Exon skipping and dystrophin restoration in patients with Duchenne muscular dystrophy after systemic phosphorodiamidate morpholino oligomer treatment: an open-label, phase 2, dose-escalation study

Summary Background We report clinical safety and biochemical efficacy from a dose-ranging study of intravenously administered AVI-4658 phosphorodiamidate morpholino oligomer (PMO) in patients with Duchenne muscular dystrophy. Method We undertook an open-label, phase 2, dose-escalation study (0·5, 1·0, 2·0, 4·0, 10·0, and 20·0 mg/kg bodyweight) in ambulant patients with Duchenne muscular dystrophy aged 5–15 years with amenable deletions in DMD. Participants had a muscle biopsy before starting treatment and after 12 weekly intravenous infusions of AVI-4658. The primary study objective was to assess safety and tolerability of AVI-4658. The secondary objectives were pharmacokinetic properties and the ability of AVI-4658 to induce exon 51 skipping and dystrophin restoration by RT-PCR, immunohistochemistry, and immunoblotting. The study is registered, number NCT00844597. Findings 19 patients took part in the study. AVI-4658 was well tolerated with no drug-related serious adverse events. AVI-4658 induced exon 51 skipping in all cohorts and new dystrophin protein expression in a significant dose-dependent (p=0·0203), but variable, manner in boys from cohort 3 (dose 2 mg/kg) onwards. Seven patients responded to treatment, in whom mean dystrophin fluorescence intensity increased from 8·9% (95% CI 7·1–10·6) to 16·4% (10·8–22·0) of normal control after treatment (p=0·0287). The three patients with the greatest responses to treatment had 21%, 15%, and 55% dystrophin-positive fibres after treatment and these findings were confirmed with western blot, which showed an increase after treatment of protein levels from 2% to 18%, from 0·9% to 17%, and from 0% to 7·7% of normal muscle, respectively. The dystrophin-associated proteins α-sarcoglycan and neuronal nitric oxide synthase were also restored at the sarcolemma. Analysis of the inflammatory infiltrate indicated a reduction of cytotoxic T cells in the post-treatment muscle biopsies in the two high-dose cohorts. Interpretation The safety and biochemical efficacy that we present show the potential of AVI-4658 to become a disease-modifying drug for Duchenne muscular dystrophy. Funding UK Medical Research Council; AVI BioPharma.

The baseline data period includes week -12, week -1 and week 1.
The post treatment data period includes week 12 to week 22.
The activity levels have been defined as follows: These parameters have been pre-defined by the Stepwatch 3.1 software.
The LSR represents activities such as sedentary tasks with short periods of walking The MSR represents intermittent walking (<2 consecutive minutes) or very slow walking The HSR represents highly active movements such as running, fast walking or walking continuously for more than 30 seconds.
During the clinical visits, the study participants and the parents reported an increase in the daily activity of most of the boys.
The combined proportion of time spent at MSR and HSR have been analysed since this variable is most likely to be responsive to treatment. We hypothesized that the daily activity may be more affected by external factors like moderate to severe disease, school holidays, weekend and bad weather conditions than be a result of the negative or positive effect of treatment in this short period of time. We therefore focused on activity during school days, which are more consistent in activities performed regularly and we will remove outlier days using the step watch diaries and the AE records.
The utility of the step watch and the variance factors for healthy boys and DMD boys have been described earlier elsewhere 11 .

Data Analysis
The baseline step activity was calculated as the mean number of steps on medium and high activity rate at baseline (week -12 (if available), week -1 and week 1) and after treatment (week 12 and week18). In the follow-up period the SAM data was collected at week 22. The same approach was used to look at the time fraction in percent spent at medium and high activity rate.

