WO2025125222A1

WO2025125222A1 - Compositions comprising viruses, viral vectors or virus-like particles

Info

Publication number: WO2025125222A1
Application number: PCT/EP2024/085471
Authority: WO
Inventors: Kristina KEMTER; Ronny Roman SCHMIDT
Original assignee: Leukocare AG
Current assignee: Leukocare AG
Priority date: 2023-12-11
Filing date: 2024-12-10
Publication date: 2025-06-19
Anticipated expiration: 2026-06-11

Abstract

The disclosure relates to stabilised liquid compositions comprising viruses, viral vectors and virus-like particles. Furthermore, the disclosure relates to methods of providing and using said compositions.

Description

COMPOSITIONS COMPRISING VIRUSES, VIRAL VECTORS OR VIRUS-LIKE PARTICLES

BACKGROUND

Viruses, viral vectors or virus-like particles are commonly used as delivery vehicles in therapeutic and prophylactic applications, e.g., in gene therapy and vaccine. While viruses have been used as vaccines already in the last century, virus like particles (VLPs) are a more recent development. Viral vectors and VLPs offer a series of advantages over traditional vaccines. In addition to inducing exceptional antibody responses, they also elicit cytotoxic T lymphocytes (CTL) that are crucial for the control of intracellular pathogens and cancer without adjuvants, a feature not observed by protein-based vaccines (Ura T. et al., 2014).

In the last two decades replication deficient viral vectors have emerged as safe and effective delivery vehicles for clinical gene transfer therapy for the treatment of genetic and infectious diseases, cancer and enzyme replacement therapy, as shown in a series of clinical studies, especially for monogenic recessive disorders, but also for some idiopathic. These clinical studies were conducted on the basis of vectors that combine low genotoxicity and immunogenicity with highly efficient delivery, including vehicles based on adeno-associated virus and lentivirus, which are increasingly enabling clinical success. Important examples for clinical treatment strategies based on viral vectors include, e.g., stem cell therapy, mucoviscidosis, haemophilia, inherited retinopathy or cystic fibrosis (Collins M. & Trascher A., 2015) Typically, the viral vectors employed in such gene transfer therapeutics include adenovirus, adeno-associated virus (AAV) among other viruses.

Adeno-associated virus (AAV) is a small non-enveloped virus that infects humans and some other primate species. AAV is not currently known to cause disease and, consequently, the virus causes a very mild immune response (Rieser R et al., 2021). AAV can infect both dividing and non-dividing cells and can incorporate its genome into that of the host cell. Moreover, episomal AAV elicits long and stable expression and, thus, AAV is suitable for creating viral vectors for gene therapy. Because of its potential use as a gene therapy vector, AAV has previously been modified (self-complementary adeno-associated virus; scAAV). Whereas AAV packages a single strand of DNA and requires the process of second-strand synthesis, scAAV packages both strands which anneal together to form double stranded DNA. This approach allows for rapid expression in the target cell. Furthermore, viral vectors are used for other cell-based therapies, wherein isolated cell populations are transduced with a trans-gene which is not comprised in the naturally occurring genotype of said cell, for example a gene encoding a chimeric antigen receptor (CAR) used in the widely emerging field of CAR-T cell therapies (Bulaklak K. et al., 2020).

Despite significant clinical and commercial successes, the cost-effective manufacturing of viral vectors yet remains a challenge (Srivastava A. et al., 2021). Especially, viral vectors may often be unstable, which can make them inconvenient or expensive for widespread use.

The efficacy or effectiveness of viruses or virus-like particles as vaccines or the efficiency of viral vectors for in vitro or in vivo gene transfer is highly dependent on their stability during the time of storage and transport since these virus related products are prone to degradation and aggregation, particularly when agitated or exposed to elevated temperatures such as room temperature (Srivastava A. et al., 2021).

To stabilise vaccines and other biologically active viral products, freeze drying (i.e., lyophilisation) is frequently used to preserve the activity of the product for storage and transport (Rieser R. et al., 2021). However, many virus derived products experience significant loss of potency during the drying process and subsequent rehydration.

As an alternative procedure to stabilise viral products, many products, including AAV, are currently stored in frozen state (Srivastava A. et al., 2021). However, as observed for drying, freeze-thaw cycles often lead to significant loss of activity. Further to the loss of activity related to drying and rehydrating, or freezing and thawing, the respective products are relatively inconvenient to handle since they cannot be administered in the form in which they are stored, but require further steps such as rehydrating or thawing to obtain the ready-to-use product.

Accordingly, there is a need to provide improved compositions for storing viruses, viral vectors or virus-like particles in liquid form.

SUMMARY

In a first aspect, the present disclosure relates to a liquid composition comprising a virus, or a viral vector or virus-like particle and (a) at least one compound according to formula (I)

wherein R1 and R2 are independently from each other selected from H, Ci-C4-alkyl, CH2CH2OH or CH₂CH(CH₃)OH, and/or

(b) threonine, and

(c) at least one buffer, and

(d) optionally at least one surfactant.

In a preferred aspect, the liquid composition is an aqueous composition.

In a highly preferred aspect, the at least one compound according to formula (I) comprises or is N-methyl-D-glucamin.

In a further aspect, the liquid compositions disclosed herein further comprise an amino acid with a negatively charged functional group, preferably glutamic acid.

In a further aspect, the liquid compositions disclosed herein further comprise at least one antioxidant, preferably an amino acid providing an anti-oxidative function, more preferably carnosine and/or methionine.

In a further aspect, the liquid compositions disclosed herein further comprise at least one polyethylene glycol, preferably polyethylene glycol 3350.

In a further aspect, the liquid compositions disclosed herein further comprise an excipient selected from inositol and preferably valine or a combination thereof.

In a further highly preferred aspect, the virus is a non-enveloped virus, preferably selected from an adenovirus or adeno associated virus.

In a further aspect, the adeno associated virus is AAV1 , AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9, or AAV10, preferably AAV2. In a further aspect the invention relates to a liquid composition comprising a virus, or a viral vector or virus-like particle and:

(a) threonine in a concentration of at least 80 mM and,

(b) valin and/or,

(c) glutamin.

In another aspect, the present disclosure relates to a composition as disclosed herein in frozen or dried form.

In another aspect, the present disclosure relates to a method of providing a liquid composition comprising a virus, or viral vector or virus-like particle as disclosed, wherein the method comprises storing the liquid composition disclosed herein in liquid form for at least 4 weeks, preferably at least 6 weeks.

In another aspect, the present disclosure relates to a method of providing a virus, or viral vector or virus-like particle to a subject in need thereof, comprising the steps of:

(i) providing a liquid composition comprising a virus, or viral vector or virus-like particle as disclosed herein;

(ii) storing the liquid composition as disclosed in liquid form for at least 4 weeks;

(iii) administering the adeno associated virus, or viral vector or virus-like particle to the subject in need thereof, preferably in a liquid composition as disclosed herein.

In a further aspect, the present disclosure relates to a therapeutic or prophylactic method of treatment, wherein a virus, or viral vector or virus-like particle is provided according to a method as disclosed herein and is subsequently administered to a subject.

In a further aspect, the present disclosure relates to a method of transfecting a cell with a virus or viral vector derived therefrom, comprising the steps of:

(i) providing a liquid composition as disclosed herein;

(ii) storing the liquid composition as disclosed herein in liquid form for at least 4 weeks;

(iii) contacting the cell with the adeno associated virus, or viral vector derived therefrom.

In a further aspect, the present disclosure relates to a process for providing a liquid composition comprising a virus, or viral vector or virus-like particle as disclosed herein, the process comprising at least:

(i) a first step comprising providing a first liquid composition comprising a virus, or viral vector or virus-like particle in a first aqueous medium; (ii) a second step of exchanging the first aqueous against an second aqueous medium comprising excipients as disclosed herein, comprising at least one polyethylene glycol, preferably polyethylene glycol 3350, to obtain a second liquid composition; and optionally

(iii) a third step of exchanging the second aqueous medium against a third aqueous medium comprising excipients as disclosed herein, which is substantially free of polyethylene glycol to obtain a third liquid composition.

DESCRIPTION OF THE DRAWINGS

Figure 1 : Analysis of the AAV2 viral genome titer (vg/mL) monitored by ddPCR and full capsid concentration (full cg/mL) analyzed by SE-HPLC-Multi-Detector Method (MD) after rebuffering by dialysis. For comparison the original, untreated Reference (AAV2 in PBS buffer at pH 7.4 + 0.001 % Poloxamer 188) was included. Arrows indicate the buffers with the highest loss of AAV2 concentration after re-buffering by dialysis.

Figure 2: Results of the nano DSF analysis of thermal unfolding of AAV2 viral vectors in the Reference and in the 12 different buffers at different pH values. The onset temperature of thermal unfolding T_on and midpoint temperature of thermal unfolding T_m are shown. The dark grey arrows represent the lowest thermal stability of the AAV2 viral vectors in the particular buffer and light grey arrows represent the highest analyzed thermal stability of the AAV2 viral vectors in the particular buffer.

Figure 3: Leakage temperatures of DNA from the AAV2 capsids T_mi and capsid rupture temperatures T_m2 of the AAV2 capsids in the different buffers at different pH values analyzed by SYBR Gold Fluorescence upon application of a thermal ramp. For comparison, the original, untreated Reference (PBs buffer pH 7.4 + Poloxamer 188) was included. The dark grey represent the lowest stability and the light grey represent the highest stabilities of the AAV2 viral vectors in the different buffers at different pH values, particularly regarding the capsid rupture temperatures T_m2.

Figure 4: % total capsids of the AAV2 viral vectors formulated in the 12 different buffers at different pH values in comparison to the original, untreated Reference formulation (PBS at pH 7.4 + 0.001 % Poloxamer 188) analyzed by SE-HPLC-Multi-Detector-Method at time-point t = 0 and after 2 weeks and 4 weeks of storage at 30 °C. Total capsid concentration at time-point t = 0 was set to 100 % and the remaining total capsid concentrations after 2 weeks and 4 weeks of storage at 30 °C, respectively were normalized to 100 % total capsids at time-point t = 0. Figure 5: % full capsids of the AAV2 viral vectors formulated in the 12 different buffers at different pH values in comparison to the original, untreated Reference formulation (PBS at pH 7.4 + 0.001 % Poloxamer 188) analyzed by SE-HPLC-Multi-Detector-Method at time-point t = 0 and after 2 weeks and 4 weeks (grey bars) of storage at 30 °C.

Figure 6: % empty capsids of the AAV2 viral vectors formulated in the 12 different buffers at different pH values in comparison to the original, untreated Reference formulation (PBS at pH 7.4 + 0.001 % Poloxamer 188) analyzed by SE-HPLC-Multi-Detector-Method at time-point t = 0 and after 2 weeks (light blue bars) and 4 weeks (grey bars) of storage at 30 °C.

Figure 7: Main effect plot of the results of the statistical regression analysis of the total particle concentration at time-point t = 0 monitored by SE-HPLC-Multi-Detector-Method (low quality fit).

Figure 8: Onset temperature of thermal unfolding T_on and midpoint temperature of thermal unfolding T_m of the AAV2 capsids in the 20 formulations after re-buffering in comparison to the original, untreated Reference formulation. The light grey represent the good stabilizing formulations and the dark grew arrows represent the less stabilizing formulations.

Figure 9 a-h: Main effect plots of the statistical regression analysis of the nano DSF data, showing the statistical significant positive effects of the excipients valine, glutamic acid, threonine and carnosine on the thermal stability of the AAV2 viral vectors in form of T_on and T_m. Good fits for nano DSF.

Figure 10: DNA leakage temperature from the AAV2 capsids T_mi and rupture temperature of the AAV2 capsids T_m2 and for some formulations an additional thermal transition T_m3 (dark grey bars) monitored by SYBR Gold DSF in the 19 DoE formulations in comparison to the original, untreated Reference formulation and the re-buffered reference formulations F20.

Figure 11 a-f: Results of the statistical regression analysis for the 19 DoE formulations with significant positive influence of valine, threonine, inositol and meglumine on DNA leakage temperature T_mi. A significant negative effect on the DNA leakage temperature T_mi was analyzed for carnosine and PEG3350.

Figure 12 a-d: Results of the statistical regression analysis for the 19 DoE formulations with significant positive influence of valine, threonine, and carnosine on the capsid rupture temperature T_m2. A significant negative effect on the AAV2 capsid rupture temperature T_m2 was analyzed for PEG3350.

Figure 13 a: Total capsid concentration (total cp/mL) before storage at 30 °C at time-point t = 0, after 2 weeks, 4 weeks, and 6 weeks of storage at 30 °C analyzed by SE-HPLC-MD in the original, untreated Reference, the 19 DoE formulations and formulation F20. The light grey arrows indicate good stabilizing formulations and dark grey arrows indicate less stabilizing formulations.

Figure 13 b: % AAV2 total capsids analyzed by SE-HPLC-MD at time-point t = 0 (100 %; dark blue bars), after 2 weeks of storage. 4 weeks of storage, and 6 weeks of storage at 30 °C in the Reference, the 19 DoE formulations and in formulation F20.

Figure 14 a-f: Results of the statistical regression analysis regarding the difference between the total capsid concentrations at t = 0 and after 4 weeks and 6 weeks of storage at 30 °C in terms of the main effect plots. For the excipients threonine and meglumine the negative slope represents only low loss of total capsid concentration during storage and so a positive influence on the stability of the AAV2 viral vectors. The positive slopes for PEG3350 represent increasing loss of total capsid concentration during storage at 30 °C and so a negative effect on the storage stability of AAV2 viral vectors.

Figure 15: Full capsid concentration (full cp/mL) before storage at 30 °C at time-point t = 0, after 2 weeks, 4 weeks, and 6 weeks of storage at 30 °C analyzed by SE-HPLC-MD in the original, untreated Reference, the 19 DoE formulations and formulation F20. The light grey arrows indicate good stabilizing formulations and dark grey arrows indicate less stabilizing formulations.

Figure 16: % AAV2 full capsids analyzed by SE-HPLC-MD at time-point t = 0 (100 %; dark blue bars), after 2 weeks of storage (light blue bars). 4 weeks of storage (grey bars), and 6 weeks of storage at 30 °C in the Reference, the 19 DoE formulations and in formulation F20.

Figure 17 a-d: Results of the statistical regression analysis regarding the difference between the full capsid concentrations at t = 0 and after 4 weeks and 6 weeks of storage at 30 °C in terms of the main effect plots.

Figure 18: Empty capsid concentration (empty cp/mL) before storage at 30 °C at time-point t = 0, after 2 weeks, 4 weeks, and 6 weeks of storage at 30 °C analyzed by SE-HPLC-MD in the original, untreated Reference, the 19 DoE formulations and formulation F20. The light grey arrows indicate good stabilizing formulations and dark grey arrows indicate less stabilizing formulations.

Figure 19: % AAV2 empty capsids analyzed by SE-HPLC-MD at time-point t = 0 (100 %; dark blue bars), after 2 weeks of storage (light blue bars). 4 weeks of storage (grey bars), and 6 weeks of storage at 30 °C in the Reference, the 19 DoE formulations and in formulation F20. For the better overview, we again normalized (%) the values for the empty capsid concentration during storage to the 100 % at time-point t = 0. In Figure 19, the % empty capsids are depicted for all DoE formulations, formulation F20 and the Reference.

Figure 20: Ratios full to total AAV2 capsids during storage at 30 °C, at time-point t = 0, after 2 weeks, 4 weeks and 6 weeks of storage at 30 °C (red bars). The arrows represent the formulations with the strongest changes in the full to total capsid ratios.

Figure 21 a-d: Results of the statistical regression analysis regarding the difference between the empty capsid concentrations at t = 0 and after 4 weeks of storage at 30 °C in terms of the main effect plots. For the excipient glutamic acid, threonine, and meglumine the negative slopes represents only low loss of empty capsid concentration during storage and so a positive influence on the stability of the empty AAV2 viral vectors. The positive slopes for PEG3350 represent increasing loss of empty capsid concentration during storage at 30 °C and so a negative effect on the storage stability of the empty AAV2 viral vectors.

Figure 22 a-d: Results of the statistical regression analysis regarding the difference between the empty capsid concentrations at t = 0 and after 6 weeks of storage at 30 °C in terms of the main effect plots. For the excipient glutamic acid, threonine, and meglumine the negative slopes represents only low loss of empty capsid concentration during storage and so a positive influence on the stability of the empty AAV2 viral vectors. The positive slopes for PEG3350 represent increasing loss of empty capsid concentration during storage at 30 °C and so a negative effect on the storage stability of the empty AAV2 viral vectors.

Figure 23: Thermal unfolding of AAV2 monitored by nanoDSF. The bars represent the onset temperature of thermal unfolding T_on in blue and the midpoint temperature of thermal unfolding T_m in red. The highest T_on values > 70 °C (dashed line at 70 °C) were observed in the formulations F_04, F_06, F_08, F_12, F_13, and F_17. The highest T_m values > 74 °C (dashed line at 74 °C) were analyzed in Formulations F_04, F_06, F_08, F_12, and F_14. Figure 24: Thermal stability of the AAV2 viral vectors analyzed by SybrGold DSF. The leakage temperature Tm1 where the DNA starts to leave the viral vector via small lacks in the protein shell is depicted (left columns). The rupture temperature Tm2, where the protein capsids is more and more damaged with increasing temperature is depicted (right columns).

Figure 251 Thermal stability of the AAV2 viral vectors. Comparison of the onset temperature of thermal unfolding T_on analyzed by nanoDSF and the rupture temperature T_m2 of the AAV2 analyzed by SybrGold DSF (left columns). The Ton values from the nanoDSF are depicted as the blue bars and the Tm2 values from the SybrGoldDSF were depicted as (right columns).

Figure 26: Stability of the AAV2 viral vectors analyzed in form of the virus genome inside the AAV2 particles per mL (vg/mL) in the 20 DoE formulations compared to the reference formulation F_21 during storage for up to 4 weeks at 40 °C monitored by ddPCR. For comparison, the titer of the positive control is represented as the dotted line in this graph. The titers of the samples are depicted as points.

Figure 27 (A-D): SEC-MALS data for the total capsid concentration/mL (A), the full capsid concentration/mL (B), the empty capsid concentration/mL (C) and the full/total capsid ratio (D) at time-point t = 0 , after 1 week of storage at 40 °C, after 2 weeks of storage at 40 °C, and after 4 weeks of storage at 40 °C.

