lumacaftor (VX-809), a CFTR corrector, was devel- oped to increase the amount of F508del CFTR that reaches the cell surface [71]. Preclinical studies in human airway epithelia derived from F508del/F508del CF patients showed that luma- caftor increased the amount of mature F508del CFTR and improved chloride secretion to approximately 15% of normal, and the addition of ivacaftor enhanced its function approxi- mately twofold [71,72]. A phase II randomized, double-blind, placebo-controlled study of lumacaftor monotherapy assessed safety, tolerability, and pharmacodynamics of increasing doses in 89 subjects homozygous for F508del CFTR. Median sweat chloride level for all subjects at the start of the study was 103.5 mmol/L and in subjects who received the highest luma- caftor dose, there was a mean decrease in sweat chloride of −8.2 mmol/L. While the study failed to demonstrate efficacy at any of the studied doses, it did provide evidence for biologic activity that set the stage for subsequent definitive trials [73]. A phase II double-blind, placebo-controlled multicohort trial evaluated escalating dosages of lumacaftor combined with iva- caftor in 160 F508del CFTR homozygous and 28 heterozygous adult patients [74]. Subjects received lumacaftor monotherapy for up to 28 days followed by lumacaftor once or twice daily combined with twice daily ivacaftor for up to an additional 28 days. The primary end points were changed in sweat chloride and safety in the treatment groups during the combination drug interval with secondary analyses comparing changes to the pla- cebo group. There was a significant decrease in sweat chloride for homozygous subjects receiving lumacaftor at the end of monotherapy period (28 days). No difference was observed in subjects during combination therapy interval (day 28 compared to day 56) within in the treatment group, which was the primary end point, or compared to placebo. There was a significant decrease in sweat chloride within the treatment group between day 1 and day 56 in homozygous subjects who received the highest daily dose of lumacaftor combined with ivacaftor (−8.9 mmol/L) or twice daily dosing of lumacaftor combined with ivacaftor (−10.3 mmol/L). Significant decreases in sweat chloride were also found in the aforementioned groups compared to placebo. There was no significant decrease in sweat chloride in heterozygote subjects who received lumacaftor–iva- caftor during any interval. The FEV1 did not change or decreased in several groups during the monotherapy interval. During com- bination therapy, the mean absolute change in FEV1 significantly increased by 7.7% for lumacaftor–ivacaftor groups compared to the placebo group. The group who received high dose lumacaf- tor combined with ivacaftor had a 5.6% increase in FEV1 between days 1 and 56 compared to placebo. No positive changes in FEV1 were seen in heterozygotes with lumacaftor–ivacaftor therapy. Twelve percent of subjects experienced chest tightness or short- ness of breath during lumacaftor monotherapy compared to none in the placebo group [74]. The combination of lumacaftor and ivacaftor was tested in two important phase III, randomized, double-blind, placebo- controlled trials (TRAFFIC and TRANSPORT) that enrolled >1100 F508del/F508del subjects (≥12 years of age) [75]. In the pooled analyses, subjects receiving active drug were 30–39% less likely to experience a pulmonary exacerbation, and when exacerbations occurred, the active treatment groups were significantly less likely to require hospitalization or IV antibiotics. Over the 24-week study, the mean percent absolute change in FEV1 from baseline was +3.3% in the lumacaftor–ivacaftor groups compared to placebo, with a mean relative increase up to 5.6% versus placebo. Similar numbers of AEs were observed in all groups, but there were more serious AEs (due to elevated liver function tests) in the lumacaftor–ivacaftor groups. There were also more reports of chest tightness, denoted as ‘respiration abnormal,’ in the lumacaftor–ivacaftor group (~11%) compared to in the pla- cebo group (~6%) [75]. The reasons for lower efficacy in the TRAFFIC/TRANSPORT studies relative to ivacaftor monotherapy in patients with gating mutations are likely due to the respon- siveness of the underlying mutations to the different interven- tions, and possibly negative interactions between ivacaftor and lumacaftor that have been proposed in in vitro studies [76–79]. The FDA approved the combination of lumacaftor– ivacaftor for patients homozygous for F508del CFTR ≥12 years of age in 2015. FDA approval was recently extended to include 6–11 year olds as an open-label trial in this age group (n = 58) demonstrated safety, reduced sweat chloride (LS mean = −24.8 mmol/L), and improved lung function as mon- itored by lung clearance index (LCI) (Table 2) [80]. A phase III, randomized placebo-controlled trial of 206 homozygous chil- dren also found that 6–11 year olds who received lumacaftor– ivacaftor for 24 weeks had improved lung function (measured by LCI and FEV1) and decreased sweat chloride [81]. A phase III open-label study is evaluating the safety and pharmacoki- netics of lumacaftor–ivacaftor in ~60 children ages 2–5 years old who are homozygous for F508del CFTR (NCT02797132). A subsequent observational study has provided evidence that prolonged lumacaftor–ivacaftor treatment (3 years) slows the rate of lung function decline and the risk of pulmonary exacerbation in patients ≥12 years old homozygous for the F508del
mutation compared with matched controls from the CFF patient registry [82].

4. Market review

CF is most common in Caucasians, with a carrier prevalence of approximately 1:30 (United States) and a disease prevalence of approximately 1:3000 live births. Although estimates are sig- nificantly lower in non-Caucasians, cases of CF have been reported in all races. In 2015, greater than 28,000 people with CF in the United States had their data contributed to the CFF patient registry, which serves as an annually updated source of population demographics, care practices, and out- comes data that inform clinicians and researchers. Amongst non-Caucasian registry participants, 4.6% were African American, 3.3% other race, and 8.5% were of Hispanic ethni- city [17]. In the United States, approximately 1000 people are diagnosed each year with CF and the total number of people with CF is estimated to be ~33,000 [83]. After the United States, the greatest number of cases occur in Canada and Western Europe. According to the World Health Organization, CF occurs in 1:2000–1:3000 live births in the EU and is rare in Asia [84]. There are an estimated 70,000 cases of CF worldwide and the number of new cases has risen every year, in part due to expanded access to genetic testing. In Europe, the number of people with CF is projected to increase 50% by the year 2025 [85]. Men and women are equally likely to have CF, but the median age of survival in several national registries is lower for females than males [86,87].

