Pharmaceutical Biotechnology
Mechanistic Studies and Formulation Mitigations of Adeno-associated Virus Capsid Rupture During Freezing and Thawing: Mechanisms of Freeze/Thaw Induced AAV Rupture

https://doi.org/10.1016/j.xphs.2022.03.018Get rights and content

Abstract

Gene therapies delivered using adeno-associated virus (AAV) vectors are showing promise for many diseases. Frozen AAV drug products are exposed to freeze-thaw (F/T) cycles during manufacturing, storage, and distribution. In this work we studied the mechanisms of AAV capsid rupture during F/T. We found that exposure to interfaces, exacerbated by F/T, and the mechanical force of excipient devitrification correlated with AAV capsid rupture during F/T. There was no impact of pH shifts, cryo-concentration, or cold-denaturation. Results were similar for AAV8 and AAV9. With these mechanistic insights we identified three formulation mitigation approaches. Addition of ≥0.0005% w/v poloxamer 188 (P188) eliminated substantial recovery losses (up to ∼60% without P188) and minimized rupture to ≤1% per F/T cycle. Elimination of exothermic devitrification events during rewarming, either by formulating with a low buffer concentration, or by adding a cryoprotectant further reduced rupture during F/T. Rupture of AAV9 was <0.2% per F/T cycle in a formulation with 1 mM phosphate, 4.4 mM dextrose, electrolytes, and 0.001% P188 at pH 7.2. Rupture of AAV8 was not detected when formulated with 4% sucrose, 100 mM salt, and 0.001% P188 at pH 7.4. These results provide insights into effective strategies for stabilizing AAVs against rupture during F/T.

Introduction

There has been an increase in the number of clinical studies evaluating adeno-associated virus (AAV) as a vector for gene therapy.1 Development of formulations that confer stability to AAV during manufacturing and extended storage is a major challenge.2,3 The route of administration may further limit the selection of buffers and excipients included in an AAV formulation. For example, it is preferred that a formulation has an extensive clinical history of being well-tolerated for ocular or intrathecal routes of administration, especially for infant patients. Frozen drug product (DP) may be used to supply early clinical studies to maximize the AAV stability and facilitate rapid deployment. The AAV formulation is typically exposed to two or three freeze/thaw (F/T) cycles during manufacturing, storage, and distribution.4 Additional F/T cycles may occur during excursions.

There have been differences in the reported outcomes of a relatively limited number of studies published on the stability of different AAV serotypes and formulations to F/T. AAV1 formulated in a phosphate-buffered saline (PBS) formulation was reported to be stable and able to generate functional transgene product in mice after 4 F/T cycles.5 In contrast, there was a loss in activity of AAV2 after F/T in a PBS formulation that was attributed to exposure to low pH conditions when frozen.2 In another study, there was up to a 75% concentration loss of AAV2 when exposed to F/T in a PBS formulation with 0.001% poloxamer 188 (P188) that was mitigated by reformulating with Tris buffer.6 In our own recently reported studies, we found that there was a minor increase in non-encapsidated (free) DNA released from ruptured AAV8 capsids exposed to 5 F/T cycles in Dulbecco's Phosphate-Buffered Saline (DPBS) with 0.001% P188 (F1).4 Capsid rupture during F/T was minimal for AAV9 in a formulation with a low sodium phosphate buffer concentration of 1 mM and 0.001% P188 (F4) and not detected for both AAV8 and AAV9 in a formulation containing the cryoprotectant, sucrose at 4% w/v with 0.001% P188 (F3).4 The potency, concentration, aggregation, and other attributes of AAV8 (in F1 and F3) and AAV9 (in F3 and F4) were unchanged after 5 F/T cycles.4 Xu et al. reported that F/T in DPBS with 0.001% P188 (F1) caused release (leakage) of DNA from AAV8, AAV3B, AAV5, AAV7, and AAVDJ, but not from AAV2.7 Addition of 10% w/v sucrose and 0.1% P188 reduced the release (leakage) of DNA from AAV8 during F/T cycles.

These reports suggest that a deeper understanding of the mechanism(s) of AAV rupture during F/T could be helpful in process and formulation mitigation strategies, and also in the selection of formulation and stability screening methods. Degradation of biologics during F/T may be a result of several factors, including: shifts in pH during cooling, cold denaturation, devitrification (crystallization of amorphous water/excipients) during rewarming, adsorption to interfaces (ice, air bubbles, or crystalline buffers, salts, and excipients), and cryo-concentration of excipients.8, 9, 10, 11, 12 Each of these factors was evaluated in this work with the objective of identifying approaches to stabilize AAV8 and AAV9 against rupture and associated release of DNA during F/T. An understanding of the mechanism of rupture during F/T can guide the selection of optimal formulation excipient types and levels. Shifts in the pH of frozen formulations were measured using a pH meter at -20°C, differential scanning fluorimetry (DSF) was used to assess the possible impact of conformational stability and cold denaturation, and differential scanning calorimetry (DSC) was used to assess the potential impact of devitrification events during rewarming. Variations of the formulations with lower and higher P188 levels were used to assess the impact of exposure to interfaces on adsorption losses and capsid rupture. Cryo-concentration was assessed by comparing formulations with different levels of ionic and non-ionic excipients.

