Safety First: Perspective on Patient-Centered Development of AAV Gene Therapy Products

Main Text

Recent publications in Molecular Therapy¹ and Human Gene Therapy² identify acute toxicity of adeno-associated virus (AAV)-mediated survival of motor neuron (SMN) or GFP expression, which warrants careful consideration to better understand the observation as well as the need to perform well-controlled studies to address this finding. However, these findings should be viewed in the context of an initial observation in a basic proof-of-concept study and not in relationship to established practices for rational and patient-centered drug development, which follow guidance from the International Committee on Harmonization (ICH) for Good Clinical Practice (GCP).³ The purpose of this commentary is to address the basis by which gene therapy medicinal products are evaluated in non-clinical studies to best inform us of the risk-benefit profile of a given product under development for a specific indication. The findings of the two publications can be summarized as preliminary studies to develop an approach to spinal muscular atrophy (SMA) using variants of AAV9. The first study tested these vectors in piglets and non-human primates (NHPs) with a finding of acute toxicity in one NHP and neural inflammation in the piglets. The second study showed a single animal with the AAV9 variant, PHP.B, which developed an acute coagulopathy. Importantly, the severe findings in a limited number of animals in these studies should not be confused with low-grade laboratory findings in clinical studies that are self-limited. Since these studies are related to variants of AAV9 vectors given systemically, the focus of this commentary will be limited to active clinical programs using AAV serotype 9, both by systemic and regional delivery, since the one conclusion of Hinderer et al.² is that, “The present results and those of another recent study utilizing a different AAV9 variant and transgene indicate that systemic and sensory neuron toxicity may be general properties of intravenous delivery of AAV vectors at high doses irrespective of the capsid serotype or transgene.”1, 2 Platform technology studies are important to understanding basic behavior of AAV-vectors in vivo. However, in order to develop therapies aimed at treating human disease, it is imperative that investigators follow the guidance agreed to by Federal Drug Administration (FDA) and European Medicines Agency (EMA) which oversee human clinical research conducted under the principles of GCP. The non-clinical and clinical data supporting nine active clinical studies should also be considered in evaluating the clinical application of AAV9 vectors.

GCP guidance is based on the ethical principles established in the Declaration of Helsinki in which the rights, safety, and well-being of trial subjects are the most important and overriding considerations in the design and conduct of research involving human subjects. Therefore, any anticipated risk should be established in light of the potential benefit for the subject. This principle is further emphasized in research involving children, the greatest proportion of those with severe genetic conditions amenable to gene therapy, where the prospect of direct benefit is required to conduct a pediatric study. Because of these principles, all available non-clinical and clinical data for an investigational product should directly support the proposed clinical use. The guidelines provide specific recommendations on how the investigator, institutional review board (IRB), independent ethics committee (IEC) and biosafety committee is informed of the non-clinical findings by way of the Investigators Brochure (IB).

