Phase 2a Study of Ataluren-Mediated Dystrophin Production in Patients with Nonsense Mutation Duchenne Muscular Dystrophy

Background Approximately 13% of boys with Duchenne muscular dystrophy (DMD) have a nonsense mutation in the dystrophin gene, resulting in a premature stop codon in the corresponding mRNA and failure to generate a functional protein. Ataluren (PTC124) enables ribosomal readthrough of premature stop codons, leading to production of full-length, functional proteins. Methods This Phase 2a open-label, sequential dose-ranging trial recruited 38 boys with nonsense mutation DMD. The first cohort (n = 6) received ataluren three times per day at morning, midday, and evening doses of 4, 4, and 8 mg/kg; the second cohort (n = 20) was dosed at 10, 10, 20 mg/kg; and the third cohort (n = 12) was dosed at 20, 20, 40 mg/kg. Treatment duration was 28 days. Change in full-length dystrophin expression, as assessed by immunostaining in pre- and post-treatment muscle biopsy specimens, was the primary endpoint. Findings Twenty three of 38 (61%) subjects demonstrated increases in post-treatment dystrophin expression in a quantitative analysis assessing the ratio of dystrophin/spectrin. A qualitative analysis also showed positive changes in dystrophin expression. Expression was not associated with nonsense mutation type or exon location. Ataluren trough plasma concentrations active in the mdx mouse model were consistently achieved at the mid- and high- dose levels in participants. Ataluren was generally well tolerated. Interpretation Ataluren showed activity and safety in this short-term study, supporting evaluation of ataluren 10, 10, 20 mg/kg and 20, 20, 40 mg/kg in a Phase 2b, double-blind, long-term study in nonsense mutation DMD. Trial Registration ClinicalTrials.gov NCT00264888


Introduction
Duchenne muscular dystrophy (DMD) results from mutations in the gene encoding dystrophin, a protein that stabilizes muscle cell membranes. The absence of normally functioning dystrophin results in contraction-induced membrane injury. Patients with DMD develop progressive proximal muscle weakness that leads to deterioration of ambulation, wheelchair dependency, and eventual respiratory and cardiac failure. Pharmacotherapy is limited to corticosteroids, which increase muscle strength in the short-term but have significant side effects and do not address the underlying cause of DMD. Strategies in clinical trials to restore dystrophin in muscle cell membranes include nonsense mutation suppression and exon skipping [1].
A nonsense mutation is a single-point alteration in DNA that results in the inappropriate presence of a UAA, UAG, or UGA stop codon in the protein-coding region of the corresponding mRNA. This premature stop codon causes the production of a truncated protein and leads to loss of protein function and consequent disease. Nonsense mutations are responsible for approximately 13% of DMD cases [2].
Ataluren (also known as PTC124) was discovered through high-throughput screening and chemical optimization to induce ribosomes to read through premature stop codons but not normal stop codons. [3] When tested in the nonsense mutation mdx mouse model of DMD, ataluren generated production of full-length, functional dystrophin protein. [3,4] We and others have shown that ataluren is active in multiple cell-based and animal disease models. [3][4][5][6][7][8][9] Phase 1 studies in healthy volunteers established the initial ataluren safety profile [10] and defined dosing regimens for achieving target trough plasma concentrations (2 to 10 mg/mL) known to be active in preclinical models [3,5].
We describe here the results from a Phase 2a clinical trial in subjects with nonsense mutation DMD that evaluated pharmacodynamic activity, as measured by immunofluorescence evidence of an increase in dystrophin production on muscle biopsy. The study also assessed additional markers of disease activity, changes in muscle strength and function, safety, and ataluren pharmacokinetics.

Methods
The protocol for this trial and supporting CONSORT checklist are available as supporting information; see Checklist S1 and Protocol S1.

Ethics
This study was performed at three sites (Children's Hospital of Philadelphia, Cincinnati Children's Hospital Medical Center, University of Utah School of Medicine) and was approved by each investigator's institutional review board (IRB). The IRBs for this study were the Children's Hospital of Philadelphia Committees for Protection of Human Subjects, Cincinnati Children's Research Foundation, University of Utah, and the Western Institutional Review Board.
The trial was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice and was registered (Identifier NCT00264888) at www.clinicaltrials.gov. A parent or guardian of each subject was required to sign an informed consent form approved by the local IRB. Subjects were also required to provide signed assent to participate.