Determination of pharmacological properties of AVI-4658
Plasma pharmacokinetics of AVI-4658 was determined from analysis of plasma taken and prepared (within 30 minutes of collection) after 1 st , 6 th and 12 th doses. 2 ml of blood was taken during predefined windows at each of the following time points: Pre-dose (within 30 minutes of start of study drug administration); and 5, 15, 30, 60 and 90 minutes; 2, 4, 6, 8, 12 and 24 hours post-dose.
Renal Clearance: Urine was collected for the 24 hours following study drug administration for estimation of renal clearance of AVI-4658, after 1 st , 6 th and 12 th doses in the following periods: Predose (within 30 minutes of start of study drug administration); 0 to 4 hours post-dose; 4 to 8 hours postdose; 8 to 12 hours post-dose and 12 to 24 hours post-dose.
Plasma PK results: Following one-hour infusions of AVI-4658 at doses of 0.5, 1, 2, 4, 10 and 20 mg/kg, plasma concentration decreased in a multiphasic manner for weeks 1, 6 and 12. Plasma concentrations returned to below the limit of quantification (BLQ) by 12 hours for the 0.5 and 1 mg/kg cohorts and were generally above BLQ at 24 hours for the 2 through 20 mg/kg cohorts. Concentration versus time profiles were similar for weeks 1, 6 and 12 and no accumulation was observed between study weeks. Pharmacokinetic parameters were thus averaged across study weeks.

RT-PCR for Exon Skipping assessment
Ten serial 7 µm sections from frozen muscle were obtained for RNA extraction and nested RT-PCR analysis using either the standard 30/35 cycles or enhanced 35/40 cycles 12 . Direct DNA sequencing from excised bands was carried out by UCL Scientific Support Services. The results are given in Supplementary Figure S1.

Dystrophin restoration assessments
RT-PCR in order to assess induction of exon 51 skipping was performed as previously described 13  Sections were then washed and mounted with Hydromount (National Diagnostic, USA). The specimens were examined by two investigators who were blinded to which sample was pre-or posttreatment. The number of dystrophin-positive fibres was subsequently counted on MANDYS106 stained slides by two independent investigators after adjusting the detection threshold for each subject so that only revertant fibres were detected in the pre-treatment biopsy 13 . Semi-Quantitative measurements of dystrophin expression levels were carried out using the Metamorph Imaging system (Universal Imaging, USA) as previously described 15 .
For Western Blotting, proteins were extracted in lysis buffer consisting of 75 mM Tris HCl pH 6.8, 1% SDS, plus a cocktail of protease inhibitors (Complete Protease Inhibitor Cocktail Tablets, Roche). Soluble proteins were resolved using a precast NuPAGE® Novex Tris-Acetate 3-8% gel (Invitrogen,CA) for 2 h at 150V and electrophoretically transferred to a Low-fluorescent PVDF membrane (GE Healthcare) for 2 hours at 30V on ice. Membranes were then blocked with 10% skimmed milk in Tris-buffered saline Tween (TBST) and probed overnight at 4°C with Dys1 (Novocastra, UK) for the detection of dystrophin protein and sarcomeric α-actinin (Sigma, UK) as a loading control. After washing, membranes were incubated with a biotinylated anti-mouse IgG (Amersham GE Healthcare) for 1h followed by incubation with horseradish peroxidise-conjugated streptavidin (DAKO) for 1h. Membranes were visualized using chemiluminescence (ECL+Plus, GE Healthcare). The control muscle biopsy from the quadriceps femoris muscle was obtained from a normal adult female. The intensity of the bands obtained was measured by Image J software. The quantification is based on relative density values (area and percent of the bands). The relative density values for all the samples (Dys1 for control and patients) and their loading protein (alpha-actinin) bands were calculated. Then for each set (Dys1 and alpha-actinin bands) the percent value for each lane was divided by the percent value in the control to get a set of density values that is relative to the amount of protein of each line versus control. Finally, the sample relative density of each lane was divided by the loading protein relative density for that same lane, and this value was then expressed as % of the control. Further details can be found under following website: http://lukemiller.org/index.php/2010/11/analyzing-gels-and-western-blots-with-image-j/

T-Cell counting in Muscle biopsies
The presence of T and B cells was assessed by immunohistochemistry in each muscle biopsy sample using antibodies to CD3, CD4 and CD8 (Dako, UK).
CD counting was performed under x40 magnification; the number of CD positive cells and the number of fibres was counted for 10-20 fields per section. Counts were independently verified for 2-4 sections to ensure reliability. The percentage of CD positive cells in each section was calculated by:

Number of CD positive cells/number of fibres × 100
Control values for each antibody were subtracted and data analysed by paired two tailed t-test. Step activity monitoring analysis

List of Supplementary Figures
Supplementary Figure 1 Table S1 shows of the Adverse Events and Serious Adverse Events.  If screening PE, vital signs, height, and labs (from Visit 1/Week -12) were performed >10 days prior to first study drug administration they must be repeated within 7 days of first study drug administration.