People with CF in the United States received health-care coverage from a variety of sources (60% covered by private insurers, 45% covered by Medicare, and 10% by Medicaid) at any given time during 2015 [17]. Approximately, 1% of adults included in the CFF registry reported no insurance, although that number may be higher as access to the registry is linked to interfacing with the health-care system [17]. The costs of CF care are substantial and are increasing with the advent of recent modulator therapies. While the cost of modulators can exceed $300,000/year in the United States, health-care systems in other countries have negotiated lower prices [88,89]. The CF drug market in 2014 (prior to the introduction of lumacaftor–ivacaftor) was estimated to be worth $1.6 bil- lion, with annual growth at that time of ~23% [90].
Costs associated with CF care go beyond access to newer drugs and include the expense of other established therapies and treatment of pulmonary exacerbations. One cost analysis of CF pulmonary exacerbations in the United States, from a single large insurance carrier, reported the mean cost per exacerbation was >$12,000, which increased to >$36,000 per exacerbation if IV antibiotics were admi- nistered [91]. In that retrospective study of 241 patients >6 years of age (2008–2013), the average number of pul- monary exacerbations was 2.9 exacerbations/person/year [91]. These costs increase with disease severity, including additional costs as CF patients become disabled and require more support to provide for themselves and their families. Finally, as most people with CF die before the age of 50, there are significant years of productive life lost that compound the societal cost estimates of the disease [92].

5. Current research goals
5.1. Outcome measures in CF clinical trials

Table 2 describes the most common primary efficacy outcome measures used in CF therapeutic trials including improvement in lung function (typically FEV1), reduction in pulmonary exacerbations, LCI, improving growth, and improving QOL- measured by the cystic fibrosis questionnaire-revised (CFQ-R) [93]. Each of these relates in some way to how a patient ‘feels, functions, or survives’ (important regulatory considerations for clinical efficacy measures). Important secondary and suppor- tive outcome measures depend somewhat on the nature of the intervention and can include diverse measures such as reduction in sweat chloride concentration (e.g. CFTR modula- tors), reduction in sputum bacterial density (e.g. inhaled anti- biotics), or reduction in sputum or blood biomarkers of inflammation (e.g. anti-inflammatories). There are numerous other biomarkers used in CF drug development, which have recently been reviewed in a series of related articles [94–96].

5.2. Curative interventions

Currently, there is no cure for CF. Established therapies target the downstream manifestations of absent CFTR function, while CFTR modulators are the first therapies to address the cause of CF by restoring mutant CFTR protein function. In addition, a recent pulmonary gene transfer study demonstrated modest clinical benefit [97]. It is hoped that future improvements in stem-cell technologies, including the use of induced pluripotent stem cells, and/or genetic editing strategies, such as clustered regu- larly interspaced short palindromic repeats (CRISPR) coupled with CRISPR-associated genes (Cas) nucleases, will provide the necessary tools to bring researchers closer to a cure [98–101].

6. Scientific rationale for CFTR-targeted therapies

The highly effective clinical benefits produced by ivacaftor in patients with gating mutations, and to a lesser extent luma- caftor–ivacaftor in F508del CFTR homozygous patients, pro- vide a strong rationale for therapeutic strategies to improve mutant CFTR function. This has led to an explosion in CFTR- focused drug development, and numerous candidate thera- pies have entered clinical trials.

7. Competitive environment

Table 3 summarizes all of the CFTR modulators in clinical development per clinical trials.gov, as of April 2017, which are discussed in detail below.

7.1. Ataluren (PTC124)

Ataluren was designed to treat CF caused by CFTR muta- tions that result from single nucleotide changes producing an in-frame PTC. This leads to rapid mRNA degradation through nonsense-mediated decay and/or translation of truncated, nonfunctional CFTR protein [102,103]. Ataluren is thought to allow the transcriptional machinery to ‘read through’ the PTC by allowing the insertion of a near cognate tRNA, resulting in generation of some full-length protein [104]. A similar mechanism of action has been demonstrated for some aminoglycoside antibiotics. Suppression of PTCs and enhanced CFTR expression were observed in gentami- cin-treated airway epithelial cells and in CF patients with PTC-mediated CF [105]. Phase II studies of ataluren provided some evidence of restored CFTR function based on lung function, nasal potential difference measurements (NPD), and cough symptoms but were complicated by the absence of placebo groups or changes in sweat chloride [106,107].