Figure 28 (A-F): Contour plots representing the significant interactions between the excipients in the optimization DoE formulations according to the most suitable concentration ranges for the highest stabilizing effect on the AAV2 total capsid concentration per mL directly after 4 weeks of storage at 40 °C. The concentrations of the excipients were represented as coded values in these contour plots.

DEFINITIONS

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs. In case of conflict, the patent specification, including definitions, will prevail.

The term "at least", as used herein, refers to the specifically recited amount or number but also to more than the specifically recited amount or number. For example, the term "at least one" encompasses also at least 2, at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, such as at least 20, at least 30, at least 40, at least 50 and so on. Furthermore, this term also encompasses exactly 1 , exactly 2, exactly 3, exactly 4, exactly 5, exactly 6, exactly 7, exactly 8, exactly 9, exactly 10, exactly 20, exactly 30, exactly 40, exactly 50 and so on. When used in the context of a compound family, such as for example amino acids, sugars, surfactants, the term “at least" refers to the presence of at least one species from that group and not only to the presence of one molecule. For example, at least one sugar refers to the presence of at least one species form the family of sugars such as sucrose.

The term "about", as used herein, encompasses the explicitly recited values as well as small deviations therefrom. In other words, an amount of sugar of "about 80 mg/ml" includes, but does not have to be exactly the recited amount of 80 mg/ml but may differ by several mg/ml, thus including for example 92 mg/ml, 84 mg/ml, 88 mg/ml, 76 mg/ml, 72 mg/ml or 68 mg/ml. The skilled person is aware that such values are relative values that do not require a complete accuracy as long as the values approximately correspond to the recited values. Accordingly, a deviation from the recited value of for example 15%, more preferably of 10%, and most preferably of 5% is encompassed by the term "about". These deviations of 15%, more preferably of 10% and most preferably of 5% hold true for all embodiments pertaining to this disclosure wherein the term “about” is used.

As used herein, the terms "a" and "an" and "the" and similar referents in the context of describing the disclosure (especially in the context of the following claims) are to be construed to cover both the singular and the plural, unless otherwise indicated herein or clearly contradicted by context.

The term "amino acid", as used herein, is well known in the art. Amino acids are the essential building blocks of proteins. In accordance with the present disclosure, the term “amino acid” refers to free amino acids which are not bound to each other to form oligo- or polymers such as dipeptides, tripeptides, oligopeptides or proteins (also referred to herein as polypeptides). The term "amino acid" includes naturally occurring amino acids, but also other amino acids such as artificial amino acids. They can be classified into the characteristic groups of excipients with non-polar, aliphatic; polar, uncharged; positively and/or negatively charged and/or aromatic R groups (Nelson D.L. & Cox M.M., "Lehninger Biochemie" (2005), pp. 122-127).

The term "amino acids that provide an osmolytic function", as used herein, relates to amino acids with that provide an osmolytic property. Such amino acids are also well-known in the art and include, for example, glycine, alanine, and glutamic acid, as well as derivatives of proteinogenic and non-proteinogenic amino acids, respectively, such as for example, betaine, carnitine, creatine, creatinine, and B-alanine. The term "amino acids that provide an anti-oxidative functional group", as used herein, relates to amino acids that provide an anti-oxidative property via (one of) their side chain(s). Such amino acids are also well-known in the art and include, for example, cysteine, histidine, tryptophan, phenylalanine, and tyrosine, as well as derivatives of proteinogenic and non- proteinogenic amino acids such as for example N-acetyl-tryptophan, N-acetyl-histidine, methionine or carnosine.

The term "amino acids that provide a buffering function" relates to amino acids that provide a buffering capacity via one or more of their functional groups. Such amino acids are also well- known in the art and include, for example, glycine, arginine, and histidine.

In the context of the present disclosure, the term "aqueous composition" refers one the one hand to water but extends on the other hand also to buffered solutions and hydrophilic solvents miscible with water, thus being able to form a uniform phase. Examples for aqueous solutions include, without being limited, water, methanol, ethanol or higher alcohols as well as mixtures thereof.

The term "drying", as used herein, refers to the reduction or removal of the liquid content present in a composition. The liquid content is considered to have been reduced if the liquid is reduced to less than 20%, such as for example less than 10%, such as for example less than 8%, more preferably less than 7%, such as less than 5% or less than 1%. Even more preferably, the liquid is reduced to 0.5% or less.

As used herein, "storing" or “storage” means that a composition is not immediately used once prepared for its intended use, for example administered to a subject or used for transfection of a cell, but is kept for a period of time under particular conditions (e.g. particular temperature, etc.) prior to the intended use.

The term "viral vector(s)", in accordance with the present disclosure, relates to a carrier, i.e. a "vector" that is derived from a virus and is able to transduce genetic material comprised in the viral vector into a target cell. "Viral vectors" in accordance with the present disclosure include vectors derived from naturally occurring or modified viruses. "Viral vectors" may be viruses derived from naturally occurring viruses by genetic modification. The term “viral vectors” may relate to multiple individual vector entities of the same vector type or multiple individual vector entities of different vector types. Accordingly a "viral vector derived from a virus XY” as used herein refers to a viral vector derived from a naturally occurring or recombinant virus type XY by genetic modification of said virus type XY.

In general, the starting materials for the development of viral vectors are live viruses. Thus, certain requirements such as safety and specificity need to be fulfilled in order to ensure their suitability for use in animals or in human patients. One important aspect is the avoidance of uncontrolled replication of the viral vector. This is usually achieved by the deletion of a part of the viral genome critical for viral replication. Such a virus can infect target cells without subsequent production of new virions. Moreover, the viral vector should have no effect or only a minimal effect on the physiology of the target cell and rearrangement of the viral vector genome should not occur. Such viral vectors derived from naturally occurring or modified viruses are well known in the art and have been described (Lukashev AN and Zamyatnin AA, 2016; Stoica L and Sena-Esteves, 2016.

The term “virus-like particle(s)” or “VLP(s)” in accordance with the present disclosure, relates to molecular assemblies that closely resemble viruses, but are non-infectious because they contain no viral genetic material. They can be naturally occurring or synthesized through the individual expression of viral structural proteins, which can then self-assemble into the viruslike structure. A "virus-like particle derived from a virus XY” as used herein refers to a viruslike particle which comprises at least one viral structural proteins said virus species XY.

One major advantage of VLPs is that they are not associated with any risk of reassembly as is possible when live attenuated viruses are used as viral vectors and VLP production has the additional advantage that it can be started earlier than production of traditional vaccines once the genetic sequence of a particular virus strain of interest has become available. VLPs contain repetitive high density displays of viral surface proteins which present conformational viral epitopes that can elicit strong T cell and B cell immune responses. VLPs have already been used to develop FDA approved vaccines for Hepatitis B and human papillomavirus and, moreover, VLPs have been used to develop a pre-clinical vaccine against chikungunya virus. Evidence further suggests that VLP vaccines against influenza virus might be superior in protection against flu viruses over other vaccines.

The presence of viral structural proteins, for example, structural proteins in the envelope or in the capsid, can result in the self-assembly of VLPs. In general, VLPs can be produced in a variety of cell culture systems including mammalian cell lines, insect cell lines, yeast, and plant cells and VLPs have been produced from different virus families including parvoviridae (e.g. adeno-associated virus), retroviridae (e.g. HIV), and flaviviridae (e.g. Hepatitis C virus). The production of AAV based VLPs is for example disclosed in WO 2010/099960 A2. Regarding the embodiments characterized in this specification, in particular in the claims, it is intended that each embodiment mentioned in a dependent claim is combined with each embodiment of each claim (independent or dependent) said dependent claim depends from. For example, in case of an independent claim 1 reciting 3 alternatives A, B and C, a dependent claim 2 reciting 3 alternatives D, E and F and a claim 3 depending from claims 1 and 2 and reciting 3 alternatives G, H and I, it is to be understood that the specification unambiguously discloses embodiments corresponding to combinations A, D, G; A, D, H; A, D, I; A, E, G; A, E, H; A, E, I; A, F, G; A, F, H; A, F, I; B, D, G; B, D, H; B, D, I; B, E, G; B, E, H; B, E, I; B, F, G; B, F, H; B, F, I; C, D, G; C, D, H; C, D, I; C, E, G; C, E, H; C, E, I; C, F, G; C, F, H; C, F, I, unless specifically mentioned otherwise.

Similarly, and also in those cases where independent and/or dependent claims do not recite alternatives, it is understood that if dependent claims refer back to a plurality of preceding claims, any combination of subject-matter covered thereby is considered to be explicitly disclosed. For example, in case of an independent claim 1 , a dependent claim 2 referring back to claim 1 , and a dependent claim 3 referring back to both claims 2 and 1 , it follows that the combination of the subject-matter of claims 3 and 1 is clearly and unambiguously disclosed as is the combination of the subject-matter of claims 3, 2 and 1. In case a further dependent claim 4 is present which refers to any one of claims 1 to 3, it follows that the combination of the subject-matter of claims 4 and 1 , of claims 4, 2 and 1 , of claims 4, 3 and 1 , as well as of claims 4, 3, 2 and 1 is clearly and unambiguously disclosed.

The above considerations apply mutatis mutandis to all appended claims. To give a nonlimiting example, the combination of claims 12, 8, 4, 3 and 1 is clearly and unambiguously envisaged in view of the claim structure. The same applies for example to the combination of claims 12, 8 and 1 etc.

DETAILED DESCRIPTION

In a first aspect, the disclosure relates to liquid composition comprising a virus, or viral vector or virus-like particle and

(a) at least one compound according to formula (I)

(b) threonine, and

(c) at least one buffer, and

(d) optionally at least at least one surfactant.

In a preferred aspect, the liquid composition disclosed herein is an aqueous composition.

Preferably the aqueous composition comprises solutes other than water in an amount of less than about 10% (v/v), less than about 5% (v/v), less than about 2% (v/v), preferably less than about 1% (v/v).

In a preferred embodiment, the composition comprises at least one compound according to formula (I) and threonine.

In a preferred embodiment, the compound according to formula (I) is N-methyl-D-glucamin also known as (2R,3R,4R,5S)-6-(Methylamino)hexane-1 ,2,3,4,5-pentol (according to LIPAC) or meglumin. These names may be used interchangeably herein. Meglumin is pharmacopoeia listed excipient, including the United States Pharmacopeia and the European Pharmacopea.

N-methyl-D-glucamin may be comprised in the composition as disclosed herein in a concentration of at least about 5 mM, at least about 10 mM, at about 20 mM, at least about 30 mM, at least about 40 mM, at least about 50 mM, at least about 60 mM, at least about 70 mM, at least about 80 mM, at least about 90 mM, at least about 100 mM, at least about 110 mM, at least about 120 mM, at least about 130 mM, at least about 140 mM, at least about 150 mM, at least about 160 mM. N-methyl-D-glucamin may be comprised in the composition as disclosed herein in a concentration of up to about 500 mM, up to about 300 mM, up to about 250 mM, up to about 220 mM, up to about 200 mM, up to about 180 mM. N-methyl-D-glucamin may be comprised in the composition as disclosed herein in a concentration of about 10 mM to about 200 mM, about 20 mM to about 250 mM, about 40 mM to about 200 mM, preferably in a concentration of about 60 mM to about 180 mM, more preferably in a concentration of about 70 mM to about 155 mM.

Threonine may be comprised in the composition as disclosed herein in a concentration of at least about 5 mM, at least about 10 mM, at about 20 mM, at least about 30 mM, at least about 40 mM, at least about 50 mM, at least about 60 mM, at least about 70 mM, at least about 80 mM, at least about 90 mM, at least about 100 mM, at least about 110 mM, at least about 120 mM. Preferably threonine is comprised in the composition in at least about 90 mM, more preferably at least about 110 mM, most preferably at least about 120 mM Threonine may be comprised in the composition as disclosed herein in a concentration of up to about 300 mM, up to about 250 mM, up to about 200 mM, up to about 180 mM, up to about 160 mM, up to about 140 mM up to about 130 mM. Threonine may be comprised in the composition as disclosed herein in a of about 20 mM to about 250 mM, preferably in a concentration of about 80 mM to about 200 mM, more preferably 100 mM to about 160 mM, most preferably in a concentration of about 100 mM to about 140 mM.

In a further preferred embodiment, the composition disclosed herein comprises at least one amino acid with a negatively charged functional group.

Amino acids with a negative charged functional group according to the present disclosure are aspartic acid and glutamic acid. Preferably, the compositions disclosed herein comprises glutamic acid.

The composition according to the disclosure may comprise an amino acids with a negatively charged functional group, preferably glutamic acid, in a concentration of at least about 2 mM, at least about 5 mM, at about 10 mM, 15 mM, at least about 18 mM, preferably at about 20 mM. The amino acids with a negative charged functional group may be comprised in a concentration of about 2 mM to about 100 mM, about 5 mM to about 50 mM, about 7 mM to about 30 mM, preferably about 8 mM to about 25 mM.

The at least one buffer comprised in the composition disclosed herein may for example, but not limited to, be selected from a phosphate, TRIS, histidine, carbonate bicarbonate, citrate, maleate, adipate, HEPES, TES, MOPS, MES or PIPES buffer. In a preferred embodiment the buffer is a phosphate buffer. In a more preferred specific embodiment, the buffer is phosphate- buffered saline (PBS). PBS is an isotonic solution that is used in many biological applications comprising about 10 mM Na2HPO4, about 1.8 mM KH2PO4. Further to the buffer, PBS comprises about 137 mM NaCI and about 2.7 mM KOI.

Generally, the composition of the present disclosure has a pH of about 5 to about 8; about 5.5 to about 8; about 6 to about 7.8; about 6.5 to about 7.8; about 7 to about 7.8, preferably of about 7.2 to about 7.6; preferably about 7.4.

The at least at least one surfactant comprised in the disclosed composition may generally be selected from an ionic or a non-ionic surfactant. Preferably, the surfactant is a non-ionic surfactant. The non-ionic surfactant may for example be a block-copolymers, preferably a poloxamer. Poloxamers are non-ionic triblock copolymers composed of a central hydrophobic chain of polyoxypropylene (polypropylene oxide)) flanked by two hydrophilic chains of polyoxyethylene (poly(ethylene oxide)). Preferably, the poloxamer is selected from poloxamer 188 (also known under the tradenames Pluronic F 68, Kolliphor P 188 or Lutrol F 68) or poloxamer 407 (also known under the tradenames Pluronic F 127, Kolliphor P 407 or Lutrol F 127). Most preferably, the non-ionic surfactant is poloxamer 188. In an alternative embodiment, the non-ionic surfactant may be a polysorbate, for example polysorbate 20, or polysorbate 80, preferably polysorbate 80 (also known under the tradename Tween 80).

The surfactant may for example be comprised in a composition according to the present disclosure in an amount of at least 0.0001 % (w/w), at least about 0.0005% (w/w), at least about 0.0007% (w/w), least about 0.0008% (w/w), or least about 0.0009% (w/w). The surfactant may be comprised in a composition according to the disclosure in an amount of 0.0001% (w/w) to about 0.01% (w/w), about 0.0005% (w/w) to about 0.006% (w/w), about 0.0007% (w/w) to about 0.005% (w/w), about 0.0008% (w/w) to about 0.002% (w/w). Most preferably the surfactant is comprised in an amount of about 0.001 % (w/w).

In a further embodiment, the composition according to the present disclosure comprises at least one pharmaceutically acceptable salt. Pharmaceutically acceptable salts include, without limitation, sodium salts, ammonium salts, potassium salts (e.g., sodium, ammonium, and potassium chloride; sodium, ammonium, and potassium acetate; sodium, ammonium, and potassium citrate; sodium, ammonium, and potassium phosphate; sodium, ammonium, and potassium fluoride; sodium, ammonium, and potassium bromide; and sodium, ammonium, and potassium iodide) and combinations hereof. Preferably, the composition comprises sodium chloride and/or potassium chloride.

The salt concentration of the pharmaceutically acceptable salt in the composition according to the disclosure may be at least about 20 mM, at least about 30 mM, at least about 40 mM, at least about 50 mM, at least about 60 mM, at least about 70 mM, at least about 80 mM, at least about 90 mM, at least about 100 mM, at least about 120 mM, at least about 130 mM. The salt concentration may be about 20 mM to about 250 mM, about 30 mM to about 220 mM, about 50 mM to about 200 mM, about 70 mM to about 180 mM, about 90 mM to about 170 mM.

The salt concentration disclosed herein may be the total salt concentration of different salts comprised in the composition as closed herein. The term “total salt concentration” describes the sum of the concentration of the salts comprised in the composition. For example, a composition that comprises 120 mM NaCI and 10 mM KCI has a total salt concentration of 130 mM.

In a further embodiment the composition according to the present disclosure comprises at least one antioxidant. The antioxidant may preferably be selected from methionine, cysteine, glutathione, tryptophan, histidine, ascorbic acid, carnosine and any derivatives of the aforementioned agents, without being limiting. Preferably the antioxidant is an amino acid providing an anti-oxidative function, more methionine.

The antioxidant, preferably methionine, may be comprised in the composition as disclosed herein in a concentration of at least about 1 mM, at least about 2 mM, at about 3 mM, at least about 4 mM, at least about 5 mM, at least about 6 mM, at least about 8 mM, at least about 10 mM, at least about 20 mM. Preferably the antioxidant is comprised in a concentration on at least about 5 mM, more probably at least 10 mM. The antioxidant may be comprised in the composition as disclosed herein in a concentration of up to about 30 mM, up to about 25 mM, up to about 20 mM, up to about 15 mM. Preferably up to about 15 mM. The antioxidant may be comprised in the composition as disclosed herein in a concentration of about 1 mM to about 25 mM, about 2 mM to about 20 mM, preferably in a concentration of about 4 mM to about 15 mM.

Alternatively, preferably when the antioxidant is carnosin, the antioxidant may be comprised in the composition as disclosed herein in a concentration of at least about 10 mM, at least about 15 mM, at about 20 mM, at least about 30 mM, at least about 40 mM. The antioxidant, preferably when the antioxidant is carnosin, may be comprised in the composition as disclosed herein in a concentration of up to about 100 mM, up to about 80 mM, up to about 60 mM, preferably up to about 50 mM. The antioxidant, preferably when the antioxidant is carnosin, may be comprised in the composition as disclosed herein in a concentration of about 10 mM to about 100 mM, about 15 mM to about 70 mM, preferably in a concentration of about 20 mM to about 50 mM.

