Gene therapies continue to hold promise as treatments for many diseases but there are numerous and unique challenges to developing them for submission as an Investigational New Drug (IND)/ Investigational Medicinal Product Dossier (IMPD) application for use in clinical trials. These applications require an optimized scientific program designed to address the scientific, regulatory and practical challenges of gene therapy development.
In a recent webinar, we discussed gene therapy preclinical areas central to an IND/IMPD submission, namely, pharmacology, toxicology and pharmacokinetics. The following is a brief summary, plus insightful Q&As, of that event.
Evolution of gene therapies
While gene therapies were conceptualized in the 1970s, it wasn’t until the early 1990s that they were successfully introduced with treatments for adenosine deaminase (ADA) deficiency and familial hypercholesterolemia. Unfortunately, the millennium ended with tragedy in two clinical trials– one in which participants developed leukemia following retroviral correction of severe combined immunodeficiency (SCID), and another in which a fatality occurred in an adenovirus trial to correct ornithine transcarbamylase (OTC). These unfortunate outcomes led to the re-engineering of delivery systems and to this day an increased industry focus on having exceptionally robust scientific programs to ensure strict regulatory and safety compliance.
What are the challenges associated with IND-enabling programs for gene therapies?
Gene therapies require tailored toxicology
Current Guidelines and evolving regulatory standards within preclinical studies can be an issue; although, a multi-disciplinary approach is pertinent in addressing to the toxicology/pharmacology of each specific evaluated therapy. Routes of exposure (e.g. intrathecal or intravenous) and pharmacokinetics (e.g. duration or concentration of exposure) will help reveal potential adverse effects. Usually, most standard approaches cannot be applied to these precision medicines – for example, gene therapy generally utilizes a single species evaluation versus the conventional two.
Each gene therapy requires a unique regulatory approach
From a regulatory perspective, engaging with the United States Food and Drug Administration (US FDA) and other regulatory agencies early on in the process of therapy development is recommended. In addition, Chemistry, Manufacturing and Control (CMC) information (an area beyond the scope of this particular webinar and blog) must be submitted to assure consistent product safety, identity, purity and potency of the investigational product.
Programs may encompass improving upon a current approved therapy or addressing an unmet need in a therapeutic indication. As there is no ‘one-size-fits-all’, working closely with regulators is essential when designing and progressing toward clinical trials. The INitial Targeted Engagement for Regulatory Advice on CBER (Center for Biologics Evaluation and Research) producTs (INTERACT) meetings at the US FDA facilitate identification of issues at an early stage; many public facing regulatory bodies have established and conduct similar programmatic meetings across the globe. This approach is especially useful in dealing with innovative investigational products with unique safety profiles. Fast track development paths within the US FDA are also of paramount importance, as they can offer expedited review and approvals when preliminary evidence suggests that the product benefits are greater than those currently available.
In vivo model selection can impact outcomes
Multiple model options exist, and species selection is dependent on anatomy, disease models and relevant interspecies signalling pathways. The most common is the use of non-diseased animal models. However, without knowledge of the true tissue expression profile, normal healthy animals may encounter issues with overexpression of certain therapies. Therefore, there are two recommended approaches to illustrate efficacy and relevant safety within the preclinical plan. Briefly, the first of these uses solely disease models; the second is a hybrid safety design using healthy and diseased animals, which allows for a control element in the design when the diseased model may have underlying health issues potentiating early morbidity. Immunosuppression regimens also need to be considered when deploying viral delivered gene therapies, of which a successful outcome is based in large part on the expertise of the scientific staff and in the special handling and/or housing of the animals.
Accounting for past exposure to viral vectors
Testing for pre-existing antibodies to adeno-associated vector (AAV) constructs is vital to understand the potential for eliciting an immunogenic response, lower transduction rates, or a potentially fatal response. Animals need to be pre-screened at a minimum two times: an early screening to ensure they are negative to past exposure to AAVs; and a second screening, just before dosing, to ensure against seroconversion. The inclusion of complementary assays and cytokine assays using serum or cerebrospinal fluid (CSF) should be considered. Lentiviral (lenti) constructs do not generally require these types of pre-existing antibody assessments.