Eligibility Criteria
Since DMD is an X-linked disease, only male subjects were enrolled. Subjects were $5 years of age and were required to have a diagnosis of nonsense mutation DMD based on a clinical phenotype present by age 5, increased serum CK, absent or diminished sarcolemmal staining with an antibody to the Cterminal portion of the dystrophin protein on muscle biopsy, and presence of a nonsense mutation in the dystrophin gene (as confirmed by gene sequencing). Complete entry criteria are provided in the Appendix S1.

Study Treatment
Eligible subjects were sequentially assigned to escalating dose levels of ataluren. The first patient was recruited on 21 December, 2005, and follow-up was completed on 03 May, 2007. Subjects were not to be enrolled at the next higher ataluren dose level until all subjects who had been treated at the previous level had completed the 28-day treatment period and a review of safety and pharmacokinetic data had indicated that dose escalation was appropriate. The study was originally planned to include two dose groups. The first (N = 6) and second (N = 20) groups of subjects received treatment with an oral suspension of ataluren TID at morning, midday, and evening doses of 4, 4, 8 mg/kg and 10, 10, 20 mg/kg, respectively. Both dose levels were well tolerated. Based on pharmacokinetic data at these dose levels, the protocol was amended and an additional 12 subjects were enrolled to receive ataluren 20, 20, 40 mg/kg. The duration of treatment was 28 days. Subjects were followed for an additional 28 days posttreatment.

Study Assessments
Excisional biopsies of the extensor digitorum brevis (EDB) muscles were performed on all subjects [11]; the muscle was obtained from one foot at baseline and the other foot on Day 28 of treatment. Immunofluorescence analysis of serial 9-mm cross sections was performed as the primary outcome measure. Immunofluorescence images were analyzed qualitatively and quantitatively, as detailed in the Appendix S1. In addition, myotubes were derived from baseline EDB biopsy specimens and cultured in vitro in the presence of a range of ataluren concentrations, including 10 mg/mL. Western blot analysis of changes in dystrophin and other muscle membrane proteins was not performed. Serum CK levels were monitored at baseline, every seven days during treatment, and at Days 14 and 28 posttreatment. Myometry and timed function tests were performed at baseline, after 28 days of treatment, and Day 28 post-treatment.
Blood samples for ataluren pharmacokinetic assessments were collected pre-dose and 1, 2, 3, and 4 hours after the morning dose; pre-dose and 1, 2, 3, and 4 hours after the midday dose; and predose and 1, 2, 3, 4, and 12 hours after the evening dose. Ataluren plasma concentrations were derived from a validated bioanalytical method [10].

Safety Monitoring
Safety monitoring included adverse events, laboratory tests, vital signs, electrocardiograms, and physical examinations. An independent data safety monitoring board, comprising two neuromuscular experts and a biostatistician, reviewed safety data during the study.

Statistics
All subjects were included in safety and compliance analyses. Subjects with both baseline and on-study measurements were included in efficacy analyses. The number and percentage of subjects with a post-treatment dystrophin response in muscle biopsy were computed for each dose group. The number and percentage of subjects who had an in vitro ataluren dystrophin response were computed. Changes in quantitative dystrophin expression, serum CK, myometry, and timed function tests were analyzed using paired t-tests. Pretreatment serum CK values were obtained at screening and baseline; the mean of each subject's screening and baseline values was used for analyses. Compliance was calculated as the proportion of actual relative to planned ataluren doses. Pharmacokinetic parameters were calculated using non-compartmental methods.
The term, ''percentage increase,'' utilized in this paper represents an increase relative to baseline in light intensity.
The statistical software used for the analyses was Statistical Analysis System (SAS) version 9.1.
Ages and body weights were generally consistent across the dose groups, albeit with a higher range at the 20, 20, 40 mg/kg dose level due to inclusion of several older, nonambulatory boys in this cohort (Table 1). Of the 38 subjects, 34 (89?5%) were ambulatory as determined by the investigators. Because of generalized muscle fragility in DMD, serum CK and transaminase values were universally abnormal at baseline. Twenty-seven of 38 subjects (71?1%) were receiving corticosteroid treatment, usually with a daily regimen of deflazacort or prednisone/prednisolone. All 3 types of premature stop codons were represented, but the UGA stop codon was predominant in all dose groups. Nonsense mutations were located across exons 6 to 70 of the dystrophin gene; no mutational hotspots (ie, specific locations with notably higher numbers of mutations) were identified.