In specific embodiments, the composition according to the present disclosure comprises at least one polyethylene glycol (PEG). Polyethylene glycols are also known under the trade name Macrogol and are characterised by their average molecular mass. For example, polyethylene glycol 3350 has an average molecular mass of 3350 g/mol. The composition according to the present disclosure comprises may for example comprise a polyethylene glycol with an average mass of about 500 g/mol to about 5000 g/mol, about 1000 g/mol to about 4500 g/mol, preferably about 3000 g/mol to about 4000 g/mol average molecular mass. In a specifically preferred embodiment, the composition comprises polyethylene glycol with an average mass of about 3350 g/mol, also known as PEG 3350.

The polyethylene glycol may be comprised in the compositions disclosed herein in an amout of at least about 0,1 mM, at least about 0,2 mM, at least about 0,3 mM, at least about 0,5 mM, at least about 0,7 mM. Polyethylene glycol may be comprised in the composition as disclosed herein in a concentration of up to about 5 mM, up to about 2 mM, up to about 1 ,75 mM, preferably up to about 1 ,5 mM. Polyethylene glycol may be comprised in the composition as disclosed herein in a concentration of about 0,2 mM to about 3 mM, about 0,5 mM to about 2 mM, preferably about 0,7 mM to about 1 ,5 mM.

In an alternative aspect of the disclosure, the compositions are substantially free of polyethylene glycol (PEG). Preferably polyethylene glycol is not comprised in compositions disclosed herein which are used for storage over an extended period of time.

In a further embodiment, the composition according to the present disclosure comprises a sugar alcohol. Preferably, the sugar alcohol is selected from a 6-carbon sugar alcohol including but not limited to mannitol, sorbitol, and inositol. Most preferably the sugar alcohol is selected from inositol.

The sugar alcohol, preferably inositol, may be comprised in the composition as disclosed herein in a concentration of at least about 10 mM, at least about 20 mM, at about 30 mM, at least about 40 mM. The sugar alcohol may be comprised in the composition as disclosed herein in a concentration of up to about 150 mM, up to about 100 mM, up to about 80 mM. The sugar alcohol, may be comprised in the composition as disclosed herein in a concentration of about 20 mM to about 120 mM, about 30 mM to about 100 mM, from about 40 mM to about 80 mM.

In a further embodiment, the composition according to the present disclosure comprises valin.

Valine may be comprised in the composition as disclosed herein in a concentration of at least about 10 mM, at least about 20 mM, at least about 30 mM, at least about 40 mM,, at least about 60 mM, at least about 70 mM. Valine may be comprised in the composition as disclosed herein in a concentration of up to about 150 mM, up to about 125 mM, up to about 100 mM, up to about 90 mM, up to about 80 mM. Valine may be comprised in the composition as disclosed herein in a concentration of about about 20 mM to about 120 mM, about 30 mM to about 100 mM, from about 35 mM to about 90 mM, from about 40 mM to about 80 mM, preferably about 30 mM to about 100 mM, more preferably from about 35 mM to about 90 mM.

In a preferred embodiment, the disclosure relates to liquid composition comprising a virus, or viral vector or virus-like particle and

(i) ) the at least one compound according to formula (I), preferably meglumin in a concentration of about 60 mM to about 180 mM, preferably about 70 mM to about 155 mM; and optionally

(ii) glutamic acid in a concentration of about 7 mM to about 30 mM, preferably about 8 mM to about 25 mM; and optionally

(iii) poloxamer 188 in a concentration of about 0.0007% (w/w) to about 0.005% (w/w), preferably about 0.0008% (w/w) to about 0.002% (w/w); and at least one buffer, and optionally a salt.

(i) threonine in a concentration of about 50 mM to about 160 mM, preferably about 60 mM to about 140 mM; and optionally

(ii) glutamic acid in a concentration of about 7 mM to about 30 mM, preferably about 8 mM to about 25 mM; and

(i) ) the at least one compound according to formula (I), preferably meglumin in a concentration of about 60 mM to about 180 mM, preferably about 70 mM to about 155 mM; and

(ii) threonine in a concentration of about 50 mM to about 160 mM, preferably about 60 mM to about 140 mM; and optionally

(iii) glutamic acid in a concentration of about 7 mM to about 30 mM, preferably about 8 mM to about 25 mM; and optionally

(iv) poloxamer 188 in a concentration of about 0.0007% (w/w) to about 0.005% (w/w), preferably about 0.0008% (w/w) to about 0.002% (w/w); and at least one buffer, and optionally a salt.

(i) the at least one compound according to formula (I), preferably N-methyl-D-glucamin in a concentration of about 60 mM to about 180 mM, preferably about 70 mM to about 155 mM;

(ii) threonine in a concentration of about 50 mM to about 160 mM, preferably about 60 mM to about 140 mM; and (iii) optionally glutamic acid in a concentration of about 7 mM to about 30 mM, preferably about 8 mM to about 25 mM.

In a highly preferred embodiment, the disclosure relates to liquid composition comprising a virus, or viral vector or virus-like particle and

(ii) threonine in a concentration of about 80 mM to about 200 mM, more preferably 100 mM to about 160 mM, most preferably in a concentration of about 100 mM to about 140 mM; and optionally

(iii) glutamic acid in a concentration of about 7 mM to about 30 mM, preferably about 8 mM to about 25 mM; and/or optionally

In a further highly preferred embodiment, the disclosure relates to liquid composition comprising a virus, or viral vector or virus-like particle and

(iii) preferably about 30 mM to about 100 mM, more preferably from about 35 mM to about 90 mM; and/or optionally

It has been found that high concentrations of threonine may stabilise a virus, or a viral vector or virus-like particle even in the absence of meglumin.

Accordingly, in a further aspect the invention relates to a liquid composition comprising a virus, or a viral vector or virus-like particle and: (a) threonine in a concentration of at least 80 mM and, in at least about 90 mM, more preferably at least about 110 mM, most preferably at least about 120 mM,

(b) valin and/or,

(c) glutamic acid

(d) optionally at least one surfactant.

Preferably the afore described composition comprises threonine in a concentration of about 80 mM to about 200 mM, more preferably 100 mM to about 160 mM, most preferably in a concentration of about 100 mM to about 140 mM.

Preferably the afore described composition comprises valin in a concentration of about 30 mM to about 100 mM, more preferably from about 35 mM to about 90 mM.

Preferably the afore described composition comprises glutamic acid in a concentration of about 5 mM to about 50 mM, about 7 mM to about 30 mM, preferably about 8 mM to about 25 mM.

The afore described composition may comprise a at least one surfactant and other excipient as disclosed for other aspects of the invention.

In a further embodiment disclosed herein, the composition disclosed herein is a composition comprising at least three different amino acids.

The amino acids comprised in the composition of the present disclosure may be selected from naturally occurring amino acids as well as artificial amino acids or derivatives of these naturally occurring or artificial amino acids.

Naturally occurring amino acids include the 20 amino acids that make up proteins (i.e. the so- called proteinogenic amino acids), i.e. glycine, proline, arginine, alanine, asparagine, aspartic acid, glutamic acid, glutamine, cysteine, phenylalanine, lysine, leucine, isoleucine, histidine, methionine, serine, valine, tyrosine, threonine and tryptophan. Other naturally occurring amino acids and amino acid derivatives are e. g. carnitine, creatine, creatinine, guanidinoacetic acid, ornithine, hydroxyproline, homocysteine, citrulline, hydroxylysine or beta-alanine. Artificial amino acids are amino acids that have a different side chain length and/or side chain structure and/or have the amine group at a site different from the alpha-C-atom. Derivates of amino acids include, without being limiting, n-acetyl-tryptophan, phosphonoserine, phosphonothreonine, phosphonotyrosine, melanin, argininosuccinic acid and salts thereof and DOPA. In connection with the present disclosure, all these terms also include the salts of the respective amino acids.

In a preferred embodiment of the composition disclosed herein comprising at least three different amino acids, the at least three different amino acids comprised in the compositions disclosed herein are not more than four different amino acids or not more than three different amino acids. Thus, the composition may comprise only three or only four different amino acids.

In a preferred embodiment of the composition disclosed herein comprising at least three different amino acids, the combination of said at least three amino acids provides at least one positively charged functional group, at least one anti-oxidative functional group, at least one osmolytic function, and at least one buffering function.

As disclosed above, methionine and/or carnosine may preferably be comprised in the disclosed composition, which are amino acids that provide an anti-oxidative functional group.

In a preferred aspect, the virus, viral vector or virus-like particle is a non-enveloped virus, or a viral vector or virus like particle derived from such non-enveloped virus. Non-enveloped viruses are typically characterised by a viral capsid comprising capsid proteins which are not enveloped by a lipid bilayer. The non-enveloped virus according to the present disclosure may be selected from adenoviruses, parvoviruses, such as adeno associated viruses, plyomaviruses, anelloviruses, caliciviruses, picornaviruses, reoviruses, astroviruses, hepeviruses, papovaviruses, and reoviruse, without being limiting. Preferably the virus is an adenovirus or adeno associated virus, most preferably an adeno associated virus (AAV).

As used herein, the reference to a certain virus type, for example adeno-associated virus (AAV), refers to the naturally occurring and recombinant forms of said virus and encompasses mutant forms of said virus.

The AAV or viral vector or virus-like particle according the present disclosure may be selected from AAV1 , AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9, or AV10. In a preferred embodiment, the AAV may be AAV2, AAV5, AAV6, AAV8, or AAV9. Most preferably, the AAV is AAV2.

In an alternative embodiment, the virus, viral vector or virus-like particle may be an enveloped virus. In contrast to non-enveloped viruses, enveloped viruses comprise a capsid comprising capsid proteins which enveloped by a lipid bilayer. Preferred enveloped viruses according to the present disclosure are lentivirus, vesicular stomatitis virus, herpes simplex virus, modified vaccinia virus Ankara or measles virus.

The viral vector or virus-like particle disclosed herein are derived from the viruses disclosed herein. Accordingly, the viral vector or virus-like particle are preferably derived from a nonenveloped virus, more preferably from an adenovirus or adeno associated virus, most preferably an adeno associated virus (AAV).

In a preferred embodiment, the viral vectors are replication-deficient viral vectors.

Replication-deficient viral vectors are viral vectors that are not capable of replicating to generate new viral particles in host cells. For example, the viral vectors can have lost their replication competence by empirical and rational attenuation processes resulting in a loss of important parts of their genome accompanied by (i) retention of their ability to infect several cell types, and (ii) retention of their immunogenicity.

Due to the lack of replication competence, replication-deficient viral vectors represent safe and robust mechanism to induce both effector cell mediated and humoral immunity. As a consequence, priming with these vectors can improve the magnitude, quality and durability of such responses, while at the same time providing an increased safety. Suitable replicationdeficient viral vectors for vaccine preparation are well known in the art. Most preferably, the replication-deficient viral vector is a viral vector derived from adenovirus or AAV.

In a further aspect, the present disclosure relates to a frozen form of the liquid compositions disclosed herein. Preferably the frozen form of the liquid compositions may be obtained by fast freezing, such as snap-freezing the liquid composition temperatures below about -50°C, below about -70°C, below about -100°C, for example by freezing in a cryogenic medium (such as e.g. liquid nitrogen).

In a further aspect, the present disclosure relates to a dried form of the compositions disclosed herein.

Suitable methods for drying include, without being limiting, lyophilisation (freeze drying), spray drying, freeze-spray drying, convection drying, conduction drying, gas stream drying, drum drying, vacuum drying, dielectric drying (by e.g. radiofrequency or microwaves), surface drying, air drying or foam drying.

Freeze-drying, also referred to as lyophilisation, is also well known in the art and includes the steps of freezing the sample and subsequently reducing the surrounding pressure while adding sufficient heat to allow the frozen water in the material to sublime directly from the solid phase to the gas phase (primary drying) followed by a secondary drying phase. Preferably, the lyophilised preparation is then sealed to prevent the re-absorption of moisture.

Spray-drying is also well known in the art and is a method to convert a solution, suspension or emulsion into a solid powder in one single process step. Generally, a concentrate of the liquid product is pumped to an atomising device, where it is broken into small droplets. These droplets are exposed to a stream of hot air and lose their moisture very rapidly while still suspended in the drying air. The dry powder is separated from the moist air in cyclones by centrifugal action, i.e. the dense powder particles are forced toward the cyclone walls while the lighter, moist air is directed away through the exhaust pipes.

Spray-drying is often the method of choice, as it avoids the freezing step and requires lower energy costs as compared to lyophilisation. Spray-drying has also been shown to be a particularly advantageous drying procedure that is suitable for biomolecules, due to the short contact time with high temperature and its special process control. Thus, because spray-drying results in a dispersible dry powder in just one step it is often favoured to freeze drying when it comes to drying techniques for biomolecules.

Spray-freeze-drying is also well known in the art and is a method that combines processing steps common to freeze-drying and spray-drying. The sample provided is nebulised into a cryogenic medium (such as e.g. liquid nitrogen), which generates a dispersion of shock-frozen droplets. This dispersion is then dried in a freeze dryer.

In a further aspect, the present disclosure relates to a method for providing a virus, or viral vector or virus-like particle derived therefrom. The method comprises at least a first step of providing a liquid composition as disclosed herein comprising a virus, or viral vector or viruslike particle and a second step of storing the liquid composition as disclosed herein in liquid form for at least about 1 week, at least about 2 weeks, preferably at least about 4 weeks, more preferably at least 6 weeks. The liquid composition as disclosed herein in liquid form for at least about 2 months, at least about 4 months, at least about 6 months, at least about 8 months, at least about 10 months, at least about 12 months.

In a preferred embodiment of the method disclosed afore, the method does not comprise a step of freezing and thawing, or dehydrating and rehydrating the liquid composition before, during, or after storing the liquid composition in liquid form for at least about 1 week, at least about 2 weeks, preferably at least about 4 weeks, more preferably at least 6 weeks.

In a preferred embodiment, the liquid composition is stored at a temperature of at least about 2°C, at least about 4°C, at least about 8°C, at least about 10°C, at least about 15°C, at least about 20°C, at least about 25°C. Preferably, the composition is stored at least about 4°C, more preferably at least about 20°C. The storing temperature disclosed herein refer to the average temperature over the disclosed storing period.

In one embodiment the liquid composition disclosed herein is stored in liquid form at a temperature of least about 2°C for at least about 1 week, at least about 2 weeks, preferably at least about 4 weeks, more preferably at least 6 weeks. In another embodiment the liquid composition disclosed herein is stored in liquid form at a temperature of least about 4°C for at least about 1 week, at least about 2 weeks, preferably at least about 4 weeks, more preferably at least 6 weeks. In another embodiment the liquid composition disclosed herein is stored in liquid form at a temperature of least about 8°C for at least about 1 week, at least about 2 weeks, preferably at least about 4 weeks, more preferably at least 6 weeks. In another embodiment the liquid composition disclosed herein is stored in liquid form at a temperature of least about 4°C for at least about 1 week, at least about 2 weeks, preferably at least about 4 weeks, more preferably at least 6 weeks.

In a further aspect, the present disclosure relates to method of providing a virus, or viral vector or virus-like particle to a subject in need thereof, comprising the steps of:

(i) providing a liquid composition as disclosed herein;

(ii) storing the liquid composition as disclosed herein in liquid form for at least about 1 week, at least about 2 weeks, preferably at least about 4 weeks, more preferably at least 6 weeks, preferably at a temperature of at least about 2°C, of at least about 4°C, of at least about 8°C, at least about 15°C, at least about 20°C;

(iii) administering the virus, or viral vector or virus-like particle to the subject in need thereof, preferably in a liquid composition as disclosed herein.

Accordingly, the disclosure also relates to a therapeutic or prophylactic method of treatment, wherein a virus, or viral vector or virus-like particle is provided according to a method disclosed herein and the virus, or viral vector or virus-like particle is subsequently administered to a subject.

The compositions disclosed herein may be used for anti-bacterial, antiviral, anti-cancer, antiallergy, vaccination and/or for gene transfer therapy for the treatment of diseases with a genetic background. The compositions may further be used for other cell-based therapies using cells which are transduced with a foreign gene which is not comprised in the naturally occurring genotype of said cell. Preferably, said transduced cell is a lymphocyte transduced with a chimeric receptor, for example a chimeric antigen receptor.

In a preferred embodiment, the composition disclosed herein is a pharmaceutical composition. In accordance with the present disclosure, the term “pharmaceutical composition” relates to a composition suitable to be administered and/or administered to a patient, preferably a human patient.

The pharmaceutical compositions can be administered to the subject at a suitable dose. The dosage regimen will be determined by the attending physician and clinical factors. As is well known in the medical arts, dosages for any one patient depend upon many factors, including the patient's size, body surface area, age, the particular compound to be administered, sex, time and route of administration, general health, and other drugs being administered concurrently. The therapeutically effective amount for a given situation will readily be determined by routine experimentation and is within the skills and judgment of the ordinary clinician or physician. The pharmaceutical composition may be for administration once or for a regular administration over a prolonged period of time. Generally, the administration of the pharmaceutical composition should be in the range of for example 1 pg/kg of body weight to 50 mg/kg of body weight for a single dose. However, a more preferred dosage might be in the range of 10 pg/kg to 20 mg/kg of body weight, even more preferably 100 pg/kg to 10 mg/kg of body weight and even more preferably 500 pg/kg to 5 mg/kg of body weight for a single dose.

The components of the pharmaceutical composition to be used for therapeutic administration must be sterile. Accordingly the compositions disclosed herein are preferably sterile composition. Sterility is readily accomplished for example by filtration through sterile filtration membranes (e.g., 0.2 pm membranes).

The various components of the composition may be packaged as a kit with instructions for use.

Accordingly, the present disclosure further relates to a kit comprising a composition comprising viral vectors obtained or obtainable by the method of the disclosure and, optionally, instructions how to use the kit.

In a further aspect, the disclosure relates to a method of transfecting a cell with a virus or viral vector derived therefrom, comprising the steps of: (i) providing a liquid composition as disclosed herein;

(iii) transfecting the cell with the virus or viral vector by contacting the cell with the virus, or viral vector derived therefrom.