As a baseline, the stability of both AAV8 and AAV9 with respect to rupture and DNA release during F/T were evaluated in four formulations. DNA released from ruptured capsids was quantitated by a fluorescent dye-based method and confirmed with size-exclusion chromatography (SEC). All base formulations included 0.001% w/v P188 to protect against adsorption losses to contact surfaces. DPBS with 0.001% P188 (F1) and TRIS-buffered saline (TBS) with 0.001% P188 (F2) were assessed as ‘off-the-shelf’ formulations that might be considered for use in very early studies when there is limited stability data available. F3 was developed for either intravenous or ocular administration of both AAV8 and AAV9, with the same buffer as F1 but with 4% w/v sucrose added as a cryoprotectant, and a corresponding reduction in salt to compensate the tonicity. F4 was developed for intrathecal delivery of AAV9. To ensure maximum tolerability, F4 was designed to have a composition and pH similar to cerebrospinal fluid (CSF), with other modifications, while maintaining stability of the vector. CSF contains sodium chloride and glucose (dextrose), low levels of potassium, calcium and magnesium electrolytes, a low level of phosphate, and bicarbonate at pH 7.3.13 For intrathecal drugs delivered to the CSF, a phosphate buffer level of ≤ 5 mM is recommended because of adverse events observed at higher phosphate buffer levels.14 F4 was designed with a low buffer concentration (1 mM), dextrose, low levels of electrolytes, with added P188, but without bicarbonate (to reduce pH shifts due to off gassing of CO2). In addition to comparing the stability of both AAV8 and AAV9 in all four formulations, a deeper evaluation of the impact of excipient levels was studied for AAV8 in F1, F2, and F3 and for AAV9 in F4. These included variations in P188 levels, and other key excipients including sucrose, electrolytes, and buffer levels. These additional studies were performed to gain a greater insight into the mechanisms contributing to capsid rupture and DNA release during F/T and the robustness of these formulations.

Section snippets

Production and Formulation of AAV8 and AAV9 Vectors

Adeno-associated virus serotype 8 (AAV8) and serotype 9 (AAV9) were produced by REGENXBIO Inc. Both vectors contained ssDNA with approximately 4 kb. Two lots of AAV8 were used with an initial level of about 2.5% (previously exposed to 4 F/T cycles in F1) and 0.8% non-encapsidated (‘free’) DNA, respectively. Two lots of AAV9 were used, containing initial levels of 1.1% and 0.3% of free DNA. The initial level of free DNA is shown in data plots and is reported as the average value measured in

Formulation Impacts on DNA Release During Freeze/thaw

The release of DNA from AAV8 and AAV9 was measured after up to 10 F/T cycles in each of the four formulations (Fig. 1). The free DNA levels presented in this work were measured using the fluorescent dye method due to its high throughput and sensitivity. DNA release during F/T cycles was highest for both AAV8 and AAV9 in F2 and F1 (Fig. 1 and Table 1). In contrast, the DNA release was very low in F4, and no increase was detected in F3, even after 10 F/T cycles. The orthogonal size exclusion

Discussion

Rupture of AAV8 and AAV9 was very low when held at room temperature or at 37°C. In contrast, freeze/thaw (F/T) caused very high levels of rupture resulting in release of DNA in certain formulations. We identified two factors that correlated with DNA release from AAV8 and AAV9 capsids ruptured during F/T. The first was exposure to the ice interface during cooling and the second was the mechanical force of devitrification during rewarming. The impact of the different potential mechanistic factors

Conclusions

Adsorption to ice during cooling and mechanical force of devitrification during rewarming were identified as leading factors correlated with AAV rupture during F/T of AAV8 and AAV9. In contrast to high rupture during F/T, rupture was low at room temperature and at 37°C. In terms of formulation screening tools, F/T of formulations with different P188 levels could be used in P188 level optimization to mitigate adsorption losses and rupture during F/T. Screening of cryoprotectant and buffer levels

Declaration of Competing Interest

All authors are, or were, employees of REGENXBIO Inc. which is a clinical-stage AAV gene therapy company. The authors declare no other conflicts of interest.

Acknowledgments

The authors acknowledge the upstream and downstream process development, manufacturing teams, and analytical teams for producing and testing the AAV materials used in this work. Funding for this project was provided by REGENXBIO Inc.

References (39)

  • JS Bee et al.

    Effects of surfaces and leachables on the stability of biopharmaceuticals

    J Pharm Sci

    (2011)
  • SA Maskarinec et al.

    Direct observation of poloxamer 188 insertion into lipid monolayers

    Biophys J

    (2002)
  • TA Khan et al.

    Key interactions of surfactants in therapeutic protein formulations: a review

    Eur J Pharm Biopharm

    (2015)
  • P Boutron et al.

    Stability of the amorphous state in the system water-glycerol-dimethylsulfoxide

    Cryobiology

    (1978)
  • N Milton et al.

    Vial breakage during freeze-drying: crystallization of sodium chloride in sodium chloride-sucrose frozen aqueous solutions

    J Pharm Sci

    (2007)
  • F Maksudov et al.

    Fluctuating nonlinear spring theory: strength, deformability, and toughness of biological nanoparticles from theoretical reconstruction of force-deformation spectra

    Acta Biomater

    (2021)
  • Y Zhang et al.

    Development of a stable lyophilized adeno-associated virus gene therapy formulation

    Int J Pharm

    (2021)
  • D Wang et al.

    Adeno-associated virus vector as a platform for gene therapy delivery

    Nat Rev Drug Discov

    (2019)
  • MA Croyle et al.

    Development of formulations that enhance physical stability of viral vectors for gene therapy

    Gene Ther

    (2001)
  • Cited by (8)

    • The ice age – A review on formulation of Adeno-associated virus therapeutics

      2023, European Journal of Pharmaceutics and Biopharmaceutics
    View all citing articles on Scopus
    View full text