Changes in Muscle Dystrophin Expression in Muscle Myotubes Cultured in vitro
Pre-treatment primary muscle cells from 35 of the 38 subjects were available for in vitro myotube culture. Myoblasts were expanded in growth medium, transferred to differentiation medium, and allowed to progress for three days. Differentiated myotubes were subsequently exposed to ataluren 10 mg/ml for nine days prior to dystrophin staining. Dystrophin levels were analyzed by quantitative immunofluoresence. The results demonstrate that when the muscle obtained from the pre-treatment biopsies were cultured in the presence of ataluren, 35/35 (100%) of the samples showed evidence of an increase in dystrophin expression in response to ataluren treatment in vitro ( Table 2).

In vivo Changes in Muscle Dystrophin Expression
Pre-and post-treatment immunofluorescence images from the EDB muscles were available from all 38 subjects. These images were assessed qualitatively by blinded reviewers who determined whether dystrophin expression was higher in the pre-treatment image, the post-treatment image, unchanged, or indeterminate. At all three dose levels, subjects demonstrated qualitative increases in the staining for dystrophin, ie, at least 2 of 3 reviewers observed higher dystrophin expression in the post-treatment image (Table 2). Overall, 2/6 (33%) subjects treated at 4, 4, 8 mg/kg, 8/20 (40%) subjects treated at 10, 10, 20 mg/kg, and 3/12 (25%) subjects treated at 20, 20, 40 mg/kg demonstrated an increase in dystrophin expression post-treatment. Figure 2 provides examples of immunofluorescence data from two subjects who were considered responders by the evaluators, and Figure S1 is representative of a non-responder. Post-treatment dystrophin expression is appropriately localized in the cell membranes of the muscle fibers.
A quantitative method for assessing the ratio of dystrophin/ spectrin expression was developed and became available for this study. Based on this quantitative analysis, a mean change from pretreatment to posttreatment of 11.0% in dystrophin expression was observed (p = 0.008, paired t-test). Of the 38 patients, 23 (61%) showed a positive change in dystrophin/spectrin expression ratio after 28 days of treatment with ataluren (Table 2, Figure 3).
Response did not appear to be dependent on age, corticosteroid use, or location or type of nonsense mutation in either method.

Changes in Serum CK Levels
Due to muscle fragility, serum CK concentrations are universally elevated in subjects with DMD. [12,13] As shown in Figure 4, the majority of subjects in each cohort had decreases in serum CK values when comparing end-of-treatment values to pretreatment values. Although no definite dose-response relationship can be discerned due to small and varying sample sizes, these changes

Changes in Clinical Measures
Collective changes in upper and/or lower extremity myometry scores (for hand grip, elbow flexion, hip abduction, and knee extension) and timed function tests (standing from supine, running 10 meters, climbing four standard stairs) were small and not statistically significant after 28 days of ataluren treatment.
Although a formal symptom survey was not used, parents and teachers of several boys anecdotally reported evidence of greater activity, increased endurance, and less fatigue during treatment.

Safety
Adverse events were mild or moderate and showed no dosedependent increase in frequency or severity. Procedural complications as a result of the muscle biopsy procedures (reported in 29 No clinically significant, treatment-emergent laboratory abnormalities were observed. One subject receiving the 20, 20, 40 mg/kg dose level experienced mild abdominal discomfort which resulted in the interruption of ataluren treatment on Day 3 to 4. The subject resumed ataluren at 15, 15, 30 mg/kg on Day 5 (upon resolution of the adverse events) and was re-escalated to 20, 20, 40 mg/kg dose level on Day 7; this dose level was well tolerated and the subject completed the study. No subject discontinued ataluren due to a drug-related adverse event.