The transfection of the cell in step (iii) of the afore disclosed method may be in vitro or in vivo.

In a further aspect, the disclosure relates to a method for providing a liquid composition disclosed herein, comprising at least:

(x) a first step comprising providing a first liquid composition comprising a virus, or viral vector or virus-like particle in a first aqueous medium;

(y) a second step of exchanging the first aqueous medium against a second aqueous medium comprising excipients comprised in the compositions disclosed herein comprising at least one polyethylene glycol, preferably polyethylene glycol 3350, to obtain a second liquid composition; and optionally

(z) a third step of exchanging the second aqueous medium against a third aqueous medium comprising excipients comprised in the compositions disclosed herein, which is substantially free of polyethylene glycol, to obtain a third liquid composition.

The aqueous medium in step (x) may for example be the medium in which the virus, or viral vector or virus-like particle is obtained when harvesting the virus, or viral vector or virus-like particle producing the virus, or viral vector or virus-like particle.

One possible way of exchanging one aqueous medium against another aqueous medium in step (y) or step (z) is by diluting an existing medium by adding the other medium. A particularly preferred method of carrying out exchange of media is via dialysis. Dialysis is a well-known method in the art wherein semipermeable dialysis membranes are used to enable diffusion of small molecule solutes across the membrane, whereby the components of the liquids are exchanged and the biomolecules are retained in the dialysis cassette dependent on the molecular weight and the applied Molecular Weight Cut Off of the dialysis membrane. An alternatively preferred method of exchanging media is the disclosed method is ultrafiltration. Buffer exchange by ultrafiltration may be performed by methods well known in the art.

According to a further aspect, the disclosure relates to a composition obtained or obtainable by the method disclosed herein.

LIST OF EMBODIMENTS

1 . A liquid composition comprising a virus, or viral vector or virus-like particle and

(a) at least one compound according to formula (I)

wherein R1 and R2 are independently from each other selected from H, C1-C4-alkyl, CH2CH2OH or CH2CH(CH3)OH, and/or

(b) threonine, and

(c) at least one buffer, and

(d) optionally at least one surfactant.

2. The composition according to embodiment 1 , wherein the at least one compound according to formula (I) comprises or is N-methyl-D-glucamin.

3. A liquid composition comprising a virus, or a viral vector or virus-like particle and:

(a) threonine in a concentration of at least 80 mM and, in at least about 90 mM, more preferably at least about 110 mM, most preferably at least about 120 mM,

(b) valin and/or,

(c) glutamic acid

(d) optionally at least one surfactant.

4. The composition according to any one of the preceding embodiments, further comprising an amino acid with a negatively charged functional group, preferably glutamic acid.

5. The composition according to any of the embodiments 1 , 2 and 4, further comprising, valine.

6. The composition according to any one of the preceding embodiments, wherein the buffer is selected from a phosphate, TRIS, histidine, carbonate bicarbonate, citrate, maleate, adipate, HEPES, TES, MOPS, MES or PIPES buffer, preferably the buffer is a phosphate buffer. 7. The composition according to any one of the preceding embodiments, having a pH of about 5 to about 8; preferably of about 7 to about 7.8; preferably about 7.4.

8. The composition according to any one of the preceding embodiments, wherein the surfactant is a non-ionic surfactant, preferably a poloxamer or a polysorbitol, most preferably poloxamer 188.

9. The composition according to any one of the preceding embodiments, further comprising at least one salt, preferably NaCI and/or KCI.

10. The composition according to any one of the preceding embodiments, further comprising at least one antioxidant, more preferably an amino acid providing an anti- oxidative function, most preferably methionine.

11. The composition according to any one of the preceding embodiments, further comprising at least one polyethylene glycol, preferably polyethylene glycol 3350.

12. The composition according to any one of the preceding embodiments, further comprising an excipient selected from inositol, valine, carnosine, and any combination thereof.

13. The composition according to any one of the preceding embodiments, comprising (i) the at least one compound according to formula (I), preferably N-methyl-D-glucamin in a concentration of about 60 mM to about 180 mM, preferably about 70 mM to about 155 mM; (ii) threonine in a concentration of about 50 mM to about 160 mM, preferably about 60 mM to about 140 mM; and (iii) optionally glutamic acid in a concentration of about 7 mM to about 30 mM, preferably about 8 mM to about 25 mM.

14. The composition according to any one of the preceding embodiments, comprising

(iv) poloxamer 188 in a concentration of about 0.0007% (w/w) to about 0.005% (w/w), preferably about 0.0008% (w/w) to about 0.002% (w/w); and/or optionally

(v) valin in a concentration of about 30 mM to about 100 mM, preferably from about 35 mM to about 90 mM and at least one buffer, and optionally a salt.

15. The composition according to any one of the preceding embodiments wherein the virus is a non-enveloped virus, preferably selected from an adenovirus or adeno associated virus.

16. The composition according to any one of the preceding embodiments wherein the virus is selected from AAV1 , AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9, or AAV10, preferably AAV2.

17. The composition according to any one of the preceding embodiments, wherein the composition is in a frozen or dried form.

18. A method of providing a liquid composition comprising a virus, or viral vector or viruslike particle derived therefrom, comprising at least a first step of providing a liquid composition according to any one of embodiment 1 to embodiment 17, and a second step of storing said composition in liquid form for at least about 4 weeks, preferably at least 6 weeks.

19. The method according to embodiment 18, wherein the method does not comprise a step of freezing and thawing, or dehydrating and rehydrating the liquid composition before, during, or after storing the liquid composition in liquid form for at least 4 weeks, preferably at least 6 weeks.

20. The method according to any one of embodiment 18 to embodiment 19, wherein the liquid composition is stored at a temperature of at least about 2°C, preferably at least about 4°C, more preferably at least about 20°C.

21 . A method of providing a virus, or viral vector or virus-like particle to a subject in need thereof, comprising the steps of:

(i) providing a liquid composition according to any one of embodiment 1 to embodiment 16; (ii) storing the liquid composition according to any one of embodiment 1 to embodiment 16 in liquid form for at least 4 weeks, preferably at least 6 weeks;

(iii) administering the virus, or viral vector or virus-like particle to the subject in need thereof, preferably in a liquid composition according to any one of embodiment 1 to embodiment 16.

22. The therapeutic or prophylactic method of treatment, wherein a virus, or viral vector or virus-like particle is provided and administered according to the method of embodiment 21.

23. A method of transfecting a cell with a virus or viral vector, comprising the steps of:

(i) providing a liquid composition according to any one of embodiment 1 to embodiment 17;

(ii) storing the liquid composition according to any one of embodiment 1 to embodiment 17 in liquid form for at least 4 weeks;

(iii) contacting the cell with the virus, or viral vector derived therefrom.

24. The method according to embodiment 19, wherein the transfection is in vitro or in vivo.

25. A process for providing a liquid composition according to any one of the preceding embodiments, comprising at least:

(i) a first step comprising providing a first liquid composition comprising a virus, or viral vector or virus-like particle in a first aqueous medium;

(ii) a second step of exchanging the first aqueous against a second aqueous medium comprising excipients according to any one of the embodiments 1 to 17, comprising at least one polyethylene glycol, preferably polyethylene glycol 3350, to obtain a second liquid composition; and optionally

(iii) a third step of exchanging the second aqueous medium against a third aqueous medium comprising excipients according to any one of the embodiments 1 to 171 , which is substantially free of polyethylene glycol to obtain a third liquid composition.

The following examples illustrate the disclosure: EXAMPLES

Example 1 : Basis Characterization of AAV2 viral vectors in different unusual buffers at different pH values in comparison to common buffers in biologies formulations at the corresponding pH values

[001] In a first step a basic characterization of AAV2 in different unusual buffers at different pH values was carried out in order to identify the influence of different buffering systems in various time-dependent and biophysical, time-independent stability indicating methods in comparison to the reference buffer phosphate and histidine buffer.

[002] In order to reduce the aggregation propensity of AAV2 in this study, all formulations of the basic characterization contained a particular concentration of NaCI for adjusting the osmolality to a physiological osmolality of approx. 300 mOsm. One formulation was prepared without NaCI for comparison.

Materials and Methods

Starting material

In this study, a recombinant Adeno-Associated Virus serotype 2 (AAV2) material containing the coding DNA for an eGFP protein produced by Sirion Biotech (Grafelfing; Germany) was used.

Table 1 : AAV systematic sample name, AAV concentrations in terms of the viral genome titer (vg/mL) analyzed by qPCR, the total capsid titer cp/mL analyzed by a Capsid ELISA and the calculated ratio total capsids/full capsids.

All preparation steps with AAV material were performed under a laminar air flow (LAF) sterile workbench. Before leaving the LAF to the instruments, all plates or cuvettes were closed. Sample Preparation

For the Basic Characterization of AAV2 viral vectors regarding the biophysical stability as well as the storage stability at elevated temperature 12 different buffers were selected at different pH values.

For comparison, the commonly applied buffers for the higher pH range in form of phosphate buffer at pH 7.4 and 7.0 and histidine buffer at pH 5.0, 5.5 and 6.0 for the lower pH ranges were included. Sodium chloride was included in all formulations except in formulation 10 to reduce aggregation of the AAV viral vectors. Poloxamer 188 was included in all formulations in order to reduce the adsorption of the AAV2 viral vectors on surfaces.

The starting material was formulated in the Reference formulation PBS buffer at pH 7.4 containing 0.001 % Poloxamer 188. Re-buffering the AAV2 preparation provided by the provider in the 12 buffers was performed by dialysis and was associated with a remarkable loss of active material particularly in the low osmolality buffer 10.

Table 2: Composition of the 12 buffers with different buffer species and/or pH values. For comparison, the commonly applied buffer systems phosphate buffer and histidine buffer were included.

Nano Differential Scanning Fluorimetry (nano DSF)

AAV samples were diluted in a 96-well plate using the respective formulations and transferred into a 364-well plate for loading into the capillary chip. 20 pl of each sample were loaded in a capillary. Replicates were created by filling two capillaries with the same sample from the same well. The chip was sealed with capillary sealing paste. The loaded capillary chip was placed into the Prometheus NT.Plex (NanoTemper). After the chip was placed into the instrument a discovery scan is started at 100 % excitation power due to the small size of the capsids. A temperature ramp with a ramp rate of 1°C/ min starting at 25 °C until 95 °C was applied using the PR.ThermControl software (NanoTemper). The stability indicating parameters T_on and T_m were determined using the PR. Stability Analysis software (NanoTemper).

SYBR Gold Differential Scanning fluoreimetry (SYBR Gold DSF)

In preparation of the measurement, the SBYR Gold DMSO stock solution with a concentration of 10.000 x was sequentially pre-diluted, first to a concentration of 400 x and secondly to a concentration of 100 x with NFW. The AAV samples were pre-diluted 1/3.6 with the respective formulation buffer in Eppendorf tubes. Subsequently, 45 pl of the AAV dilutions were pipetted in triplicates into a 96-well PCR plate (PerkinElmer) and 5 pl of the SYBR Gold 100 x predilution was added to reach a final SYBR Gold concentration of 20 x and a final dilution factor of 4 for the AAV material. The plate was sealed with an adhesive sealing sheet (Thermo Fisher Scientific). After short centrifugation at 200 rpm, the 96 well plate was placed into a qPCR cycler (BioRad). Through the BioRad CFX Manager software the fluorophore channel “FRET” was selected and the following thermal ramp was set: 28 - 86 °C, 0.5 °C/min and 90 s hold time. The analysis was also performed in the software by calculating the first derivative of the genome ejection curve and determining the DNA leakage temperature T_mi and the capsid rupture temperature T_m2 through the inflection points.

Dynamic Light Scattering (PLS)

In order to analyse the size of the AAV particles (hydrodynamic radius) and the particle size distribution (polydispersity) in the two different starting materials a DLS analysis was performed. The first step was the filtration of DPBS (used as diluent) and Aqua B. Braun (used to maintain the temperature of the sample during measurement) through an Anotop-Filter with a 0.02 pm pore size. The AAV material batches was measured in triplicates undiluted and with the following dilution factors: 2, 5, 10 pL of the samples were pipetted into a single use cuvette, a Lid is applied before measurement and the outer rim of the cuvette is filled with 400 pl of the filtrated Aqua B. Braun. Subsequently, the cuvette was placed into the DynaPro NanoStar (Wyatt) for the DLS measurement. The parameters used for measurement are shown in Table 3. Measurements were performed under the control of the software Dynamics 8.1.2.144 (Wyatt).

Table 3: Parameters used in the Dynamics software for DLS measurement.

Digital droplet Polymerase Chain Reaction (ddPCR) ddPCR was used for the determination of the AAV concentration in form of the viral genome titer (vg/mL), an orthogonal parameter to the full capsid titer analyzed by the SE-HPLC-Multi- Detector- Method.

Step I DNase I treatment:

Initial digestion of DNA material was performed to remove host cell DNA from manufacturing and/or leaked DNA from the capsids during storage at 30 °C with DNase I (Merck). The DNase I stock solution (10.000 ll/rnl) was diluted with Incubation Buffer 10 (Merck) and NFW(Ambion) to a DNase I concentration of 0.625 ll/pl. Subsequently, 20 pL of the DNase I dilution (0.625 ll/pl) was then mixed with 5 pl of the AAV samples in a 96-well PCR plate resulting in an initial 5-fold dilution of the AAV samples. The final concentration of DNase I in the reaction mixture was 0.5 ll/pl. A negative control was prepared by mixing 5 pl of NFW with 20 pl of DNase I dilution (0.625 ll/pl). One positive control was prepared by mixing 5 pl of freshly thawed AAV reference material with 20 pl of DNase I dilution (0.625 ll/pl) and another positive control was prepared by pipetting 5 pl AAV sample to 20 pl of NFW. The plate was sealed with a microseal “B” sealing foil (Bio-Rad) and incubated in a thermocycler at 37 °C for 30 min.

Digestion of DNA material outside the AAV particles by DNase I was stopped by the addition of the DNase I inhibitor EDTA (Invitrogen). The EDTA stock solution (0.5 M) was diluted to a concentration of 0.05 M. 25 pl of the diluted EDTA (0.05 M) were added to all wells containing 25 pl of the DNase I digestion reaction mixture (including positive controls and the negative control). The final EDTA concentration for the stop reaction was 0.025 M and the AAVs were further 2-fold diluted. Subsequently, the PCR plate was sealed with a microseal “B” sealing foil and incubated in the thermocycler at 80 °C for 20 min. Step II Proteinase K treatment:

The extraction of viral ssDNA was performed through digestion of the protein capsids by the further treatment of the AAV samples with proteinase K.

A 20 mg/ml Proteinase K stock solution (Qiagen) was diluted with DPBS (Gibco) to a concentration of 2 mg/ml. 50 pl of the diluted proteinase K (2 mg/ml) were added to the wells containing 50 pl of the reaction mixture of step I. The final concentration of proteinase K in the reaction mixture was 1 mg/ml. The AAV samples were further 2-fold diluted in this step. The PCR plate was sealed with a microseal “B” sealing foil and incubated in the thermocycler with the following steps: 56 °C for 2h, 95°C for 30 min and 4°C for 10 min.

Step III Sample dilution with Salmon Sperm DNA buffer

[003] After the first two pre-treatment steps, the extracted DNA samples were further diluted sequentially with salmon sperm DNA (Invitrogen). A salmon sperm DNA buffer was prepared by diluting the salmon sperm DNA stock solution (10 mg/ml) with NFW to a concentration of 2 pg/ml.

The prepared serial dilutions of the extracted DNA samples are shown in Table 4.

Table 4: Serial dilutions of the extracted viral DNA with salmon sperm DNA.

Dilution 5 and 6 were chosen for ddPCR measurement in order to get the right number of copies of DNA per reaction between 100 and 100.000. The preparation of master mix for PCR reaction is shown in Table 5. Table 5: Composition of master mix for the PCR reaction.

In a 96-well ddPCR plate, 15 pl of master mix were mixed with 10 pl of samples. The ddPCR plate was sealed with a piercable alufoil from using a plate sealer and vortexed briefly on the corners. The plate was then centrifugated at 1000 rpm for 2 min at room temperature.

Generation of droplets was done in the Automated Droplet Generator by using the droplet generation oil for Probes and Auto DG Catridges. Afterwards, the ddPCR plate was placed into a PCR thermocycler and the following PCR program was applied (Table 6). Table 6: Programm used in the thermocycler for PCR reaction.

After the PCR reaction, the ddPCR plate was placed into the Droplet Reader for fluorescence reading with the QuantaSoft software. Size Exclusion (SE) - High Performance Liquid Chromatography (HPLC) - Multi-Detector- Method (MD)

For this method the UltiMate 3000 HPLC system was used, equipped with a Wyatt AAV column WTC-050N5 with a bead size of 5 pm, a diameter of 4.6 mm and a column length of 300 mm. Table 7: The composition of running buffer is shown in the following table.

The reagents were weighed in with a scale and dissolved in HPLC water. After complete dissolution the pH value was measured with a pH-meter and adjusted to 7.4 with 10 M NaOH or 10 M HCI. In a volumetric flask the volume was adjusted with HPLC water to the desired quantity. The buffer was sterile filtrated with stericups (0.22 pm; Merck).

The AAV samples were measured undiluted in duplicates in a 96-well plate with a V-Bottom (Greiner). In order to check the column resolution 2 injections of BSA (Sigma) 1 mg/ml were analyzed before the other samples were injected.

The following conditions were applied: Flow rate of 0.3 ml/min, Column oven temperature of 25°C, Sample temperature of 5°C, Injection volume 35 pL.

[004] The following 3 detectors were applied: UV 260/280 nm, Multi-angle light scattering (DAWN), dRI (Optilab).

After the measurement the column was cleaned using 0.5 M Na2SC>4 pH 2.7 and subsequently stored in a running buffer + NaNs solution. Chromatograms were evaluated using the Chromeleon Chromatrography Studio software and the Conjugate Analysis package in the ASTRA software of Wyatt.

Results

Evaluation of the concentration of the AAV2 viral vectors after dialysis

The concentration of the AAV2 viral vectors after re-buffering by dialysis was analyzed by monitoring the viral genome titer (vg/mL) using digital droplet PCR and the full capsid concentration (full cg/mL) using SE-HPLC-Multi-Detector method. For comparison we included the AAV2 viral vectors in the original, untreated Reference (PBS buffer pH 7.4 + 0.001 % Poloxamer 188).

