Characterization and Release Testing for AAV Therapies

Flora: As discussed in the Q+A portion of the webinar, most AAV products are suitable for AUC. The main considerations that can sometimes cause issues are the concentration and volume of available sample, and the presence of possible interfering components/residuals. For new products being tested at CRL, we ask our Clients to complete a quick questionnaire that is reviewed by an AUC specialist that allows us to identify and address any potential issues prior to testing. We also run feasibility testing on any new product types to confirm suitability.

Karen: Acceptable thresholds of partially filled capsids for a given product will depend on what is actually packaged into the capsid (truncated therapeutic transgene, DNA from transfection plasmids, host cell DNA, etc.) and how this impacts safety and efficacy.

Karen: US FDA recommends determining the ratio of full:empty particles¹, but there is not a standard acceptable ratio or threshold for % full. Sponsors need to be able to justify the percent of empty or partially filled capsids in their product and try to minimize them with sufficient downstream processing. Product-specific acceptance criteria will depend on the analytical method used and the product’s dose, indication, and route of administration (ROA)².

In the FDA advisory committee meeting on AAV toxicity³, reviewers highlighted the importance of consistency in impurity levels of a given product until sufficient data is available to set a limit: “For early-phase development, sponsors should, at a minimum, measure the levels of impurities in clinical vector lots and demonstrate similar purity between lots used in IND-enabling preclinical studies and clinical lots. And for licensure and for release of post-licensure lots, manufacturers are required to establish and justify appropriate limits for impurities in their product, based on clinical and manufacturing experience.”

¹FDA Guidance for Industry “Chemistry, Manufacturing, and Control (CMC) Information for Human Gene Therapy Investigational New Drug Applications (INDs)”, January 2020.   

²OTP Town Hall: Gene Therapy Chemistry, Manufacturing, and Controls, April 2023   

³Food and Drug Administration (FDA) Cellular, Tissue, and Gene Therapies Advisory Committee (CTGTAC) Meeting #70, “Toxicity Risks of Adeno-associated Virus (AAV) Vectors for Gene Therapy (GT)”, September 2-3, 2021.

Karen: Characterization of empty/partial/full capsid content should begin as early as possible in development since this attribute will directly impact the dose determination and potency. At a minimum, AAV lots used for IND-enabling preclinical studies should be well characterized for capsid content so they can be compared to future clinical lots. Also, since there is high likelihood that new analytical methods will become available later in the development lifecycle, it is important to keep retain samples from early preclinical and clinical lots so they can be tested with the new methods for the purpose of comparability.

Karen: Off the shelf reference material is available at Charles River in both empty and full capsid format. This reference material is prepared using a triple transient transfection process for wild type AAVs (AAV1, AAV2, AAV5, AAV6, AAV8, and AAV9) with ubiquitous CMV promoter and GFP reporter. FDA supports the use of such viral vector reference materials to calibrate internal reference standards, and then using those internal standards as controls for routine use in analytical assays and validation of assays².
See here for more information: AAV reference material

²OTP Town Hall: Gene Therapy Chemistry, Manufacturing, and Controls, April 2023   

Flora: Our methods all include a 2-hour hold time to allow the centrifuge chamber to reach a stable vacuum, and so that the rotor and AUC cells can equilibrate to the set temperature. The length of time taken for the spin stage of the method depends on a number of factors, such as rpm, matrix viscosity/density, and the empty/full profile of the sample.

Karen:  Since viral transduction is a biological process, transduction efficiency must be measured using a relevant bioassay instead of direct physical measurement. There is not currently a single analytical technique that can emulate the complete biological pathway of the vector binding to the cell membrane, being transported to the nucleus, and releasing the DNA. While there are methods available to evaluate these steps individually for the purpose of capsid screening and optimization (such as cell microarray technology), the actual transduction efficiency of a final drug product requires transduction in a relevant cell line followed by DNA extraction and analysis.

Karl: Aggregated samples are immediately detected in our cryo-EM assay, as the method evaluates data on the actual per-particle level. Highly aggregated (or structurally damaged) particles are not evaluated in a Full/Empty/Intermediate perspective, as it is impossible to assign a meaningful value to a (partially) damaged particle. Intact particles that touch each other in solution, and i.e. form particle rafts, are detected and evaluated.

Karl: Our cryo-EM analysis solution typically tolerates up to 50% host cell protein (HCP) contamination. AAV particles should generally be the major particle type within the analyzed sample.

Karl: The price per sample is typically comparable or lower than that of a corresponding Analytical Ultracentrifugation (AUC) procedure.

Flora: For AUC, the SEDFIT software can calculate the estimated molecular weight of each species, and that information can be used to help provide guidance on the transgene size of partial species. Although it should be noted that this is only an estimated MW, so, as the question suggests, this should be used in comparison with the values generated for the empty and full capsid. 

Karl: Our cryo-EM solution analyzes contrast differences inside each AAV particle that directly correlate with the amount of nucleic acid material (DNA) present in each AAV particle. Approximate per-particle transgene size determination can be achieved but should be set-up and validated on a per sample-family basis to generate meaningful results.

Flora: The serotype and expected genome size of the product are proprietary and we are therefore not able to provide this information. However, as noted in the presentation, AUC is suitable for all AAV serotypes and genome sizes.

Flora: SV-AUC is not a new method for Charles River Labs - the Biophysical Characterization team has been offering it as a service for many years. During that time CRL has performed over 40 qualifications/validations of AAV products for SV-AUC. 

Karl: The cryo-EM AAV characterization technology was recently added to the Charles River portfolio as a next-generation analytical tool for Cell and Gene Therapy applications. The analytic procedure will be offered as a routine test method under Charles River GMP soon.

Karen: Reference material can be customized for specific capsids, promoters, reporter genes, and manufacturing reagents.
See here for more information: CRL AAV products