Taking precautions against viral shedding
Certain aspects of preclinical study designs need to be considered when utilizing viral elements of delivery. One of these pertains to the timing of dose administration. When cohort dosing or split initiation is separated by more than 2 to 3 days, the dosed and non-dosed animals need to be housed separately to avoid cross contamination of shed (excreted) viral particles via, for example, feces, urine or saliva which may impact to results interpretation. In addition, appropriate personal protective equipment (PPE) for technicians and appropriate hygiene practices for room entry and egress should be employed to ensure the safety of the technical staff and to prevent cross contamination between animals and rooms.
Identifying the optimum routes of administration of a gene therapy
As well as the traditional systemic routes of administration, gene therapies are frequently administered via the CNS or ocular system for a more targeted delivery, depending on the specific disease indication being considered for treatment.
At Covance, we inject around 50–100 microliters of viral vector by intravitreal, subretinal or suprachoroidal routes. These injections can expose the subject to smaller viral loads of around 108–109 copies per ml as the target is proximal to the injection. Whilst leakage from the site of injection can occur, e.g. into the optic nerve, ocular delivery can be optimised, and risks minimized, by working with specialised veterinarian ophthalmologists. Further, we partner with Ocular Services On Demand (OSOD)– a world-class team of vision scientists to ensure exceptional study design, execution and results interpretation. Functional changes in the reactivity of the eye can be assessed during the live phase and specialized tissue collections post-life can evaluate the biodistribution within the eye as well as assess any changes in histology.
Central nervous system (CNS) administration
CNS administration takes advantage of the CSF antegrade and retrograde flow for viral vector distribution. An easier and clinically relevant access point is within the lumbar spine below the level of the conus medullaris but within the area of the cauda equina. A blunt needle is inserted into the lumbar intrathecal space between vertebrae (below third lumbar vertebrae) of the flexed spine. Cerebral spinal fluid is removed and the same dose volume of the test material formulation is gently injected over the course of at least 1 minute. Once the needle is removed, the animal is placed in a Trendelenburg position for improved distribution. Other routes include injecting into the cisterna magna which places the material in the proximity of the fourth ventricle and central canal, and injecting the intracerebral ventricular area which targets the lateral ventricles, and specific parenchymal regions where stereotaxic MRI guided dose administration is utilized to ensure targeted delivery. Post-dose clinical observations should include evaluating for pain or distress (headaches) and potential neural deficits due to increased pressure upon injection; with respect to the latter, analyzing injection rates and volumes is crucial to minimizing the impact.
Designing biodistribution assessments
Biodistribution needs to be monitored at intervals generally ranging from 1–12 months and is key to understanding the distribution of product in multiple tissue types. Even with targeted delivery, viral particles can become widely distributed; thus multiple systemic tissues should also be evaluated. The gonads should also be analyzed as concentration here may influence potential transfer to offspring. In the brain (flash-frozen or fixed), the right side is often used for biodistribution and the left side for histology. There is also a known risk of severe toxicity and degeneration of the dorsal route ganglia (DRGs) after viral vector therapy. It is a labour-intensive task to analyze the regions within the DRG, for example: the cervical, thoracic and lumbar regions, however these can be collected and analyzed for transgene product as well as histopathology. Distal peripheral nerves as well as ascending tracts within the spinal cord should also be evaluated histologically to determine extent of changes.
At Covance, we streamline the journey of developing unique precision medicine products. We have demonstrated expertise in the gene therapy space, helping scientists overcome the challenges and study design considerations relating to toxicology, CMC testing, biodistribution analysis, biomarkers and transitioning rapidly into first in human trials and beyond.