As shown in Fig, 1 a remarkable loss of AAV2 titers was analyzed in all re-buffered AAV2 formulations, particularly in the low osmolality histidine buffer at pH 5.5 (buffer 10) containing the AAV2 viral vectors solely in aggregated forms as analyzed by DLS (data not shown). In buffer 06 and buffer 12, the recovery of the AAV2 viruses was also remarkably decreased but to a lesser extent in comparison to the low osmolality buffer 10. In contrast to the original, untreated Reference, the results of the viral genome titer analysis by ddPCR was remarkably decreased compared to the determination of the full capsid concentration using the SE-HPLC- Multi-Detector-Method (MD) in the re-buffered formulations. The analyzed concentrations of the AAV2 viral vectors after dialysis was more or less comparable in buffers 01-05, 07-09 and in buffer 11. Both methods revealed comparable trends between the re-buffered formulations.

Nano DSF

Thermal unfolding of the cAAV2 capsid proteins was monitored by measuring the intrinsic tryptophan fluorescence under application of the thermal ramp between 25 °C and 95 °C. Two critical characteristic temperatures were analyzed for thermal unfolding of the capsid proteins, the onset temperature of thermal unfolding T_on and the midpoint temperature of thermal unfolding T_m. For comparison the AAV2 viral vectors in the untreated reference formulation (Reference; PBS buffer at pH 7.4 + 0.001 % Poloxamer 188) was included.

Interestingly, the thermal stability of AAV2 viral vectors was in most buffers higher in comparison to the untreated Reference, particularly in terms of T_m as shown in Fig 2. The second interesting fact is the analyzed normal thermal unfolding profile in the low osmolality formulation 10 containing only AAV aggregates analyzed by DLS (data not shown). In the PIPES buffers at pH 7.4 a higher thermal stability of AAV2 was observed compared to PBS buffer at the same pH value of 7.4 + Poloxamer 188 underlining the dependence of the thermal stability of AAVs on the kind of buffer. In PIPES buffer at pH 7.0 a further increase in the thermal stability of AAV2 was observed. In addition, the highest thermal stability of the AAV2 capsids particularly regarding the T_m values was analyzed in buffer 06 (citrate buffer at pH 6.0, buffer 07 (maleate buffer at pH 5.0), buffer 11 (histidine buffer at pH 5.5 and buffer 12 (histidine buffer at pH 6.0) at lower pH values underlining the dependence of the thermal stability of AAV capsids on the pH value and on the kind of buffer. The increase of the thermal stability of the AAVs in histidine buffer with increasing pH values between buffer 09 (pH 5.0), buffer 11 (pH 5.5) and buffer 12 (pH 6.0) was in line with the analysis of thermal unfolding of monoclonal antibodies in these buffers at these pH values. Table 8: Onset temperatures of thermal unfolding T_on and midpoint temperatures of thermal unfolding T_m with standard deviations SD of AAV2 formulated in 12 different kinds of buffers at different pH values measured with nanoDSF.

In summary, at lower pH values of the buffers (citrate buffer at pH 6.0, maleate buffer at pH 5.0, histidine buffer at pH 5.5, pH 6.0) and at high pH values in PIPES buffers at pH 7.4 and 7.0, the AAVs seem to have the highest thermal stability particularly regarding the T_m values monitored by nano DSF in form of intrinsic tryptophan fluorescence under application of a thermal ramp.

SYBR Gold DSF

In additions to the analysis of thermal unfolding by nano DSF, the thermal stability of the AAVs in different buffers at different pH values was analyzed by SYBR Gold DSF. SYBR Gold is a fluorescence dye that increases the fluorescence yield upon specific binding to nucleic acids. Using this method the release of ssDNA from the AAV capsids is commonly monitored by the increase of the SYBR Gold fluorescence upon application of a thermal ramp on the AAV viral particles from 28 to 86 °C. In the results, this method revealed 2 thermal transitions with two midpoints of the thermal transitions T_mi associated with the leakage of DNA from the capsid and Tm2 associated with the complete capsid rupture.

In line with the thermal stability analyzed by nano DSF, the thermal stability of the AAV2 viral vectors in the Reference formulation revealed the lowest thermal stability in terms of both, the leakage temperatures of DNA from the AAV2 capsids T_mi and capsid rupture temperatures Tm2 of the AAV2s. As shown in Fig. 3, the thermal stability regarding the DNA release from the AAV2 capsids was higher in all buffers compared to the Reference. Interestingly, one of the highest thermal stabilities was monitored by SYBR Gold fluorescence regarding the leakage temperature and the rupture temperature in the low osmolality histidine formulation at pH 5.5 (buffer 10) containing only AAV2 aggregates suggesting a thermal stabilization of the capsids by aggregation. High thermal stability was also analyzed in Tris buffer at pH 8.0 (buffer 05) and in citrate buffer at pH 6.0 (buffer 06). In line with thermal unfolding, the thermal stability of AAVs in PIPES buffer at pH 7.4 was higher in comparison to the Reference in PBS at the similar pH of 7,4 + Poloxamer 188. The thermal stability was further increase in PIPES buffer at pH 7.0 particularly regarding the rupture temperature T_m2. In the case of AAV2 in buffer 09 and buffer 12 the two T_m values coincided in the thermal profiles of DNA release. In PIPES buffer at pH 7.4, maleate buffer pH 5.5, Adipate buffer pH 5.0 and in Histidine buffer pH 5.5 low osmolaliity and histidine buffer pH 5.5 high osmolality the leakage temperature T_mi and the capsid rupture temperature T_m2 shifted closer together.

Table 9: Numerical values for T_mi and T_m2 measured with SYBR Gold DSF.

The onset temperatures of thermal unfolding T_on monitored by nano DSF are in the most cases in the range of the capsid rupture temperature T_m2 monitored by SYBR Gold DSF. A remarkable dependency of both transition temperatures on the kinds of buffers and pH values was observed in SYBR Gold DSF. In all cases of buffers, the DNA leakage from the capsids started at much lower temperatures compared to the rupture of the capsids or the onset of thermal unfolding monitored by nano DSF and was also strong dependent on the kind of buffer and the pH values in the buffer. In summary, comparable trends in the thermal stability of the AAV2 viral vectors in the different buffers at different pH values were monitored by SYBR Gold DSF in comparison nano DSF. The lowest stability of the AAV2 regarding DNA leakage and capsid rupture was monitored in the original, untreated Reference. Both transition temperatures were highly dependent on the kind of buffers as well as the pH vales but with slightly different trends or dependencies between the two thermal events. DNA leakage already starts at temperatures high below the onset temperatures of thermal unfolding T_on and the capsid rupture temperature T_m2, respectively.

Storage stability of the AA V2 viral vectors at 30 °C in the different buffers at different pH values The storage stability of the AAV2 viral vectors formulated in the 12 buffers was analyzed at indicated time points at t = 0, after 2 weeks and after 4 weeks of storage at 30 °C using the SE-HPLC-Multi-Detector-Method. Using Size Exclusion Chromatography, the biomolecules are eluted in the order according to their hydrodynamic volume and so to their size. Using the three detectors, the MALS-, dRI- and UV-detectors, this method can analyze the molecular mass of the protein capsid, the molecular mass of the DNA and the molecular mass of the combination of DNA and the protein capsid. As result the molecular masses are converted by the conjugate analysis software from Wyatt into the total capsid concentration (total cp/mL), the full capsid concentration (full cp/mL), and the empty capsid concentration (empty cp/mL).

In contrast to the thermal stability indicating methods, the storage stability of the AAV2 viral vectors at 30 °C in the original, untreated Reference formulation, seems to be not reduced in comparison to the other buffers (see Fig. 4). The % total capsids analyzed after 2 weeks and 4 weeks storage at 30 °C in the original, untreated Reference formulation (PBS buffer at pH 7.4 + 0.001 % Poloxamer 188) was reduced to 89 % after 2 weeks and 86 % after 4 weeks of storage at 30 °C, respectively.

In buffers, 06 (citrate buffer at pH 6.0), 08 (adipate buffer at pH 5.0), 09 (histidine buffer at pH 5.0), 11 (histidine buffer at pH 5.5) and 12 (histidine buffer at pH 6.0) the best storage stability of the AAV2 viral vector total capsids was analyzed suggesting, that lower pH values are required for storage stability at elevated temperature. The reduction of the total capsids in these buffers was analyzed to be 83 % in buffer 06, 84 % in buffer 08, 85 % in buffer 09, 83 % in buffer 11 and 87 % in buffer 12 after 2 weeks of storage and 82 % in buffer 06, 86 % in buffer 08, 89 % in buffer 09, 91 % in buffer 11 and 90 % in buffer 12 after 4 weeks of storage at 30 °C. In the first period of storage of 2 weeks at 30 °C the % total capsids was remarkably reduced in these formulations (67 % in buffer 01 , 76 % in buffer 02, 72 % in buffer 03, 71 % in buffer 04, 67 % in buffer 05). Only a slight reduction of the total capsid concentration was observed during the further storage time in these formulations (65 % in buffer 01 , 74 % in buffer 02, 69 % in buffer 03, 72 % in buffer 04, 66 % in buffer 05). The lowest stability of the AAV2 viral vectors was analyzed in the low osmolality histidine buffer at pH 5.5 (buffer 10) in terms of the total capsid concentration (35 % and 48 % decrease of the total capsids after 2 weeks and 4 weeks, respectively in comparison to t = 0).

The analysis of the % full capsids during storage for up to 4 weeks at 30 °C revealed the same trends comparable to the trends of the % total capsids. The % full AAV2 capsids in the Reference was only slightly reduced after 2 weeks of storage to 98 %, and slightly stronger after 4 weeks of storage to 84 %. As mentioned before for the % total capsids, the % full capsids were only slightly reduced after 2 weeks of storage at 30 °C in the lower pH buffers to 93 % in buffer 06, 94 % in buffer 08, 94 % in buffer 09, 91 % in buffer 11 and 97 % in buffer 12 in comparison to the 100 % at t = 0. After 4 weeks of storage at 30 °C the loss of full capsids was slightly higher (75 % in buffer 06, 81 % in buffer 08, 83 % in buffer 09, 80 % in buffer 11 and 79 % in buffer 12). The results support the conclusion that the storage stability in the lower pH buffers was increased in comparison to the higher pH buffers.

As shown in Fig. 5. the remaining % of full capsids after 2 weeks of storage at 30 °C was lower in the higher pH buffers, i.e., 80 % in buffer 01 , 88 % in buffer 02, 86 % in buffer 03, 84 % in buffer 04 and 81 % in buffer 05. After 4 weeks of storage at 30 °C the decrease of the % full capsids was more pronounced in comparison to the Reference and the lower pH buffers, i.e., 73 % in buffer 01 , 73 % in buffer 02, 70 % in buffer 03, 73 % in buffer 04 and 67 % in buffer 05, underlining the hypothesis that higher pH values in the buffers resulted in lower total and full AAV2 capsid stabilities. The full AAV2 viral capsids showed the lowest stability during storage at 30 °C in the low osmolality histidine buffer at pH 5.5 with the remarkable decrease of the % full capsids to 39 % after 2 weeks and no further decrease after 4 weeks at 30 °C (40 %).

Empty capsids in the respective buffers during the course of storage at 30 °C is shown in Fig. 6. After 2 weeks of storage at 40 °C a remarkable decrease of the % empty capsids was analyzed in all buffers, may be due to the lack of capsid stabilizing DNA. These observation was more pronounced in the Reference (38 %), in buffer 01 (38 %), buffer 02 (40 %), buffer 03 (32 %), buffer 04 (33 %) and buffer 05 (28 %) at higher pH values. In the in general better stabilizing buffers at lower pH values, i.e., buffer 06 (53 %), buffer 07 (55 %), buffer 08 (55 %), buffer 09 (57 %), buffer 11 (61 %) and buffer 12 (60 %) this effect was not so pronounced indicating a more stable empty capsid at lower pH values. As expected, the highest decrease of the % empty capsids was analyzed for the AAV2s formulated in the osmolality buffer 10 after 2 weeks at 40 °C (27 %).

After 4 weeks of storage, the % empty capsids increased remarkably in all buffers to different extents. This could be a result of the DNA release from the full capsids during the further storage at 30 °C resulting in an increase of the empty capsids. Interestingly, the strongest increase of the empty capsids in this context was analyzed in the low pH buffers 06, 07, 08, 09, 11 , and 12 may be due to the less reduction of the % empty capsids after 2 weeks of storage. In the Reference and in the higher pH buffers 01 , 02, 03, 04 and 05 this increase was less pronounced that could be a result of the stronger reduction of the empty capsids after 2 weeks of storage at 30 °C. The ratios between the % empty capsids after 2 weeks of storage and after 4 weeks of storage at 30 °C remained more or less unchanged for all buffers.

In summary, buffers at low pH values seem to have a higher stabilizing efficiency on the AAV2 viral particles analyzed by SE-HPLC-MD at indicated time-points during storage at 30 °C. These results are partially in line with the thermal stability results. Only in the original, untreated Reference formulation, the stability of the AAV2 particles seems to be higher compared to the AAV2 particle in PIPES or phosphate buffer at pH 7.4 and in Tris buffer at pH 8.0 suggesting a high negative influence of the process of re-buffering on the storage stability of the AAV2 viral vectors at elevated temperature. The stability of the AAV2 viral vectors at 30 °C in the original, untreated formulation was more or less comparable with histidine buffer at pH 6.0 (buffer 12).

Example 2: Development of stabilizing formulations for AAV2 viral vectors under application of a Screening DoE approach.

In order to develop stabilizing formulations for liquid AAV2 formulations, a screening DoE matrix (Table 10) was designed. In orderto better analyse the influence of the single excipients on the stability of AAV2 viral vectors, a base buffer PBS + 0.001 % Poloxamer 188, pH 7.4 was included. The selected excipients added to this base buffer were amino acids such as the hydrophobic amino acid valine, the negatively charged amino acid glutamic acid, the polar amino acid threonine acting as osmolytic compounds and conformational stabilizers and the antioxidative oxygen scavenging amino acid methionine. Carnosine was included as a peptide composed of histidine and B-alanine, a well-known antioxidant and a reagent against Maillard reactions between positively charged amino acid residues (lysine) and reducing sugars (glucose). Inositol and meglumine are sugar alcohols acting as osmolytic compounds and conformational stabilizer for proteins. PEG3350 was included in order to reduce adsorption of AAV2 viral vectors on different surfaces.

Table 10: Screening DoE matrix for the development of liquid stabilizing AAV2 formulations.

[005] The AAV2 starting material was re-buffered in these 19 DoE formulations and for comparison in formulations 20 (PBS buffer, pH 7.4 + 0.001 % Poloxamer 188) using the buffer exchange kit PD MidiTrap G-25 from Cytiva. The re-buffered AAV2 formulations were analysed by time-independent, biophysical stability indicating methods, e.g., nanoDSF and SYBR Gold DSF. Additionally, these samples were stored for up to 6 weeks at 30 °C and analysed at indicated time-points using the before mentioned SE-HPLC Multi-Detector-Method.

Materials and Methods

Starting material

Tablel 1 : AAV systematic sample name, AAV concentrations in terms of the viral genome titer (vg/mL) analyzed by qPCR at Sirion, the total capsid titer cp/mL analyzed by a Capsid ELISA at Sirion and the calculated ratio total capsids/full capsids.

All preparation steps with the AAV material were performed under a laminar air flow (LAF) sterile workbench. Before leaving the LAF to the instruments all plates or cuvettes were closed.

The AAV2 starting material was re-buffered in 19 DoE formulations and for comparison in formulations 20 (PBS buffer, pH 7.4 + 0.001 % Poloxamer 188) using the buffer exchange kit PD MidiTrap G-25 from Cytiva. Rebuffering was performed by gravity elution. Equilibration was performed with 3 X 5 ml of the respective formulation buffer. Subsequently, 1 .0 ml of sample was added to the column. Elution was then performed with 1.5 ml of the formulation buffer followed by sterile filtration the AAV2 samples in the 19 formulations with a 0.22 pm filter

Nano Differential Scanning Fluorimetry (nano DSF)

AAV samples were diluted in a 96-well plate using the respective formulations and transferred into a 364-well plate for loading into the capillary chip. 20 pl of each sample were loaded in a capillary. Replicates were created by filling two capillaries with the same sample from the same well. The chip was sealed with capillary sealing paste. The loaded capillary chip was placed into the Prometheus NT.Plex (NanoTemper). After the chip was placed into the instrument a discovery scan is started at 100% excitation power due to the small size of the capsids. A temperature ramp with a ramp rate of 1°C/ min starting at 25 °C until 95 °C was applied using the PR.ThermControl software. The stability indicating parameters T_m and T_on were determined using the PR. Stability Analysis software.

SYBR Gold Differential Scanning fluoreimetry (SYBR Gold DSF)

In preparation of the measurement, the SBYR Gold DMSO stock solution with a concentration of 10.000 x was sequentially pre-diluted, first to a concentration of 400 x and secondly to a concentration of 100 x with NFW. The AAV samples were pre-diluted 1/3.6 with the respective formulation buffer in Eppendorf tubes. Subsequently, 45 pl of the AAV dilutions were pipetted in triplicates into a 96-well PCR plate (PerkinElmer) and 5 pl of the SYBR Gold 100 x predilution was added to reach a final SYBR Gold concentration of 20 x and a final dilution factor of 4 for the AAV material. The plate was sealed with an adhesive sealing sheet (Thermo Fisher Scientific). After short centrifugation at 200 rpm, the 96 well plate was placed into a qPCR cycler (BioRad). Through the BioRad CFX Manager software the fluorophore channel “FRET” was selected and the following thermal ramp was set: 28 - 86 °C, 0.5 °C/min and 90 s hold time. The analysis was also performed in the software by calculating the first derivative of the genome ejection curve and determining T_m through the inflection points. Size Exclusion (SE) - High Performance Liquid Chromatography (HPLC) - Multi-Detector- Method (MD)

For this method the UltiMate 3000 HPLC system was used, equipped with a Wyatt AAV column WTC-050N5 with a bead size of 5 pm, a diameter of 4.6 mm and a column length of 300 mm.

Table 12: Composition of the SE-HPLC running buffer.