The results of a randomized, double-blind, placebo-con- trolled phase III trial of ataluren in patients ≥6 years of age with PTC-mediated CF found no difference in FEV1 or any secondary end points, including NPD (NCT00803205) [108]. A post-hoc analysis, which excluded subjects who were receiv- ing chronic inhaled tobramycin, showed a mean relative change in FEV1 of +5.7% from baseline and a reduction in pulmonary exacerbation events (40%) in the ataluren group compared to placebo. The authors proposed that aminogly- cosides interfered with ataluren’s mechanism of action by disrupting ribosomal read through. There was an increase in acute kidney injury in the ataluren-treated group (15% in drug and <1% in placebo) [108]. A subsequent phase III multicenter, randomized, double-blind, placebo-controlled trial in patients not receiving inhaled aminoglycoside ther- apy failed to meet its primary (FEV1) and secondary (pul- monary exacerbation) end points and the sponsor (PTC Therapeutics) chose not pursue approval for ataluren to treat CF caused by PTCs (NCT02139306) [109]. 7.2. Cavosonstat (N91115) This drug is an inhibitor of S-nitroglutathione reductase (GSNOR), which is thought to reduce degradation of F508del CFTR protein and help improve its stability [110]. In primary human bronchial epithelial cells (HBEs), GSNOR expression has been reported to be elevated in cells from people homozy- gous for F508del CFTR compared to non-CF donors in vitro. Furthermore, siRNA knockdown of GSNOR in mutant F508del CFTR CFBE41o− cells increased the expression of both mature and immature forms of CFTR protein [111]. The proposed mechanism of action for cavosonstat is that by blocking GSNOR activity, S-nitrosoglutathione (GSNO) levels increase, which in turn promotes nitrosation of the chaperone proteins Hsp70/Hsp90 organizing protein (HOP). It has been suggested that this modification of HOP makes it less likely to associate with F508del CFTR, thereby reducing CFTR degradation [110,111]. Phase I studies (SNO-2 and SNO-5) in healthy adult subjects examined safety, tolerability, pharmacokinetics, food effect, and impact of co-therapy with a CYP3A4 inducer (NCT02013388 and NCT02500667) [110,112]. These results prompted trials of cavosonstat in adults homozygous for F508del CFTR. The first was a phase Ib, double-blind, rando- mized, placebo-controlled, dose-ranging study (SNO-4) asses- sing pharmacokinetics, safety, and tolerability as well as exploratory end points including FEV1, sweat chloride, and candidate biomarkers (NCT02275936). The 51 subjects with CF received oral twice a day cavosonstat or placebo for 28 days followed by a 2-week washout period. There were three serious adverse events (SAEs) related to pulmonary infections, which were not considered related to the study drug. There were AEs in all groups, with a greater number [66] in the 200-mg group than in the placebo group [40], most of which were mild to moderate. There was no change in FEV1 from baseline within the treatment groups, but there was a statistically significant, albeit small, within-dose group reduction in mean sweat chloride concentration (−4.1 mmol/L) in the 200-mg twice a day cavosonstat group compared to pre- treatment [110]. Two phase II studies of cavosonstat have recently been completed in CF patients receiving treatment with other CFTR modulators. The first (SNO-6) was a rando- mized, double-blind, placebo-controlled, study evaluating cavosonstat added to lumacaftor–ivacaftor therapy compared to lumacaftor–ivacaftor alone in adults homozygous for F508del CFTR in 138 adults (NCT02589236). No clear improve- ment in lung function or change in sweat chloride was detected, but there was a statistically significant mean increase in BMI of 0.17 kg/m2 in the cavosonstat arm [113]. The second phase II study (SNO-7) assessed the efficacy and safety of cavosonstat and ivacaftor versus ivacaftor monother- apy in 19 adults with one F508del CFTR and 1 gating mutation responsive to ivacaftor (NCT02724527). Similar to SNO-6, there were no clear benefits observed on lung function or sweat chloride in the cavosonstat-treated arm, but there was a mean increase in BMI of 0.2 kg/m2, although statistical analysis was not provided [110,114]. Based on the results of these studies, the company (Nivalis) has halted development of cavosonstat to modulate CFTR [115]. 7.3. CTP-656 CTP-656 is a CFTR potentiator, generated by chemical modifi- cation of ivacaftor with deuterium. It is under development for CF patients with Class III gating mutations. Deuteration replaces carbon–hydrogen bonds with stronger carbon–deu- terium bonds, which has been shown to slow metabolism and alter accumulation of toxic metabolites for some drugs [116]. CTP-656 was developed to provide the benefits of ivacaftor, currently given twice a day, in a once daily dosing form. Two phase I trials of CTP-656 were completed in 2016 [117]. The first was a crossover comparison of a single dose of ivacaftor versus CTP-656 followed by a double-blind, placebo-controlled ascending dose study in subjects who received drug or pla- cebo for 7 days (NCT02599792). AEs, safety, tolerability, and pharmacokinetics were assessed in 37 adult subjects. The second phase I study of 15 adult males evaluated food effects on the bioavailability, pharmacokinetics, and safety after a single dose of CTP-656 (NCT02680249). CTP-656 had a slower clearance compared to ivacaftor and pharmacokinetics sug- gesting it could be evaluated in subsequent trials with once daily dosing [117]. A phase II randomized, double-blind, pla- cebo-controlled trial began in December of 2016 evaluating pharmacokinetics and the various doses of CTP-656 versus placebo and open-label ivacaftor on sweat chloride, as well as secondary end points such as FEV1 and QOL (NCT02971839). This study is sponsored by Concert Pharmaceuticals and has a planned enrollment of 40 adult CF subjects who have been stable on ivacaftor for at least 3 months and possess at least 1 of the following mutations: G551D, G1244E, G1349D, G178R, G551S, S1251N, S1255P, S549N, and S549R. 7.4. FDL169 FDL169 is a small molecule under development by Flatley Discovery Labs to correct F508del CFTR. In vitro studies in primary HBEs show that FDL169 can improve F508del CFTR activity in the presence of ivacaftor and cAMP-elevation (forskolin) [118]. No further increase in F508del CFTR activity was observed in cells exposed to FDL-169 combined with lumacaftor in vitro, and FDL169 did not improve open channel probability. These studies suggested less loss of F508del CFTR activity in cells treated with FDL-169 com- bined with ivacaftor for >24 h, compared with lumacaftor combined with ivacaftor [118,119]. Additional in vitro stu- dies suggest that FDL169 alone can increase CFTR matura- tion and cell surface expression in F508del HBEs, although quantification was not provided [120]. Three phase I trials of FDL-169 were recently completed. The first was a phase I, randomized, double-blind, placebo-controlled, dose-escala- tion trial to determine the safety, tolerability, and pharma- cokinetics in healthy adult males, including food effects (NCT02359357). An additional phase I nonrandomized, open label, dose escalation study evaluated pharmacoki- netics, safety, and tolerability in eight adult female subjects (NCT02680418). Another phase Ib nonrandomized, open- label study evaluated bioavailability of different formula- tions of FDL169 and pharmacokinetics in two cohorts of healthy male and female adults, and a third cohort of male and female patients with CF (not limited to specific CFTR mutations) (NCT02767297) [118]. Results were not accessible for these phase I studies. A subsequent phase I randomized, double-blind, placebo-controlled, parallel study of 24 adults with CF homozygous for F508del CFTR will evaluate safety and pharmacokinetics of multiple doses of FDL 169 and reportedly begin in mid-2017 (NCT03093714). According to the CFF website, a phase IIa study to evaluate safety of FDL169 in F508del CFTR is also planned to begin in Europe [121].