Q: Could you comment on the inclusion and exclusion criteria based on pre-screening antibody titers and what assays are typically used for neutralizing antibodies?
A: Most humans and animals test subjects have experienced some exposure to specific serotypes of adeno-associated viruses (AAV). In test animals, we aim for ≤1:10 titers. It is important to understand pre-screening protocols and what a laboratory defines as a ‘negative background’ and how relevant that is to your situation. Sometimes we must utilize seropositive animals as controls, as there may not be enough seronegative animals in the current supply for a specific AAV serotype. What’s crucial is knowing the cut-offs for the assays and the situation with each animal.
Q: Can you control the copy number? What range of copy number is typical?
A: Copy number is dependent on dose and route of administration, although 108–1014 is typical, and for eye or brain 108 is generally adequate. Higher copy numbers usually equate to greater biodistribution of the virus.
Q: How does Covance address shedding samples in infectivity assays?
A: Most assays we have conducted have shown positive or negative shedding, and not an assessment of infectivity of those samples. The nonsterile sample (e.g. feces, urine, etc) is the challenge, and in vitro plaque assay formation is particularly difficult to keep clean.
Q: What is the stance at Covance on readouts for biodistribution – DNA, RNA or protein? Or a combination?
A: Typically, readouts for gene therapy biodistribution has used DNA, RNA or protein – or the activity of that protein. DNA via qPCR is the most common method. However, in consideration of RNA and protein, it may be difficult to discern between inter-species RNA and establish if the measured activity results from, for example, the humanized recombinant versus native species enzyme. To differentiate between the two isoforms, an examination of the mRNA indicates that the target gene is being transcribed. It is vital to differentiate between endogenous protein and exogenous product using techniques like electroretinography (ERG), electroencephalogram (EEG) and Doppler to evaluate changes in targeted function if possible.
In some programs, we are able to identify the DNA within tissue, but not the RNA or protein, so establishing an argument on the mRNA or protein readout will help convince regulatory bodies that your product is likely to function correctly.
Q: What is your opinion on tissue distribution?
A: It is important to be as comprehensive as possible. We recommend collecting approximately 14-20 tissues at necropsy. For example, if one is administering viral vectors into the brain, then samples of tissues from the brain, spinal column or spinal cord are clearly required. It’s important to collect and analyze tissues that may experience systemic exposure such as the liver, kidneys, intestine, eye and draining lymph nodes. Also it is paramount to check distribution and/or expression in the gonads – as questions related to germline cell transfer will allow one to be prepared to address that concern.
Q: What PK time points do you recommend for AAV IM or IV routes of administration?
A: Given the IgM wave generally increases after the first 6–8 days, equating to the rapid decline of AAV vectors with the IgG wave response producing full clearance of circulating AAV at around 10 days; addressing the PK of the viral capsid is more akin to a vaccination response. As AAV and most viruses used to deliver gene therapies are non-replicative, there is no continued expansive viral production once administered. This attribute will lower expression of the product, however, biodistribution will persist for longer periods of time. We don’t recommend sampling on the day of administration, as the virus needs to transduce cells and expression needs to be established. However, after the first week a time-point may be collected and then followed up with either weekly or monthly intervals for the study duration and is therapy specific, especially if the transgene produces a product that enters the circulation
AAV adeno-associated virus
ACO Antisense Oligonucleotide
ADA adenosine deaminase
CBER Center for Biologics Evaluation and Research
CMC Chemistry, Manufacturing and Control
CSF cerebrospinal fluid
DNA deoxyribonucleic acid
DRG dorsal route ganglia
FDA Food and Drug Administration
IND investigational new drug
INTERACT The INitial Targeted Engagement for Regulatory Advice on CBER (Center for Biologics Evaluation and Research) producTs (INTERACT)
MRI magnetic resonance imaging
OSOD Ocular Services On Demand
OTC ornithine transcarbamylase
RNA ribonucleic acid
SCID severe combined immunodeficiency