[006] The AAV samples were measured undiluted in duplicates in a 96-well plate with a V- Bottom (Greiner). In order to check the column resolution 2 injections of BSA (Sigma) 1 mg/ml were analyzed before the other samples were injected.

[007] The following conditions were applied: Flow rate of 0.3 ml/min; Column oven temperature of 25°C; Sample temperature of 5°C; Injection volume 35 pL. The following 3 detectors were applied: UV 260/280 nm; Multi-angle light scattering (DAWN) ; dRI (Optilab).

After the measurement the column was cleaned using 0.5 M Na2SC>4 pH 2.7 and subsequently stored in a running buffer + NaNs solution. The chromatograms were evaluated using the Chromeleon Chromatrography Studio software and the Conjugate Analysis package in the ASTRA software of Wyatt.

Results

Re-buffering using PD MidiTrap columns from Cytiva with gravity elution

Before starting the whole study, the AAV2 viral vectors formulated in the Reference (PBS buffer at pH 7.4 + 0.001 % Poloxamer) had to be re-buffered into the 19 DoE formulations and in the original reference formulation F20. In order to limit the AAV2 losses observed during rebuffering by dialysis, we decided to apply another re-buffering method based on PD MidiTrap columns with gravity elution. In order to evaluate the re-buffering success, we analyzed the re-buffered samples after preparation at time-point t = 0 using the SE-HPLC-Multi-Detector-Method. The results represented simultaneously the data describing the starting situation before all experiments at time-point t = 0 (see the section concerning the storage stability of AAV2 viral vectors under accelerated aging conditions). We analyzed the total capsid concentration (total cp/mL), the full capsid concentration (full cp/mL) and the empty capsid concentration (empty cp/mL) at time-point t = 0 using the SE-

Taking into account the 1.5-fold dilution of the starting material concentration (1.55 x 10¹³ cp/ml) analyzed by SE-HPLC-Multi-Detector-Method to a theoretical AAV2 concentration of 1.033 x 10¹³ cp/ml, we calculated the recovery of capsids in the different formulations directly after re-buffering based on the SE-HPLC-Multi-Detector-Method data and the % particle losses during re-buffering. In Table 13, the results for the total capsid concentration are summarized. It is obvious from this table, that the particle recovery is clearly dependent on the formulation condition. This theoretically eluted total AAV2 particle concentration was then set to 100 % and the percent of total AAV2 capsid concentrations in the re-buffered solutions at t = 0 was calculated (Table 13).

The strongest loss of total AAV2 capsids after re-buffering was observed for formulation 10 with a loss of 54.61 % of AAV2 total particles. It is well known from literature that AAV particle severely tends to adsorb on surfaces such as the column resin of the desalting columns.

In contrast, in formulation F03 no loss of the total capsid concentration was observed and in formulation F04 only a low loss of total particles of 0.32 % was observed. An AAV2 particle loss smaller than 5 % after re-buffering at time-point t=0 was observed for formulations F03, F04, F07, F13, F14, F16, F17, F18 and F19. All of these formulations contained PEG3350 indicating a positive influence of PEG3350 against adsorption of AAV2 particles on the column resin. In formulations F01 , F02, F05, F06, F08, F09 and F15 a particle loss after re-buffering between 7 % to 10 % at time-point t = 0 was analyzed. Interestingly, these formulations did not contain PEG3350 but each of them contained either meglumine or inositol or both. This suggested a stabilizing effect of these two excipients against the loss of AAV2 particles during re-buffering. Formulations F11 , F12 and F20 showed a high particle loss between 18 to 30 % at time-point t=0 after re-buffering (Table 13).

Table 13: Total capsid concentration of AAV2 in 19 DoE formulations and in the reference formulation (formulation 20) after re-buffering at time-point t = 0. The percentage of AAV2 particles after re-buffering was calculated by normalization of the total AAV2 capsid concentration to the analyzed concentration in the Reference taking into account the theoretical concentration in 1.5 ml elution volume from the re-buffering columns. The coloured cells define the stabilizing efficacy of the formulations during re-buffering. Green represents a low loss of AAV2 particles and red represents a remarkable loss of AAV2 particles.

The results of the statistical regression analysis of the data showed a significant positive influence of PEG3350 on the total capsid concentration after re-buffering, indicating a positive effect of PEG3350 against AAV2 particle adsorption on the column resin. The main effect plot for the positive influencing PEG3350 on the total capsid concentration of AAV2 in the DoE formulations is shown in Figure 7 (low quality of fit).

Comparable results were observed for the full capsid concentration and the empty capsid concentration (data not shown).

In summary, the SE-HPLC analysis of the AAV2 particles after re-buffering in the 10 DoE formulations and into the reference formulation using the PD MidiTrap columns with gravity elution revealed a formulation dependent loss of particles during re-buffering. The qualitative and statistical analysis of the data at time-point t = 0 showed a positive influence of the excipient PEG3350 on the recovery of the AAV2 particles after re-buffering.

Nano DSF

The thermal stability of the AAV2 particles in the 20 formulations was analysed by changes in the intrinsic tryptophan fluorescence upon thermal unfolding of the AAV2 capsids using nano DSF. For comparison, the original, untreated Reference formulation was analysed. In Figure 8, the onset temperatures of thermal unfolding T_on and the midpoint temperatures of thermal unfolding T_m of AAV2 in the different formulations are depicted.

As it is obvious from the figure, the AAV2 capsids in formulation 20 revealed the lowest thermal stability in comparison to all other formulations (T_on = 59.31 °C and T_m = 66.93 °C; Table 14). Interestingly, in the same formulation composition of the original, untreated Reference the AAV2 viral vectors revealed a higher stability (T_on = 66.48 °C and T_m = 70.55 °C) indicating a negative influence of re-buffering using the columns into formulation F20 (PBS buffer pH 7.4 + 0.001 % Poloxamer 188). In contrast to formulation 20, the other formulations had a positive effect on the thermal AAV2 capsid stability during re-buffering resulting in higher thermal stability monitored by nano DSF to different extent.

Table 14: Thermal stability of the AAV2 particles in the 19 DoE formulations compared to the original, untreated Reference and formulation 20 (re-buffered in the reference formulation using the same method).

The AAV2 viral vectors in formulation F01 , F04, F08, F09, F11 , F16, F19 showed the best stabilizing efficiency against thermal unfolding of the AAV2 capsids with T_on of 71.90 °C in F01 , 71.25 °C in F04, 71.37 °C in F09, 71.82 °C in F11 , 71.66 °C in F16 and 71.02 °C in F19. The corresponding T_m values were analyzed to 74.67 °C in F01 , 74.10 °C in F04, 74.13 °C in F08, 73.94 °C in F09, 74.74 °C in F11 , 74.56 °C in F16 and 73.96 °C in F19.

All these formulations contained the excipients valine and/or glutamic acid and/or threonine and/or carnosine. The results of the statistical regression analysis of the thermal stability of

AAV2 capsids in the 19 DoE formulations revealed significant positive effects of the excipients valine, glutamic acid, threonine and carnosine on the onset temperature of thermal unfolding Ton and the midpoint temperature of thermal unfolding T_m depicted by the main effect plots and the forest plots (see Figure 9). Similar to the Reference, the AAV2 particles in formulation 06 showed one of the lowest thermal stability with a T_on value of 66.25 °C and a T_m value of 70.10 °C. This formulation only contained Inositol and PEG3350 in the base buffer and not the significant positive influencing excipients. Also a low thermal stability of AAV2 capsids particularly regarding the T_on value was analyzed in formulations 05 and 12 with T_on values of 68.84 °C and 68.39 °C, respectively and T_m values of 71.95 °C and 71.35 °C, respectively. Formulation 05 contained glutamic acid and methionine in combination with Inositol and meglumine in the base buffer and formulation 12 contained valine and methionine in the base buffer. In all other formulations, a medium thermal stability of the AAV2 viral vectors was analyzed. The statistical regression analysis could not identify excipients with statistical significant negative effect on the thermal stability of the AAV2 capsids.

In summary, the nano DSF data revealed the lowest thermal stability in formulation 20 and in the Reference. The qualitative and statistical analysis of the data showed a positive influence of the amino acids valine, glutamic acid, threonine and of the dipeptide carnosine on the thermal stability of the AAV2s.

SYBR Gold DSF

As described before, SYBR Gold DSF monitors the DNA release from the AAV2 capsids as a function of increasing temperature. The SYBR Gold fluorescence increases with increasing binding of the fluorescence dye on the DNA molecules as a result of DNA leakage from the intact capsid associated with the first thermal transition T_mi and DNA release as a result of capsid rupture associated with the second thermal transition T_m2 in the fluorescence profile upon application of a thermal ramp.

In comparison the AAV2 particles formulated in different buffers at different pH values, the DNA leakage temperature in these DoE formulations containing the similar base buffer PBS at pH 7.4 + 0.001 % Poloxamer 188 was more or less comparable in all formulations may be due to the similar pH value and the similar kind of buffer in the formulations (red bars). Nevertheless, some differences in the DNA leakage temperature from the AAV2 capsids were observed between the formulations.

In line with thermal unfolding monitored by nano DSF the lowest stability of the AAV2 viral vectors was analyzed in formulation F20 and in the original, untreated Reference formulation. The DNA leakage temperatures T_mi from the AAV2 capsids was analyzed to be 59 °C in the Reference and 59.5 °C in formulations F20. The AAV2 capsid rupture temperature T_m2 was analyzed to be 64.5 °C in the Reference and 63.5 °C in formulation F20. It is obvious from Figure 10 that the DNA leakage proceeds at remarkably lower temperature than the capsid rupture. As before mentioned the capsid rupture temperature is more or less in the range of the onset temperature of thermal unfolding T_on monitored by nano DSF. The slight differences between the DNA leakage temperatures T_mi indicate only a small influence of the excipients on the DNA leakage temperature T_mi in contrast to the pH values and the kind's of buffer. However, thermal stability of the AAV2 particles regarding the DNA leakage temperatures T_mi was slightly influenced by the formulations and the containing excipients. In contrast to thermal unfolding monitored by nano DSF, the best stabilizing formulations regarding DNA leakage from the more or less intact capsids are partially different to the best formulations in the context of thermal unfolding T_on and T_m. The highest DNA leakage temperatures T_mi were observed in formulations F01 (61 °C), F02 (61 °C), F04 (61.17 °C), F05 (61.33 °C), F08 (61.50 °C), F09 (61 °C), and F15 (61.50 °C). All these formulations contained the excipients valine and/or threonine and/or inositol and/or meglumine. Despite the Reference formulation and Formulation F20 without any additional excipients, the DNA leakage temperature T_mi was very low in formulations F03 (60.5 °C), F06 (60 °C), F07 (60.5 °C), F10 (59 °C), F11 (60 °C), F12 59.5 °C), F13 (60 °C), F14 (60 °C), F16 (60 °C), F17 (60 °C), F18 (60.5 °C), and F19 (60 °C). The center point formulations F03, F07, and F19 contained a combination of all excipients in half-maximal concentrations. In general, all of these formulations with low T_mi contained the excipients carnosine and/or PEG3350 in most cases a combination thereof (F03, F07, F14, F16, F17, and F19) or did not contain particular combinations of the before mentioned positive influencing excipients. Interestingly, in formulation F05, F15, F17, F18, and F19 we identified an unusual third thermal transition in the range of the Tm value analyzed by nano DSF. This transition was not included into the statistical regression analysis.

The results of the statistical regression analysis in form of the main effect plots revealed a strong significant positive influence of valine, threonine, inositol, and meglumine on DNA leakage monitored by SYBR Gold DSF om form of T_mi. In contrast, a strong significant negative effect was analyzed for the excipients carnosin and PEG3350 on DNA leakage (Figure 11). In comparison to the thermal unfolding results, these results could indicate probably another mode of action of the stabilization of the AAV2 viral vectors against DNA leakage with increasing temperature.

In contrast, regarding the capsid rupture temperatures T_m2, the AAV2 capsids in formulations F01 (68.67 °C), F02 (68 °C), F04 (68.50 °C), F08 (68.5 °C), F09 (68.50 °C), F11 (68.50 °C), F13 (67.83 °C), F16 (67.83 °C) showed the highest stabilities against capsid rupture. These formulations contained the excipients glutamic acid and/or threonine and/or carnosine and were partially identified as best performing formulations in the stabilization of AAV2 against thermal unfolding in particular regarding the onset temperature of thermal unfolding T_on monitored by nano DSF. This result substantiates the hypothesis, that the capsid rupture temperature is associated with the onset temperature of thermal unfolding T_on. The lowest stability of the AAV2 capsids in the context of the capsid rupture temperature was analyzed in formulations F05 (66 °C), F06 (64 °C), F12 (65 °C), F15 (66.5 °C), F17 (66 °C), and F18 (66 °C). Most of these formulations contained PEG3350 or did not contain the before mentioned positive influencing excipients.

Table 15: SYBR Gold DSF results of AAV2 particles in the 19 DoE formulations compared to the untreated Reference and formulation 20 (re-buffered in the reference formulation).

The statistical regression analysis of the capsid rupture temperatures Tm2 of the AAV2 viral vectors in the 19 DoE formulations revealed a statistically significant positive influence of the excipients glutamic acid, threonine, and carnosine on the capsid rupture. The identification of partially the same positive influencing excipients on the capsid rupture and on the onset temperature of thermal unfolding substantiated the correlation between these two thermal events. Both temperatures are measures for the conformational stability of AAV2 viral capsids.

In summary, SYBR Gold DSF revealed slightly differences between thermal unfolding and DNA release from the AAV2 capsids. The DNA leakage temperature Tm1 was positively influenced by the amino acids valine and glutamic acid in line with thermal unfolding and by the sugar alcohols inositol and meglumine different to thermal unfolding. In contrast to thermal unfolding, a significant negative influence of carnosine and PEG3350 on the DNA leakage temperature was analyzed. Moreover, regarding the AAV2 capsid rupture temperature T_m2, in line with thermal unfolding, a significant positive effect of glutamic acid, thereonine and carnosine was analyzed. For the capsid rupture temperature T_m2 a significant negative effect of PEG3350 was analyzed. DNA leakage starts before the onset of thermal unfolding T_on and the capsid rupture T_m2 where the capsids remained more or less correct folded. The start of thermal unfolding is in the range of capsid rupture. In some cases of formulations a third transition in the range of T_m was observed.

Timedependent storage stability of A AV2 in the 19 DoE formulations

In order to evaluate the storage stability of AAV2 formulated in 19 screening DoE formulations under accelerated aging conditions in comparison to the reference formulation (20), a timedependent storage stability study was performed for up to 6 weeks at 30 °C. At indicated timepoints after 2, 4 and 6 weeks of storage at 30 °C, the content of total, full and empty capsids was analysed by SE-HPLC-Multi-Detector-Method.

Due to the dilution and the loss of capsids during re-buffering, the absolute values of the total capsid concentration are decreased in comparison to the original, untreated Reference (Figure 13A). To get a better overview, the total capsid concentrations of the AAV2 vectors after rebuffering calculated as before mentioned, were set to 100 % and the analyzed total particle concentration after 2 weeks, 4 weeks and 6 weeks of storage were calculated as % recovery of total capsids after the indicated storage time-pints (Figure 13B).

The figures show that after 2 weeks of storage the loss of the total AAV2 particles remained more or less unchanged or was slightly reduced in some cases with a recovery between 93- 100 %. Only in the case of formulation F20 already after 2 weeks of storage at 30 °C the total capsid concentration was reduced to 89 %. In general, the re-buffered original reference formulation F20 had the lowest stabilizing efficiency on the AAV2 particles during storage at the elevated temperature (see below) on the total capsid concentration.

[008] After 4 weeks of storage of the AAV2 particles in the 19 DoE formulations and in formulation F20, a remarkable decrease of the % total AAV2 capsids was observed dependent on the formulation compositions. The % of total AAV2 capsids during storage was calculated by normalization of the analyzed total AAV2 capsid concentrations after 2 weeks, 4 weeks and 6 weeks of storage at 30 °C to the total AAV2 capsid concentrations at time-point t = 0 set to 100 % in each particular formulation.

In the DoE formulations F01 , F03, F04, F05, F07, F08, and F09, the highest recovery of the total capsids between 89 to 93 % was observed. Interestingly, in the original, untreated Reference formulation the highest recovery of the total capsids was found to be 98.71 % after 4 weeks of storage at 30 °C, may be due to the lack of re-buffering in comparison to the other formulations particularly the similar formulation F20 after re-buffering. All good performing DoE formulations contained threonine and/or meglumine.

In formulations F02, F10, F12, F20 the lowest stabilizing efficiency against storage for 4 weeks at 30 °C was observed (71 % -82 % total AAV2 capsids) for the total AAV2 capsids. The stabilizing efficiency of the total AAV2 capsids was additionally low in formulations F11 , F13, F14, F15, F16, F17, F18, and F19 after storage for 4 weeks at 30 °C.

Regarding the storage stability after 6 weeks of storage at 30 °C, in the most cases no further reduction of the % total capsids was observed. In the original untreated Reference, the highest recovery of 92.64 % was found. In the other formulations the recovery of % total capsids was more or less unchanged between storage for 4 weeks and storage for 6 weeks. So, we have seen that the main degradation in the formulations takes place between 2 weeks and 4 weeks of storage at 30 °C.

The statistical regression analysis of the differences of the total capsid concentration at timepoint t = 0 and after 4 weeks and 6 weeks storage at 30 °C revealed only very low to moderate quality of the fit. Nevertheless, influences of particular excipients on the storage stability of the total AAV2 capsids could be derived. Significant positive effects of the excipients threonine and meglumine were determined with negative slopes in the main effect plots in Figure 14. In contrast to re-buffering, PEG3350 showed significant negative influence on the storage stability with positive slopes in the main effect plots in Figure 14. Concerning the stability of full AAV2 capsids during storage for up to 6 weeks at 30 °C we observed more or less comparable trends. The analyzed absolute full capsid concentrations depicted in Figure 15 revealed a stabilizing efficiency of the original, untreated Reference formulation and of the DoE formulations F01 , F03, F04, F05, F06, F07, F08, F09, F10. The strongest decrease in the full capsid concentration was observed more or less for all formulations during storage between 2 weeks of storage and 4 weeks storage at 40 °C. In formulation 20, the full capsid concentration was remarkably reduced after 4 weeks of storage at 30 °C.