7.5. GLPG1837

This is a candidate CFTR potentiator targeted for patients with Class III CFTR mutations. The first phase I randomized, double- blind, placebo-controlled trial included administration of sin- gle doses or multiple ascending doses for 14 days to healthy adults to evaluate safety, tolerability, pharmacokinetics as well as how food and CYP3A induction affected these parameters (NCT02325037). This first-in-human trial of GLPG1837 planned to enroll 64 people and concluded in 2015. Two subsequent open-label phase I studies of GLPG1837 in healthy adult males were completed in 2015. One evaluated the oral bioavailability of two preparations (oral suspension and oral tablet) and food effects in 12 subjects, and the second study examined drug– drug interactions between GLPG1837 and midazolam in 24 subjects (NCT02562859 and NCT02562950). No results from these trials were available. Two phase II trials of GLPG1837 were concluded in late 2016. The first was a phase IIa open label, single group trial (SAPHIRA I) of 26 adults with CF with at least 1 G551D mutation who received 1 month of twice-daily oral GLPG1837 (NCT02707562). Recent topline results were reported on the Galapagos’ website [122]. In this trial, 25 of the 26 subjects on ivacaftor were required to stop ivacaftor treatment for 39 days, including a 7-day washout period at the start of the trial. The primary outcomes were safety and toler- ability, with secondary outcomes of change in sweat chloride, FEV1, and pharmacokinetics. Two subjects dropped out between day 15 and day 29, 1 for elevated creatine phospho- kinase (CPK) and 1 for patient’s decision, leaving 24 subjects to complete the trial (69% had F508del as their second CFTR mutation). The company reported a statistically significant reduction in subjects’ sweat chloride from a mean of 99 mmol/L after ivacaftor washout (pre-GLPG1837) to 66 mmol/L on ~day 22. Summary graphs of the data showed that FEV1 pre-ivacaftor washout was 70–75%, decreased to <70% during the ivacaftor washout, and then returned to approximately baseline FEV1 on day 29 of GLPG1837 treat- ment. The number of subjects contributing to these results was not cited, and statistical analysis was not included. There were three SAEs reported in two subjects including distal intestinal obstruction syndrome during screening, an increase in CPK mid-trial, and a pulmonary exacerbation resulting in hospitalization at the end of trial [122]. A small phase IIa open- label study (SAPHIRA II) included six subjects with the S1251N Class III CFTR mutation who received twice-daily oral GLPG1837 at one dose for 2 weeks, followed by a higher dose for two additional weeks (NCT02690519). Preliminary results indicated no SAEs, which was the primary end point. Four of the five subjects had a decrease in sweat chloride of ≥15 mmol/L and one subject had a >50 mmol/L decrease on day 29 of receiving GLPG1837 [123,124].

7.6. GLPG2222

GLPG2222 (also known as ABBV/GLPG2222) is a candidate corrector for use in people with F508del CFTR. In vitro studies suggest that GLPG2222 increases CFTR cell surface expression in CFBE41o− cells, as HRP-tagged F508del CFTR. A phase I randomized, double-blind, placebo-controlled study of a sin- gle dose and then 2 weeks of ascending doses of GLPG2222 was completed in 2016 (NCT02662452). This first-in-human trial examined safety, tolerability, pharmacokinetics, and potential CYP3A4 interactions in 40 healthy adult males. Preliminary data reported did not reveal any SAEs and pro- vided favorable pharmacokinetics [125]. An ongoing phase II randomized, double-blind, placebo-controlled trial (ALBATROSS) in subjects with one F508del CFTR mutation and one Class III mutation, already receiving ivacaftor, will primarily evaluate safety and tolerability as well as secondary end points of sweat chloride, FEV1, and QOL. Subjects will receive GLPG2222 + ivacaftor or placebo + ivacaftor for 1 month (NCT03045523).

7.7. GLPG2451

A phase I study of a new CFTR potentiator compound, GLPG2451, alone as well as in combination with GLPG222 is reported to be recruiting subjects for a randomized, double- blind, placebo-controlled trial (NCT02788721) [126]. This study began in 2016 to assess safety, tolerability, and pharmacoki- netics of single doses as well as ascending doses in 56 healthy adult female subjects.