2.
Visit 3/Week 1 PE is to be performed within 2 hours before study drug administration, 30 minutes after study drug administration and before discharge.

3.
Screening ECG, ECHO, and PFTs are to be performed within 30 days of the first study drug administration. 4.
ECG to be performed within 8 hours following study drug administration and must be interpreted by medically qualified personnel prior to discharge from the study site. 5.
SAM will be worn for 7 days during baseline (note that up to 10 days prior to study drug start is allowed for obtaining the baseline) ; 7 days, once a month during the treatment period (Visits 3-14/Weeks 1-12); and 7 days once every month during the follow-up period (Visits 15-18/ Weeks 14-26). 6.
PEs during Visits 4-14/Weeks 2-12 are to be performed 30 minutes after study drug administration and before discharge. 11.
Additional PEs may be done at the Medical Doctor's (Investigators) discretion throughout a subject's participation. 13.
Obtain a muscle biopsy from the contralateral bicep of the screening muscle biopsy (or alternative) 14.
To be reviewed before and after all study drug administrations. If this assessment has not been performed prior to signing consent, previously performed genetic testing may be used to qualify the subject for this study.   Supplementary table S4: 6 min walk test listings per cohort 6 min walk test was performed as described in the study protocol and manual. Below the 6 min walk test values in metres, empty cells indicate that the test was not performed (due to adverse events, protocol amendments, loss of ambulation due to natural disease progression or poor cooperation of the child). Baseline was defined as the 6 min walk distance at week -1. SD= standard deviation. For the baseline the walked distance in 6 min is listed and in order to reflect the variability better we have listed the 6 min walk test as, in changes from Baseline (Distance (Week X) -Distance (Week -1)). -13 -

Supplementary table S5: Northstar scores mean values listings per cohort and study week
The Northstar score (ordinal scale) was collected as described in the study protocol and manual. Empty cells: test was not performed (due to adverse events, protocol amendments, loss of ambulation due to natural disease progression or poor cooperation by the child). No data was obtained in week 6 for cohort 1, in accordance with the first version of the clinical trial protocol (Version 0.0). Baseline was defined as the score at week -1. Week 12 13 34 3 5 Week 18 11 31 3 7 Week 22 12 34 3 4 Week 26 11 32 2 3

Supplementary table S6: Myometry mean values listings per cohort
Myometry testing was performed as described in the study protocol and manual. Values in Newton: empty cells indicating that the test was not performed (due to adverse events, protocol amendments, loss of ambulation due to natural disease progression or poor cooperation by the child). Baseline was defined as week -1 , Subjects who had a longer screening period had an additional pre-treatment test at week -12. Although the mean of each cohort was calculated, due to high variability of the functional ability at start, variability in treatment response and low number of subjects per cohort, no functional conclusion can be draw . SD= standard deviation. Step activity monitoring analysis The baseline step activity was calculated as the mean number of steps on medium and high activity rate at baseline (week -12 (if available), week -1 and week 1) and after treatment (week 12 and week1 8). In the follow-up period the SAM data was collected at week 22. The same approach was used to look at the time fraction in percent spent at medium and high activity rate. The table below shows the analyzed and extracted data as described above of the SAM. Empty cells indicate that the test was not performed (due to adverse events, protocol amendments, loss of ambulation due to natural disease progression or poor cooperation of the child). SD = standard deviation; no of steps = mean number of steps per day at medium (med) or high activity level as defined above. A) RNA was extracted from pre or post-treated muscle sections and used as template for RT-PCR analysis. Exon 51 skipping was observed in all patients although for lower cohorts the production of the skipped product was variable and generally only observed using enhanced RT-PCR conditions. B) Direct sequencing of the skipped products for each patient confirmed correct exon skipping. Patients are grouped by cohort according to the dose received Ol d protocol -no week 1 no 6 mi nute wa l k tes t pl a nned.