Pharmacokinetics
Ataluren was rapidly absorbed, with median T max values after the morning dose ranging from 1?5 to 2?5 hours. The 10, 10, 20 mg/kg and 20, 20, 40 mg/kg dose levels produced trough plasma concentrations of ataluren that were in or above the range known to be active in nonclinical animal models (,2 to ,10 mg/mL); high intersubject variability and substantial overlap between these two dose groups was observed. Trough plasma concentrations for the 4, 4, 8 mg/kg dose level were near the lower end of the range and did not consistently produce levels that were in the target range ( Figure 5). The 24hour concentration of one subject in the 20, 20, 40 mg/kg dose group was excluded from all pharmacokinetic analyses because his value at the 24-hour timepoint on Day 1 (171 mg/mL) was as an outlier based on the Grubbs statistical outlier test [14].
The potential effect of concomitant corticosteroid use on the pharmacokinetics of ataluren was investigated by stratifying the subjects by corticosteroid use (either prednisone/prednisolone, deflazacort, or no corticosteroid) and comparing dose-normal-ized area under the concentration-time curve through 24 hours (AUC 0-24 ) values on Day 27 ( Figure 6). Significant overlap was observed among the dose groups, with similar mean AUC 0-24 values, suggesting the absence of a significant effect of corticosteroids on the pharmacokinetics of ataluren. While the data analysis indicates that there is no apparent effect of corticosteroid use on ataluren pharmacokinetics, the small sample size must be considered in interpretation of these data.