7.8. PTI-428

PTI-428 is candidate compound categorized as an amplifier, which is proposed to increase the amount of immature CFTR protein substrate. In vitro data reported by the company Proteostasis suggest that PTI-428 increases normal CFTR mRNA and chloride transport by CFTR in HBEs. It was also reported that when combined with ivacaftor or lumacaftor– ivacaftor, PTI-428 further augments in vitro chloride ion trans- port in HBEs from CF patients with several different genotypes, although a statistical analysis was not provided [127,128]. A phase I randomized, placebo-controlled, double-blind, cross- over study of single and multiple ascending doses of PTI-428 is anticipated to enroll approximately 88 healthy female con- trols, assessing safety, tolerability, pharmacokinetics, drug– drug interactions, and exploratory end points (NCT02846142). Preliminary results provided on the company’s website indicate no initial safety concerns in the PTI-428 doses tested to date, administered for up to 7 days. The results describe that some subjects receiving PTI-428 had an approxi- mately twofold increase in nasal CFTR mRNA, although the number of subjects included in these results and statistical analyses were not provided [128]. A phase I/II study of PTI-428 in subjects with CF began recruitment in 2016 (NCT02718495). This study differs from other CFTR modulator trials as it recruits CF patients independent of their CFTR mutation. It is designed to assess safety, tolerability, and pharmacokinetics of single and multiple ascending doses of PTI-428 in approxi- mately 36 adults, some of whom may be on lumacaftor– ivacaftor or ivacaftor. Preliminary results on the company’s website suggest no initial safety concerns, although the num- ber of subjects enrolled to date has not been reported [128].

7.9. QBW251

QBW251 is a candidate CFTR potentiator. Recent in vitro results describe that QBW251 combined with lumacaftor sus- tained F508del CFTR expression and function above lumacaf- tor alone [129]. A phase II trial of QBW251 to assess safety, tolerability, and pharmacokinetics of single and ascending doses was completed in 2015 (NCT02190604). This rando- mized, double-blind, placebo-controlled trial included 118 adult subjects in 4 cohorts, including healthy controls and CF patients who had at least 1 Class III–VI mutation and some patients homozygous for F508del CFTR [129]. This first-in- human study also evaluated FEV1, LCI, and QOL. Interim reports describe that in 22 CF patients with residual function mutations, those who received QBW251 every 12 h had an increase in adjusted mean FEV1 of 7.31%, compared to those who received placebo. They also reported lower doses of QBW251 decreased mean sweat chloride over placebo but only provided results from four treated subjects. No efficacy was reported in the 12 F508del CFTR homozygous subjects receiving QBW251. The report noted that the drug was well tolerated but does not provide details regarding AEs in the 40 CF subjects administered QBW251 [129].