To get a better overview, the full capsid concentrations of the AAV2 vectors after re-buffering calculated as before mentioned, were set to 100 % and the analyzed total particle concentration after 2 weeks, 4 weeks and 6 weeks of storage were calculated as % recovery of total capsids after the indicated storage time-pints (Figure 16). As already described before for the total capsids, after 2 weeks of storage the % full capsids remained more or less unchanged (93 % - 100 %).

After 4 weeks of storage, a stronger decrease of the % full capsids was analyzed to a different extent dependent on the excipient composition in the formulations. In line with the results for the total capsids except for formulation F06 and F10 in formulations F01 , F03, F04, F05, F06, F07, F08, F09, and F10 the % full capsids was only very slightly decreased (between 89 % - 93 %). The lowest recovery of the % full capsids was analyzed in formulation F02, F12, and F20 (between 75 % - 79 %). This was also partially in line with the corresponding results for the total capsids. As expected, in formulation F20 the lowest % of full capsids was observed after 4 weeks of storage at 40 °C.

For the excipient threonine the negative slope represents only low loss of full capsid concentration during storage and so a positive influence on the stability of the full DNA containing AAV2 viral vectors as shown in Figure 17. The positive slopes for PEG3350 represent increasing loss of full capsid concentration during storage at 30 °C and so a negative effect on the storage stability of the full DNA containing AAV2 viral vectors.

The statistical regression analysis revealed also in the case of the full capsid analysis a very low to moderate fit quality. The main effect plots show the difference of the full particle concentration at time-point t = 0 and after 4 weeks and 6 weeks of storage at 30 °C. However, the statistical regression analysis identified a strong significant positive influence of threonine on the stability of the full capsids with a negative slope in the main effect plots representing the low loss of full particles. A significant negative influence was observed for PEG3350 with a positive slope in the main effect plots indicating the higher loss of full particles upon storage.

In the case of the stability of the empty capsids, some differences compared to the total and full capsids are obvious from Figures 18 and 19. In general, the empty capsids seem to be less stable compared to the full capsids, may be due to the lack of DNA that could have a stabilizing effect on the stability of the protein capsid. The strongest decrease of the empty capsid concentration was observed in formulation F20, but also in formulation F02, F06, F10and F12 (Figure 19).

As shown in Figure 19, the % empty capsids was considerably reduced in formulation F06, F12, F10, and F20 already after 2 weeks of storage (67 % - 74 %). In contrast, the recovery of the empty capsids was high after storage for 2 weeks at 30 °C in formulations F03, F04, F05, F07, F08, F09, F11 , F15, F17, F18 (94 % - 100 %).

After 4 weeks of storage, the empty AAV2 particles were most stable in formulations F01 , F03, F04, F05, F07,F 08, and F09 (81 % - 92 %). In these formulations after 6 weeks of storage at 30 °C, the % empty capsids remained more or less unchanged (81 % - 97 %). The least stable empty AAV2 particles were analyzed in formulations F02, F06, F10, F12, and F20 (43 % - 70 %). In contrast to the before mentioned stabilizing formulations, the % empty capsids partially further decreased in these formulations (F06 to 56 %, F10 to 35 %, F12 to 51 %).

As a result, the full to empty ratio very strong changed during the course of storage at 30 °C in these formulations (figure 20). In therapeutic AAV formulations, the full to empty ratio should not change upon storage.

The statistic regression analysis for the empty capsid concentration revealed also moderate to low fit qualities. But, the evaluation of significant effects was nevertheless possible. In contrast to the total and full capsids, the analysis identified other excipients with positive influence on the empty capsids (Figure 21). The main effect plots in Figure 21 show the difference between the concentration of empty capsids at time-point t = 0 and after 4 weeks of storage at 30 °C with a negative slope indicating the lowest loss of empty particles (Figure 21) for glutamic acid, threonine and meglumine. The addition of PEG3350 resulted again in a negative influence of the stability of empty particles with a positive slope in the main effect plot, indicating a higher loss of particle during storage for 4 weeks at 30 °C. Regarding the storage stability of the empty capsids we can see a qualitative correlation to the analysis of the conformational (thermal) stability of the AAV2 capsids in form of the positive influence of the excipients glutamic acid and threonine (T_on, T_m and T_m2 in nano DSF and SYBR Gold DSF). The positive influence of meglumine could indicate the qualitative correlation of the stabilizing influence on the total and full capsids and on the DNA leakage temperature in SYBR Gold DSF. The negative influence of PEG3350 was also analyzed for the DNA leakage temperature and the capsid rupture temperature in SYBR Gold DSF.

The same significant influences of glutamic acid, threonine, meglumine and PEG3350 were also identified after 6 weeks of storage, may be due to the more or less retention of the AAV2 particles between 4 weeks and 6 weeks of storage in the best stabilizing formulations for the empty capsids.

Example 3: Further development of stabilizing formulations for AAV2 viral vectors under application of an Optimization DoE approach based on the results of Example 2. Optimization DoE Formulation Matrix

The design of the optimization DoE formulation matrix was based on the results of the screening DoE matrix in Example 2 of this patent application. The osmolytic amino acids Threonine, Glutamic acid, and Valine revealed a statistically significant positive stabilizing effect on the thermal stability of the AAV2 viral vectors monitored by nanoDSF and SybrGoldDSF. The sugar alcohol Meglumin showed a statistically significant stabilizing effect on the leakage temperature T_mi of the DNA from the AAV2 capsids analyzed by SybrGoldDSF. In addition, Meglumin showed a statistically significant stabilizing effect on the storage stability of the AAV2 capsids in terms of the total AAV2 capsid concentration per mL and the full capsid concentration per mL under accelerated storage conditions at 30 °C for up to 6 weeks. In order to analyze the most suitable concentration ranges of these excipients and possible interactions between these excipients, an optimization DoE formulation matrix with the above mentioned four excipients as factors at a minimal and maximal concentration level using an A optimal Design as calculated. 20 DoE formulations were analyzed for their stabilizing efficacy on the AA2 viral vectors and compared them with the commonly used buffer formulation (PBS pH 7.4 + 0.001 % Poloxamer 188 (Formulations F_21). This buffer formulation was the background buffer for all 20 DoE formulations and the excipient mixtures were added to this buffer. In Table 1 , the compositions and concentrations of excipients in these 20 DoE formulations and in the reference formulation F_21 are summarized.

Table 16: Optimization DoE formulation matrix with 20 DoE formulations and a reference formulation F_21 (PBS pH 7.4 + 0.001 % Poloxamer 188). This buffer was the background buffer in all DoE formulations containing various concentrations and amounts of excipients. The excipient concentration is indicated in mM. F_02, F_10, and F_20 are the center point fomrulations containing all excipients in half maximal concentrations for the validation of the results.

The thermal stability of the formulated AAV2 viral vectors was analysed using nanoDSF and SybrGoldDSF. In addition, the AAV2 viral vectors formulated in the above mentioned formulations were stored for up to 4 weeks under accelerated conditions at 40 °C. The stability of the AAV2 viral vectors during accelerated storage was analyzed at indicated time-points during storage by ddPCR in form of the virus titer (virus genome/mL) and by SEC-MALS in form of the AAV2 capsid concentration (total particles/mL, full particles/mL, empty particles/ml, and ration full/total particles).

Materials and methods

The starting material was a mixture of the AAV2 preparations used in Examples 1 and 2 of this patent application both formulated in the above mentioned standard buffer for AAVs PBS + 0.001 % poloxamer 188 pH 7.4. This pool of two AAV2 preparations was re-buffered in the 20 DoE formulations by dialysis. For comparison, the AAV2 viral vectors in the standard formulation were also re-buffered in the same formulation by dialysis in order to apply the same stress conditions during sample preparation. After this re-buffering step, the resulting material was aliquoted in the desired volumes for further analysis and storage experiments. The aliquots were frozen at - 70 °C until their application in the various experiments.

Nano Differential Scanning Fluorimetry (nano DSF), SYBR Gold Differential Scanning fluoreimetry (SYBR Gold DSF), Digital droplet Polymerase Chain Reaction (ddPCR), Size Exclusion (SE) - High Performance Liquid Chromatography (HPLC) - Multi-Detector-Method (MD) were performed in accordance with Example 1.

Results

1. Nano DSF

Thermal stability of AAV2 viral vectors was monitored by changes in the intrinsic tryptophan fluorescence of the protein capsid upon application of a thermal ramp between 25 °C and 95 °C. The highest thermal stability of AAV2 re-buffered in the 20 DoE formulations by dialysis in terms of the onset temperature of thermal unfolding (T_on) was analyzed in the formulations in the following order F_06 > F_12 > F04 > F_08 > F_13 > F_17 (Table 17). The corresponding onset temperatures T_on were observed to be between 70-71 °C (dotted line at 70 °C in Figure 23) all containing high concentrations of threonine in combination with variable concentrations of valine, and/or glutamic acid and/or meglumine (Figure 23). This trend was also observed for the midpoint temperature of thermal unfolding T_m monitored by intrinsic tryptophan fluorescence. The midpoint temperatures of thermal unfolding T_m were analyzed to be between 74.35 °C and 75.06 °C for the formulations in the following order F_12 > F_06 > F_08 > F_14 > F_04 (Table 16; dotted line at 74 °C in Figure 23). These formulations contained high concentrations of Threonine, and/or low to high concentrations of Glutamic acid and/or low to high concentrations of Valine.

Table 17: Thermal unfolding analysis of the AAV2 viral vectors in the 20 DoE formulations and the reference formulation F_21. The analyzed thermal unfolding parameters are the onset temperature of thermal unfolding T_on and the midpoint temperature of thermal unfolding T_m.

All formulations without threonine, F_05, F_19, F_01 , F_15, F_16, and F_07 revealed a very low thermal stability more or less comparable to F_21 with onset temperatures of thermal unfolding T_on between 65.49 °C and 68.04 °C and midpoint temperatures of thermal unfolding T_m between 70.52 and 71.98 °C.

Medium thermal stability regarding the onset temperature of thermal unfolding T_on was analyzed in the formulations with the following order, F_09 (68.43 °C) < F_20 (68.47 °C) < F_10 (69.05 °C) < F_11 (69.38 °C) < F_02 (69.45 °C) < F18 (69.59 °C) < F03 (69.90 °C) = F_14 (69.90 °C) all containing threonine in variable concentrations. The thermal stability was increased in the formulations with increasing threonine content. For the midpoint temperature of thermal unfolding T_m a similar trend was analyzed with increasing thermal stability in the following order, F_20 (72.47 °C) < F_10 (72.96 °C) < F_09 (73.22 °C) < F_11 (73.26 °C) < F_02 (73.37 °C) < F_03 (73.60 °C) < F_17 (73.78 °C) < F_13 (73. 82 °C) < F_18 (73.86 °C), also containing Threonine in variable concentrations.

In summary, the thermal stability of the AAV2 viral vectors in all formulations was higher in comparison to the common AAV2 formulation F_21 (PBS pH 7.4 + 0.001 % Poloxamer 188) without further excipients. The addition of threonine seemed to have the highest stabilizing effect on the thermal stability of AAV2 analyzed by nanoDSF. All formulations with the highest amount of threonine showed the best stabilizing efficacy on the AAV2 viral vectors against thermal unfolding. Intermediate thermal stability of AAV2 was analyzed in formulations containing intermediate and low concentrations of threonine. The formulations with the lowest thermal stability F_21 , F_95, and F_19 did not contain threonine.

The statistical evaluation of the DoE data supported this observation. The analyzed excipients threonine, meglumine and valine revealed the highest, statistically significant stabilizing efficacy for AAV2 viral vectors against thermal stress. The regression analysis revealed a very high quality of the fit with a R² value of 0.0.9503 and a p value < 0.001.

In an optimization design, predictors effects are best described in interaction with each other, as they do not exist alone in the design space (i.e. the formulations composition) nor they behave linearly (due to quadratic effects representing curvature). For this reason, the interpretation of effects is given by analysis of contour plots (not shown).

From the contour plots it can be derived, that high concentrations of Meglumin in combination with low concentrations of Valine are efficient in stabilizing the AAV2 against thermal stress in terms of the onset temperature T_on monitored by nanoDSF. The highest stabilizing effect regarding the onset temperature of thermal unfolding T_on of AAV2 can be achieved at high concentrations of Valine in combinations with mid/low and mid/high concentrations of Meglumin (not shown). The best stabilizing effect in the combination of Threonine and Meglumin, can be achieved at high concentrations of Threonine in combinations with mid/low to mid/high concentrations of Meglumin (not shown).

The interactions between glutamic acid and valine, glutamic acid and meglumin and Threonine and Valine are not significant. Regarding the midpoint of thermal unfolding T_m, the regression analysis revealed a statistically high significant stabilizing effect of the amino acids threonine and valine on the thermal transition temperature of AAV2 monitored by nanoDSF. The regression analysis revealed a very high quality of the fit with a R² value of 0.0.9711 and a p value < 0.001 .

For the interpretation of these calculated effects of the excipients particularly of their interactions in the formulation mixture, contour plots for T_m (not shown) were calculated.

The best stabilizing effect on thermal unfolding of the AAV2 can be achieved in the presence of high concentrations of Threonine and low concentrations of Meglumin (not shown). Regarding the combination of Threonine and glutamic acid, the best stabilizing effect can be achieved at high concentrations of Threonine in combination with low to medium high concentrations of glutamic acid (not shown).

The results of the statistical analysis of the DoE results for the nanoDSF data substantiated our qualitative conclusions above.

SybrGold DSF

In line with the results of Example 2 for the leakage temperatures T_mi of the AAV2 viral vectors analyzed by SybrGoldDSF, the temperatures when the nucleic acid starts to leave the vector through small lacks in the protein capsids due to the application of a temperature ramp was not remarkably influenced by the presence of the excipients. This could be a result of the similar background buffer composed of PBS buffer + 0.001 % poloxamer 188 pH 7.4. In Example 1 of this patent application, a remarkable differences between the leakage temperatures dependent on the buffer and pH of the composition was observed.

Table 18: Thermal unfolding of the AAV2 viral vectors monitored by SybrGoldDSF. Two thermal transition were analyzed in the SybrGold fluorescence after binding of the fluorescence dye SybrGold to the DNA. T_mi can be assigned to the leakage temperature, where the DNA begins to leave the capsids and T_m2 can be assigned to the rupture temperature of the capsids, where the protein capsids are damaged with increasing temperature.

In this experiment, the lowest leakage temperature was analyzed for the AAV2 in formulations F_21 without excipients in line with the results in Example 2 for this formulation. The highest T_mi was analyzed in formulation F_04 (60.50 °C), one of the formulations with high concentrations of Threonine and/orMeglumin, and/or Valine. In formulations 15 (60.25 °C), 11 (60.00 °C), and 12 (60.00 °C) also relatively high T_mi values were analyzed (see Figure 24).

The trends of the rupture temperature T_m2 of the AAV2 capsids in the different formulations analyzed by SybrGoldDSF is strong in line with the onset temperatures of thermal unfolding analyzed by nanoDSF (Figure 25). The rupture temperature of the AAV2 protein capsid is strongly influenced by the excipients in the DoE formulations. The observed trends are comparable to the trends in the analysis of thermal unfolding. In general, the rupture temperature is smaller than the onset temperature of thermal unfolding (Figure 25).

The highest thermal stability in terms of the rupture temperature T_m2 in SybrGold DSF of the AAV2 viral vectors was analyzed in formulations F_04 (68.50 °C), F_06 (68.00 °C), F_12 (68.00 °C), F_08 (67.75 °C, and F_14 (67.50 °C) partially in line with the nanoDSF results. These formulations all contained the highest concentration of threonine and low to high concentrations of meglumin. In contrast, the lowest rupture temperatures T_m2 were observed in formulation F_15 (63.50 °C), F_01 (64.00 °C) and in formulation F_21 (64.00 °C), F_05 (64.50 °C), F_16 (64.50 °C), F_19 (64.50 °C), and F_07 (65.00 °C) all without threonine. The remaining formulations showed a medium stability regarding the rupture temperature T_m2, with increasing stability related to increasing threonine concentrations.

The statistical evaluation of the SybrGold results for the 20 DoE formulations substantiated the qualitative observations mentioned above.

Regarding the regression analysis of the leakage temperature T_mi of the AAV2 viral vectors in the 20 DoE formulations, the excipients Threonine, Megumine and Valine showed a statistically significant stabilizing effect. The regression analysis revealed a medium quality of the fit with a R² value of 0.73 and a p value < 0.001 may be due to the low differences between the formulations for T_mi.

For the interpretation of the regression results particularly for the identification of the most suitable concentration ranges for interactions between the excipients in the formulation mixture achieving the best stabilizing effect on the leakage temperature T_mi contour plots were generated.

The calculated interactions are depicted in the contour plots and show the concentration ranges of the two excipients suitable for maximal stabilization regarding leakage of the DNA from the AAV2 particles T_mi during application of a temperature ramp.

The contour plots (not shown) revealed two statistically significant interactions between the excipients. High concentrations of meglumin and low to medium concentrations of glutamic acid are positively influencing the stability of the viral vectors regarding the leakage temperature T_mi (not shown). Furthermore, high concentrations of valine and threonine show a significant positive effect on the leakage temperature T_mi (not shown). The before mentioned rupture temperature T_m2 of the AAV2 in the 20 DoE formulations is positively influenced by partially the similar excipients analyzed for the onset temperature of thermal unfolding T_on monitored by nanoDSF. For Meglumin the regression revealed even a destabilizing effect. The regression analysis revealed a very high quality of the fit with a R² value of 0.93 and a p value < 0.001. In contrast to T_mi, the differences of T_m2 between the DoE formulations are more pronounced.

In for the regression analysis of the rupture temperature pf AAV2 in the 20 DoE formulations, particularly Threonine showed a stabilizing effect on T_m2. The interactions between the excipients Threonine and Meglumin showed the best stabilizing effect with high concentration of Threonine in combination with low or the highest concentration of Meglumin. This observation is substantiated by the contour plot calculated for Threonine and Meglumin and the rupture temperature T_m2 of the AAV2 viral vectors analyzed in the 20 optimization DoE formulations using the best coded model (not shown).

The contour plot showed the best stabilizing effect of Threonine at high concentrations and of Meglumin at high and low concentrations. The evaluated interactions between the excipients Threonine/Glutamic acid and ThreonineA/aline are not significant.