The IB provides information on the characteristics of the investigational product and the non-clinical studies which support the clinical use. Therefore, a comparison of the cumulative findings from IBs which support active AAV9 studies to the current non-clinical studies by Hordeux et al.¹ and Hinderer et al.² is warranted. The findings in the two publications are related to “an attempt to develop PHP.B for human gene therapy”¹ and therefore an evaluation of efficacy and toxicity in nonhuman primates (NHPs) was done using AAV9 and two variants⁴ which were anticipated to enhance CNS gene transfer following systemic delivery of the vector. The findings of the paper in Human Gene Therapy are the subject of a commentary by Flotte and Buning⁵ and, in reference to the study in Molecular Therapy, there are two key observations. First, a well-designed set of experiments confirmed the interesting finding that the enhanced activity of the PHP.B variant is limited to the mouse strain in which it was developed. The second finding is significantly limited by the use of a single animal per group and, using GFP expression as the marker of gene transfer at two dose levels, modeled on the prior experience in humans with AAV-SMN.⁶ In the initial dose group, GFP expression was detected at 21 days after delivery and therefore it is not surprising that gene expression would have begun even sooner in the higher dose group, since plus/minus strand annealing allows for rapid transgene expression when AAV dose is at the level of dose saturation. Specific assays for anti-capsid or anti-GFP immune response are not presented. The rapidity of the response may be related to immune activation of compliment or oxidative damage in hepatocytes due to impurities in the vector preparation. Immuno-toxicity can in fact be managed by immune modulation which would allow for incremental dosing.7, 8 An additional challenge is that the vector production and quantification was accomplished by a different method than established for other clinical programs which followed Good Laboratory Practice (GLP) guidance, especially related to a qualified assay for characterization of vector quantity. Therefore, reliance on the results of these studies should be limited to understanding the means by which an acute response leading to coagulopathy can be explained in NHPs and the unique finding of strain specific murine receptor for AAV PHP.B. In contrast, nine active clinical programs are supported by non-clinical data showing product safety in excess of the clinical dose used in those clinical studies, often in multiple species. The ascending dose study design used in most early phase studies is required by regulators in order to establish the most efficacious dose in the absence of dose limiting toxicity. Table 1 lists the study number in ClinicalTrials.gov, indication, and supporting non-clinical data. Among these studies, the recent publication in NEJM by Mendell et al.⁶ confirms not only the safety of AVXS-101 (AAV9-SMN) therapy but also a striking clinical benefit in a vulnerable patient population under six months of age. The findings of this study and the others underway confirm the important underlying principle of GCP: that risk-benefit must be in favor of a clinically meaningful benefit to participants in the study.

Table 1.

Listing of Active AAV9 Studies in Clinicaltrials.gov

Study Number	Title	AAV9 Vector	Indication	NHPs/Canine	Rodents
NCT02240407	Re-administration of Intramuscular AAV9 in Patients with Late-Onset Pompe Disease	AAV-GAA	Pompe Disease	28	280
NCT02725580	Batten CLN6 Gene Therapy	AAV-CLN6	Batten Disease	n/a	n/a
NCT03381729	Study of Intrathecal Administration of AVXS-101 for Spinal Muscular Atrophy	AVXS-101 AAV-SMN	SMA	n/a	n/a
NCT02362438	Intrathecal Administration of scAAV9/JeT-GAN for the Treatment of Giant Axonal Neuropathy	scAAV9/JeT-GAN	GAN	14	180
NCT03368742	Microdystrophin Gene Transfer Study in Adolescents and Children with DMD	SGT-001	DMD	16/21	500
NCT03362502	A Study to Evaluate the Safety and Tolerability of PF-06939926 Gene Therapy in Duchenne Muscular Dystrophy	PF-06939926	DMD	n/a	n/a
NCT03306277	Gene Replacement Therapy Clinical Trial for Patients with Spinal Muscular Atrophy Type 1	AVXS-001	SMA	4	403
NCT02716246	Phase I/II Gene Transfer Clinical Trial of scAAV9.U1a.hSGSH for Mucopolysaccharidosis (MPS) IIIA	scAAV9.U1a.hSGSH	MPSIII A	n/a	n/a
NCT03315182	Phase I/II Gene Transfer Clinical Trial of AAV9.CMV.hNAGLU for Mucopolysaccharidosis (MPS) IIIB	rAAV9.CMV.hNAGLU	MPSIII B	8	n/a

The regulatory framework which requires the rigorous review of proof-of-concept data and product specific toxicology and biodistributoin studies is the basis for which clinical research with AAV vectors can be accomplished by emphasizing safety as the first priority.⁹ It is important for all those in the field of gene therapy to be vigilant regarding findings related to product safety. An additional tenant of GCP guidance is that the medical care provided to subjects who volunteer for clinical research studies be provided by a qualified physician. In my own experience of providing medical supervision for over 50 individuals enrolled in a variety of clinical studies using AAV-mediated gene transfer, each program requires a detailed knowledge of the supporting literature, the IB, and the product characteristics (Chemistry, Manufacturing, and Control [CMC]), all of which establish the dose, potency, and safety by the proposed route of administration. In the current era of clinical implementation of AAV-mediated gene transfer we must rely on all those in the field to conduct the most informative basic and clinical investigation with the common goal of bringing much needed innovative medical therapy to patients who have a high burden of disease. By following a patient-focused and rational drug development path, the gene therapy community will be successful in meeting the long anticipated expectation of targeted precision medicine using gene therapy.