7.10. QR-010

QR-010 is a single stranded, chemically modified RNA oligo- nucleotide being tested in people with CF caused by the F508del CFTR mutation. Its mechanism has not been well established, but it is a 33mer oligonucleotide that encodes a stretch of the wild-type human CFTR sequence that spans the region of F508del. It is postulated to restore the three nucleo- tides missing in the F508del mRNA, thereby providing a tem- plate for translation of a wild-type CFTR protein. Preclinical in vitro studies in HBEs show that QR-010 increased CFTR-specific current, and in F508del CFTR mice, there was improvement in CFTR activity by NPD (Table 2) and salivary secretion (another in vitro assay of CFTR function) [130]. The company (ProQR) also evaluated in vitro diffusion of QR-010 into a mucus layer from control and CF HBEs, with QR-010 penetration into both mucus substrates reported. Orotracheal administration of QR- 010 in transgenic mice with a phenotype of mucus obstruction and neutrophilic inflammation resulted in detectable systemic levels of QR-010 [131]. A phase Ia open label, nonrandomized, multicenter, exploratory study evaluating the effects of nasal administration of QR-010 on ENaC and CFTR by NPD in sub- jects with CF and one or two F508del CFTR mutations was completed in 2016 (NCT02564354). The study included 10 subjects who were homozygous and 8 who were heterozy- gous for F508del CFTR. After 4 weeks of QR-010 dosing (2.5 mg three times a week), subjects homozygous for F508del CFTR saw a significant increase in total chloride response with a mean change in NPD of −4.1 mV (p < 0.05), and a significant reduction in ENaC activity (p < 0.05). There was no measurable effect in subjects with a single F508del CFTR mutation [132,133]. A phase I/II randomized, double- blind, placebo-controlled, dose-escalation trial of oral inhala- tion of QR-010 in people homozygous for F508del CFTR is ongoing to test safety and tolerability (NCT02532764).Preliminary results from early dose cohorts (single ascending doses) reported no dose-limiting toxicity [133]. The second part of the phase IIb study testing multiple escalating doses (three times a week for four weeks) is ongoing. 7.11. Riociguat (BAY63-2521) This drug is a soluble cyclic (sc) guanosine 3',5'-monopho- sphate (GMP) stimulator that increases nitric oxide (NO). It is already FDA approved in the United States to treat subgroups of patients with WHO Class IV pulmonary hypertension due to chronic thromboembolic disease and is marketed under the name Adempas® by Bayer [134]. Preclinical studies suggest that raising cGMP increases CFTR expression on the cell sur- face of rat enterocytes and may improve other molecular abnormalities in CF [135,136]. There is an ongoing phase II randomized, double-blind, placebo-controlled trial to assess multiple dosages of riociguat in adults homozygous for F508del CFTR who are not on lumacaftor–ivacaftor (NCT02170025). A second cohort will enroll subjects receiving concomitant lumacaftor–ivacaftor. This study will evaluate safety, tolerability, and efficacy (e.g. change in sweat chloride, lung function) for approximately 42 subjects for up to 28 days and is anticipated to conclude in 2018. 7.12. Tezacaftor (VX-661) Tezacaftor is a CFTR corrector thought to increase the amount of functional F508del CFTR that reaches the cell surface [76]. When tezacaftor is combined with ivacaftor, additive in vitro effects are reported [137]. An open-label, nonrandomized, crossover, multidose phase I study of tezacaftor + ivacaftor + ciprofloxacin versus ivacaftor + ciprofloxacin in 34 healthy adults evaluated pharmacokinetics, safety, and tolerability (NCT02015507). This study was completed in 2014, but results have not been published. In a phase II randomized, double- blind, placebo-controlled study of ascending doses of tezacaf- tor monotherapy, study drug in combination with ivacaftor or placebo was tested in 128 people with CF who were ≥12 years old and homozygous for F508del CFTR. Another 18 people with F508del CFTR/G551D received tezacaftor in addition to ivacaftor compared with ivacaftor monotherapy. These studies primarily evaluated safety, tolerability, and changes in sweat chloride with additional secondary outcome measures of effi- cacy and pharmacokinetics. After 28 days of drug or placebo, within-group sweat chloride was modestly reduced in subjects homozygous for F508del CFTR who received tezacaftor mono- therapy or those dosed with tezacaftor and ivacaftor, the latter by −5.7 mmol/L (p < 0.05). The combination of tezacaftor and ivacaftor increased the mean absolute change in FEV1 by 4.8% (p = 0.01) compared to placebo, but there was not a signifi- cant increase in FEV1 in the tezacaftor monotherapy group. In the cohort with the F508del CFTR/G551D genotype, tezacaftor and ivacaftor reduced sweat chloride by −7 mmol/L (p = 0.053) and increased FEV1 by 4.6% (within-group comparisons, p = 0.012). The number of AEs was reported to be similar in the active drug and placebo groups [137,138]. Two phase II studies are recruiting adults homozygous for F508del CFTR, examining mucociliary clearance, gastrointestinal pH, sweat chloride, and CT scan imaging (NCT02508207 and NCT02730208). The largest trial to date of tezacaftor (EVOLVE) is a phase III trial completed in January 2017, which evaluated tezacaftor every 24 h and ivacaftor every 12 h versus placebo in approxi- mately 509 subjects with F508del/F508del ≥12 years of age (NCT02347657). The trial evaluated efficacy (changes in FEV1), and secondary end points including the number of pulmonary exacerbations, changes in weight and BMI, QOL, and changes in sweat chloride between baseline and week 24 [139]. Subjects treated with tezacaftor–ivacaftor had a mean abso- lute increase in their FEV1 of 3.4% from their baseline and the placebo group decreased by 0.6%, resulting in a mean treat- ment effect on FEV1 of 4.0% (p < 0.001). There was also a reduction in pulmonary exacerbations with 122 exacerbations occurring in the placebo group and 78 in the tezacaftor– ivacaftor group resulting in a hazard ratio of 0.65 (p = 0.005) for the treatment group. The safety data reported that most AEs in the trial were of mild or moderate severity for both the placebo and tezacaftor–ivacaftor group. Eighteen percent of the subjects in the placebo group and 12% of the subjects in the tezacaftor–ivacaftor group experienced SAEs, with approximately 3% of subjects in each group discontinuing study treatment due to AEs. Composite respiratory AEs (including wheezing, shortness of breath, and chest discom- fort) were 16% in placebo group and 13% in tezacaftor–iva- caftor group, with less chest tightness reported with tezacaftor–ivacaftor than observed with previously lumacaf- tor–ivacaftor [75,139,140]. The sponsor is seeking FDA approval for tezacaftor–ivacaftor for F508del homozygotes based on these results [139]. A phase III open-label trial of tezacaftor–ivacaftor is recruiting pediatric subjects 6–11 years old with F508del CFTR-mediated CF to evaluate safety, toler- ability, and pharmacokinetics (NCT02953314). A phase III randomized, double-blind, placebo-controlled, parallel group study of the tezacaftor–ivacaftor combination versus placebo in people ≥12 years old with CF with one copy of F508del CFTR mutation and a second mutation not likely to respond to tezacaftor and/or ivacaftor was completed in August 2016 (NCT02516410). A planned interim analysis of data from ~150 people showed no improvement in lung function and the study was stopped for