149th ENMC International Workshop and 1st TREAT-NMD Workshop on: “Planning Phase I/II Clinical trials using Systemically Delivered Antisense Oligonucleotides in Duchenne Muscular Dystrophy”

3.1. Other AON backbones or co-administration with compounds for increasing muscle uptake of AONs 

A number of speakers presented experimental work on different AON backbone to increase muscle targeting. Annemieke Aartsma-Rus showed that locked nucleic acids (LNA) have very high affinity but also high self annealing properties and reduced specificity compared to 2OMePS. Both she and Matthew Wood commented that the peptide nucleic acid (PNA) AONs uptake in cell cultures is significantly affected by their uncharged backbone, while their in vivo behavior is much better. Matthew Wood also presented data on PNAs conjugated with different compounds (arginine rich peptide for example), aimed at increasing the muscle uptake, although no significant increase in their efficacy was demonstrated in vivo.

Steve Wilton compared modified PMOs (peptide conjugate) to improve muscle targeting efficiency in mdx mice. Different acute and chronic regimens were evaluated, with AONs being administered intraperitoneally (IP). The results of these studies suggested that peptide conjugated PMOs targeted muscle much more efficiently than standard PMOs (between 5 and 10 times more efficiently); however increased toxicity of the peptide modified PMO was observed, in contrast to the good tolerability of the unmodified PMO. Recently, he also started to work on positively charged PMOs. He finally discussed that the different backbone modification while affecting the efficiency of uptake, do not appear to change the efficacy of splicing, i.e., the hierarchies of efficacy of individual AONs obtained with 2OMePS backbone is recapitulated by other chemistry such as PMOs [9].

Qi Lu indicated that modified PMOs (Vivoporter) were at least 3 times more effective compared to ordinary PMOs. He also showed ongoing work using PMOs and polymer to increase muscle uptake following IV delivery.

Dominic Wells showed the effect of delivering PMOs to the heart in combination with diagnostic ultrasound and microbubbles. Microbubbles are indeed routinely use in radiology and cardiology, and the use of diagnostic ultrasound focused to the heart can induce microbubble bursting and this appears to facilitate PMO targeting to cardiomyocytes. Indeed, this delivery method resulted in significant improvement in exon skipping in the heart of mdx mice, although not to levels seen in skeletal muscle. These results might point towards differences in trafficking and/or endothelial barriers between skeletal and cardiac muscles. Judith van Deutekom also showed that the administration of repeated high dosage 2OMePS over a short period or low dosage over a long period was indeed capable of inducing cardiac exon skipping in mdx mice.

3.2. Toxicological and delivery considerations for systemic delivery trials 

Sjef de Kimpe summarised the toxicology work performed to take the 2OMePS AON into clinical trials, but also more in general the issue of toxicology and issues related to AON administration. AON toxicity in general can be divided into hybridisation dependent and hybridisation independent toxicity. The first indicates the effect that an AON can have on the target (too much of a good thing, which is not a concern for Duchenne application), but also off-target. Such sequence off-target RNA does not need to have a 100% homology as for some backbone chemistries still hybridise significantly despite mismatches. The hybridisation independent toxicology relates to unknown motifs present in the AON sequence which might trigger unspecific reactions (such as toll receptor-mediated immune response or complement activation). The hybridisation independent toxicology relate generally to the backbone structure or (un)known sequence motifs. An example is the CpG motif that can cause a prominent immunostimulatory response via activation of toll-like receptor, TLR9. Such motifs can be avoided in the sequence selection for AONs. Much of the hybridisation independent effects are influenced by the backbone AON structure. For example first generation AONs, phosphorothiated DNA oligonucleotides, interacted with proteins of the coagulation cascade causing a transient increase in APTT, but without clinical evidence of enhanced bleeding. In toxicity studies rodents are especially sensitive to immunostimulatory effects of first generation oligonucleotides. These do not appear in monkeys. Many of such effects noticed with first generation AONs are ameliorated or even absent in second generation AONs, like the 2OMe modifications [10]. Another advantage of 2OMe is the extended terminal half life (∼4 wks), making once a week or less frequent administration feasible. Moreover, 2OMe modified AONs can be administered subcutaneously (as an alternative to intravenous infusions) and this has been employed clinically with 2nd generation AONs. Toxicology studies by Prosensa showed that the 2OMe AON for exon 51 is non-mutagenic and well tolerated in rodents and non-human primates.

In the clinic, phosphorothioate AONs in general have a good safety record with >10 different sequences administered to >3000 humans treated with at least 100,000 doses [11]. In some studies administration of elevated systemic doses by 2h infusion have been associated with low grade fever and transient elevation of the Partial Thromboplastin Time (PTT), a good dose-dependent marker of toxicity. To date, two phosphorothioate oligonucleotides have been approved by the FDA and many (with or without 2nd generation modifications) are in phase II/III trials (for further information see websites of ISIS, Macugen, Genta, Dynavax, Coley, Santaris and Prosensa).