Discussion
This Phase 2a, short-term study conducted from Dec 2005 to May 2007 evaluated whether ataluren, a first-in-class investigational new drug, can restore dystrophin production in the muscle cells of patients with DMD whose disease is caused by a nonsense mutation. This therapeutic approach holds considerable promise for such patients, who currently lack dystrophin-targeted, disease-  modifying treatment options. Overall, 34% (13/38) of the subjects in this study showed evidence of an increase in dystrophin production on EDB muscle biopsy by immunofluorescence, as assessed qualitatively by independent reviewers who were blinded to timepoint and dose level. Correct localization of post-treatment dystrophin at the sarcolemma was observed. Revertant fibers were excluded from the dystrophin analysis. When analyzed quantitatively, 61% (23/38) of the subjects showed a post-treatment increase in dystrophin. These data documented pharmacodynamic proof-of-concept for ataluren in DMD and supported the conduct of a now-completed Phase 2b randomized, placebocontrolled, multiple dose, long-term efficacy and safety study (conducted from Feb 2008 to Dec 2009;ClinicalTrials.gov, NCT00592553), and for the confirmatory Phase 3 randomized placebo-controlled study (2013-ongoing; ClinicalTrials.gov, NCT01826487). Increases in in vivo dystrophin expression were not dose-dependent or correlated with demographic or genetic factors. It is unclear why some of the subjects showed increases in dystrophin production while others did not. Among other factors, the limited duration of drug exposure may account for the variability in the findings. Pharmacokinetic data showed that although there was overlap of the ataluren plasma concentrations substantially among the three dose groups, the 4, 4, 8 mg/kg dose did not consistently achieve the target range based on preclinical models. Further study of the relationship between ataluren concentration and activity is required. Nonetheless, the results of the in vitro myotube experiment (ie, increases in dystrophin expression in 100% [35/35] of samples) indicate that even nonresponders by muscle biopsy in this study had the potential to respond at the cellular level.
This study highlights the challenges and limitations of currently available methods to assess changes in dystrophin expression. The value of muscle biopsy dystrophin expression as a biomarker in therapeutic trials targeting small increases in dystrophin levels has been questioned, because specimens only provide local information on muscle quality that may not reflect the state of all muscles. [15] Typically, a bilateral muscle is sampled from one side of the body before treatment and from the other side after treatment; thus, any differences between the left and right muscles introduce variability. Furthermore, a DMD muscle is often heterogeneous with respect to fibrofatty replacement of muscle, so that sampling error during biopsy may cause additional variability. We aimed to mitigate potential sampling-related variability by excising and evaluating the entire EDB muscle. The EDB muscle is distally located, little used, and therefore unlikely to demonstrate substantial fibrofatty replacement. Analysis of changes in dystrophin expression is hampered also by the lack of a sensitive, robust, and reproducible assay for quantifying low levels of dystrophin.
In this study, we initially performed a qualitative analysis of dystrophin expression. Subsequently, a quantitative analysis was applied to the same images that were examined qualitatively. Quantification of very low levels of dystrophin expression in immunostaining images of muscle samples presents numerous technical challenges. [16] Different methods (eg immunostaining and Western blot) applied to the same specimens result in inconsistent dystrophin levels. [17,18] Western blotting analysis was not conducted in this study because there was insufficient biopsy material remaining, post-immunostaining, to perform Western blot analysis. Although the individual patient results were not always concordant and the overall results of positive changes in dystrophin expression amongst the two methods were not entirely consistent, both the qualitative and quantitative methods showed post-treatment increases in dystrophin in this study (Table 2). Collectively, these data support ataluren's activity as a dystrophin restoration therapy in patients with nmDMD.
With regard to the other measure of pharmacodynamic activity in this study, serum CK reductions were observed in 84% (31/37) of subjects during ataluren administration. These changes were statistically significant at the 10, 10, 20 mg/kg and 20, 20, 40 mg/ kg dose levels, but not at the 4, 4, 8 mg/kg dose level. However, using CK as a pharmacodynamic endpoint in DMD is also problematic. Although CK levels decreased in this study on treatment, changes in serum CK levels are highly variable and difficult to interpret. Decreases in serum CK levels may reflect a treatment effect or reduction in activity, while increases may indicate lack of a treatment effect or exercise-related stress to the body. [15,19] Given the limitations in current quantitative dystrophin assays and CK, a validated biomarker for monitoring of drug activity in DMD is needed to facilitate drug development in this disease. Ultimately, clinically meaningful outcome measures such as the 6-minute walk test are required to evaluate therapeutic benefit in DMD [20,21].
In this short-term study clinical improvements were not observed in timed function tests and myometry, which is not unexpected for a 28-day treatment with a drug intended to increase dystrophin production. Anecdotal reports of symptomatic effects at home and in the classroom were encouraging. Ataluren was generally well-tolerated in this study. Mild treatmentemergent adverse events of transient gastrointestinal complaints were observed at all three dose levels and appeared consistent with background symptoms commonly observed in clinical trials. No clearly dose-dependent increases in frequency or severity were evident. Episodes of nausea and vomiting were primarily related to anesthesia administered at the time of muscle biopsies. No drugrelated serious adverse events were reported. No subject discontinued ataluren due to an adverse event.
Mean values for maximum concentration (C max ) and AUC 0-24 for ataluren plasma concentrations on Days 1 and 27 showed no difference in ataluren exposures in boys receiving or not receiving corticosteroids in DMD. At the 10, 10, 20 mg/kg and 20, 20, 40 mg/kg dose level, mean trough plasma concentrations achieved or exceeded target levels active in the mdx mouse model of DMD. Considerable intersubject variability was observed in ataluren exposure at all dose levels. The presence of a nonsense codon in a transcript can affect the level of the mRNA, although the precise ''rules'' governing nonsensemediated mRNA decay (NMD) have not been completely defined. In different cell lines harboring the CFTR W1282X mutation, NMD efficiency was shown to correlate with response to the stop codon readthrough compound gentamicin, as measured by restored chlorine channel activity. [22] Down regulating NMD by reducing UPF1 or UPF2 levels increased chloride channel activity in response to gentamicin in cell culture, consistent with the notion that transcript levels may act as a disease modifier. The applicability of modifying NMD for DMD has yet to be investigated.
The activity, safety, and pharmacokinetic findings from this initial study of ataluren in nonsense mutation DMD warranted advancement of this new drug candidate into larger, placebocontrolled testing. Given its potential to safely restore dystrophin production and thereby modify the course of this severely disabling disease affecting pediatric males, efforts to develop ataluren as a treatment for nonsense mutation DMD should remain an important priority for the DMD research community.