futility [141]. The EXPAND trial is a phase III, double-blind, placebo-controlled, crossover study of CF patients ≥12 years old with an F508del CFTR mutation and a second CFTR mutation projected to have partial function (NCT02392234). Two hundred and thirty-four subjects received tezacaftor–ivacaftor, ivacaftor monotherapy, or placebo for 8 weeks, followed by a washout period and then crossover to a different intervention for eight additional weeks [142]. Investigators found that compared to placebo, the tezacaftor–ivacaftor group and the ivacaftor monotherapy group had mean absolute changes in FEV1 of 6.8% (p < 0.001) and 4.7% (p < 0.001), respectively, as well as improvements in CFQ-R of 11.1 (p < 0.001) and 9.7 (p < 0.001), respectively. When subjects received tezacaftor–ivacaftor, compared to iva- caftor monotherapy, there was a 2.1% (p < 0.001) absolute increase in FEV1. Safety results were reportedly similar to those in the EVOLVE trial [142]. An additional phase III randomized, double-blind study of tezacaftor–ivacaftor compared to ivacaftor monotherapy is enrolling CF patients ≥12 years old with an F508del mutation and a second gating mutation that is known to be responsive to ivacaftor (NCT02412111). 7.13. VX-440 VX-440 is a ‘next-generation’ corrector developed to provide additive F508del CFTR modulator activity with a proposed mechanism distinct from lumacaftor or tezacaftor [143]. In vitro VX-440 and VX-152 (described below) facilitated the processing and trafficking of F508del CFTR in CF HBEs, although no quanti- fication was provided. VX-440 or VX-152 monotherapy increased in vitro chloride transport in CF HBEs. The triple combination of tezacaftor, ivacaftor, and VX-440 or VX-152 reportedly increased chloride transport, even above that seen with lumacaftor–iva- caftor, although no statistical analysis was included [143]. The combination of VX-440, tezacaftor, and ivacaftor is being eval- uated in a phase II, randomized, double-blind, placebo-con- trolled trial of adults with CF who are heterozygous for F508del CFTR with a second CFTR mutation with minimal func- tion or F508del CFTR homozygotes (NCT02951182). Preliminary 4-week safety and efficacy results from subjects with F508del and a minimal function mutation were released on the Vertex website. Similar types of AEs were observed in drug and placebo groups, except two subjects receiving the triple drug developed elevated liver enzymes (6–8× upper limit of normal), which later normalized [144]. Data from 47 subjects receiving triple drug compared to triple placebo reported a mean absolute within- group change of up to 12% versus 1.4% in FEV1, −33.1 versus +1.6 mmol/L in sweat chloride, and 20.7 points versus 2.2 points in CFQ-R. Preliminary 4-week results for 26 F508del homozygous subjects reported a mean absolute within-group change in FEV1 of 9.5% and in sweat chloride −31.3 mmol/L for those who received VX-440 + tezacaftor + ivacaftor compared to 2.5% and +2.1 mmol/L for those who received placebo + tezacaftor + ivacaftor. Analysis of the safety data in F508del homozygotes was not yet reported. 7.14. VX-152 VX-152 is another next-generation corrector being evaluated in a similar fashion to and with reported comparable in vitro effects as VX-440 [143]. In an ongoing phase II, randomized, double-blind, placebo-controlled trial, an anticipated 60 sub- jects with 1 F508del CFTR and a minimal function mutation or those homozygous for F508del CFTR will receive combinations of VX-152, tezacaftor, and ivacaftor (NCT02951195). Preliminary trial data on the company’s website indicated that most side effects were mild or moderate and one subject stopped VX-152 due to pneumonia; however, the number of subjects included in this safety analysis was not provided. Two-week preliminary results from 21 subjects with an F508del CFTR and a minimal function mutation showed a mean absolute within-group change of up to a 9.7% in FEV1 and −19.6 mmol/L in sweat chloride in the triple drug group, compared to −0.9% and 1.0 mmol/L, respectively, in the triple placebo group. Preliminary results from 14 subjects homozygous for F508del CFTR showed that those who received VX- 152 + tezacaftor + ivacaftor had mean absolute within-group changes of up to 7.3% in FEV1 and −20.9 mmol/L in sweat chloride compared to −1.4% in FEV1 and +3.4 mmol/L in sweat chloride in the placebo + tezacaftor + ivacaftor group [144]. 7.15. VX-659 VX-659 is a next-generation F508del CFTR corrector being examined in a phase I, randomized, double-blind, placebo- controlled study of an anticipated ~130 healthy subjects and adults with CF heterozygous for F508del CFTR with a second minimal function mutation (NCT03029455) to assess bioavail- ability, safety, and pharmacokinetics of combinations of VX- 659, tezacaftor, and ivacaftor. Preliminary 2-week safety reports on Vertex’s website from 12 CF subjects describe most AEs as mild to moderate. There is a suggestion of improvement from the subjects’ baseline in some efficacy measurements, but detailed protocol info and statistical ana- lysis was not provided [144]. 8. Potential drug development challenges 8.1. Multiple CFTR modulators under development Despite the high costs of CFTR modulator therapy in the United States, 89% of the eligible 1071 people ≥6 years old who possess a G551D mutation were prescribed ivacaftor and 74% (or a total of 154 additional people) with another gating muta- tion were receiving ivacaftor therapy in 2015 (CFF patient registry statistics). Furthermore, approximately 45% of the eligible CF population homozygous for F508del CFTR were prescribed lumacaftor–ivacaftor soon after it was approved by the FDA, and these numbers are expected to rise as lumacaftor–ivacaftor was not available to individuals outside of clinical trials until midyear 2015 [17]. With numerous investigational drugs cur- rently being tested to improve CFTR function in ongoing or recently completed trials, there will be significant competition to recruit subjects for larger and pivotal phase III studies. This is particularly challenging for studies requiring subjects not already receiving other CFTR modulator therapies. In many cases, companies have chosen to position their compounds as synergistic to current therapies or emphasized benefits such as reduced drug–drug interactions, improved pharmacokinetics, or more consistent bioavailability regardless of food consumption. With decreasing numbers of CFTR modulator-naïve subjects available to participate in clinical trials in the United States, some companies have shifted their studies of new compounds to countries where CFTR modulators are less available [145]. This may limit opportunities for subjects in the United States and elsewhere to participate in clinical trials of novel agents to treat CF and is a significant challenge for the development of new mono and multidrug combinational therapies not already well into development. 9. Conclusion The advent of CFTR modulator therapies is transforming the care and outcomes of CF patients. Highly effective (e.g. iva- caftor) and modestly effective (e.g. lumacaftor-ivacaftor) thera- pies are being expanded to younger and more healthy CF populations, providing the opportunity to fundamentally change the trajectory of CF disease and potentially prevent the onset of disease progression. With a full pipeline of CFTR modulator strategies under development for >90% of CF patients, it appears likely that highly effective modulator therapies will ultimately be successfully expanded to most CF patients.