Regarding PMOs, several AONs have been studied from a toxicology perspective in different animal species. In terms of off target effects, PMOs have been used extensively and successfully to knock down gene expression of individual genes during development in a variety of animal models, especially zebrafish, typically with a high degree of selectivity [11]. AVI has studied the toxicology of a number of PMOs in clinical trials (such as cMYC, West Nile Virus, Hepatitis C virus PMOs, for further information visit http://www.avibio.com/devNeugene.html). Eleven clinical trials performed on several hundred patients with dosage up to 300mg SC daily per 14 days have not revealed any toxicity.

One of the aspects discussed by Janet Rose Christensen and Sjef de Kimpe was on how much toxicology the regulatory authorities will require for new AONs coming to the market, considering the (sequence independent) predictability of pharmacokinetics and toxicity within a chemical class of AON.

Clearly information related to generic toxicology for a comparable duration to the proposed study is important, also in consideration of the fact that if effective these AONs will need to be administered for life. However the necessity to perform extensive toxicology studies in animals was questioned, considering that not all animal data are necessarily predictive of potential problems in the human, and that the specific sequences used in these AONs are human specific.

Regarding pharmacokinetic studies, Prosensa has performed studies in mdx mice showing that while higher peaks of 2OMePS are achieved following IV administration, the biodistribution of the AONs after SC administration was similar. Also regarding PMOs, the SC mode of delivery (or even IM) might represent an acceptable mode of delivery and this has already been used in other trials (http://www.avibio.com/devNeugene.html).

Excellent reviews on the applications and safety of AONs can be also found in [12], [13].

3.3. Planned systemic trials using AONs 

Judith van Deutekom and Jan Verschuuren illustrated the planned systemic exon 51 2OMePS study, which should start early 2008. This will be an ascending repeated dose study, the primary outcome being safety and tolerability, and as secondary outcome pharmacokinetics, assessment of effects at the molecular level and on muscle strength. At least 3 patients per group will be studied in the dose escalation studies, and the recruitment of each group will only start after evaluation of the safety data on the previous cohort. A muscle biopsy will be performed 2–4 weeks after last injection, the decision on which muscle to biopsy has not been taken yet and might be influenced by muscle MRI data. The age range of patients that will be recruited is between 5 and 16 years. The manufacturing and toxicology package of the exon 51 2OMePS is underway.

Francesco Muntoni discussed the planned AVI and MDEX (the UK Consortium) systemic delivery trial of the exon 51 PMO. It is proposed for this to be a repeated dose studies in 9 ambulant DMD patients (3 per cohort) with ascending dose between cohorts. Patients will receive a weekly systemic administration of a given dosage, followed by a muscle biopsy of the biceps three weeks after the last PMO administration. In addition to safety parameters and dystrophin restoration, patients will be studied using muscle MRI, myometry and functional scales to assess a possible improvement.

Alessandra Ferlini presented the Italian Consortium for non-viral-antisense research (ICoN). Part of the consortium coordinated by the University of Ferrara is a nanotechnology group from the University of Turin (Michele Laus), the CNR of Bologna (Luisa Tondelli), and a SME (ICE) which, in collaboration with the University of Ferrara (Alessandro Medici and Alessandra Ferlini) is set up to produce and purify AONs. The strategy of this consortium is to identify PMMA nanoparticles that could package AONs and target them to skeletal and cardiac muscles. In particular, the group is testing in patients’ cells mutation-specific AONs targeting small mutations occurring in skippable exons.

Louis Garcia outlined the plans for French consortium which is considering use of the AAV-U7 viral vector to deliver antisense to muscle. This work derives from the impressive preclinical rodent and canine studies which demonstrated the proof of principle of this approach [14]. AAV vector targeting exon 51 has already been produced and others are in production. The main obstacle for the clinical application of this approach are related to AAV production and issues related to immunological problems and systemic administration, but the ongoing French AAV trial for gamma sarcoglycanopathy will provide data to inform further development in this area (for further information on the AAV trial on sarcoglycanopathy visit http://www.genethon.fr/index.php?id=49&L=1).

3.4. Other therapeutic developments using AONs 

Annemieke Aartsma-Rus and Steve Wilton discussed the pros and cons of multiple exon skipping [15], [16]: the main advantage of this approach is that inducing skipping of a relatively large stretch of exons with combinations of AONs would be applicable to a significant population of DMD patients. For example the combined targeting of exon 42–55 would restore the reading frame in approximately 50% of all patients, while 45–55 would be applicable to ∼30% of cases, therefore considerably reducing the number of exons requiring AON development and toxicology testing. However multiexon skipping can be rather complicated because of the induction of multiple intermediate product of splicing; in addition a number of exons are co-transcriptionally spliced so that in some cases it is not possible to induce the skipping of one and not others that are also part of the same splicing reaction[16]. Wilton presented data which showed that in some areas of the gene targeting one exon is invariably associated by the skipping of the neighboring exon. He also showed that combinations of multiple AONs for overlapping exonic targets could make a big difference in the efficiency of exon skipping [16], [17]. The discussion confirmed the difficulties in identifying a priori the behavior of AONs; this was also confirmed by Ian Graham who is using hexamer hybridisation arrays to systematically assess the ability of software programs such as for example ESE-finder to predict active AONs, but this proved to be an almost impossible task. The collaborative strategy used by the MDEX consortium to identify and validate by various independent models the sequence used for the clinical trial in UK was presented. In particular 8 AON sequences targeting exon 51 were tested in two different chemistries and in three different preclinical models: cultured human muscle cells and explants (wild-type and DMD), and local in vivo administration in transgenic mice harbouring the entire human DMD locus [18] This study describes a rational collaborative path for the preclinical selection of AONs for evaluation in future clinical trials.

