Systemic delivery of AAV for neuromuscular gene therapy

15 Dec 2010

Evry, France

Meeting report

As part of their preclinical work within TREAT-NMD, TREAT-NMD partner Généthon arranged a one-day meeting of experts to discuss issues surrounding the preclinical evaluation of adeno-associated virus (AAV) vectors as a mechanism for systemic delivery of neuromuscular disease therapies. This is a critical step when moving from the research and development phase towards clinical studies. The many challenging questions surrounding mode of administration, animal models, biosafety issues and immune responses have meant there is still a lack of consensus between experts, and the primary goal of the meeting was an attempt to establish common ground. The meeting was split into four round-table sessions:

  • Round-table 1: Non-regulatory preclinical studies: route and dose rationales, safety pharmacology studies. What should be covered before toxicology studies? Moderator: Yves Fromes (Généthon, F). Panel: Jerry Mendell (Ohio State Univ., USA), George DICKSON (Royal Holloway – Univ. London UK), Patrick AUBOURG (Inserm, F), Amit Nathwani (UCL Cancer Institute, UK), Sophie LUCAS (Afssaps, F)
  • Round-table 2: Regulatory toxicology and biodistribution of the product: How to choose the best model(s) and protocol(s) in animals to forecast worst-case tox effects? (in the frame of the guidelines). This may include the AAV serotype issue (species tropism per serotype) Moderator: Valder Arruda (UPenn, USA). Panel: Patrick Gonin (IGR, F), Serge RICHARD (CERB, F), Didier CAIZERGUES (Généthon, F), Sophie LUCAS (Afssaps, F), Marc SONNEMANS (AMT, NL)
  • Round-table 3: Surgical (isolated limb local perfusion) potential side-effects and local toxicity: What is the best animal setting for a pertinent surgical model? Moderator: Jerry Mendell (Ohio State Univ., USA). Panel: Thomas Voit (Inst. de Myologie, F), Yves Fromes (Généthon, F), Jude SAMULSKI (UNC, USA), Louis CHICOINE (Ohio State Univ., USA), Kevin FLANIGAN (Ohio State Univ., F), Sumio Fukui (Paris, F)
  • Round-table 4: Immune response / immunotoxicology. What do we know about the potential immune-related side effects on efficacy and safety? Immune suppression? Moderator: Maria Cristina Galli (Inst. Sup. di Sanita, EMA). Panel: Jean DAVOUST (INSERM, F), Anne GALY (Généthon, F), Olivier Boyer (CHU, Rouen/F), Florence salmon (AMT, NL), Luk Vandenberghe (UPenn, USA)

Conclusions from the meeting

Animal models: The consensus is that regulatory toxicology studies should be performed in rodents, whereas other preclinical animal studies may be performed in other animal models. A summary of the relevance of different models (mdx mouse, grmd dog, non-human primate and pig) was provided.

Proof of concept/functionality data/evaluation of oncogenicity: Although proof of concept has to be based on the detection of a functional improvement in the animal model, this needs to be evaluated in the context of the drug in question. In some cases gene expression may be adequate for proof of concept, but functional improvement will have to be seen in a subsequent step. Questions over the percentage of dystrophin expression required, the lack of criteria for the evaluation of efficacy, and how to define the length of follow-up required to prove absence of oncogenicity were also discussed.

Route of administration (intramuscular, intravenous, loco-regional) - choice / immune reactions: Overall, the choice of the intramuscular (im) versus intravenous (iv) route of administration should be decided in light of the disease to be treated. In view of the necessity to treat the whole body and potential immunological problems associated with an im approach, the loco-regional /systemic administration approach should usually be the route of choice. Patients/parents and patient organisations should also be involved in this discussion as there is the possibility that an early phase im trial could prevent a patient being eligible for a later trial as they may have been ‘immunised’ against the AAV by the im trial. It was noted that immune response is a very serious issue and there is therefore much to justify planning of gene therapy treatments combined with immunosuppressive treatment.

Toxicology issues: Due to problems meeting all criteria for in vivo toxicology testing in cases where e.g. species-specific vectors would be required, toxicology testing is very challenging, and there are a number of highly controversial issues that even the regulators (EMA/FDA) may not completely coincide on. The EMA reflection paper GTWP/125459/2006 appears to support the following interpretations:

  • Same route and method of administration – unless otherwise justified.
  • The dosing should mimic the clinical use with appropriate safety margins
  • Only one species used: the most relevant one for expected toxicological effects, choice scientifically justified.
  • Single or repeated dose toxicity studies: depending on the prolonged functionality?
  • Use of a serotype specific for the animal model, in pivotal non-clinical studies, rather than the human serotype that will be used in clinical trial, could be acceptable.

However, these are still controversial areas, and the recommendation is therefore to discuss these issues as early as possible with the regulatory agencies and propose scientifically solid testing protocols which can be easily defended.

For full details of the meeting outcomes and the meeting report, please contact Otto Merten of Généthon.