Despite the qualitatively comparable trends between the T_on and T_m2 values, there are some differences in the biophysical background and in the statistical evaluation of the data for these two stability indicating parameters of AAV2.

ddPCR

In this experiment, the AAV2 titer (AAV2 concentration in terms of the nucleic acid concentration inside the viral vectors [virus genome/mL] upon storage under accelerated aging conditions for up to 4 weeks at 40 °C by monitoring the nucleic acid concentration was analysed via ddPCR in comparison to the untreated positive control. In Figure 26, the analyzed titer of the positive control, in other words the starting titer, is depicted as dotted line.

The comparison of the determined titers directly after dialysis at time-point t = 0 with the titer of the positive control indicated a partial loss of initial AAV2 titer as a result of sample preparation dependent on the formulation composition, particularly in the reference formulation F_21 , F_13, F_14, F_15, F_16, F_18, F_19, F_20. The highest loss of initial AAV2 titer was analyzed in the reference formulation F_21.

In contrast, the highest retention of the initial AAV2 titer at time-point t = 0 directly after rebuffering by dialysis was analyzed in formulation F_04, F_06, F_11 , and F_12, and to a minor degree in formulations F_01 , F_03, F_09, and F_10. These formulations contained low to high concentrations of Glutamic acid and/or low to high concentrations of Threonine and/or low to high concentrations of Valine. Most of the formulations also contained medium to high concentrations of Meglumin.

In contrast, the least stable formulations with the highest difference between the nucleic acid titer at time-point t = 0 were formulations F_21 without further excipients, F_05, and F_07 all without threonine and in the case of F_05 with the lowest concentration of Meglumin. AAV2 in the other formulations showed a medium stability in terms of the AAV2 nucleic acid titer.

The statistical evaluation of the DoE data supported the before mentioned observations about Threonine and Meglumin observation. The analyzed excipients threonine and meglumine revealed the highest, statistically significant stabilizing efficacy for AAV2 viral vectors during storage stress. The regression analysis revealed a medium quality of the fit with a R² value of 0.72 and a p value < 0.001.

For the interpretation of the regression results particularly for the identification of the most suitable concentration range for interactions between the excipients in the formulation mixture achieving the best stabilizing effect during re-buffering by dialysis at time-point t = 0 contour plots were generated.

Regarding the interactions between the excipients Glutamic acid and Threonine, medium-high to high concentrations of Glutamic acid in combination with the highest concentration of Threonine led to the highest stabilizing effect on the AAV2 virus titer analyzed be ddPCR directly after sample preparation. For the combination of Glutamic acid and Valine both components at the highest concentrations revealed the best stabilizing efficiency on the AAV2 directly after sample preparation at time-point t = 0 . The highest concentration of Valine and Threonine can achieve the highest retention of the virus titer during sample preparation via dialysis.

The positive trend of the stabilizing efficiency of formulations F_04, F_06, F_11 , and F_12 and to a minor degree in formulations F_01 , F_03, F_09, and F_10 was also observed until 2 weeks of storage at 40 °C. After 4 weeks of storage a remarkable reduction of the analyzed AAV2 titer was observed in all formulations. But, in formulations F_04 and F_06 and to a minor extent in formulation F_02 and F_03 the best recovery of the virus titer after 4 weeks of storage at 40 °C was analyzed.

In contrast, the least stable formulations with the highest difference between the nucleic acid titer at time-point t = 0, and after 1 week, 2 weeks and 4 weeks of storage at 40 °C were formulations F_21 without further excipients, formulations F_05, and F_07 all without threonine and meglumin, and in the case of F_05 without threonine and with the lowest concentration of Meglumin. AAV2 in the other formulations showed a medium stability in terms of the AAV2 nucleic acid titer.

It can be concluded that the excipients Threonine and Meglumin are very important for the retention of the AAV2 stability during storage for up to 4 weeks at 40 °C. The regression analysis of the DoE data was performed for the analyzed virus titer using ddPCR after 2 weeks of storage at 40 °C, because after 4 weeks of storage at 40 °C, the results for the center point formulations with the similar excipient composition were very variable. Because these center point formulations contain the similar excipient composition, the results for these formulations are expected to be very similar. For this reason, the regression analysis with highly variable center point formulation makes no sense.

The regression analysis of the AAV2 virus titers in the 20 DoE formulations during storage for 2 weeks at 40 °C revealed that the excipients Threonine, Meglumin and Valine showed a statistically significant stabilizing effect on the AAV2 viral vectors during storage. The regression analysis revealed a high quality of the fit with a R² value of 0.84 and a p value < 0.001.

For the interpretation of the regression results particularly for the identification of the most suitable concentration range for interactions between the excipients in the formulation mixture achieving the best stabilizing effect during re-buffering by dialysis at time-point t = 2 weeks of storage at 40 °C contour plots (not shown) were generated.

The contour plots showed that the highest Meglumin concentration in combinations with medium-high to high concentrations of Glutamic acid can lead to the highest stabilizing efficiency regarding the retention of the virus titer during storage for 2 weeks at 40 °C (not shown). Regarding the combination of Threonine and Valine, high concentrations of Threonine and Valine seem to be the most promising combinations of these excipients regarding the retention of the virus titer during storage for 2 weeks at 40 °C (not shown).

SEC-MALS

In addition to the analysis of the nucleic acid titer of the AAV2 viral vectors by ddPCR, the concentration of AAV2 capsids was analysed by SEC-MALS during storage at accelerated aging conditions, for up to 4 weeks at 40 °C. Using this method the quantification of the total AAV capsid concentration per mL, the full capsid concentration containing nucleic acid per mL and the concentration of empty capsids without nucleic acid per mL is possible. This method is a common quality control method during AAV manufacturing and storage. The stabilizing efficiency of the formulations was already observable at time-point t = 0 directly after AAV2 sample preparation in line with the ddPCR results.

In formulation F_04 the complete retention of the total, full and empty capsid concentration was observed already after sample preparation end re-buffering at time-point t = 0 in line with the ddPCR results. Also in formulations F_06, F_03, F_11 , F_12, F_01 , and F_02, high concentration of particles were observed directly after sample preparation by re-buffering using dialysis at time-point t = 0. These are the similar formulations with the best stabilizing efficiency during re-buffering by dialysis analyzed by ddPCR for the determination of the virus genome. These formulations all contained high or medium concentration of Meglumin, and/or high and low concentrations of Glutamic acid, and/or high and low concentrations of Threonine, and/or high and low concentrations of Valine. Interestingly, in the reference formulation F_21 without further excipients a remarkable loss of particle concentrations already after samples preparation at time-point t = 0 was observed. In formulations F_18, F_07, F_20, F_09, F_16, and F_13 also a remarkable loss of AAV2 capsids was analyzed at time-point t = 0. Most of these formulations did not contain Meglumin and/or Glutamic acid, and/or Threonine, and/or Valine. In the other formulations, a medium loss of particle concentration depending on the formulation composition was observed. Table 19: SEC-MALS analysis of the AAV2 viral vectors in the 20 DoE formulations in comparison to the reference formulation F_21 and the AAV2 untreated standard at time-point t = 0, directly after sample preparation. The analysis revealed the total capsid concentration per mL, the full capsid concentration per mL and the empty capsid concentration per mL. An important additional parameter is the full to total ratio.

The regression analysis of the total AAV2 capsid concentration in the 20 DoE formulations during at t = 0 revealed that the excipients Threonine and Meglumin showed a statistically significant stabilizing effect on the AAV2 viral vectors during storage. The regression analysis revealed a high quality of the fit with a R² value of 0.82 and a p value < 0.001. For the interpretation of the regression results particularly for the identification of the most suitable concentration range for interactions between the excipients in the formulation mixture achieving the best stabilizing effect during re-buffering by dialysis at time-point t = 0 contour plots were generated.

The analysis of the contour plots (not shown) showed the lowest concentration of Glutamic acid in combination with the highest concentration of Threonine resulted in a good recovery of the total capsids after dialysis. The same effect was determined for the lowest concentrations of Glutamic acid and highest concentrations of Meglumin.

Regarding the mixture of Glutamic acid and Valine, both excipients at lowest concentration and at highest concentration can result in a high recovery of the AAV2 viral vectors after sample preparation via dialysis. Moreover, the combinations of Threonine/Valine and Meglumin/Valine can lead to the highest recovery of AAV2 viral vectors at their respective highest concentrations after sample preparation.

Comparable results were analyzed by regression of the DoE data for the full particle concentration at t = 0, directly after sample preparation. The regression showed the stabilizing effect of Threonine and Meglumin. The resulting fit was of high quality with a R² value of 0.82 and a p value p < 0.001.

As expected from the data, the contour plots for t = 0 directly after sample preparation showed similar interactions to the total capsid concentration for the full capsid concentration in comparison to the total capsid concentration.

In general, during the course of storage under accelerated aging conditions at 40 °C, the capsid concentrations analyzed by SEC-MALS remarkably decrease already after one week of storage (Figure 27), what is not in line with the AAV2 titer concentration analyzed by ddPCR. During the further course of storage for up to 2 weeks at 40 °C, a moderate decrease of the capsid concentration was observed in all formulations, of course composition dependent (Figure 27). A dramatic decrease of the capsid concentrations was observed in all formulations after 4 weeks of storage at 40 °C (Figure 27).

A similar trend of stabilization to the values at t = 0 directly after sample preparation can be observed for the 2 weeks of storage as well as for the 4 weeks of storage at 40 °C. The formulations F_04, F_06, F_03, and to a minor extent formulations F_02, F_11 , F_12 showed the highest capsid concentration after 2 weeks as well as after 4 weeks of storage at 40 °C. All formulations contained high or medium concentration of Meglumin, and/or high and low concentrations of Glutamic acid, and/or high and low concentrations of Threonine, and/or high and low concentrations of Valine. Comparable to t = 0, the highest loss of AAV2 capsids was observed in the reference formulation F_21 without further excipients (Table 20, 21).

Partially in line with the results at t = 0, in formulations F_05, F_07, F_09, F_15, F_16, F_18, F_19 also a remarkable loss of AAV2 capsids was analyzed after two weeks and 4 weeks of storage at 40 °C. Most of these formulations did not contain Meglumin and/or Glutamic acid, and/or Threonine, and/or Valine. In the other formulations, a medium loss of particle concentration depending on the formulation composition was observed (Table 20, 21).

Table 20: SEC-MALS analysis of the AAV2 viral vectors in the 20 DoE formulations in comparison to the reference formulation and the AAV2 untreated standard at time-point t = 2 week of storage at 40 °C, directly after sample preparation. The analysis revealed the total capsid concentration per mL, the full capsid concentration per mL and the empty capsid concentration per mL. An important additional parameter is the full to total ratio.

Regarding the full to total ratio, this parameter remained more or less unchanged during the first 2 weeks of storage at 40 °C (Figure 27 D, and Table 21). After 4 weeks of storage, the full to total ratio was remarkably reduced in formulations F_01 , F_05, and F_15, indicating a loss of full capsids in these formulations (Table 21 full particles/mL). These three formulations did not contain meglumin. In other formulations with low stability, the total, full, and empty capsids were reduced during storage in the comparable order of magnitude. Thus, the full to total ratio remained more or less unchanged (Table 21).

Table 21 : SEC-MALS analysis of the AAV2 viral vectors in the 20 DoE formulations in comparison to the reference formulation and the AAV2 untreated standard at time-point t = 4 week of storage at 40 °C, directly after sample preparation. The analysis revealed the total capsid concentration per mL, the full capsid concentration per mL and the empty capsid concentration per mL. An important additional parameter is the full to total ratio.

The regression analysis of the total AAV2 capsid concentration in the 20 DoE formulations during after 4 weeks of storage at 40 °C monitored by SEC-MALS revealed that the excipients Threonine, Meglumin and Valine showed a statistically significant stabilizing effect on the AAV2 viral vectors during storage. The regression analysis revealed a very high quality of the fit with a R² value of 0.89 and a p value < 0.001 .

For the interpretation of the regression results particularly for the identification of the most suitable concentration range for interactions between the excipients in the formulation mixture achieving the best stabilizing effect during storage for 4 weeks at 40 °C contour plots (see Figure 28) were generated.

The analysis of the contour plots for the total capsid concentration after 4 weeks of storage at 40 °C, revealed significant interactions between the excipients Glutamic acid/Threonine, Glutamic acid/Meglumin, Glutamic acid/Valine, and Threonine/Valine. The plots show for the interaction between Glutamic acid and Threonine, that medium to low concentration of Glutamic acid in combination with high concentration of Threonine can read to the best recovery of the total capsids during storage at 40 °C. For all other interactions, the highest concentrations of the respective excipients can result in the highest stability of the AAV2 total capsids during storage at 40 °C.

Regarding the results for the full AAV2 capsid concentrations, similar trends qualitatively to the total AAV2 capsid concentrations were observed.

The regression analysis of the full AAV2 capsid concentration in the 20 DoE formulations during after 4 weeks of storage at 40 °C monitored by SEC-MALS revealed that the excipients Threonine, Meglumin and Valine showed a statistically significant stabilizing effect on the full AAV2 viral capsids during storage. The regression analysis revealed a very high quality of the fit with a R² value of 0.97 and a p value < 0.001 .

For the interpretation of the regression results particularly for the identification of the most suitable concentration range for interactions between the excipients in the formulation mixture achieving the best stabilizing effect during storage for 4 weeks at 40 °C the contour plots were generated. The interactions between all excipients were identified to be significant. In general, comparable trends were observed for these interactions between the excipients for the stabilization of the full AAV2 particles to the total AAV2 particles.

Similar to the full particle concentration, medium to low Glutamic acid in combination with the highest concentration of Threonine can lead to the best stabilization of the full AAV2 particles. Moreover, for the combination of Glutamic acid and Mglumin, low concentrations of Glutamic acid and the highest concentration of Meglumin can result in the highest stability of the full AAV2 vectors. Regarding the remaining combinations Glutamic acid/Valine, Threonine/Meglumin, ThreonineA/aline, and Meglumin Valine, the highest concentrations of both respective excipient can result in the best stabilizing efficiency of the formulation for the full AAV2 viral vectors during storage for up to 40 °C.

Some of the interactions identified here for the SEC-MALS results during storage, were also identified concerning the other biophysical methods nanoDSF and SybrGoldDSF. Comparable interactions were also identified in the ddPCR virus titer analysis during storage at 40 °C.

REFERENCES

Bulaklak K, Gersbach CA. “The once and future gene therapy”. 2020. Nat Commun 11 : 5820. Srivastava A, Mallela KMG, Deorkar N, Brophy G. Manufacturing Challenges and Rational Formulation Development for AAV Viral Vectors. 2021. J Pharm Sci 110: 2609-2624.

Collins M, Thrasher A. “Gene therapy: progress and predictions”. 2015. Proc Biol Sci. 282 (1821)

Lukashev AN and Zamyatnin AA ’’Viral Vectors for Gene Therapy: Current State and Clinical Perspectives”. 2016. VFront Mol Neurosci.;9: 56.

Rieser R, Menzen T, Biel M, et al. Systematic Studies on Stabilization of AAV Vector Formulations by Lyophilization. 2022. J Pharm Sci 111 : 2288-2298.

Stoica L and Sena-Esteves M “Adeno Associated Viral Vector Delivered RNAi for Gene Therapy of SOD1 Amyotrophic Lateral Sclerosis”. 2016v. Front Mol Neurosc. 9: 56

Ura T, Okuda K, Shimada M. Developments in Viral Vector-Based Vaccines. 2014. Vaccines (Basel) 2: 624-641.

Claims

1. A liquid composition comprising a virus, or viral vector or virus-like particle and

(a) at least one compound according to formula (I)

wherein R1 and R2 are independently from each other selected from H, C1-C4-alkyl, CH2CH2OH or CH₂CH(CH₃)OH, and/or

(b) threonine, and

(c) at least one buffer, and

(d) optionally at least at least one surfactant.

2. The composition according to claim 1, wherein the at least one compound according to formula (I) comprises or is N-methyl-D-glucamin.

3. The composition according to any one of the preceding claims, further comprising an amino acid with a negatively charged functional group, preferably glutamic acid.

4. The composition according to any one of the preceding claims, wherein the buffer is selected from a phosphate, TRIS, histidine, carbonate bicarbonate, citrate, maleate, adipate, HEPES, TES, MOPS, MES or PIPES buffer, preferably wherein the buffer is a phosphate buffer.

5. The composition according to any one of the preceding claims, having a pH of about 5 to about 8; preferably of about 7 to about 7.8; and most preferably of about 7.4.

6. The composition according to any one of the preceding claims, wherein the surfactant is a non-ionic surfactant, preferably a poloxamer or a polysorbitol, and most preferably poloxamer 188.

7. The composition according to any one of the preceding claims, further comprising at least one salt, preferably NaCI and/or KCI.

8. The composition according to any one of the preceding claims, further comprising at least one antioxidant, preferably an amino acid providing an anti-oxidative function, more preferably methionine.

9. The composition according to any one of the preceding claims, further comprising at least one polyethylene glycol, preferably polyethylene glycol 3350.

10. The composition according to any one of the preceding claims, further comprising an excipient selected from inositol, valine, carnosine and any combination thereof.

11. The composition according to any one of the preceding claims, comprising (i) the at least one compound according to formula (I), preferably N-methyl-D-glucamin in a concentration of about 60 mM to about 180 mM, preferably about 70 mM to about 155 mM; (ii) threonine in a concentration of about 50 mM to about 160 mM, preferably about 60 mM to about 140 mM; and (iii) optionally glutamic acid in a concentration of about 7 mM to about 30 mM, preferably about 8 mM to about 25 mM.

12. The composition according to any one of the preceding claims wherein the virus is a non-enveloped virus, preferably selected from an adenovirus or adeno associated virus.

13. The composition according to any one of the preceding claims wherein the virus is selected from AAV1 , AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9, or AAV10, preferably AAV2.

14. The composition according to any one of the preceding claims, wherein the composition is in a frozen or dried form.

15. A method of providing a liquid composition comprising a virus, or viral vector or viruslike particle derived therefrom, comprising at least a first step of providing a liquid composition according to any one of claim 1 to claim 13, and a second step of storing said composition in liquid form for at least about 4 weeks, preferably at least 6 weeks.