10. Expert opinion

One can envision a day in which CF patients are assigned highly effective, genotype-directed therapies at the time of diagnosis. Indeed, evidence from open-label studies of ivacaf- tor in young CF patients possessing gating mutations (age 2–5 years) suggests that exocrine pancreatic function can potentially be improved [146]. Thus, the benefits may expand beyond pulmonary disease. If highly effective CFTR modula- tors become available for the majority of CF patients, CF care may ultimately shift to intense monitoring for disease devel- opment and the addition of currently approved symptom- based therapies on an individualized basis. Caution is clearly warranted before adopting this approach, however, as the systematic application of symptom-based therapies is built upon evidence-based clinical trials that are responsible for the steady improvement in CF outcomes and survival over the past six decades [147].

Current CF care is complex and demanding. It has been estimated that the average adult CF patient spends approxi- mately 2 h daily to maintain health, with additional time and therapies required during pulmonary exacerbations [148,149]. It is also clear that adherence to prescribed therapies is negatively impacted by increasing numbers of therapies [150]. Thus, there will be great interest in the CF community to determine the relative benefit of traditional, symptom-based therapies in the context of effective CFTR modulator treatment. Unfortunately, this is also likely to be complex, as the relative benefit of a given therapy is dependent upon a number of patient-specific factors, such as underlying structural lung disease, infection with various CF pathogens, inflammation, and so forth. Utilizing the power of large and detailed patient registries may be critical in assigning the relative benefits of traditional therapies to specific patient profiles. Ultimately, these decisions will be in the hands of the health-care provider and patient/ family, but provided with appropriate tools and data, it is our opinion that patient-specific therapy can be optimized to max- imize clinical benefit and QOL [151].

Despite the excitement and tangible benefits demon- strated by CFTR modulators to date, not all approaches have been successful (e.g. ataluren for CF caused by PTCs, cavoson- stat for F508del-related disease), currently only ~50% of CF patients are candidates for CFTR modulators, and a minority of these are candidates for highly effective modulator therapies. Thus, tools will be needed to help identify subjects with rare mutations who may benefit from available CFTR modulators (but are not candidates for traditional clinical trials), and to help optimize modulator therapy when several options exist. Recent advances with patient-derived model systems (e.g. organoids grown from rectal biopsies) suggest that these may be used to quantify variable modulator effects and pre- dict patients likely to benefit [152,153].

It is likely that novel CFTR therapies independent of CFTR modulators would be of benefit for a sizeable number of CF patients. For example, there will likely be a small percentage (but significant number) of CF patients unable to benefit from CFTR modulators due to side effects, intolerance, or unrespon- sive CFTR genotypes. For these patients, CFTR gene transfer would potentially provide benefit in a genotype agnostic approach. Furthermore, being able to ‘add on’ CFTR gene transfer to the lungs of CF patients receiving systemic mod- ulator therapy could be of value, particularly for patients not demonstrating full or desired pulmonary benefit. To date, over
20 CFTR gene transfer clinical trials have been performed, albeit with limited success. The fundamental biology of the human airways makes successful gene therapy for CF challen- ging [154,155]. Delivery has been attempted using inhalation of various viral and nonviral vectors to target the respiratory epithelium [156–160]. The luminal surface of the human air- way is typically resistant to gene transfer, and obstructive CF mucus is a barrier to target cell access [154,155]. In addition, airway cell turnover points toward the need for repeated dosing (in the absence of long-term progenitor cell correc- tion), and viral vectors can elicit a host immune response that limits subsequent dosing efficacy [154].

The most recent sizable clinical trial testing airway gene therapy in CF patients reported modest results in 2015 [97]. This phase IIb study, conducted by the UK Cystic fibrosis Gene Therapy Consortium, evaluated the effects of monthly treat- ments with an aerosolized plasmid DNA–liposomal complex (containing a wild-type CFTR gene) versus placebo over 1 year in 116 stable CF patients ≥12 years old. The primary end point was the change in FEV1, and a number of secondary exploratory end points were included. The placebo group had a 3.7% decline in their FEV1 over the course of the study, while the treatment group FEV1 was essentially unchanged. There was no improvement in QOL measures or difference in number of antibiotic courses received, the latter of which the authors used as a surrogate marker for number of CF exacerbations. There were trends in improvements for several of the explora- tory biomarkers, but most did not meet statistical significance [97]. While the results of the trial were modest, they are the first CFTR gene transfer study to demonstrate a clinical treatment difference between the active and control arms and provide some momentum for future gene transfer improvements.

In addition to drugs that modulate CFTR, there are numer- ous drugs in the pipeline targeting other key elements that contribute to CF pathogenesis. These include drugs to modify inflammation, inhibit ENaC, improve mucociliary clearance, new anti-infectives as well as drugs focused on enhancing nutrition and addressing GI dysfunction [161–167]. These therapies are genotype agnostic and likely to be of relevance to CF patients for many years to come. Novel therapies such as stem cells are also entering CF clinical trials. One example (CEASE-CF) is a phase I, single center, open-label study that is evaluating single escalating doses of an allogenic human mesenchymal stem cell (MSCs) infusion in 15 CF adults with well-characterized CFTR mutations (NCT02866721). This study began in 2016 and will evaluate safety, tolerability, pulmonary exacerbations, FEV1, QOL, sputum microbiology, and candi- date inflammatory biomarkers for up to 1 year. A recent randomized, placebo-controlled phase II study in COPD assessed safety and efficacy of four monthly IV infusions of MSCs in 62 subjects, with no changes in clinical end points observed (PFTs, number of exacerbations, or QOL). Despite the lack of clinical efficacy, it was reassuring that there were no SAEs and similar numbers of AEs were observed in both the placebo and MSC groups over 2 years [168].

The current landscape of CF therapeutic development is robust, and CFTR modulators in particular have demonstrated the potential to profoundly benefit CF patients with respon- sive CFTR mutations. This success has not only created excite- ment in the field but also challenges to new drug development that will need to be addressed through future clinical trials. It is possible that highly effective CFTR modula- tor therapies will be available for the majority of CF patients over the next several years (e.g. combinations of F508del CFTR modulators that are sufficient to benefit patients with one or two copies of F508del CFTR), but there will be a continuing need for tools to link patients with rare mutations to beneficial treatments independent of traditional clinical trials. In addi- tion, gene transfer of CFTR has the potential to benefit patients in a genotype agnostic fashion, and recent trials suggest that clinically meaningful benefits are within grasp. Future CF health-care providers will be faced with a different treatment paradigm, initiating CFTR-directed therapies well before the onset of progressive lung disease and determining when to apply additional symptom-based therapies on a patient-specific basis. The CF health-care community welcomes VX-561 these changes and foresees a bright future for the patients that they treat.