Multiple exon skipping is also being explored by Louis Garcia using the above mentioned AAV technology.

3.5. Choice of outcome measure and biomarkers for future clinical studies 

Michelle Eagle presented her experience of strength measurement and relationship with functional outcome in ambulant DMD boys treated with corticosteroids. The functional measure used was the North Star assessment tool, a modification of the widely used Hammersmith Functional Scale which was validated for DMD boys from the age of 4 and has been in use since 2004 in the entire of UK [19]. Muscle strength was measured using hand-held myometry (Cytec); forced vital capacity, a reliable measure of respiratory muscle strength, can be measured from the age of 4–4.5 years.

Volker Straub discussed the role of muscle MRI and summarized the result of an ongoing study in Newcastle where DMD boys are scanned with contrast at day 0, and again at day 7, after they have performed (on day 3) an exercise test. These studies are ongoing; Terry Partridge commented on T2 weighted images in the dystrophic dogs treated with PMOs: 2 weeks after the administration of the AONs there was a very significant reduction of the pathological T2 signal in the thigh muscles, and this could therefore represent a helpful technique for the monitoring of muscle protection following administration of AONs.

Gert Jan van Ommen introduced the concept of integrative genomics and multidimensional analytical tools for understand pathogenesis of pathological processes, but also monitor response to treatment and characterize potential off site targets. He first explained the power of meta-analysis of published data; indeed unbiased analysis of published gene profiling studies of different muscular dystrophies allows the identification of classes and groups of genes differentially regulated, which allow assigning a severity of the disease process. For example, the differential regulation of muscle contraction genes is seen only in the most severe disease forms of muscular dystrophy, remodeling of the extracellular matrix is seen in conditions of intermediate severity, while genes involved in inflammation are a feature in even milder muscular dystrophies. He proposed that these differentially regulated genes could be used as biomarkers of individual conditions. In addition the expression of genes previously shown to be elevated in muscular dystrophies can be shown to partly or completely returned to wild-type expression levels following therapeutic intervention; reductions in inflammation and fibrosis can also be monitored using this techniques [20]. Finally, genetic profiling also allows one to evaluate if off target effect of antisense oligonucleotides can be observed [20]; so far the use of exon 23 in mdx mice resulted in no aberrant splicing of other genes present in the arrays studied.

3.6. Cost/scale relationship for antisense oligonucleotides for systemic trials 

Sjef de Kimpe presented the Prosensa plans to pursue the systemic delivery of exon 51, and will start a program on a second exon. 2OMePS AONs can be purchased from 6 suppliers world wide at clinical grade. The final cost of 2OMePS AONs, will largely depend on the cost of the single building blocks (amidites) and this again depends on the volumes worldwide produced (and required for any oligonucleotide incorporating 2OMe building blocks).

Janet Rose Christensen indicated that AVI is the only licensed site for producing clinical grade PMOs. The assembly of these AONs is more complex than 2OMePS AONs, and this is reflected in higher costs. However, the final cost will depend on the scaling up required; AVI has the capacity to produce sufficient exon 51 PMO at clinical grade to take this as a medicinal product into the market. AVI and MDEX consortium are also considering an exon 53 systemic delivery study as the next target.

3.7. Ethical issues 

Volker Straub, Nick Catlin, and Elizabeth Vroom discussed the various ethical aspects which investigators, parents and their organisations need to consider whenever approaching children for experimental medicine.

The prevailing sense was that of a careful risk benefit assessment of each approach especially in the early phases of the study, but followed by a rapid expansion to include as many affected individuals as possible. The parental organisations are particularly keen to see collaborative efforts between different consortia, in respect of the individual program of research and intellectual properties, so that if effective, this treatment is available to as many affected individuals as possible. These organisations are also interested to be involved in the lobbying for the approval of therapies once demonstrated to be effective.

The meeting finished with a discussion on possible areas of collaborations between the different consortia, which will take advantage of the recently funded by the European Community (EU) TREAT-NMD network of excellence. Amongst the primary aims of TREAT-NMD is the plan for dissemination of effective treatments to the various state members, and also beyond the limits of the EU borders, and AONs are one of the flagship projects identified. Several follow up meetings to take these collaborative efforts further are being planned and one on outcome measure in trial design for DMD has already been held in July 2007.

3.8. The workshop organisers 

Prof. Francesco Muntoni1, Prof. Kate Bushby1 and Prof. Gertjan van Ommen2; England 1 and The Netherlands2.

Participant list:

Websites:

For a list of the members of the Dutch and MDEX Consortia, and for up to date information on the status of both trials, visit: