Compositions and methods for stabilizing live-attenuated flaviviruses

Abstract

This disclosure relates to stabilizers for compositions comprising live-attenuated flaviviruses, methods for stabilizing live-attenuated flaviviruses, and stabilized live-attenuated flavivirus vaccine compositions.

Description

[0001] COMPOSITIONS AND METHODS FOR STABILIZING LIVE-ATTENUATED

[0002] FLA VI VIRUSES

[0003] TECHNICAL FIELD

[0004] The disclosure relates to compositions and methods for stabilizing live-attenuated flaviviruses and to stabilized live-attenuated flavivirus vaccine compositions.

[0005] BACKGROUND

[0006] Vaccination is one of the greatest achievements of medicine and has protected millions of people from the effects of devastating diseases. Vaccination is widely used to prevent and treat infection by bacteria, viruses and other pathogens. Several different approaches are used in vaccination, including the administration of killed pathogens, live-attenuated pathogens and inactive pathogen subunits. Live- attenuated vaccines consist of pathogens that have lost their virulence but that are still able to replicate within the host. Live-attenuated vaccines are considered the most successful of all human vaccines because they are able to generate both humoral (antibody) and cellular immune responses by mimicking natural infections and to confer long-term immunity after one or two immunizations (Burke, 1999). The antigens of such vaccines are bacteria (tuberculosis (BCG), typhoid) or most often viruses (measles, mumps, rubella, polio, yellow fever, varicella and rotavirus).

[0007] Live-attenuated viral vaccine potency correlates with their replication potential and any factors that inactivate the virus can reduce the vaccine efficacy. The limited stability of viruses in aqueous media is well known and, except the oral polio vaccine, all live-attenuated viral vaccines are freeze-dried. These lyophilized dosage forms can often be stored at 2-8°C.

[0008] Freeze-drying offers many advantages over other drying processes: the low temperatures used during this process allow avoiding high destabilizing drying temperatures, the shelflife of the freeze-dried product is significantly enhanced, the dried product can be easily reconstituted and the aseptic processing operation meets the finished product sterility requirements without the stress of a terminal sterilization step. Freeze-drying is performed to increase the stability of the vaccine but the freezing and the dehydration steps themselves can be stressful for the virus (Hansen 2015). Various additives have been identified that can help stabilize the virus during and after lyophilization (see for instance Burke, 1999 and Hansen, 2015) as most live-attenuated viruses are thermo-sensitive, losing approximatively 1.0 log after 1 week at 37°C under freeze-dried form. Typically, a vaccine formulation will contain one or more of the following components: a bulking agent (e.g., a sugar), a stabilizer (e.g., a sugar or a protein) and a buffer.

[0009] But live viral vaccines are a particular formulation challenge because of the macromolecular complexity of intact viruses (Burke, 1999) and there is a need to improve the stability of both existing and new live-attenuated viral vaccines.

[0010] Flaviviruses are among the most labile viruses. They are small, enveloped, positive-strand RNA viruses that are generally transmitted by infected mosquitoes and ticks. Flaviviruses consist of a lipoprotein envelope surrounding a nucleocapsid composed of the capsid protein and a single-stranded, positive-sense RNA. In between a 5’ untranslated (5’ UTR) and a 3’ untranslated regions (3’ UTR), the RNA encodes, from the 5’ end to the 3’ end, three structural proteins, namely a capsid protein (C protein), a premembrane / membrane protein (prM / M protein), an envelope protein (E protein) and non-structural (NS) proteins, namely NS1, NS2A, NS2B, NS3, NS4A, NS4B and NS5 proteins. The Flavivirus genus of the Flaviviridae family includes approximately 70 viruses, many of which, such as yellow fever (YF), dengue (DEN), Japanese encephalitis (JE), and tick -borne encephalitis (TBE) viruses, are major human pathogens. Different approaches have been used in the development of vaccines against flaviviruses. For instance, in the case of yellow fever virus, six countries are producing live-attenuated vaccine compositions from substrains derived from the 17D strain which was attenuated through several passages in mouse and chicken tissues (Barrett, 2017). Another approach to attenuation of flaviviruses for use in vaccination involves the construction of chimeric flaviviruses, which include components of two (or more) different flaviviruses. For example, the ChimeriVax™ technology employs the yellow fever 17D virus capsid and nonstructural proteins (yellow fever backbone) to deliver the envelope proteins (prM and E) of other flaviviruses (see, e.g., WO9837911 or WO03101397). Other chimeric flaviviruses employ a dengue backbone (see, e.g., WO9640933 or W00160847). These chimeric flavivirus technologies have been used to develop vaccine candidates against dengue, Japanese encephalitis (JE), West Nile (WN), and St. Louis encephalitis (SLE) viruses (see, e.g., Lai, 2003).

[0011] SUMMARY

[0012] Stabilizers including sorbitol and / or proteins like gelatin or serum albumin, including recombinantly produced proteins, have been used for flaviviruses (see for instance Burke 1999, Table pages 56-57, Hansen 2015, Table 1, Clenet 2019, Table 2, Wiggan 2011, and Stamaril or CY110b formulations described below and in the experimental section). PVP 10 and trehalose were also added more specifically for formulations of a live-attenuated yellow fever virus under prilling freeze dried micropellets, which require a high dry content of the formulation above at least 20%.

[0013] There is thus a need for flavivirus stabilizers, especially efficient during freeze-drying process, which can be used at an industrial scale in the field of vaccination, and which limits the number of required excipients to those essential for stabilization.

[0014] Unexpectedly, the present disclosure provides new formulations to stabilize live-attenuated flaviviruses, for example during a freeze-drying process, comprising PVP 10 and trehalose, whilst having a reduced dry content not exceeding 20%, and which may be devoid of sorbitol and devoid of protein.

[0015] The present disclosure thus relates to a stabilizer for compositions comprising at least one or more live-attenuated flaviviruses, which comprises at least:

[0016] - a buffer;

[0017] - trehalose;

[0018] - CaCh;

[0019] - a Polyvinylpyrrolidone (PVP) having an average molecular weight from about 2 500 to about 40 000 Da;

[0020] - L-Proline;

[0021] - a poloxamer; and

[0022] - urea, wherein said stabilizer has a percentage of dry content which is above about 5% and below about 20%, for example below 18%, inter alia below 15% or below 13%. In some embodiments, the buffer is a TRIS buffer, the PVP has an average molecular weight of about 10 000 Da, and / or the pol oxamer is P407.

[0023] In other embodiments, the stabilizer is devoid of sorbitol, or is devoid of lysine, or is devoid of albumin, or is devoid of sorbitol and albumin, oris devoid of sorbitol, lysine and albumin. According to an embodiment, the stabilizer of the disclosure does not comprise sorbitol, nor lysine, nor protein.

[0024] In still other embodiments of the stabilizer, CaCL is present at a concentration below about 0.7%(w / v), i.e. below about 7 g / L. In other embodiments, the PVP is present at a concentration below about 2%(w / v), i.e. below about 20 g / L.

[0025] In still other embodiments, the stabilizer comprises L-Proline at a concentration below about 2%(w / v), i.e. below about 20 g / L, and / or comprises urea at a concentration below about 1.5%(w / v), i.e. below about 15 g / L and / or the poloxamer is present at a concentration below about 0.5%(w / v), i.e. below about 5 g / L.

[0026] In some embodiments, the stabilizer of the disclosure comprises:

[0027] TRIS buffer, at pH = 8, for example at a concentration from about 0.8 to about 2.5 g / L; trehalose, at a concentration from about 50 to about 150 g / L;

[0028] CaCh at a concentration from about 0.6 to about 5 g / L; PVP 10 at a concentration from about 5 to about 15 g / L, L-Proline at a concentration from about 5 to about 15 g / L, P407 at a concentration from about 0.02 to about 1 g / L and urea, at a concentration from about 3 to about 10 g / L; and may not comprise sorbitol.

[0029] In another aspect, the disclosure relates to a vaccine composition, which can be either in liquid or dry form. A liquid vaccine composition according to the disclosure comprises at least one or more live-attenuated flaviviruses and a stabilizer according to the present disclosure, as defined above.

[0030] A solid vaccine composition, or dry vaccine composition, according to the disclosure is a lyophilizate of a liquid vaccine composition according to the disclosure. Stated another way, the solid vaccine composition is a lyophilizate of a composition comprising at least one or more live-attenuated flaviviruses and a stabilizer according to the present disclosure. The lyophilizate is, in some embodiments, obtained by freeze drying.

[0031] In some embodiments, a liquid or solid vaccine composition according to the disclosure comprises a stabilizer comprising:

[0032] TRIS buffer; trehalose;

[0033] CaCh;

[0034] PVP10;

[0035] L-proline;

[0036] P407; and urea.

[0037] In still another aspect, the disclosure relates to a method for stabilizing at least one or more live-attenuated flaviviruses, comprising combining at least one or more live-attenuated flaviviruses with a stabilizer according to the present disclosure.

[0038] In some embodiment, the method further comprises drying, e.g., freeze-drying, the one or more live-attenuated flaviviruses combined with the stabilizer.

[0039] Another aspect of the disclosure is a method for preparing a vaccine composition comprising at least one or more live-attenuated flaviviruses, which comprises at least: a) culturing the one or more live-attenuated flaviviruses; b) harvesting the one or more cultured live-attenuated flaviviruses; and c) combining the harvested one or more live-attenuated flaviviruses with a stabilizer according to the present disclosure.

[0040] In some embodiment, the method further comprises:

[0041] - filling the vaccine composition and drying the filled vaccine composition, e;g., by freeze-drying; or

[0042] - drying the vaccine composition, e;g., by freeze-drying, and filling the dry vaccine composition. In still another aspect, the disclosure is also directed to a dry or solid vaccine composition obtainable by this method.

[0043] The present disclosure further relates to a vaccine kit comprising a first container containing a dry vaccine composition according to the present disclosure and a second container containing an aqueous solution for reconstituting the vaccine.

[0044] In some embodiments of the stabilizers, vaccine compositions, methods and vaccine kits of the disclosure, the at least one or more live-attenuated flaviviruses comprise live-attenuated yellow fever virus.

[0045] BRIEF DESCRIPTION OF THE FIGURES

[0046] Figures 1A-1B: CCID50 infectious titer in Logic loss after DP (drug product) thermal stresses for different formulations compared to Stamaril and Wiggan formulations. Values after 3 years at 5°C were predicted from kinetic models of each formulation, except for Stamaril (*). FIGURE 1 A: Comparison of Formulation A and Formulation B to Stamaril and Wiggan formulations. FIGURE IB: Comparison of formulations Phase I and Phase III to Stamaril and Wiggan formulations.

[0047] Figure 2: Cake appearance by microscopy of 6 formulations, namely FO-S20#1 (top, left); #2 (top, middle); #3 (top, right); #8 (bottom, left); #9 (bottom, middle) and #14 (bottom, right).

[0048] Figure 3: Composition of different formulations and cake appearance.

[0049] Figure 4: Statistical analyses of ML (machine-learning) model for infectious titers at t-zero, after 14 days at 37°C and after 90 days at 25°C, quantitatively estimating impact of concentration of each excipient. SHAP stands for SHapley Additive exPlanations.

[0050] Figure 5: Overlay of experimental data (open symbols - infectious titer in Log loss) at 5°C, 25°C and 37°C for DP formulated from pH 7.5 (triangles), pH 8.0 (circles) and or pH 8.5 (squares) and kinetic model (lines) developed using several batches built with selected BUI formulation at pH 8.0. Dashed lines represent 95% prediction intervals of stability predictions.

[0051] Figure 6: A. Long-term stability (infectious titer in Log loss) for a vYF demo batch in BUI formulations predicted by kinetic models (lines) at 5°C , 25°C and 37°C. Experimental data used for kinetic modeling are displayed as filled symbols. At 5°C, titer loss prediction is shown with predictive band (95% PB as dashed lines). B. Stability after reconstitution (infectious titer in Log loss) for vYF batches in BUI formulations, predicted by kinetic models (lines) at 5°C, 22°C and 25°C and experimental data (two technical batches, as squares and triangles; and one demo batch as circles).

[0052] Figure 7: Stability predictions of vYF infectious titers (CCID50) based on best kinetic models developed for phase III demo batch (left, panel A) and phase I clinical batch (right, panel B). 4 years stability predictions at 5°C including a 4 days excursion at 40°C after 2-years were illustrated, including predictive bands representing 95% PI (dashed lines).

[0053] Figure 8: Sero-neutralizing antibody titers measured by pPRNT50 in sera collected at D21 from hamsters inoculated with YF-VAX or with vYF in phase I, phase III or BU2 formulations

[0054] Statistical method: one factor ANOVA from mixed model with product as fixed effect. P- value < 0.05 = * ; P-value < 0.01 = **).

[0055] Figure 9: A. Long-term stability (infectious titer in Log loss) for three vYF DP production batches in phase III formulation predicted by kinetic models (lines) at 5°C, 25°C and 37°C. Experimental data used for kinetic modeling are displayed as filled symbols (as squares, triangles and circles). Titer loss prediction is shown with predictive band (95% PB as dashed lines). B. Long-term stability (infectious titer in Log loss) for three vYF DP production batches in phase III formulation predicted by kinetic models (lines) at 5°C with 3 days of excursion at 40°C. Experimental data used for kinetic modeling are displayed as filled symbols (as squares, triangles and circles). Titer loss prediction is shown with predictive band (95% PB as dashed lines).

[0056] DETAILED DESCRIPTION

[0057] The present disclosure provides formulations that stabilize a composition comprising live- attenuated flaviviruses, and particularly the very fragile live-attenuated yellow fever viruses. Formulations that stabilize even very labile live-attenuated flaviviruses are provided in the present disclosure.

[0058] In the following sections, various exemplary compositions and methods are described in order to detail various embodiments. It will be obvious to one skilled in the art that practicing the various embodiments does not require the employment of all or even some of the specific details outlined herein, but rather that concentrations, times and other specific details may be modified through routine experimentation. In some cases, well-known methods or components have not been included in the description.

[0059] Miscellaneous definitions

[0060] As used herein, "a" or "an" can mean one or more than one of an item, component or element. As used herein, "about" can mean up to and including plus or minus five percent, for example, about 100 mM can mean 95 mM and up to 105 mM.

[0061] It is understood that aspects and embodiments of the present disclosure described herein include “having,” “comprising,” “consisting of,” and “consisting essentially of’ aspects and embodiments. The words “have” and “comprise,” or variations such as “has,” “having,” “comprises,” or “comprising,” will be understood to imply the inclusion of the stated element(s) (such as a composition of matter or a method step) but not the exclusion of any other elements. The term “consisting of’ implies the inclusion of the stated element(s), to the exclusion of any additional elements. The term “consisting essentially of’ implies the inclusion of the stated elements, and possibly other element(s) where the other element(s) do not materially affect the basic and novel characteristic(s) of the disclosure. It is understood that the different embodiments of the disclosure using the term “comprising” or equivalent cover the embodiments where this term is replaced with “consisting of’ or “consisting essentially of’.

[0062] The concentration expressed in percentage of weight per volume (% w / v) corresponds to the amount of solute, in gram, in 100 mL of solution (g in 100 mL).

[0063] Within the scope of the present disclosure, the term “live” is used in its conventional meaning, a live virus being a virus which has not been inactivated, i.e. a virus capable of replicating on permissive cells.

[0064] The expression “live-attenuated flavivirus”, as used herein, has the common meaning known by a person skilled in the art, i.e. a live flavivirus which induces reduced or no clinical signs in animals or humans compared to the corresponding live wild-type flavivirus. A “stabilizer” of live-attenuated flaviviruses typically reduces inactivation of the said flaviviruses. The stabilizing effect of a formulation may typically be shown by the reduction of the loss of potency and / or infectious titer of the live-attenuated flaviviruses during stresses (for instance shift in pH, changes in temperature, surface interactions, extraneous impurities, etc.) in presence of the formulation compared to the loss of potency of the live-attenuated flaviviruses in absence of the formulation or in presence of another formulation. In vitro or in vivo assays to monitor potency loss of live-attenuated flaviviruses are known.

[0065] For instance, the potency or infectious titer of the live-attenuated flavivirus may be measured as CCID50. As used herein, “CCID50” refers to cell culture infectious dose 50%, i.e. the amount of a virus sufficient to cause a cytopathic effect in 50% of inoculated replicate cell cultures, as determined in an end-point dilution assay in monolayer cell cultures.

[0066] The potency or infectious titer of the live-attenuated virus may also be measured as plaqueforming unit. A “plaque-forming unit” (PFU) as used herein, refers to the smallest quantity of a virus suspension that will produce a plaque in monolayer cell cultures.

[0067] As used herein, “YFV” relates to a yellow fever virus”, whereas the term “vYF” denotes a Vero cells-adapted yellow fever virus, i.e. a yellow fever virus adapted to grow on Vero cells. “Vero cells-adapted yellow fever virus” and “yellow fever virus adapted to grow on Vero cells” are intended to be interchangeable expressions.

[0068] Within the scope of the present disclosure, a virus adapted to grow on Vero cells is a virus which has undergone at least 3 successive passages on Vero cells. By “passage”, one may understand any step in which the virus undergoes at least one replication cycle in Vero cells, for example any step of transfection, amplification or cloning of the virus in Vero cells.

[0069] As used herein, the terms "virus chimera,", "chimeric virus," "flavivirus chimera" and "chimeric flavivirus" can mean a chimera having at least 2 different viruses represented in a construct. For example a construct of a flavivirus chimera has 2 different flaviviruses represented by including non-structural and structural elements from each of flavivirus. Examples of flavivirus can include, but are not limited to, dengue virus, West Nile virus, Japanese encephalitis virus, St. Louis encephalitis virus, tick-borne encephalitis virus, yellow fever virus, Zika virus and any combination thereof. As such, examples of contemplated flavivirus chimeras can include, but are not limited to, a dengue-dengue, dengue-Zika, or a yellow fever-dengue chimera.

[0070] As used herein, the term "dengue-dengue chimera" can mean at least two different dengue virus serotypes make up the dengue-dengue chimera.

[0071] In the context of the present disclosure, "a dry vaccine composition" or “a vaccine composition which is in dry form” or “in solid form” is intended to mean a composition of which the residual water content is less than or equal to 3%, and which is ready to be reconstituted with an aqueous solution in order to be used as a vaccine or directly in dry particulate form.

[0072] In the context of the present disclosure, the “dry content” of a liquid solution or composition is the amounts of solids which are left after drying the liquid solution or composition; it is given as a percentage of the original weight of the liquid solution or composition. It is calculated by adding the weight of the solids (in g / L) present in the composition and dividing the total by 10.

[0073] Stabilizer for vaccine compositions

[0074] In one aspect, the disclosure relates to a stabilizer for compositions comprising one or more live-attenuated flaviviruses, which comprises:

[0075] - a buffer;

[0076] - trehalose;

[0077] - CaCh;

[0078] - PVP;

[0079] - L-Proline;

[0080] - a poloxamer; and

[0081] - urea, wherein the stabilizer has a percentage of dry content which does not exceed about 20%. The dry content of the stabilizer may be advantageously above about 5%.

[0082] In some embodiments, the dry content of the stabilizer is comprised from about 5% to about 20%, or from about 5% to about 18%. In some specific embodiments, the dry content of the stabilizer is above about 5% and below about 15% or above about 5% and below 13%. In particular, the PVP has an average molecular weight from about 2 500 to about 40 000 Da, for example about 10 OOODa.

[0083] According to some embodiments, the stabilizer is devoid of sorbitol. The stabilizer can thus stabilize at least one or more live-attenuated flaviviruses, for example, live-attenuated yellow fever virus, without sorbitol.

[0084] According to other embodiments, the stabilizer of the disclosure is devoid of Lysine, either L-Lysine or D-Lysine. The stabilizer can thus stabilize at least one or more live -attenuated flaviviruses, for example, live-attenuated yellow fever virus, without lysine.

[0085] According to still another embodiment, the stabilizer of the disclosure is devoid of protein, either naturally occurring proteins, or recombinant proteins, including albumin or gelatin. The stabilizer can thus stabilize at least one or more live-attenuated flaviviruses, for example, live-attenuated yellow fever virus, in absence of proteins, for example, in absence of albumin.

[0086] The stabilizer, according to some embodiments of the disclosure, does not comprise albumin. Accordingly, the stabilizer does not comprise lactalbumin, human serum albumin, recombinant human serum albumin (rHA), bovine serum albumin, mammalian serum albumin, recombinant bovine serum albumin, or other serum albumins or albumin gene family members, or modified albumin. Modifications can be made to albumin by any suitable means, for example, by recombinant DNA technology, by post-translational modification, by proteolytic cleavage and / or by chemical means. Those substitutions and alterations to albumin that provide essentially equivalent stabilizing function to serum albumin without substitutions and alterations, are also proteins which are absent in the stabilizer as defined according to some embodiments of the disclosure.

[0087] The last three exclusions, as described above, can be combined. Namely, a stabilizer can be devoid of sorbitol and Lysine, for example, devoid of sorbitol and L-Lysine, or can be devoid of sorbitol and protein, for example, devoid of sorbitol and albumin, or can be devoid of lysine and protein, for example, albumin.

[0088] In various embodiments, the stabilizer of the disclosure does not comprise sorbitol, nor lysine, nor protein.

[0089] In certain embodiments, the stabilizer is devoid of sorbitol, devoid of Lysine, either L-Lysine or D-Lysine, and devoid of protein, for example, devoid of albumin. According to an embodiment, the stabilizer does not comprise sorbitol, nor L-lysine, nor albumin, nor any other protein.

[0090] According to still another embodiment, the stabilizer as described does not comprise sucrose or saccharose. According to this embodiment without sucrose, the stabilizer may also not comprise sorbitol; or may not comprise lysine, or may not comprise albumin or any other protein. In certain embodiments, the stabilizer may not comprise sucrose, nor sorbitol, nor L-lysine, nor albumin, nor any other protein.

[0091] In some embodiments, the stabilizer does not comprise more than 1 or 2 additional excipients (water excluded), in addition to the buffer solution, trehalose, CaCh, PVP, L-Proline, poloxamer and urea, and none of these potential additional excipients is sorbitol, protein or lysine.

[0092] The stabilizer as defined above can be useful in stabilizing live viruses, for example, live- attenuated viruses (e.g. live-attenuated flaviviruses). The disclosure is thus also directed to a vaccine composition, comprising the stabilizer as defined above and at least one or more live-attenuated flaviviruses.

[0093] The stabilizers, or the vaccine compositions comprising one or more live-attenuated flaviviruses, disclosed herein, and comprising: trehalose; CaCh; proline; a PVP having an average molecular weight from about 2 500 to about 40 000 Da; a poloxamer and urea, may further comprise one or more buffer solutions or buffers or buffering media, which are used interchangeably in stabilizer or vaccine composition.

[0094] In various embodiments, stabilizers or vaccine compositions can be described that typically include a physiologically acceptable buffer. A variety of suitable physiologically acceptable buffers are available, including, but not limited to, buffers containing phosphate, TRIS, MOPS, HEPES, bicarbonate, other suitable buffers known and combinations of buffers. In addition, adjusting salt concentrations to near physiological levels (e.g., saline or 0.15 M total salt) may be optimal for parenteral administration of compositions to prevent cellular damage and / or pain at the site of injection. Furthermore, as carbohydrate concentrations increase, salt concentrations can be decreased to maintain equivalent osmolarity to the formulation.

[0095] In some embodiments, a buffering media with pH greater than 6.0 to about pH 10.0 is contemplated. In particular, a buffering media with pH greater than 6.8 is contemplated; some live-attenuated flaviviruses are unstable at low pH. In particular, a buffering media with a pH from about 6.8 to about 10.0 is contemplated. In particular, a buffering media with a pH from about 7.0 to about 9.0 is contemplated. More particularly, a buffering media with a pH from about 7.6 to about 8.7 is contemplated. In particular, a buffering media with a pH from about 8.1 to about 8.7 is contemplated. In particular, a buffering media with a pH of about 8.0 is contemplated.

[0096] The stabilizers, or the vaccine compositions comprising the stabilizer and one or more live- attenuated flaviviruses, disclosed herein, may further comprise one or more buffers or buffering media with a pH from about 7.0 to about 9.0 or from about 7.3 to about 8.7 or from about 7.5 to about 8.5 or about 8.0.

[0097] The buffer can include TRIS (tris(hydroxymethyl)amino- methane), HEPES (2-(4-(2- hydroxyethyl)-l-piperazinyl) ethanesulfonic acid), monosodium glutamate (MSG), Tricine, or Mcllvaine (citrate / phosphate) buffer, with or without salt, or any other suitable buffer. In particular, the buffer is selected from TRIS, HEPES and MSG buffer.

[0098] More particularly, the buffer or buffer solution is TRIS.

[0099] In various embodiments, the stabilizers, or the vaccine compositions comprising one or more live-attenuated flaviviruses and the stabilizer, disclosed herein, further comprise a TRIS buffer.

[0100] The stabilizers, or the vaccine compositions comprising one or more live-attenuated viruses or flaviviruses, disclosed herein, and comprising: trehalose; CaCh; proline; a PVP having an average molecular weight from about 2 500 to about 40 000 Da; a poloxamer and urea, may further comprise one or more buffers or buffering media selected from TRIS, HEPES and MSG buffer, with a pH from about 7.0 to about 9.0, from about 7.3 to about 8.7, from about 7.5 to about 8.5 or about 8.0.

[0101] In some embodiments, the stabilizers, or the vaccine compositions comprising one or more live-attenuated flaviviruses, disclosed herein, and comprising: trehalose; CaCE; proline; a PVP having an average molecular weight from about 2 500 to about 40 000 Da; a poloxamer and urea, further comprise a TRIS buffer with a pH from about 7.0 to about 9.0, from about 7.3 to about 8.7, from about 7.5 to about 8.5 or about 8.0.

[0102] In particular, the stabilizers, or the vaccine compositions comprising one or more live- attenuated flaviviruses, disclosed herein, further comprise a TRIS buffer with a pH from about 7.5 to about 8.5 or about 8.0. The stabilizers or the vaccine compositions disclosed herein can include buffers having concentrations of about 4.0 to 50.0 mM. In particular, the buffer concentration is about 5.0 to 30.0mM. More particularly, the buffer concentration is about 6.0 to 25.0 mM.

[0103] In particular, the stabilizers, or the vaccine compositions comprising one or more live- attenuated flaviviruses, disclosed herein, may further comprise one or more buffers or buffering media selected from TRIS, HEPES and MSG buffer having concentration of about 6 mM to about 25mM, or about 8.3 mM.

[0104] In particular, the stabilizers, or the vaccine compositions comprising one or more live- attenuated flaviviruses, disclosed herein, further comprise a TRIS buffer having concentration of about 6mM to about 25mM, or about 8.3mM. In various embodiments, the stabilizers, or the vaccine compositions comprising one or more live-attenuated viruses or flaviviruses, disclosed herein, further comprise a TRIS buffer having a concentration of about 0.8 g / L to about 2.5 g / L, or about 1 g / L.

[0105] In some embodiments, the stabilizers, or vaccine compositions comprising one or more live- attenuated flaviviruses and the stabilizer, disclosed herein, may further comprise one or more buffers or buffering media selected from TRIS, HEPES and MSG buffer having a concentration of about 6mM to about 25mM, or about 8.3mM, with a pH from about 7.5 to about 8.5 or about 8.0.

[0106] In particular, the stabilizers, or the vaccine compositions comprising one or more live- attenuated flaviviruses and the stabilizer, disclosed herein according to the disclosure, further comprise a TRIS buffer having concentration of about 6mM to about 25mM, or about 8.3mM, or of about 0.8 g / L to about 2.5 g / L, or about 1 g / L with a pH from about 7.5 to about 8.5 or about 8.0.

[0107] Trehalose (e.g., D-trehalose dihydrate) may be used as a bulking agent. Trehalose can be present at an appropriate concentration for stabilizing live-attenuated flaviviruses, in combination with the other excipients, whilst ensuring a dry matter content below 20%. In some embodiments, the concentration of trehalose is below about 19% (w / v) corresponding to 190 g / L, for example below about 18%. In some embodiments, trehalose is present at concentrations ranging from about 1% to about 19% (w / v), from about 5% to about 18% (w / v), from about 5% to about 15% (w / v) or from about 5% to about 12% (w / v), or trehalose can be present at a concentration of about 10% (w / v) or of about 12% (w / v), corresponding to a concentration of about 100 g / L or about 120 g / L.

[0108] Calcium dichloride (CaCh) (e.g., calcium dichloride dihydrate (CaCL 2H2O)) may be used as complexing agent. It is present at an appropriate concentration for stabilizing live- attenuated flaviviruses, in combination with the other excipients, whilst avoiding potential reactogenicity. In some embodiments, the concentration of calcium dichloride is below about 0.7%, corresponding to about 7 g / L. In some embodiments, calcium dichloride can be present at concentrations ranging from about 0.01% to about 2.0% (w / v), from 0.05 % to about 1.0% (w / v), or from about 0.06% to about 0.5% (w / v), or at about 0.1% (w / v), about 0.13% (w / v) or about 0.15% (w / v), corresponding to concentrations of about 1 g / L, about 1.3 g / L and 1.5 g / L, respectively.

[0109] Urea may be used as stabilizing agent. Urea may be present at an appropriate concentration for stabilizing live-attenuated flaviviruses, in combination with the other excipients. In some embodiments, the concentration of urea is below about 1.5%, corresponding to about 15 g / L. In some embodiments, urea can be present at concentrations ranging from about 0.1% to about 5% (w / v), from about 0.2% to about 2% (w / v), or from about 0.3% to about 1% (w / v), or urea can be present at a concentration of about 0.5% (w / v), corresponding to about 5 g / L.

[0110] Proline is an amino acid that may be used as an osmolyte. In various embodiments, the proline may be L-proline. The proline may be present at an appropriate concentration for stabilizing live-attenuated flaviviruses, in combination with the other excipients. In some embodiments, the concentration of proline is below about 2%, corresponding to about 20 g / L. In some embodiments in accordance with the disclosure, proline concentration can be from about 0.1% to about 8% (w / v), from about 0.3% to about 3% (w / v), or from about 0.5% to about 1.5% (w / v), or about 1% (w / v), corresponding to about 10 g / L.

[0111] In accordance with these embodiments, proline concentration can be from about 0.1% to about 2% (w / v), from about 0.5% to about 1.5% (w / v), or about 1% (w / v).

[0112] Polyvinylpyrrolidone, also known as PVP, polyvidone or povidone, is a water-soluble polymer made from the monomer N-vinylpyrrolidone. PVP chains are of varying lengths and PVP fractions are supplied in molecular weights from 2 500 Daltons (Da) to 3 000 000 Daltons. The molecular weight of the fraction is defined in terms of the average molecular weight. The designation 2.5, 8, etc. represents the mean molecular weight divided by 1 000 Daltons. Thus PVP10 (also known as Povidone K-17) corresponds to a PVP with a mean / average molecular weight of 10 000 Da (with a molecular weight distribution of about 10 000±1000 Da). PVP which are suitable in the compositions of the disclosure have an average molecular weight ranging from 2 500 to 40 000 Da. In some embodiments PVP 2.5 (with an average molecular weight of about 2 500 Da), PVP 8 (with an average molecular weight of about 8 000 Da), PVP 10 (with an average molecular weight of about 10 000 Da), PVP 30 (with an average molecular weight of about 30 000 Da) or PVP 40 (with an average molecular weight of about 40 000 Da) are used in the stabilizers or compositions of the disclosure. In particular, the stabilizers or the compositions comprise PVP 10 (with an average molecular weight of about 10 000 Da).

[0113] In accordance with these embodiments, PVP (for example, PVP 10) may be used as stabilizer and / or glass transition enhancer. PVP may be present at an appropriate concentration for stabilizing live-attenuated flaviviruses, in combination with the other excipients. In some embodiments, the concentration of PVP, for example, PVP10, is below about 2%, corresponding to about 20 g / L. In some embodiment, PVP, inter alia PVP 10, can be present at concentrations ranging from about 0.1% to about 8% (w / v), or from about 0.3% to about 3% (w / v), or from about 0.5% to about 1.5% (w / v), or about 0.7% (w / v), corresponding to a concentration of about 7 g / L.

[0114] In the context of the present disclosure, it has been shown that a higher concentration of PVP 10 may enhance the stabilization of the live-attenuated flaviviruses. However, higher concentrations may not be acceptable for regulatory reasons in some countries. The above- mentioned ranges of values are thus reduced with regard to optimal concentrations, in order to fit different regulatory constraints. It should, however, be kept in mind that higher concentrations of PVP (e.g., PVP10) can be envisaged by a skilled person.

[0115] In some embodiments, poloxamer is Poloxamer 407 (P407), also known as Pluronic F127. P407 is a polyoxyethylene-polyoxypropylene block copolymer that was previously used in a formulation to stabilize live-attenuated dengue serotype 2 PDK-53 virus (Wiggan, 2011). In accordance with these embodiments, the P407 may be used as a surfactant. The poloxamer may be present at an appropriate concentration for stabilizing live-attenuated flaviviruses, in combination with the other excipients. In some embodiments, the concentration of poloxamer (e.g., P407) may be below about 0.5%, corresponding to about 5 g / L. In some embodiments, the poloxamer (e.g., P407) can be present at a concentration ranging from about 0.001% to about 0.5% (w / v), or from about 0.002% to about 0.3% (w / v), or from about 0.002% to about 0.1% (w / v), or about 0.005%, corresponding to a concentration of about 0.05 g / L.

[0116] In particular, the stabilizer can include trehalose having a concentration of about 1% to 19% (w / v); CaCh having a concentration of about 0.01% to about 2% (w / v); proline having a concentration of about 0.1% to about 8% (w / v), urea having a concentration of about 0.1% to about 5% (w / v), PVP10 having a concentration of about 0.1% to about 8% (w / v) and P407 having a concentration of about 0.001% to about 0.5% (w / v). The dry content of the stabilizer may be above about 5% and less than about 20%, above about 5% and less than about 18%, above about 5% and less than about 15%, above about 5% and less than about 13%, or about 13% of the stabilizer.

[0117] In particular, the stabilizer can include trehalose having a concentration of about 5% to about 18% (w / v); CaCh having a concentration of about 0.05% to about 1.0% (w / v); proline having a concentration of about 0.3% to about 3% (w / v), urea having a concentration of about 0.2% to about 2% (w / v), PVP10 having a concentration of about 0.3% to about 3% (w / v) and P407 having a concentration of about 0.002% to about 0.3% (w / v). The dry content of the stabilizer may be above about 5% and less than about 20%, above about 5% and less than about 18%, above about 5% and less than about 15%, above about 5% and less than about 13%, or about 13% of the stabilizer.

[0118] In particular, the stabilizer can include trehalose having a concentration of about 5% to about 15% (w / v); CaCh having a concentration of about 0.06% to about 0.5% (w / v); proline having a concentration of about 0.5% to about 1.5% (w / v), urea having a concentration of about 0.3% to about 1% (w / v), PVP10 having a concentration of about 0.5% to about 1.5% (w / v) and P407 having a concentration of about 0.002% to about 0. 1% (w / v). The dry content of the stabilizer may be above about 5% and less than about 20%, above about 5% and less than about 18%, above about 5% and less than about 15%, above about 5% and less than about 13%, or about 13% of the stabilizer.

[0119] In particular, the stabilizer can include trehalose having a concentration of about 5% to about 12% (w / v); CaCh having a concentration of about 0.06% to about 0.5% (w / v); proline having a concentration of about 0.5% to about 1.5% (w / v), urea having a concentration of about 0.3% to about 1% (w / v), PVP10 having a concentration of about 0.5% to about 1.5% (w / v) and P407 having a concentration of about 0.002% to about 0. 1% (w / v). The dry content of the stabilizer may be above about 5% and less than about 20%, above about 5% and less than about 18%, above about 5% and less than about 15%, above about 5% and less than about 13%, or about 13% of the stabilizer.

[0120] In particular, the stabilizer can include trehalose having a concentration of about 5% to about 15% (w / v); CaCh having a concentration of about 0.01% to about 2.0% (w / v); proline having a concentration of about 0.1% to about 8% (w / v), urea having a concentration of about 0.1% to about 5% (w / v), PVP10 having a concentration of about 0. 1% to about 8.0% (w / v) andP407 having a concentration of about 0.001% to about 0.5% (w / v). The dry content of the stabilizer can be above about 5% and less than about 20%, above about 5% and less than about 18%, above about 5% and less than about 15%, above about 5% and less than about 13%, or about 13% of the stabilizer.

[0121] In particular, the stabilizer can include trehalose having a concentration of about 1% to about 19% (w / v); CaCh having a concentration of about 0.06% to about 0.5% (w / v); proline having a concentration of about 0.1% to about 8% (w / v), urea having a concentration of about 0.1% to about 5% (w / v), PVP10 having a concentration of about 0.1% to about 8.0% (w / v) and P407 having a concentration of about 0.001% to about 0.5% (w / v). The dry content of the stabilizer may be above about 5% and less than about 20%, above about 5% and less than about 18%, above about 5% and less than about 15%, above about 5% and less than about 13%, or about 13% of the stabilizer.

[0122] In particular, the stabilizer can include trehalose having a concentration of about 1% to about 19% (w / v); CaCh having a concentration of about 0.01% to about 2.0% (w / v); proline having a concentration of about 0.5% to about 1.5% (w / v), urea having a concentration of about 0.1% to about 5% (w / v), PVP10 having a concentration of about 0.1% to about 8.0% (w / v) and P407 having a concentration of about 0.001% to about 0.5% (w / v). The dry content of the stabilizer may be above about 5% and less than about 20%, above about 5% and less than about 18%, above about 5% and less than about 15%, above about 5% and less than about 13%, about 13% of the stabilizer.

[0123] In particular, the stabilizer can include trehalose having a concentration of about 1% to about 19% (w / v); CaCh having a concentration of about 0.01% to about 2.0% (w / v); proline having a concentration of about 0.1% to about 8% (w / v), urea having a concentration of about 0.3% to about 1% (w / v), PVP10 having a concentration of about 0.1% to about 8.0% (w / v) and P407 having a concentration of about 0.001% to about 0.5% (w / v). The dry content of the stabilizer may be above about 5% and less than about 20%, above about 5% and less than about 18%, above about 5% and less than about 15%, above about 5% and less than about 13%, or about 13% of the stabilizer.

[0124] In particular, the stabilizer can include trehalose having a concentration of about 1% to about 19% (w / v); CaCh having a concentration of about 0.01% to about 2.0% (w / v); proline having a concentration of about 0.1% to about 8% (w / v), urea having a concentration of about 0.1% to about 5% (w / v), PVP10 having a concentration of about 0.5% to about 1.5% (w / v) and P407 having a concentration of about 0.001% to about 0.5% (w / v). The dry content of the stabilizer may be above about 5% and less than about 20%, above about 5% and less than about 18%, above about 5% and less than about 15%, above about 5% and less than about 13%, or about 13% of the stabilizer.

[0125] In particular, the stabilizer can include trehalose having a concentration of about 1% to about 19% (w / v); CaCh having a concentration of about 0.01% to about 2.0% (w / v); proline having a concentration of about 0.1% to about 8% (w / v), urea having a concentration of about 0.1% to about 5% (w / v), PVP10 having a concentration of about 0.1% to about 8.0% (w / v) and P407 having a concentration of about 0.002% to about 0.1% (w / v). The dry content of the stabilizer may be above about 5% and less than about 20%, above about 5% and less than about 18%, above about 5% and less than about 15%, above about 5% and less than about 13%, or about 13% of the stabilizer.

[0126] In particular, the stabilizer can include trehalose having a concentration of about 10% (w / v); CaCh having a concentration of about 0.13% (w / v); proline having a concentration of about 1.0% (w / v); urea having a concentration of about 0.5% (w / v); PVP10 having a concentration of about 0.7% (w / v) and P407 having a concentration of about 0.005% (w / v). The stabilizer and vaccine compositions, of the disclosure, may also comprise a buffer or buffering solution. Suitable buffers, suitable concentrations and pH have been detailed above.

[0127] In some embodiments, the buffer comprises or is TRIS, at a concentration from about 0.8 g / L to about 2.5 g / L in the stabilizer, at a pH from about 7.5 to about 8.5, or of about 8.0.

[0128] In the context of the present disclosure, the concentration of CaCh may indifferently refer to the concentration of CaCh per se, or to the concentration of CaCh, 2H2O. Similarly, the concentration of trehalose may indifferently refer to the concentration of trehalose per se, or to the concentration of trehalose 2H2O.

[0129] In some embodiments, the stabilizer for compositions comprising at least one or more live- attenuated flaviviruses, according to the disclosure, further comprises a dextran having an average molecular weight from about 1 000 to about 70 000 Da, for example, a dextran having an average molecular weight about 10 000 Da.

[0130] In accordance with these embodiments, the dextran may be used as stabilizer and / or glass transition enhancer and can be present at concentrations ranging from about 0.05% to about 10% (w / v), from about 0.5% to about 5% (w / v), from about 1% to about 3% (w / v), or about 2% (w / v).

[0131] As provided, in various embodiments, a stabilizer according to the disclosure does however not comprise more than 1 or 2 additional excipients (water excluded), in addition to the buffer solution, trehalose, CaCh, PVP, L-Proline, poloxamer and urea, and none of these potential additional excipients is sorbitol, protein or lysine. In some embodiments, none of these potential additional excipients is sorbitol, protein, lysine or sucrose.

[0132] The stabilizer may, in some embodiments, not comprise sucrose.

[0133] A minimal list of components, ensuring an adequate stabilization of viruses, for example, of live-attenuated flaviviruses, are provided in the present disclosure. Minimizing the list of components may be of great interest, as it can reduce the costs, the risks of contamination and the risks of interaction. A vaccine composition according to the present disclosure comprises the stabilizer according to the disclosure, and one or more viruses (e;g., live-attenuated flaviviruses), and may not comprise further components (e.g., excipients for stabilizing the viruses). The composition may, however, comprise one or more preservers and / or other excipients not aimed at stabilizing a virus, where appropriate.

[0134] In some embodiments, a vaccine composition according to the disclosure does not comprise sorbitol, nor lysine, nor protein; for example, does not comprise sorbitol, L-Lysine and albumin. In some embodiments, a vaccine composition according to the disclosure does not comprise sorbitol, protein, lysine nor sucrose.

[0135] In particular embodiments, stabilizers or vaccine compositions according to the disclosure comprise:

[0136] - trehalose in a concentration from about 5% to about 15% (w / v);

[0137] - CaCh in a concentration from about 0.06% to about 0.5% (w / v);

[0138] - urea in a concentration from about 0.3% to about 1% (w / v);

[0139] - proline in a concentration from about 0.5% to about 1.5% (w / v);

[0140] - a PVP, having an average molecular weight of about 10 000 Da, in a concentration of about 0.5% to 1.5% (w / v); and

[0141] - P407 in a concentration from about 0.002% to about 0. 1% (w / v), with optionally one or more buffers or buffering media (e.g., TRIS buffer).

[0142] In some embodiments, stabilizers contemplated herein are capable of stabilizing live- attenuated viruses (e.g., flaviviruses), which means they are capable of enhancing the stability of, reducing and / or preventing deterioration and / or inactivation of live-attenuated viruses (e.g., flaviviruses) and vaccine compositions compared to known stabilizers. In some embodiments stabilizers contemplated herein are capable of enhancing the stability of, reducing and / or preventing deterioration and / or inactivation of live-attenuated viruses and vaccine compositions compared to, for example, Stamaril formulation, CY110b formulation or Wiggan formulation. Certain stabilizers or compositions disclosed herein provide stability of aqueous viruses (e;g., flaviviruses) for up to 2 hours, up to 3 hours, up to 4 hours and greater than 4 hours at or about 37°C. Certain stabilizers or compositions disclosed herein reduce the aqueous virus infectious titer loss (for instance, in CCID50) to less than 1.0 Logic for up to 2 hours, up to 3 hours, up to 4 hours, up to 6 hours, or up to 12 hours at or about 37°C. Certain stabilizers or compositions disclosed herein provide stability of aqueous viruses for up to 1 day, at or about room temperature (e.g., 25° C), or for up to 12 or 3 days, at or about 22°C. Certain stabilizers or compositions disclosed herein reduce the aqueous virus infectious titer loss (for instance, in CCID50) to less than 1.0 Logic for up to lOh at or about room temperature. Certain stabilizers or compositions disclosed herein reduce the aqueous virus infectious titer loss (for instance, in CCID50) to less than 1.0 Logic for up to 1 day, up to 2 days, up to 3 days, or up to 5 days, at or about 5°C. Aqueous viruses may be bulk (purified or not, concentrated or not) aqueous vaccine composition comprising one or more live-attenuated viruses and / or dry vaccine composition comprising one or more live- attenuated viruses reconstituted with an aqueous solution, i.e., after reconstitution.

[0143] Embodiments contemplated herein provide increased protection of a live-attenuated virus from for example, freezing and / or thawing, and / or elevated temperatures. In certain embodiments, stabilizers or compositions herein can stabilize, reduce and / or prevent deterioration and / or inactivation of live-attenuated viruses (e.g., flaviviruses), during freeze- drying process. In particular, stabilizers or compositions herein can reduce inactivation of live-attenuated viruses to less than 1.0 Logic CCID50 titer loss during freeze-drying. In certain embodiments, stabilizers or compositions herein can stabilize, reduce and / or prevent deterioration and / or inactivation of dehydrated or dry vaccine compositions comprising one or more live-attenuated viruses (e.g., flaviviruses), either at refrigeration temperature (e.g., about 5°C), in room temperature conditions (e.g., about 25° C), or even at about 37°C or 40°C. In particular, stabilizers or compositions herein can reduce inactivation of live- attenuated viruses in dehydrated or dry vaccine compositions to less than 1.0 Logic CCID50 titer loss, during 1 or 3 or 5 years at about 5°C and / or during 3 months at about 25° C and / or during one week or 14 days at about 37°C and / or during 4 days at about 40°C. In some embodiments, stabilizing one or more live-attenuated viruses as contemplated herein can include obtaining a 10%, a 20%, a 30%, a 40% or a 50% or more reduction in potency loss of the live-attenuated viruses (e.g., flaviviruses), from freezing and / or thawing, and / or elevated temperatures when formulated by stabilizers and methods disclosed herein.

[0144] Stabilizers, compositions and methods disclosed herein can facilitate the storage, distribution, delivery and administration of viral vaccines in developed and under developed regions. The stabilizers and vaccine compositions according to the disclosure may be useful for protecting live-attenuated viruses (e.g., flaviviruses), during lyophilization cycle, even during lyophilization cycle of less than about 40 hours, even those of about 25 hours or less. These stabilizers and vaccine compositions are indeed compatible with aggressive lyophilization cycles of about 25 hours or less.

[0145] The stabilizers of the disclosure may be used for compositions or vaccine compositions comprising any one or more live-attenuated viruses (for example, one or more live- attenuated flaviviruses, e.g. at least one live-attenuated flavivirus), potentially in combination with a further virus, either flavivirus or not, or live-attenuated or not.

[0146] In particular, the stabilizers disclosed herein can be used for compositions or vaccine compositions comprising one or more live-attenuated flavivirus (e.g., dengue virus, West Nile virus, yellow fever virus, Japanese encephalitis virus, St. Louis encephalitis virus, tick- borne encephalitis virus, zika virus), potentially in combination with one or more further live-attenuated viruses.

[0147] In particular, the stabilizers disclosed herein can be used for compositions comprising at least one or more enveloped viruses. In particular, the stabilizers disclosed herein can be used for vaccine compositions comprising at least one or more flaviviruses, potentially in combination with one or more further enveloped viruses. For instance, the stabilizers disclosed herein can stabilize, in addition to the one or more flaviviruses, one or more viruses with membrane envelopes (e.g., enveloped viruses) of the Togavirus, Coronavirus, Rhabdovirus, Filovirus, Paramyxovirus, Orthomyxovirus, Bunyavirus, Arenavirus, Retrovirus, Hepadnavirus, Herpesvirus or Poxvirus families.

[0148] In certain embodiments, compositions can contain one or more live-attenuated Flaviviruses (e.g., dengue virus, West Nile virus, yellow fever virus, Japanese encephalitis virus, St. Louis encephalitis virus, tick-borne encephalitis virus, zika virus, any chimera thereof).

[0149] Formulations disclosed herein are capable of stabilizing, for example, any serotype or strain of flavivirus. In certain embodiments, flaviviruses include enveloped viruses, for example, dengue virus (e.g., serotypes 1-4), yellow fever virus, West Nile virus and Zika virus.

[0150] The compositions described herein can comprise one or more live-attenuated viruses including, but not limited to, attenuated yellow fever viruses (such as 17D), attenuated Japanese encephalitis viruses, (such as SA 14-14-2), attenuated dengue viruses (such as DEN-2 / PDK-53 or DEN-4A30) or recombinant chimeric flaviviruses (such as ChimeriVax™ YF -Dengue virus or dengue-dengue chimeras).

[0151] In particular, the stabilizers of the disclosure may be used for compositions comprising any one or more live-attenuated flaviviruses, potentially in combination with another virus, including, but not limited to, Togavirus, Coronavirus, Rhabdovirus, Filovirus, Paramyxovirus, Pneumovirus, Orthomyxovirus, Bunyavirus, Arenavirus, Retrovirus, Hepadnavirus, Picornavirus, Reovirus, Parvovirus, Papovavirus, Adenovirus, Herpes virus, or Poxvirus.

[0152] In particular, the stabilizers may be used for compositions comprising at least one or more live-attenuated viruses which are selected from the group consisting of flavivirus, chimeric flavivirus, dengue virus, and combinations thereof. More particularly, the live-attenuated viruses are directed to flaviviruses or chimeric flaviviruses. In particular, the stabilizers of the disclosure may be used for compositions comprising at least one or more live-attenuated viruses which comprise yellow fever virus. In particular, the stabilizers may be used for compositions comprising at least one or more live-attenuated viruses, wherein the live- attenuated virus is yellow fever virus. More particularly, the stabilizers may be used for compositions comprising at least one or more live-attenuated viruses, wherein the live- attenuated viruses comprise or is a yellow fever virus adapted to grow on Vero cells.

[0153] In various embodiments, stabilizing one or more live-attenuated flaviviruses (including chimeric flaviviruses) (e.g., live-attenuated yellow fever virus), can include reducing potency loss of the one or more live-attenuated viruses or flaviviruses upon freezing, thawing, freeze drying, at refrigeration temperatures, at room temperature, at about 25°C and at about 37°C, as compared to a composition not containing the stabilizers disclosed herein. In particular, stabilizers or compositions herein can reduce inactivation to less than 1.0 Logio CCID50 titer loss of live-attenuated viruses or flaviviruses during freeze-drying and / or in dehydrated or dry vaccine compositions, during 1 or 3 or 5 years at about 5°C and / or during 3 months at about 25° C and / or during one week or 14 days at about 37°C and / or during 4 days at about 40°C. In particular, stabilizers or compositions herein can reduce inactivation to less than 1.0 Logio CCID50 titer loss of aqueous live-attenuated flaviviruses for up to 2 hours, up to 3 hours, up to 4 hours and up to 6 hours at or about 37°C. In particular, stabilizers or compositions herein can reduce inactivation to less than 1.0 Logio CCID50 titer loss of aqueous live-attenuated flaviviruses for up to lOh at or about room temperature. In particular, stabilizers or compositions herein can reduce inactivation to less than 1.0 Logio CCID50 titer loss of aqueous live-attenuated flaviviruses for up to 1 day, up to 2 days, or up to 3 days at or about 5°C.

[0154] Stabilizers according to the present disclosure have moreover a reduced dry content whilst minimizing the number of required excipients for a conventional freeze-dried product; the compositions may also limit foam during process and reduce the viscosity of the formulation, which can thus increase its processability.

[0155] In some embodiments, the stabilizers and compositions ensure a formulation scalable and compatible with an industrial lyophilization cycle of less than about 25 hours.

[0156] Stabilized vaccine compositions and vaccine kits

[0157] Certain embodiments relate to a vaccine composition comprising at least one or more live- attenuated flaviviruses and a stabilizer of the disclosure.

[0158] In particular, the vaccine composition is a stabilized vaccine composition comprising at least one or more live-attenuated flaviviruses and a stabilizer of the disclosure.

[0159] The compositions, for example, vaccine compositions, comprising live-attenuated flaviviruses and the stabilizer according to the disclosure are in some embodiments in liquid form (e.g., aqueous form). The disclosure, however, also concerns dry form of these compositions. Formulations that improve thermal viral stability and prevent freezing, thawing and / or drying inactivation can improve products that are liquid, dry micropellets or freeze-dried (also called lyophilized), prepared by suitable methods.

[0160] In certain embodiments, the vaccine composition according to the disclosure is in liquid form or in dry form.

[0161] In some embodiments, the stabilized vaccine composition is in liquid form or in dry form.

[0162] In particular, the vaccine composition described herein is in dry form and is obtainable by freeze-drying a vaccine composition according to the disclosure in liquid form.

[0163] In particular, the stabilized vaccine composition described herein is in dry form and is obtainable by freeze-drying a stabilized vaccine composition as described in liquid form.

[0164] In particular, the vaccine composition as disclosed is in freeze-dried form. In particular, the vaccine composition as disclosed, when in freeze-dried form, is not in the form of dry micropellets. Stabilizers for dry micropellets form indeed require a high dry content, which is not the case for the stabilizers according to the disclosure.

[0165] In particular, the stabilized vaccine composition according to the disclosure is in freeze-dried form. In some embodiments, the stabilized vaccine composition is in freeze-dried form but not in the form of dry micropellets.

[0166] In particular, the vaccine composition or the stabilized vaccine composition according to the disclosure which is in dry form is in freeze-dried form and not in dry micropellets form.

[0167] Other embodiments of the disclosure concern a vaccine kit comprising a first container containing the dry vaccine composition as described herein and a second container containing an aqueous solution for reconstituting the vaccine. In some embodiments, the vaccine kit comprises a first container containing the stabilized dry vaccine composition as described herein and a second container containing an aqueous solution for reconstituting the vaccine. The dry vaccine composition or the stabilized dry vaccine composition in the first container of the kit may be in freeze-dried form, in particular in freeze-dried form which is not dry micropellets. The second container may be a sodium chloride pre-filled syringe for reconstitution, or a water pre-filled syringe for reconstitution.

[0168] After reconstitution of the dry vaccine composition or stabilized dry vaccine composition, a liquid composition is obtained, which comprises the components of the stabilizer as defined. Depending on the volume of liquid used for reconstitution, the concentrations of the components of the stabilizer as defined may be modified by a factor corresponding to the ratio between the volume before drying and the volume after reconstitution; the respective proportions of the components of the stabilizer in relation to each other remaining however unmodified. In an embodiment, the concentrations of the components of the stabilizer are divided by a factor of around 1.67, corresponding to the lyophilization of 0.3 mL of a liquid vaccine composition as disclosed, and reconstitution in a volume of 0.5 mL, corresponding to the usual volume of a vaccine dose.

[0169] The vaccine composition or kit as described, either in liquid form or in a dry form obtained after lyophilization, comprises the components of the stabilizer as defined, namely such a composition comprises TRIS buffer, trehalose, CaCL, PVP10, L-proline, P407, and urea. In accordance with all these embodiments, the vaccine compositions or the vaccine kits may comprise one or more live-attenuated flaviviruses.

[0170] The vaccine compositions or the vaccine kits may comprise one or more live-attenuated flaviviruses (e.g., dengue virus, West Nile virus, yellow fever virus, Japanese encephalitis virus, St. Louis encephalitis virus, tick-borne encephalitis virus, zika virus), potentially in combination with one or more further live-attenuated viruses.

[0171] In particular, the vaccine compositions or the vaccine kits disclosed herein comprise at least one or more live-attenuated flaviviruses, and also comprise one or more further live- attenuated viruses which are enveloped viruses. The vaccine compositions or the vaccine kits may thus further comprise one or more viruses with membrane envelopes (e.g., enveloped viruses) of the Togavirus, Coronavirus, Rhabdovirus, Filovirus, Paramyxovirus, Orthomyxovirus, Bunyavirus, Arenavirus, Retrovirus, Hepadnavirus, Herpesvirus or Poxvirus families. In certain embodiments vaccine compositions or the vaccine kits contain one or more further enveloped RNA viruses of the Togavirus, Coronavirus, Rhabdovirus, Filovirus, Paramyxovirus, Orthomyxovirus, Bunyavirus, Arenavirus, or Retrovirus families.

[0172] In certain embodiments, compositions or the vaccine kits contain one or more live, attenuated Flaviviruses (e.g., dengue virus, West Nile virus, yellow fever virus, Japanese encephalitis virus, St. Louis encephalitis virus, tick-borne encephalitis virus, zika virus, any chimera thereof).

[0173] The vaccine compositions or the vaccine kits described herein can comprise one or more live-attenuated viruses including, but not limited to, for example any serotype or strain of flavivirus. In certain embodiments, flaviviruses include enveloped viruses, for example, dengue virus (e.g., serotypes 1-4), yellow fever virus, West Nile virus and Zika virus.

[0174] The vaccine compositions or the vaccine kits described herein can comprise one or more live-attenuated viruses including, but not limited to, attenuated yellow fever viruses (such as 17D), attenuated Japanese encephalitis viruses, (such as SA 14-14-2), attenuated dengue viruses (such as DEN-2 / PDK-53 or DEN-4 A30) or recombinant chimeric flaviviruses (such as ChimeriVax™ YF-Dengue virus or dengue-dengue chimeras).

[0175] In particular, the vaccine compositions or the vaccine kits as disclosed comprise at least one or more live-attenuated viruses which are selected from the group consisting of flavivirus, chimeric flavivirus, dengue virus, and combinations thereof. More particularly, the live- attenuated viruses are directed to flaviviruses or chimeric flaviviruses. In particular, the vaccine compositions or the vaccine kits as disclosed comprise at least one or more live- attenuated viruses which comprise yellow fever virus.

[0176] In particular, the vaccine compositions or the vaccine kits of the disclosure comprise one or more live-attenuated flaviviruses, wherein the live-attenuated flavivirus is yellow fever virus. More particularly, vaccine composition or the vaccine kits comprises one or more live- attenuated flaviviruses, wherein the live-attenuated flavivirus comprises or is a yellow fever virus adapted to grow on Vero cells.

[0177] The vaccine composition or kit according to the present disclosure may correspond to a single dose for injection, optionally after reconstitution in case of dry composition, or may correspond to multiple doses.

[0178] Methods for stabilizing live-attenuated viruses

[0179] The disclosure also relates to a method for stabilizing at least one or more live-attenuated flaviviruses, comprising combining at least one or more live-attenuated flaviviruses with a stabilizer as disclosed.

[0180] In some embodiments, the method for stabilizing at least one or more live-attenuated flaviviruses, comprises combining a composition comprising at least one or more live- attenuated flaviviruses, and potentially further live-attenuated viruses or other viruses, with a stabilizer of the disclosure, in order to obtain a stabilized composition of one or more live- attenuated flaviviruses, potentially comprising further live-attenuated viruses or other viruses. The disclosure also relates to a method for stabilizing at least one or more live- attenuated flaviviruses, and potentially further live-attenuated viruses or other viruses, comprising combining a viral harvest comprising at least one or more live-attenuated flaviviruses, and potentially further live-attenuated viruses or other viruses, with a stabilizer as described. This method may comprise combining a viral harvest comprising at least one or more live-attenuated flaviviruses with a stabilizer of the disclosure in order to obtain a stabilized bulk aqueous composition comprising said one or more live-attenuated flaviviruses. This method may comprise combining a purified viral harvest comprising one or more live-attenuated flaviviruses, and potentially further live-attenuated viruses or other viruses, with a stabilizer as disclosed. This method may comprise combining a concentrated viral harvest comprising one or more live-attenuated flaviviruses, and potentially further live-attenuated viruses or other viruses, with a stabilizer of the disclosure. This method may comprise combining a purified and concentrated viral harvest comprising one or more live- attenuated flaviviruses and potentially further live-attenuated viruses or other viruses, with a stabilizer as disclosed. This method may comprise combining a purified and concentrated viral harvest comprising one or more live-attenuated flaviviruses, and potentially further live-attenuated viruses or other viruses, with a stabilizer of the disclosure in order to obtain a stabilized bulk aqueous vaccine composition comprising said one or more live-attenuated flaviviruses, with potentially further live-attenuated viruses or other viruses.

[0181] The stabilizer can be added to the harvest of the viral culture, for example, after purification and / or concentration of the harvest, by any suitable methods.

[0182] Additionally, methods and compositions disclosed herein can include freeze-drying or other dehydrating methods for the composition. In accordance with these methods and compositions, the methods and compositions decrease inactivation of the flavivirus against freeze-drying stress and / or decrease inactivation of the freeze-dried or partially or wholly dehydrated live-attenuated flavivirus. The stabilization method according to the present disclosure may further comprise drying the aqueous composition by a method selected from the group of foam-drying, spray-drying or freeze-foam-drying, freeze-drying, spray -freeze- drying or prilling and then freeze-drying. All these drying methods are well-known to those skilled in the art. In particular, the method does not comprise a step of prilling and then freeze-drying.

[0183] In some embodiments, the methods and compositions comprise a step of freeze-drying the vaccine composition, wherein this step is carried out by a lyophilization cycle of less than about 40 hours, a cycle during about 25 hours or less, or a lyophilization cycle during about 23 hours or less. Exemplary lyophilization cycles are provided in the experimental section. In particular, the method for stabilizing at least one or more live-attenuated flaviviruses, comprising combining at least one or more live-attenuated flaviviruses (e.g., dengue virus, West Nile virus, yellow fever virus, Japanese encephalitis virus, St. Louis encephalitis virus, tick-borne encephalitis virus, zika virus), and potentially further live-attenuated viruses, with a stabilizer as disclosed, further comprises drying the one or more live-attenuated flaviviruses, and potentially further live-attenuated viruses or other viruses, in particular by freeze-drying. In various embodiments, the method for stabilizing at least one or more live- attenuated flaviviruses, comprises combining at least one or more live-attenuated flaviviruses with a stabilizer of the disclosure, in order to obtain a stabilized composition of one or more live-attenuated flaviviruses, and further comprises drying the stabilized one or more live-attenuated flaviviruses, in particular by freeze-drying. The drying step may be carried out by a cycle during 40 hours or less, 25 hours or less, or 23 hours or less.

[0184] The disclosure also relates to a stabilized composition of least one or more live-attenuated flaviviruses obtainable by a method as described herein.

[0185] In some embodiments, the further live-attenuated viruses can be one or more viruses with membrane envelopes (e.g., enveloped viruses) of the Togavirus, Coronavirus, Rhabdovirus, Filovirus, Paramyxovirus, Orthomyxovirus, Bunyavirus, Arenavirus, Retrovirus, Hepadnavirus, Herpesvirus or Poxvirus families.

[0186] In certain embodiments compositions contain one or more enveloped RNA viruses of the Togavirus, Flavivirus, Coronavirus, Rhabdovirus, Filovirus, Paramyxovirus, Orthomyxovirus, Bunyavirus, Arenavirus, or Retrovirus families. In other embodiments, the live-attenuated viruses can comprise one or more enveloped, positive strand RNA virus of the Togavirus, Flavivirus, Coronavirus, or Retrovirus families. In certain embodiments, the live-attenuated viruses can contain one or more live, attenuated Flaviviruses chosen from dengue virus, West Nile virus, yellow fever virus, Japanese encephalitis virus, St. Louis encephalitis virus, tick-borne encephalitis virus, zika virus and any chimera thereof.

[0187] The live-attenuated flaviviruses comprise for example any serotype or strain of flavivirus. In certain embodiments, flaviviruses include enveloped viruses, for example, dengue virus (e.g., serotypes 1-4), yellow fever virus, West Nile virus and Zika virus.

[0188] The one or more live-attenuated flaviviruses may include, but are not limited to, attenuated yellow fever viruses (such as 17D), attenuated Japanese encephalitis viruses, (such as SA 14-14-2), attenuated dengue viruses (such as DEN-2 / PDK-53 or DEN-4 A30) or recombinant chimeric flaviviruses (such as ChimeriVax™ YF-Dengue virus or dengue-dengue chimeras). In certain embodiments, methods for stabilizing at least one or more live-attenuated flaviviruses as disclosed are contemplated, wherein the at least one or more live-attenuated viruses are selected from the group consisting of flavivirus, chimeric flavivirus, dengue virus and combinations thereof. In particular, the live-attenuated flaviviruses are directed to flaviviruses stricto sensu or chimeric flaviviruses. In particular, a method for stabilizing at least one or more live-attenuated flaviviruses according to the disclosure is contemplated, wherein the at least one or more live-attenuated flaviviruses comprise yellow fever virus. In particular, a method for stabilizing at least one or more live-attenuated flaviviruses as disclosed is contemplated, wherein the live-attenuated flavivirus is yellow fever virus.

[0189] More particularly, a method for stabilizing at least one or more live-attenuated flaviviruses according to the disclosure is contemplated, wherein the live-attenuated flaviviruses comprise or is a yellow fever virus adapted to grow on Vero cells.

[0190] Methods for preparing a vaccine composition

[0191] The disclosure also relates to a method for preparing a vaccine composition comprising at least one or more live-attenuated flaviviruses, which comprises at least: a) culturing the one or more live-attenuated flaviviruses; b) harvesting the one or more cultured live-attenuated flaviviruses; and c) combining the harvested one or more live-attenuated flaviviruses with a stabilizer as disclosed.

[0192] The culturing step can be done on any culturing system, for instance on eggs or on cell culture. For example, the culturing step may be done on cell culture, for instance, on mammalian cell culture, for instance, on Vero cells. In particular, the one or more live- attenuated flaviviruses may comprise or be a virus adapted to grow on Vero cells, in particular a yellow fever virus adapted to grow on Vero cells.

[0193] The method for preparing a vaccine composition comprising at least one or more live- attenuated flaviviruses, which comprises at least: a) culturing the one or more live-attenuated flaviviruses; b) harvesting the one or more cultured live-attenuated flaviviruses; and c) combining the harvested one or more live-attenuated flaviviruses with a stabilizer as disclosed; may further comprise a step of drying the vaccine composition, e.g., by freeze-drying. This method may further comprise a step of purifying the harvested one or more live- attenuated flaviviruses. The method may also comprise one or more steps of filtration. The method may also comprise one or more steps of checking the absence of contamination.

[0194] The method further comprising a step of drying the vaccine composition, e.g., by freeze- drying, may further comprise a step of filling the vaccine composition. The filling step may be before or after the drying step.

[0195] In particular, this method further comprises:

[0196] - filling the vaccine composition and drying the filled vaccine composition, e.g., by freeze-drying; or

[0197] - drying the vaccine composition, e.g., by freeze-drying, and filling the dry vaccine composition.

[0198] The filling can be into vials or syringes or any other container suitable for vaccines.

[0199] When the method further comprises filling the vaccine composition and drying the filled vaccine composition, e.g., by freeze-drying, the vaccine composition that is filled is in aqueous form. A cake may be obtained after drying the filled vaccine composition.

[0200] In particular, the method further comprises filling the vaccine composition and drying the filled vaccine composition, in particular by freeze-drying.

[0201] According to some embodiments, the method of preparing a vaccine composition according to the present disclosure does not comprise freezing and drying the aqueous solution in the form of regular spherical particles or micropellets.

[0202] The disclosure also relates to a stabilized dry vaccine composition obtainable or prepared by any of the methods as described herein.

[0203] The method for preparing a vaccine composition of the present disclosure may further comprise the step of reconstituting the dry vaccine composition with an aqueous solution. After reconstitution, such stabilized vaccines can be administered by a variety of routes, including, but not limited to intradermal administration, subcutaneous administration, intramuscular administration, intranasal administration, pulmonary administration or oral administration. A variety of devices are known in the art for delivery of the vaccine including, but not limited to, syringe and needle injection, bifurcated needle administration, administration by intradermal patches or pumps, needle-free jet delivery, intradermal particle delivery, or aerosol powder delivery.

[0204] In the method for preparing a vaccine composition as disclosed, the vaccine composition comprises at least one or more live-attenuated flaviviruses, including, but not limited to dengue virus, West Nile virus, yellow fever virus, Japanese encephalitis virus, St. Louis encephalitis virus, tick-borne encephalitis virus, zika virus; potentially in combination with one or more further live-attenuated viruses.

[0205] In the method for preparing a vaccine composition according to the disclosure, the vaccine composition can contain one or more further viruses with membrane envelopes (e.g., enveloped viruses) of the Togavirus, Coronavirus, Rhabdovirus, Filovirus, Paramyxovirus, Orthomyxovirus, Bunyavirus, Arenavirus, Retrovirus, Hepadnavirus, Herpesvirus or Poxvirus families. In certain embodiments compositions contain one or more further enveloped RNA viruses of the Togavirus, Coronavirus, Rhabdovirus, Filovirus, Paramyxovirus, Orthomyxovirus, Bunyavirus, Arenavirus, or Retrovirus families.

[0206] The vaccine composition comprises for example any serotype or strain of flavivirus. In certain embodiments, flaviviruses include enveloped viruses, for example, dengue virus (e.g., serotypes 1-4), yellow fever virus, West Nile virus and Zika virus.

[0207] The vaccine composition can comprise one or more live-attenuated flaviviruses including, but not limited to, attenuated yellow fever viruses (such as 17D), attenuated Japanese encephalitis viruses, (such as SA 14-14-2), attenuated dengue viruses (such as DEN-2 / PDK- 53 or DEN-4 A30) or recombinant chimeric flaviviruses (such as ChimeriVax™ YF-Dengue virus or dengue-dengue chimeras).

[0208] In certain embodiments, methods for preparing a vaccine composition comprising at least one or more live-attenuated flaviviruses, according to the disclosure are contemplated, wherein the at least one or more live-attenuated flaviviruses are selected from the group consisting of flavivirus, chimeric flavivirus, dengue virus, and combinations thereof. In particular, the live-attenuated flaviviruses are directed to flaviviruses stricto sensu or chimeric flaviviruses. In particular, a method for preparing a vaccine composition comprising at least one or more live-attenuated flaviviruses, as disclosed is contemplated, wherein the at least one or more live-attenuated flaviviruses comprise yellow fever virus. In particular, a method for preparing a vaccine composition comprising at least one or more live-attenuated flaviviruses, as disclosed is contemplated, wherein the live-attenuated flavivirus is yellow fever virus. More particularly, a method for preparing a vaccine composition comprising at least one or more live-attenuated flaviviruses, as disclosed is contemplated, wherein the live-attenuated flaviviruses comprise or is a yellow fever virus adapted to grow on Vero cells.

[0209] EXAMPLES

[0210] Materials and methods

[0211] Excipients

[0212] Sucrose, and carboxymethyl cellulose (CMC) were purchased from Acros (Geel, Belgium), and trehalose from Pfanstiehl (Zug, Switzerland). Potassium monophosphate anhydrous, urea, maltose, TRIS, L-proline, dextran 10 and Ficoll 400 from Merck (Darmstadt, Germany), raffinose from Alfa Aesar (Lancashire, United Kingdom), Povidone K-17 (PVP10) was purchased from Ashland (Covington, U.S.A.). L-Lysine, monohydrate was from Evonik (Essen, Germany). Chloride sodium dihydrate and dichloride calcium were obtained from VWR (Leuven, Belgium). Sorbitol was obtained from Roquette Freres (Lestrem, France). L-Proline was purchased from Fragon (Rotterdam, The Netherlands). Poloxamer 407 sample was received from BASF (Ludwigshafen, Germany). Recombinant HSA (rHA) was obtained from Novozymes (Franklinton, NC, U.S.A.). Ready to use mixtures of amino acids (non-essential amino acids, essential amino acids) were from ThermoFisher Scientific at 100C concentration.

[0213] All excipients stock solutions were solubilized in TRIS buffers and the pH was adjusted to 8.0.

[0214] Virus titration - Infectious titers xCELLigence® analyses

[0215] The xCELLigence® Real Time Cell Analysis Instrument (ACEA Biosciences Inc., San Diego, California, USA) is a real-time cellular biosensor, which measures the net adhesion of cells to high-density gold electrode arrays printed on custom-designed 96-well E-plates. Cells adhere to the plate surface and influence the electrical impedance across the array, which is measured and recorded by the xCELLigence® software. The impedance values are converted by the software into the Cell Index (CI). In the absence of cells, the Cell Index will be zero, and as cells adhere to the array, the Cell Index increases. The strength of cellular adhesion is influenced by cell viability and real-time cell CI analysis can be used for determining viral infectious titers (Charretier, 2018).

[0216] The viral suspensions were diluted in MEM GlutaMAX (GIBCO Ref: 42360-024) + 1% FCS at 1 / 4 on a serial basis (around 8 dilutions) and each titration was composed of at least 2 independent serial basis dilutions. One hundred microliters of each virus dilution were added into wells containing Vero cells seeded in 96-well E-plates three days before the assay (20000 cells / well). A YF-Vax® reference (WSL YF-Vax® FDV02509) with known CCID50 infectious titer is used to create a standard curve.

[0217] After a 7 days incubation period at +36±1°C, in a 5±1% CO2°C atmosphere, Cell Index measures are recorded every 15 minutes during 192h. The software provides an electronic record of the experimental details and grouping of wells for calculating averages (±standard deviation). The data are then exported and analyzed in Excel, in order to calculate the titer of each sample via the equation of the standard curve taking into account the dilution factor.

[0218] CCID50 determinations

[0219] The virus infectious titration was determined by a CCID50 assay. The virus, for instance, the YFV, was titrated in 96-well microtiter plates using Vero cells infected with different virus dilutions. The viral suspensions were diluted in IMDM (THERMOFISHER SCIENTIFIC) + 4% FCS at 1 / 4 on a serial basis (around 8 dilutions) and each titration was composed of at least 2 independent serial basis dilutions. Samples with high virus content were pre-diluted on a serial basis of 1 / 10 to obtain the first dilution which will be tested on cells. One hundred microliters of each virus dilution were added into 10 wells containing Vero cells seeded in flat bottom 96-well plates three days before the assay (8000 cells / well). After a 7 to 10 days incubation period at +36±1°C, in a 5±1% CO2 atmosphere, the number of wells presenting a cytopathic effect was determined by microscopic observation. Positive wells, i.e., wells containing at least one plaque stained in black were counted and the final titer was calculated using the Least Square regression method. The titer was expressed in 50% cell culture infective doses per dose (CCIDso / dose) or per milliliter (CCIDso / mL). Based on experimental reproducibility results, infectious titers were obtained at ± 0.2 Logic CCID50.

[0220] Freeze-drying process

[0221] Freeze-drying was performed on a GEA Lyovac™ FCM10 freeze-dryer. Formulations were randomly distributed on shelves. Vials were equilibrated at 5°C and frozen at -50°C in 30 min. Samples were kept at -50°C for 2 hours before applying vacuum (0.045 mbar). For primary drying, shelves reached -45°C in 1500 min and held for 3h. Secondary drying was performed at 25°C, reaching this temperature in 3500 min. under 0.015 mbar. Temperature and pressure were held for 1440 min. The vials were stoppered under nitrogen at 0.8 mbar.

[0222] Process to prepare micropellets

[0223] The prilling process and freeze-drying process for obtaining micropellets are as disclosed in the publication Clenet et al, 2019. The particle size distribution of micropellets was measured by laser light diffraction as disclosed in Clenet et al, 2019.

[0224] Glass transitions by DSC

[0225] Thermal analysis of solutions and freeze-dried powders was performed using a power compensation differential scanning calorimeter (DSC) equipped with an Intracooler II. Approximatively 10 pL of solution (bulk) or 2 mg of dried powder were used. The sample was sealed in an aluminum pan (with hole for powders) and an empty pan was used as the reference. Cooling and heating rates of 5°C / min were used. All glass transition temperatures were reported as the midpoint temperature of the heat capacity step associated to the glass transition. Liquid samples were cooled to -60°C to ensure temperature stability and sample equilibration, and scanned for the first time to 25°C. Dried samples were cooled to 20°C and then heated to 135°C. The second heating scan was used to determine glass transitions. Glass temperatures were given at ± 2°C, based on experimental reproducibility results.

[0226] Stability monitoring and stress programs

[0227] Freeze-dried products were incubated in a cold chamber at 5°C for up to 1 year and in incubators at 25°C and 37°C for up to 6 months. The cold chamber was maintained at ± 3°C and incubators at ± 1°C. To assess Controlled Temperature Chain (CTC) labeling, an experimental temperature excursion was performed on freeze-dried products (maintained 4 days at 40°C). During these forced degradation studies, vYF was sampled after various times and analyzed by CCID50 as expressed by Log loss. Solid forms were reconstituted just before analyses with 0.5mL of 0.4% NaCl solution. One year CCID50 data were used for stability modeling. AKTS-Thermokinetics, TIBCO-Spotfire® and Excel software were used to build graphical representations.

[0228] Freeze-thaw cycles on aqueous bulk vaccine composition were automatically performed under controlled from -60°C to +5°C by using the Mecour block piloted by the CM3- Freeslate robot.

[0229] Kinetic-based modeling and stability predictions

[0230] AKTS-Thermokinetics software (version 5.02, AKTS AG, Advanced Kinetics and Technology Solutions, Siders, Switzerland) was used to screen and compare kinetic models of loss of infectious titer as a function of time and temperature and predict the loss of solid form infectious titer. CCID50 data obtained from freeze-dried products samples after months of forced degradation were used to inform the modeling method. The general modeling procedure has been previously described in detail (Clenet, 2018). Briefly, a large variety of models based on the truncated Sestak-Berggren equation were analyzed and evaluated according to a least-square regression analysis. The best model was identified according to the higher AIC (Akaike information criterion) and BIC (Baysian information criterion) weighted scores and the lower sum of residual squares (RSS) value. Finaly, 95% percentile confidence intervals were calculated for predictions using bootstrap analysis (resampled 1000 times with replacement). The selected kinetic model was used to predict the course of vYF infectious titer loss during storage under isothermal conditions up to 3 years (long term stability) and during a temperature excursion mimicking CTC requirement (4 days at 40°C). The AKTS software was used to build long-term stability representations, including predictive bands (CI95%).

[0231] EXAMPLE 1 — Stability of a live-attenuated yellow fever virus in known formulations

[0232] 1.1 / Live- attenuated vYF vYF (also called YFV TV3112 strain) was used as an exemplary yellow fever virus. Other yellow fever viruses can be used alternatively. vYF is a live-attenuated YFV that was obtained after 8 passages on Vero cells of the YFV 17D-204 strain used in the commercialized vaccine YF-VAX®. The RNA sequence of the YFV 17D-204 strain is available in Genbank (accession number X03700). The RNA sequence of the YFV 17D-204 derived YF-VAX® strain may be represented by the RNA sequence SEQ ID NO. 1. The RNA sequence of vYF is represented by SEQ ID NO. 2. 1.2 / Known stabilizers for live-attenuated flaviviruses

[0233] 1.2.1 / Stamaril formulation

[0234] Stamaril formulation is the stabilizer used in the live-attenuated yellow fever vaccine commercialized under the name of Stamaril™.

[0235] Table 1: Stamaril formulation

[0236] 1.2.2 / CYllOb formulation

[0237] CYl lOb formulation is the stabilizer used in the live-attenuated chimeric yellow fever- dengue vaccine commercialized under the name of Dengvaxia®.

[0238] Table 2: CYl lOb formulation

[0239] 1.2.3 / Wiggan formulation

[0240] Another flavivirus stabilizer (Wiggan stabilizer) containing P407, rHA and trehalose in phosphate-buffered saline is described in Wiggan, 2011. Wiggan stabilizer detailed composition is KH2PO4 (0.44 mg / mL), ISfeHPCU, 2H2O (2.1 mg / mL), Trehalose, 2H2O (150 mg / mL), P407 (20 mg / mL) and rHA (10 mg / mL).

[0241] 1.2.4 / stabilizer specifically designed for micropellet form

[0242] 1.3 / Stability results in known stabilizers for freeze-drying

[0243] Final Bulk Product (FBP), i.e. bulk aqueous vaccine composition before freeze-drying, comprising vYF live-attenuated virus is formulated with each one of the 3 known stabilizers for freeze-drying. Virus infectious titer of 5.0 Logic CCIDso / dose was targeted. Freeze-dried products were then obtained using a conservative lyophilization cycle (i.e., a cycle of about 63 hours) and reconstituted with NaCl 0.4% solution just before analysis. Infectious titers were determined by CCID50.

[0244] High infectious titer losses were observed after freeze-drying process when vYF was formulated in known flavivirus stabilizers: more than 2.5 Logic CCIDso / dose with CY110b stabilizer, 1.51 Logic CCIDso / dose with Stamaril stabilizer and 1.16 Logic CCIDso / dose with Wiggan stabilizer.

[0245] To assess Controlled Temperature Chain (CTC) labeling, an experimental temperature excursion was performed (freeze-dried product maintained 4 days at 40°C). The cakes collapsed after 4 days at 40°C with CYl lOb and Stamaril stabilizers, while an acceptable cake appearance was obtained with Wiggan stabilizer. vYF can be considered as a very unstable flavivirus and a formulation that would increase its stability would also increase the stability of other flaviviruses.

[0246] EXAMPLE 2 — Initial screenins of excipients

[0247] Most live attenuated viruses are thermo-sensitive, losing approximatively 1.0 log after 1 week at 37°C under freeze-dried form. They often require complex formulations comprising numerous excipients such as salts, polysaccharides, amino acids, cryoprotectants, tonicity modifiers and surfactants.

[0248] Several usual excipients, with different functionalities were thus individually evaluated at least at a low and a high concentration in lOmM TRIS buffer, pH 8.0 as an exemplary buffer. 72 vYF FBP formulations targeted at 7.0 LoglO CCID50 / mL were prepared. LoglO CCID50 gap from target of vYF in presence of individual excipients at different concentrations (C1<C2<C3) was measured by xCELLigence®. A graphical visualization of results was used to quickly identify the impact of excipients on vYF infectious titer. This graphical view merged all infectious titers data in a single scatter plot and helped to instantly highlight the excipients capable of protecting vYF. A horizontal line represents the average value and dotted lines 1 standard deviation (s.d.). Excipients of interest were easily identified above the average. Best excipients were easily identified at 1 s.d. above the average and higher.

[0249] The results showed that, from the amino acids tested, L-Proline is of interest as it demonstrated a positive effect on vYF infectious titer at low and high concentrations. L- Arginine is also of interest but only at high concentration. L-Cysteine and L-lysine should be good candidates at low concentration. L-alanine also showed a stabilizing effect at high concentration. Interestingly, classical mixture of essential (EAA or AAE) and non-essential amino acids (NEAA or AANE) induced a dramatic loss of infectious titer.

[0250] For the proteins, rHA clearly showed a positive effect and was selected for further investigations.

[0251] Salts did not demonstrate any stabilizing effect, except MSG but at high concentration.

[0252] From the polymers, dextran 10 and PVP10 showed a protective effect.

[0253] Most of the sugars showed a stabilizing effect. The best stabilization is given by sucrose, lactose and raffinose. Maltose at low concentration (20 mM), sorbitol at high concentration (820 mM) and trehalose could also be selected for further investigations.

[0254] In addition, turbidity evaluated by UV spectroscopy at 350 nm indicated a negative impact of CMC, Ficoll and lactose. With this physical parameter, a positive effect could be reached in presence of salts (KC1, Na2SO4, NaCl, CaCL and KC1), trehalose and sorbitol at high concentration (data not shown).

[0255] EXAMPLE 3 — Designs of experiments (DoE) by combinins promising excipients

[0256] Aiming to maximize thermal stability and protection of vYF against freeze-drying process, formulations mixing best excipients were evaluated by using designs of experiments. This screening of formulations was built in a 20 mM TRIS-based formulation, pH 8.0 as an exemplary buffer. The formulations were automatically built at 9 mL scale in 3x4 vial racks with the CM3-Freeslate robot. JMP® software was used to build and analyze designs of experiments (DoE). LEA database and Spotfire were used to facilitate data management and exploration. CYl lOb stabilizer, Stamaril stabilizer and Wiggan stabilizer were used as controls. Freeze-dried products were obtained following a conservative lyophilization cycle and reconstituted with water just before analysis. Infectious titers were determined for the dry vaccine composition just after freeze-drying and for the dry vaccine stored after 1 week at 37°C. They were targeted at 6 LoglO CCID50 / mL and data were expressed as Log loss due to lyophilization and Log loss due to 1 week at 37°C, by xCELLigence®.

[0257] The majority of the formulations induced an important loss of vYF infectivity, down to below limit of quantification, indicating a high fragility of vYF (data not shown).

[0258] Infectious titers were confirmed by CCID50 on 15 DP formulations, leading to the identification of 2 compositions clearly loosing less than 1 LoglO CCID50 after freeze- drying and 1 week incubation at 37°C .

[0259] The results showed that the known flavivirus stabilizers CYl lOb, Stamaril or Wiggan stabilizers, do not stabilize the vYF virus, giving a CCID50 titer loss of more than 1.6 LoglO. The Log loss on the dry vaccine after 1 week at 37°C was not measured for the formulation with Stamaril stabilizer, as the CCID50 titer loss just after reconstitution was already more than 2 LoglO.

[0260] Two formulations demonstrated a strong stabilizing effect on vYF virus with a total (after freeze-drying + 1 week storage of dry vaccine at 37°C) CCID50 titer loss below 1 LoglO:

[0261] - Formulation A: TRIS buffer (20mM), D-Trehalose, 2H2O (150 mg / mL), CaC12, 2H2O (1.325 mg / mL which can be rounded up to 1.3 mg / mL), rHA (0.8 mg / mL), Dextran 10 (20 mg / mL), D-Sorbitol (30 mg / mL), Urea (2.5 mg / mL)

[0262] - Formulation B: TRIS buffer (20mM), D-Trehalose, 2H2O (150 mg / mL), CaC12, 2H2O (1.325 mg / mL which can be rounded up to 1.3 mg / mL), L-Proline (3.2 mg / mL), L-Lysine base (0.3 mg / mL), PVP10 (20 mg / mL), Sorbitol (30 mg / mL), Urea (2.5 mg / mL), P407 (5 mg / mL).

[0263] These formulations were developed to be compatible with micro-pellet product formats.

[0264] EXAMPLE 4 — Stability under forced degradation conditions and long term stability predictions for Formulation A and Formulation B freeze-dried vYF DP, with a target CCID50 titer of 5 Logio / dose, formulated with Formulation A, Formulation B, Stamaril formulation or Wiggan formulation, freeze-dried, were incubated in a cold chamber at 5°C for up to 1 year and in incubators at 25°C and 37°C for up to 6 months. To assess Controlled Temperature Chain (CTC) labeling, an experimental temperature excursion was performed on DP (maintained 4 days at 40°C). vYF was sampled after various times and titer loss (expressed by Logic loss) was analyzed by CCID50. Solid forms were reconstituted just before analyses with 0.5mL of 0.4% NaCl solution.

[0265] Accelerated stability study results showed better behavior for vYF formulated in Formulation A or Formulation B compared to vYF formulated in Stamaril or Wiggan formulations. It was illustrated by taking into account VVM14 criteria (Vaccine Vial Monitors (WHO E6 / IN5 specifications) corresponding to 14 days at 37°C, 90 days at 25°C and > 3 years at 5°C), showing significantly lower Logic losses for Formulation A and Formulation B after 3 months at 25°C, 14 days at 37°C; and also taking into account CTC labeling criteria, showing significantly lower Logic losses for Formulation A and Formulation B after 4 days at 40°C (Figure 1A, average titer loss in Logic CCID50 are indicated below each bar). Indeed the average titer loss (in Logic CCID50) for 3 years at 5°C is -0.61 for Formulation A, -0.62 for Formulation B and -0.90 for Wiggan formulation. The average titer loss (in Logic CCID50) for 3 months at 25°C is -0.73 for Formulation A, -0.80 for Formulation B , -1.34 for Stamaril formulation and -1.86 for Wiggan formulation. The average titer loss (in Logic CCID50) for 14 days at 37°C is -0.87 for Formulation A, -1.07 for Formulation B, -1.49 for Stamaril formulation and -2.05 for Wiggan formulation. Finally, the average titer loss (in Logic CCID50) for 4 days at 40°C is -0.66 for Formulation A, -0.79 for Formulation B, -1.50 for Stamaril formulation and -1.16 for Wiggan formulation.

[0266] Given the high costs of rHA in Formulation A, Formulation B was selected for the following steps.

[0267] EXAMPLE 5 — Stability of Dengue chimeras in Formulation B

[0268] Live-attenuated dengue chimeras were generated for each of the 4 dengue serotypes. Chimeric dengue viruses from each dengue serotype were obtained with the backbone of a YF 17D virus in which the sequences encoding the premembrane and envelope proteins have been replaced with the corresponding sequences of the respective dengue serotype (see WO9837911). Tetravalent bulk aqueous vaccine candidate (with dengue chimeras of the 4 dengue serotypes) was diluted in Formulation B and targeted at 5.5 Logic CCIDso / dose for each dengue serotype. The delta form titer target was -0.38 Logic CCID50 for dengue serotype 1, -0.15 Logic CCID50 for dengue serotype 2, -0.18 Logic CCID50 for dengue serotype 3 and -0.17 Logic CCID50 for dengue serotype 4, showing a protective effect of Formulation B on the dengue chimeras in liquid form. This FBP was then submitted to freeze-drying with a conservative cycle. A protective effect of Formulation B on the 4 serotypes was observed on DP with less than 0.5 Logic loss in CCID50 after lyophilization (no loss for dengue serotypes 1, 3 and 4, and 0.46 Logic loss in CCIDso for serotype 2).

[0269] EXAMPLE 6 — Formulation B limits for industrialization

[0270] Formulation B was initially designed with a view to stabilizing live-attenuated flaviviruses such as vYF under micropellet form (see Clenet et al, 2019). In view of the good results obtained regarding stabilization of vYF in Formulation B, this Formulation B (also referred to as formulation phase I) was selected for further developments and industrialization. At industrial scale for freeze-drying, some issues with the Formulation phase I were however detected:

[0271] - Formation of foam during the process, especially when reconstituting lyophilizate product, and when transferring liquids,

[0272] - Duration of lyophilization cycle needed above 24 hours, which is not compatible for industrial scale which requires cycle below 24 hours,

[0273] - High Viscosity giving rise to processability and cleaning problems.

[0274] It was thus determined to improve Formulation B with regard to at least these 3 parameters, namely:

[0275] - Inhibiting foam formation,

[0276] - Compatibility with short lyophilization cycle (below 24h) and

[0277] - Reduction of viscosity, while simultaneously preserving the stabilizing properties without loss of antigenicity and the properties necessary for use in vaccination, namely:

[0278] - Prior to reconstitution, DP exhibiting a shelf-life of at least 3 years under refrigerated storage conditions (5 ± 3°C) and compliant with VVM14 criteria (14 days at 37°C, 90 days at 25°C and > 3 years at 5°C) and CTC labeling (4 days at 40°C);

[0279] - Reconstituted DP solution compatible with injectable route (osmolality > 280 mOsmo / kg, free of visible particles), stable up to 6h under refrigerated storage conditions (2-8°C) after reconstitution,

[0280] - Formulation and process scalable from 10 mL to 5-8 L, - Reduction of the number of required excipients and dry content of formulation.

[0281] Different Designs of Experiments (DoE) were thus elaborated and conducted to try to solve these issues while maintaining a high stabilizing effect, especially during lyophilization, in order to develop a Phase III formulation.

[0282] Specifically, in a first step, in order to reduce dry content in phase III formulation, a reduction of the highly concentrated excipients present in phase I formulation was tested, including trehalose and sorbitol.

[0283] Furthermore, elimination of few other excipients was tested to limit number of required excipients for a commercialized formulation.

[0284] Secondly, to limit foam during process, reduction of concentration of surfactant (P407) was also tested down to 0.01 g / L.

[0285] Then, after identification of minimal number of excipients required for both process and stability of vYF, their concentrations were optimized.

[0286] EXAMPLE 7 — Designs of experiments (DoE) to identify required excipients and their best concentration.

[0287] It was decided in this first set to keep the TRIS buffer, Trehalose and L-Lys, as core elements, and to identify additional excipients which are necessary and sufficient, amongst those present in the Phase I formulation.

[0288] Consequently, in a TRIS buffer solution at pH 8.0 including Trehalose (lOOg / L) and L-Lys (0.3g / L), formulations mixing excipients of phase I formulation were evaluated in a large range of concentration through a fractional factorial design of experiment. Freeze-dried products were obtained using a conservative lyophilization cycle and reconstituted with NaCl solution just before analysis. Infectious titers were determined for the drug product (DP) at time-zero and after incubation for 2 weeks at 37°C. The glass transitions temperatures Tg’ and Tg were also determined in °C.

[0289] For this DoE, 6 continuous factors were tested, with PVP10 (10-50 g / L), L-Proline (1-30 g / L), CaCh, 2H2O (1-10 g / L), Sorbitol (5-55 g / L), Urea (1-10 g / L) and P407 (0.01-5 g / L). CCID50 results

[0290] Best results with less than 1.0 Log CCID50 lost after 2 weeks at 37°C (in dry form after lyophilization) were observed for formulations containing low concentration of sorbitol and P407 and high concentration of CaCl2and urea, namely:

[0291] 1- PVP10 (10 g / L), L-Proline (Ig / L), CaCl2, 2H2O (10 g / L), Sorbitol (5 g / L), Urea (10 g / L) and P407 (0.01 g / L); and

[0292] 2- PVP10 (50 g / L), L-Proline (30 g / L), CaCh, 2H2O (10 g / L), Sorbitol (5 g / L), Urea (10 g / L) and P407 (0.01 g / L).

[0293] For CCID50 results at t-zero, no lack of fit was observed with this design. Main statistical parameters and list of excipients inducing a significant impact on vYF infectious titer at T- zero and after 14 days at 37°C and a statistical analysis data pointed out significant effects including interactions between several excipients.

[0294] While urea and PVP10 exhibited positive effect, L-proline showed negative effect. Interestingly, interaction was shown between sorbitol and urea and between CaCl2and P407, while sorbitol and P407 did not show significant effect.

[0295] Then, in practice, sorbitol removal will significantly reduce dry content and P407 concentration minimization may limit foam, in these circumstances, urea concentration should be increased and CaCl2concentration be kept low.

[0296] For CCID50 results after 2 weeks at 37°C, no lack of fit was observed with this design and a statistical analysis of data pointed out positive effect of L-proline and urea and a negative effect of sorbitol. Interestingly, again, interaction was shown between sorbitol and urea.

[0297] Then, in practice, removing sorbitol will lead to a stronger positive effect of urea that can be increased in content to protect vYF.

[0298] Taken together, these results demonstrate that, in a TRIS buffer solution at pH 8.0 containing Trehalose (lOOg / L) and L-Lys (0.3g / L), sorbitol can be removed or strongly limited without significant negative impact on infectious titer of vYF at t-zero or after 2 weeks at 37°C after lyophilization. In such a condition, in view of the interaction between sorbitol and urea, concentration of urea should be increased to stabilize vYF, in the absence of sorbitol. These results did not show significant impact for P407, even at high concentration. Then, its concentrations can be reduced to be lower than 1 g / L or even lower than 0.5 g / L. This result is of high interest to limit foam during process.

[0299] Regarding CaCh, results show a slight positive impact at t-zero in presence of P407.

[0300] While PVP10 showed a positive impact at t-zero (lyoprotectant), it looked neutral during thermostability study.

[0301] Interestingly, L-proline exhibited a negative impact at t-zero but a positive one after a thermal stress of DP (thermostabilizer).

[0302] Glass transition temperatures results

[0303] Regarding impact of excipients on glass transition temperatures, very good fits were obtained for this design (data not shown). PVP10 exhibited a positive impact, enhancing both Tg’ and Tg. Removing sorbitol will have a positive impact as this excipient strongly decrease Tg’ and Tg. While L-proline and urea exhibited a stabilizing effect, they also decrease glass transition temperatures. PVP10 showed a positive impact on both Tg’ and Tg. Interestingly, CaCh showed a negative impact on Tg’ while CaCh showed a positive one on Tg. In this formulation plan, P407 did not show any significant effect on the whole range of concentration.

[0304] Then, addition of L-proline and urea should be moderated to limit decrease of glass transition temperatures. This effect can be compensated by addition of PVP10, keeping in mind final targets of Tg’ > -36°C and Tg > 50°C.

[0305] Based on the combined results of stabilization and impact on glass transition temperature, it can be concluded that, in a TRIS buffer solution at pH 8.0 containing Trehalose (lOOg / L) and L-Lys (0.3g / L), sorbitol can be removed or strongly limited without significant negative impact on infectious titer of vYF at t-zero or after 2 weeks at 37°C, after lyophilization. In such a condition, considering interaction between sorbitol and urea, concentration of urea should be increased to stabilize vYF.

[0306] However, addition of L-proline and urea should be moderated to limit decrease of glass transition temperatures. This effect can be compensated by addition of PVP10, keeping in mind final targets of Tg’ > -36°C and Tg > 50°C. EXAMPLE 8 — Designs of experiments (DoE) to test or confirm the possibility of removing sorbitol and L-Lysine.

[0307] Further to the previous results, various concentrations of excipients were tested by limiting the presence of sorbitol and P407 and increasing concentration of urea. Furthermore, to limit the number of required excipients, potential removal of L-lysine was also tested.

[0308] It was observed that, in TRIS buffer 20mM at pH =8.0, with high concentration of L-proline (20 g / L) and PVP10 (30 g / L), low concentration of P407 (0.01 g / L) and without sorbitol, increasing amount of urea, from 0 g / L to 10 g / L, induced a stabilization of vYF during thermal stress. Overall data obtained during this formulation plan suggested a stabilizing effect of urea at the tested concentrations of 5 g / L and 10 g / L, in this combination. Alternatively, removal of L-lysine could be an option while a complete removal of CaCh does not appear recommended.

[0309] Together, results obtained at that stage confirmed that removing sorbitol while increasing content of L-proline and urea with a low concentration of P407 would lead to protective effect of vYF, also in the absence of L-Lysine.

[0310] EXAMPLE 9 Additional designs of experiments (DoE) to better define the effects of some excipients and the ranges of concentrations.

[0311] A. To better understand the effect of CaCh, L-lysine and L-proline, a central composite DoE was developed in a matrix corresponding to 20mM TRIS buffer solution at pH 8.0 containing Trehalose (lOOg / L), PVP10 (20 g / L), urea (5 g / L) and P407 (0.05 g / L). L-Proline was tested from 3 to 30 g / L; L-lysine in the range from 0 to 5 g / L and CaCh in the range from 0 to 12 g / L.

[0312] In such a matrix, a positive effect of L-proline was observed, especially without CaCh at t- zero.

[0313] An optimal concentration of CaCh and L-lysine was suggested at 7 g / L and 2.7 g / L, respectively.

[0314] Interestingly, while L-lysine exhibited both slight lyoprotective and thermostabilizing effect at 2.7 g / L, CaCh at 7 g / L showed a thermostabilizing effect only.

[0315] Considering that reactogenicity was observed in an animal model using CaCh at elevated concentration (7 g / L) but in a different composition (data not shown), it was however decided to limit its concentration at 1.3 g / L, as in phase I formulation, only as a precautionary measure, not for stabilization reasons and contrary to the optimal concentration deduced from this DoE.

[0316] Moreover, in view of the low positive impact exhibited by L-lysine; it was decided to remove it for further screening of formulations. In this plan, only L-proline showed a clear stabilizing effect both at t-zero and after 2 weeks at 37°C.

[0317] This plan thus allows the refining of the formulation by removal of L-Lysine and setting the CaCh at the highest acceptable concentration (as a precautionary measure), namely 1.3 g / L.

[0318] B. To better understand the effect of CaCh, L-proline and urea, a full factorial DoE was developed in a matrix corresponding to 20mM TRIS buffer solution at pH 8.0 containing Trehalose (lOOg / L), PVP10 (20 g / L), and P407 (0.05 g / L).

[0319] In such a matrix, limiting concentration of CaCh was chosen, aligned with low concentration required to prevent reactogenicity effect in vivo. Clear stabilizing effect was observed for L- proline and urea, both at t-zero and after 2 weeks at 37°C. Interestingly, CaCh should not exhibit negative effect when L-proline and urea are not present at too high concentrations, e.g., around 10 g / L and 5 g / L, respectively.

[0320] This observation validates the choice of a low concentration of CaCh at 1.3 g / L or lower required for phase III formulation.

[0321] At that stage, stabilizing effect of L-proline and urea observed on infectious titer must be mitigated by their negative impact on glass transition temperatures. Then, to secure a commercialized freeze-drying short cycle (< 24 hours), it was decided to give priority to L- proline and urea if possible not higher than around 20 g / L and not higher than around 10 g / L, respectively, at least in the conditions of this combination.

[0322] In any event, for further screenings, considering comments of the Japanese regulator about PVP in injectable products, formulations with PVP concentration above 20 g / L were removed from the development of phase III formulation, for regulatory reasons, unlinked to the stabilization.

[0323] C. To challenge the need for PVP10 and CaCh, another DoE was developed without these excipients, mixing L-proline (0.2 - 10 g / L), urea (1.0 - 5.5 g / L) and P407 (0.05 - 2.05 g / L) in a 20mM TRIS buffer solution at pH 8.0 including Trehalose (100 g / L). In such a matrix avoiding PVP10 and CaCL, stabilizing effect was confirmed for L-proline and urea at both t-zero and after a thermal stress (2 weeks at 37°C). Trehalose also exhibited a stabilizing effect wile P407 showed a best impact at 1.3 g / L.

[0324] In the best formulation suggested by the statistical analysis of this DoE (140 g / L trehalose, 10 g / L L-proline, 1.3 g / L P407, 10 g / L urea), Tg’ value would be very close to -36°C and dry content of the formulation is still high. The dry content could be reduced by limiting concentration of trehalose but this would lead to decrease glass transition temperatures. Furthermore, with a concentration of surfactant at 1.3 g / L, foam issue is expected to appear during process. Taken together, such a formulation stabilizing vYF looked not realistically implementable for a commercial product.

[0325] These results thus confirm the importance of CaCL and PVP in the Phase III formulation, in order to solve the problems mentioned at the beginning, including a limited concentration of P407.

[0326] Furthermore, polymers being enhancers of glass transition temperatures and considering that PVP 10 already showed a lyoprotective effect, it was decided to keep it in the final formulation, at a lower concentration than 20 g / L (phase I). This point was aligned with consideration of Japan regulators.

[0327] Considering both positive results obtained from DoEs to stabilize vYF and process constrains for a commercial vaccine, as well as precautionary measures, a suitable phase III formulation can meet the following requirements:

[0328] - TRIS buffer (< 2.5 g / L) pH 8.0 Trehalose around 100 g / L

[0329] - PVP10 < 20 g / L

[0330] - CaCl2< L5 g / L

[0331] - L-proline [5-15 g / L]

[0332] - Urea [3-10 g / L]

[0333] - P407 < 1 g / L

[0334] No sorbitol nor L-lysine EXAMPLE 10 Selection of formulations for phase III.

[0335] A formulation plan (see Table 3 below) was designed to select lead and back-up formulations for phase III. Here, two formulations including sucrose were tested, considering that based on scientific literature, mixtures of trehalose and sucrose can improve cake appearance.

[0336] Table 3: formulation plan 2021-FO-S20 for selecting phase III formulation.

[0337] Results presented in Table 3 and FIG.2 suggested that adding sucrose did not show significant impacts in term of cake appearance. Formulations with highest Tg’ value (#3, #2, #8) did not show significant collapse area on images using a conservative lyophilization cycle (63 hrs).

[0338] Furthermore, Tg’ and Tg values were not improved by adding sucrose, compared with pure trehalose-based formulations. Formulations without PVP10 showed good results in term of thermostability, however with the lowest Tg’ values.

[0339] Formulation #14 appeared of interest with high infectious titers, but the elevated P407 content is likely to lead foam during scale up of process. For this reason, this formulation was not selected. To conclude, formulations #1 and #2 ensure approximately all previously defined constrains with a good thermostability, then were selected as Lead and Back-up 1 (BUI) candidates, respectively. To secure project development, formulation #9 was also selected as a Back- up2 (BU2) formulation.

[0340] To evaluate compatibility of such best formulations (Lead, BUI and BU2) with aggressive lyophilization cycles, the following lyophilization cycle (25 hours, Table 4) was tested:

[0341] Table 4: lyophilization cycle of 25 hours

[0342] Cake appearance is shown in FIG.3. Significant micro-collapses were observed in cakes of Lead formulation, while back-up formulations looked more homogeneous, suggesting that Lead formulation composition could potentially not be entirely compatible with aggressive lyophilization cycles, required at industrial scale, although this point was not tested. This observation correlated with low level of TRIS buffer, namely Ig / L corresponding to 8.3 mM, in back-up formulations, this excipient decreasing Tg’. Furthermore, limiting concentration of PVP10 at 7 g / L instead of 20 g / L of phase I formulation looked more favourable to prevent miscibility issue of polymer at negative temperatures during freezing step of lyophilization. Taken together and considering low immunogenicity developed by BU2 compared to BUI formulation (data not shown), BUI was preferred and selected for phase III. EXAMPLE 11 —Confirmation of the requirements for each individual constituents of the stabilizing compositions, acceptable range of concentrations and pH.

[0343] A. To align the selection of BUI composition with QbD (Quality by Design) methodology, the following formulation plans including DoEs were performed (see table 5 below), using determination of vYF infectious titers after lyophilization process (t-zero), a 4-days incubation at 40°C, a 14 days incubation at 37°C and a 3-months incubation at 25°C for thermostability. These plans challenged the impact of absence of one or several excipients in the selected BUI formulation and then focused on impact of L-proline, urea and P407 concentrations.

[0344] Formulations #7, #9, #10 and #12 with very low dry content (1.7%, 1.8%; 1.5% and 1.1% respectively) were not lyophilizable (empty vials after freeze-drying).

[0345] Best formulation (least Log lost) was formulation #1 (phase III formulation, previously designed BUI, dry content of 12.4%).

[0346] Formulations without P407 (#2, #5, #8, #10, #12, #15 and #17) led to huge loss of infectious titers during freeze-drying process, confirming the lyoprotective effect of this surfactant.

[0347] Absence of TRIS, PVP10, urea, L-proline or trehalose showed at least 1 Log loss after freeze drying, demonstrating their need to protect vYF.

[0348] Absence of TRIS buffer (formulation #3) led to a product far from the target and thermosensitive, justifying the interest of this excipient.

[0349] Reduction of all concentrations by 2 (formulation #4) induced an infectious titer at the target (5 Logio / dose) but very fragile during thermal stress (- 2 Log loss), confirming optimal selection of concentration for the phase III formulation (#1), although this formulation remains of interest.

[0350] PVP10 and CaCh exhibited less impact.

[0351] Table 5: New formulation DoE plan

[0352]

[0353]

[0354] Results were analyzed by machine learning using Chem Assistant software, leading to extra tree models with the following acceptable statistical scores:

[0355] A SHAP (Shapley Additive exPlanations) analysis highlighted the strength of each excipient on CCID50 values (FIG. 4).

[0356] According to this analysis, trehalose, CaCL and P407 appeared to be the most important excipient needed to protect vYF during freeze-drying process, with a minimum concentration around 40 g / L, 0.6 g / L and 0.02 g / L, respectively. Other excipients appeared less impacting as their absence would lead to infectious titer loss close to the variability of the analytical technique (0.2 Log CCID50).

[0357] At 14 days at 37°C and 90 days at 25°C, models showed positive impact for P407, trehalose, L-proline, urea and TRIS, suggesting a minimum concentration of 0.02 g / L, 50 g / L, 5 g / L, 3 g / L and 0.8 g / L, for P407, trehalose, L-proline, urea and TRIS respectively.

[0358] These results confirm the choice of these excipients and their respective minimum concentration to protect vYF.

[0359] B. range of pH

[0360] To evaluate the impact of a pH shift on vYF stability, a formulation plan (see Table 6 below) at ± 0.5 unit of pH from the target was built, then an accelerated stability study was applied.

[0361] Table 6: Formulation plan and results

[0362] Ranging pH from 7.5 to 8.5, infectious titers of vYF at t-zero and after thermal stress were the same, with less than 1.0 Log losses after 90 days at 25°C, 14 days at 37°C and 4 days at 40°C. These experimental results are well included within predictive bands of kinetic model developed using representative batches of BUI1 formulation (pH 8.0), showing a comparable behavior between batches manufactured from pH 7.5 to 8.5 (FIG. 5).

[0363] These results confirm the robustness of the selected BUI formulation to stabilize vYF at ± 0.5 unit of pH around the selected one.

[0364] EXAMPLE 12 — Compatibility of the formulation with lyophilization cycle and with scale- up.

[0365] Aggressive Lyophilization cycle.

[0366] Starting from a conservative cycle taking 63 hrs, a more aggressive one (23 hrs) was first tested to control compatibility of the selected phase III formulation.

[0367] Cake was in place with acceptable appearance, without visible collapse.

[0368] Infectious titers for freeze-dried product obtained with aggressive lyophilization cycle were at 4.7 Log CCID50 / d and 4.1 Log CCID50 / d, at t-zero and after 14 days at 37°C, respectively. These results showed infectious titer close to the target (5.0 Log CCID50 / d) and less than 1.0 Log loss after the thermal stress, confirming that formulation selected for phase III is compatible with an aggressive lyophilization cycle (~24 hrs).

[0369] It is to be noted that a thermal stress for 4 days at 40°C (CTC label criteria) did not induce collapse of cake. This result indicates that drug product can experience high excursion of temperature, and is aligned with high Tg (> 50°C) exhibited by the selected phase III formulation.

[0370] Scale up to 5 L and 8 L using phase III formulation (technical and demo lot)

[0371] A technical batch at 5 L scale was targeted at 5.0 Logio / dose with an overage of 0.4 Log, mimicking GMP process, however, with a conservative lyophilization cycle, kept at 63 hrs. By another side, a demo batch at 8 L was targeted at 5.0 Logio / dose with an overage of 0.5 Log, mimicking GMP process with an optimized lyophilization cycle (43 hrs).

[0372] The overage corresponds to the quantity of virus added to compensate for the loss of infectious titer during lyophilization.

[0373] At FBP stage, no infectious titer loss was observed before and after 1 and 2 filtrations, leading to targeted titer at ± 0.2 Log. Determination of key physical parameters of FBP was done on a technical batch, showing a viscosity of 1.6 mPa.s, a surface tension of 42.6 mN / m and a Tg’ of -33°C. vYF characterization was performed at DP and reconstituted DP stages, comparing two technical batches and the demo batch.

[0374] All DP criteria were found acceptable before and after reconstitution. In addition, less than 1.0 Logio was lost when DP was exposed at thermal stresses linked with CTC label or VVM14 criteria (i.e., 4 days at 40°C, 14 days at 37°C or 90 days at 25°C). Only one technical batch lost 1.1 Log after a 90 days exposure at 25°C. However, all other results, especially for demo batch showed acceptable thermal stability with less than 1.0 Log lost in all stress conditions, justifying the selection of this BUI formulation for phase III.

[0375] EXAMPLE 13 — Confirmation of the phase III formulation- stability modelins.

[0376] Applying AKM (Advanced Kinetic Modeling) and associated good modeling practices, kinetic models were developed to estimate long-term stability of vYF in the selected BUI formulation. FIG. 6 shows stability predictions in Log loss (infectious titer) up to 3 -years in freeze-dried format (A, left panel) and after reconstitution of the freeze-dried product (B, right panel). Using experimental stability data available to date (i.e., 1-year), less than 1.0 Log loss is expected for DP after 3 years and 4 days at 5°C, in solid and liquid states, respectively.

[0377] It was thus confirmed that phase III formulation (BUI) as defined above, stabilized vYF as satisfactorily as and even slightly better than the phase I formulation (see inter alia FIG. IB comparing phase I and phase III formulations), but with a smaller number of excipients and in the absence of foam during reconstitution.

[0378] In term of impact of temperature excursions, the model developed with available experimental data to date predicts a better stability for phase III formulation (FIG. 7), with less than 1.0 Log lost for DP stored 2 years at 5°C, then experienced 4 days at 40°C (CTC labeling criteria). Furthermore, due to elevated Tg value (> 50°C) for the phase III formulation, cake appearance of freeze-dried product is not impacted by a such excursion of temperature.

[0379] Selected BUI FBP formulation is stable at least 3 months at 5°C

[0380] To evaluate the stability of the BUI formulation per se and its stabilizing effect on vYF, an accelerated stability study was applied on a freshly built BUI FBP at 5°C and 45°C for 3 months. First, physico-chemical attributes (turbidity, color, Tg’) of BUI FBP were monitored. Second, aged-BUl FBP were used to build DP and evaluate its ability to still protect vYF during freeze-drying process and after 14 days at 37°C.

[0381] The results are reported in the below Table 7.

[0382] Table 7:

[0383] They showed that physico-chemical attributes of BU 1 FBP were not impacted by a such accelerated stability study. Regarding infectious titer of vYF at DP stage, targeted values were obtained at t-zero, even by using BU 1 FBP kept 3 -months at 5°C or at 45°C, indicating that BUI formulation still protect vYF during freeze-drying process. However, in more drastic storage conditions of BUI FBP (i.e., at 45°C up to 3 months), infectious titer drops observed on DP were higher than accepted, with more than 1.5 Log CCID50 loss. By the opposite, using BUI FBP kept up to 3 months at 5°C, stability of DP looked as expected with close to 1.0 Log CCID50 loss after 14 days at 37°C.

[0384] These results suggest that the selected BU 1 formulation (FBP) can be stored at least 3 months at 5°C, then use for receiving vYF bulk for DP process. Conclusions:

[0385] For Phase III clinical study, the composition of Yellow Fever vaccine stabilizer and freeze- drying cycle were optimized in order to achieve conventional freeze-dried format compatible with industrial-scale manufacturing. This optimization allowed to:

[0386] • Reduce dry content of formulation whilst minimizing number of required excipients for a conventional freeze-dried product;

[0387] • Limit foam during process and reduce viscosity of the formulation (processability);

[0388] • Ensure scalability from 100 mL up to 8 L;

[0389] • Secure a formulation scalable and compatible with an industrial aggressive lyophilization cycle;

[0390] • Secure thermostability of vYF.

[0391] Phase I and Phase III formulations are compared below in Table 8 with respect to these components and physical properties.

[0392] Table 8:

[0393] A stability plan was appropriately designed under recommended storage condition (5±3°C) and accelerated conditions (25±2°C, 37±2°C) on representative large-scale batches in order to develop a kinetic model for long-term stability predictions of vYF vaccine. This kinetic model may also be used to predict stability of products under any time-temperature conditions (long-term storage, during shipment or temperature excursions etc.). EXAMPLE 14 — Immunogenicity and safety of a live-atenuated yellow fever vaccine bulk in phase III formulation.

[0394] Immunogenicity

[0395] The Syrian golden hamster (Mesocricetus auratus) was shown to develop symptoms similar to human disease after infection with the Jimenez YFV strain adapted after 1 passage in monkeys followed by 10 passages in hamsters. Moreover, this animal species has been reported to raise high neutralizing antibody titers after immunization with the YF-17D vaccine strain. Based on these properties, the model was previously used for the evaluation of immunogenicity and protective efficacy of YF egg-based vaccines. The hamster model has moreover been selected to assess the immunogenicity and / or protective efficacy of the new Vero cell-based vYF vaccine candidate all along its development.

[0396] 4 groups with 15 4-week-old female Syrian golden hamsters per group were injected subcutaneously (SC) in the flank at DO with the vYF live-attenuated vaccine (half of the human dose), in the following stabilizing formulations:

[0397] - vYF live-attenuated vaccine bulk in phase I formulation;

[0398] - vYF live-attenuated vaccine bulk in phase III formulation;

[0399] - vYF live-attenuated vaccine bulk in BU2 formulation (corresponding to formulation #9 in 2021-FO-S20, Table 3)

[0400] - YF-VAX®.

[0401] Post vaccination, the clinical signs were observed. The immunogenicity of the vaccines was evaluated by titration of serum neutralizing (SN) antibodies at day 21 post -vaccination. The animals were euthanized at D28.

[0402] The titration of the administered vaccine doses is given below in table 9.

[0403] Table 9; titration of vaccine doses: The functional neutralizing antibodies present in the serum of the immunized animals were titrated at D21 after injection.

[0404] Briefly, the heat-inactivated sera were serially 2-fold diluted in IMDM (THERMOFISHER SCIENTIFIC) + 4% fetal calf serum (FCS) starting from 1 :5. YF-17D Stamaril® virus grown on Vero cells was diluted in order to obtain 4000 pPFU / mL in IMDM and incubated 90 minutes with 2-fold diluted serum samples (v / v). The virus / serum mixture was then added to Vero cells in 96-well plates and incubated for 45 + / - 2 hours. After incubation, cells were fixed with 85% acetone before immunostaining. Plates were blocked with PBS + 0.05% Tween 20 + 2.5% skim milk and incubated first with an anti -flavi virus monoclonal antibody 4G2 (RD BIOTECH®), and second with a goat anti-mouse IgG HRP conjugate. Finally, plates were stained with the Trueblue™ chromogen. Plaques were counted with Viruscope reader from Microvision™.

[0405] The final seroneutralizing antibody titer is calculated using the least square method and correspond to the inverse of the dilution demonstrating a neutralization of 50% of virus plaques. The LOD of the assay was 10, corresponding to the first reciprocal dilution in the final volume.

[0406] For the statistical analysis, an ANOVA model by time with product as fixed factor was used. All analyses were performed on Graph Pad Prism v7®. A margin of error of 5% was used for effects of the factors.

[0407] Results:

[0408] In all groups, no clinical signs, including local reactions at the injection site, were observed during the study.

[0409] The correlate of protection for Yellow Fever vaccines is based on the induction of detectable neutralizing antibodies in previously YF naive subjects (World Health Organization. Requirements for Yellow Fever Vaccine. WHO Technical Report Series Report No. 978 > Annex 5).

[0410] Neutralizing antibodies specific for YFV were quantified by pPRNTso in serum samples from vaccinated hamsters at day 21 post-vaccination. Group geometric mean and individual titers are depicted in Figure 8. Neutralizing antibody titers achieved levels far above the correlate of protection threshold, i.e., 10 PRNT50 in all hamsters of the groups vaccinated with YF-VAX, vYF phase I formulation and vYF phase III formulation, with mean titers at 3.8 Log pPRNTso for vYF phase III formulation and at 4.0 Log pPRNTso for vYF phase I formulation. In the group immunized with vYF BU2 formulation, neutralizing antibodies were detected in only 9 / 15 hamsters, with mean titers at 3.2 Log pPRNTso.

[0411] The neutralizing antibody titers are slightly higher for the YF-VAX control group, but the virus preparation was done extemporaneously and with a different process than the vYF batches and may not be directly compared.

[0412] No significant difference was observed between the groups immunized with vYF phase I and phase III.1 formulations (p=0.467; ANOVA). The neutralizing antibody GMT is significantly lower in the group immunized with vYF in phase BU2 formulation than in the groups immunized with vYF in phase I and phase III formulations (p=0.008 and p=0.048 respectively, ANOVA).

[0413] Discussion and conclusion:

[0414] The hamster study aimed to assess the immunogenicity of two vYF drug products freeze- dried batches composed of vYF bulk diluted in two formulation buffer candidates (formulation phase III and formulation BU2). For that purpose, the vaccine formulation candidates were compared to the drug product manufactured from the same bulk batch in phase I formulation buffer.

[0415] High neutralizing antibody response were induced in 100% of hamsters vaccinated with vYF phase I and phase III formulations, and the geometric mean titers were not significantly different. On the opposite, in the group immunized with vYF BU2 formulation, only 9 / 15 hamsters have quantifiable neutralizing antibody response, and the GMT is significantly lower than the vYF phase I formulation.

[0416] Based on these results BU2 was thus not pursued.

[0417] Safety

[0418] Safety of the phase III formulation was evaluated on immuno-deficient A129 mouse model, which is knocked out (KO) for type I IFN receptors and suitable to assess potential viscerotropic and neurotropic symptoms caused by yellow fever viruses. 3 groups with 12 6- to 9-week-old A129 mice per group were injected at DO in SC with 200 pl of following formulations: vYF live-attenuated vaccine bulk (4 LogCCID50) in phase I formulation; vYF live-attenuated vaccine bulk (4 LogCCID50) in phase III formulation; or YF-VAX® (4 LogCCID50) positive control.

[0419] A negative control group of 6 mice were injected with phase III formulation buffer.

[0420] Serum samples were collected at days 4, 6 and 11 post -vaccination for viremia evaluation by YF-NS5 qRT-PCR (Mantel 2008). At day 6 and 11 post-vaccination, 6 mice per group were euthanized and liver, spleen and brain were sampled for evaluation of the viral load by qRT-PCR. Morbidity (i.e., clinical signs evaluated through an established scoring and body weight monitoring) as well as mortality were followed all along the study.

[0421] Neither mortality nor morbidity was recorded after injection of the vYF DP lots or with YF- VAX control vaccine.

[0422] Viremia and viral load in organs induced by vYF DP in phase III formulation were significantly lower than, or equivalent to, that induced by the vYF DP in phase I formulation. Comparison of the viremia and viral load in spleen, brain and liver after vYF vaccination in phase I and phase III formulation is reported in Table 10.

[0423] Table 10: comparison of the vYF in phase III formulation to vYF in phase I formulation

[0424] In conclusion, the viral loads in spleen, liver and brain after vYF vaccination in phase III formulation are inferior to phase I formulation. They are also inferior to those obtained with YF-VAX vaccination, confirming that the vYF DP in phase III formulation was able to fulfill the attenuation criteria in the neuro- and viscero-tropism A129 mouse model.

[0425] EXAMPLE 15 — Confirmation of a live-atenuated yellow fever vaccine stability in phase III formulation in DP production lots.

[0426] Three production batches of vYF DP in phase III formulation were produced.

[0427] FIG. 9 shows stability predictions in Log loss (infectious titer) up to 5-years in freeze-dried DP, using experimental stability data available to date (i.e., 1-year). Less than 1.0 Log loss (0.5702 Log loss) is expected for DP after at least 5 years at 5°C (FIG. 9A).

[0428] Even after a three days excursion at 40°C, less than 1.0 Log loss (0.6875 Log loss) is expected for DP after 5 years at 5°C (FIG. 9B).

[0429] REFERENCES

[0430] - Non patent references

[0431] - Barrett ADT. Yellow fever live attenuated vaccine: A very successful live attenuated vaccine but still we have problems controlling the disease. Vaccine. 2017;35(44):5951-5955.

[0432] - Burke CJ, Hsu TA and Volkin DB. Formulation, Stability, and Delivery of Live Attenuated Vaccines for Human Use. Critical Reviews in Therapeutic Drug Carrier Systems. 1999; 16(1): 1-83.

[0433] - Charretier C, Saulnier A, Benair L, Armanet C, Bassard I, Daulon S, Bernigaud B, de Sousa ER, Gonthier C, Zorn E, Vetter E, Saintpierre C, Riou P and Gaillac D. Robust real-time cell analysis method for determining viral infectious titers during development of a viral vaccine production process. Journal of Virological Methods. 2018; 252:57-64.

[0434] - Clenet D. Accurate prediction of vaccine stability under real storage conditions and during temperature excursions. European Journal of Pharmaceutics and Biopharmaceutics, 2018;125:76-84. -Clenet, D. et al. A spray freeze dried micropellet based formulation proof-of- concept for a yellow fever vaccine candidate. European Journal of Pharmaceutics and Biopharmaceutics, 2019;142:334-343.

[0435] - Eggers J and Villermaux E. Physics of Liquid Jets. Reports on Progress in Physics. 2008;71(3): l-79.

[0436] - Hansen LJJ, Daoussi R, Vervaet C, Remon JP and De Beer TRM. Freeze- drying of live virus vaccines: A review. Vaccine. 2015 ;33 (42): 5507-5519.

[0437] - Lai CJ and Monath TP. Chimeric Flaviviruses: Novel Vaccines against Dengue Fever, Tick-Borne Encephalitis, and Japanese Encephalitis. Advances in Virus Research. 2003;61 :469-509.

[0438] - Mantel N, Aguirre M, Gulia S, Girerd-Chambaz Y, Colombani S, Moste C and Barban V. Standardized quantitative RT-PCR assays for quantitation of yellow fever and chimeric yellow fever-dengue vaccines. Journal of Virological Methods. 2008;151(l):40-6.

[0439] - Wiggan ON, Livengood JA, Silengo SJ, Kinney RM, Osorio JE, Huang CYH and Stinchcomb DT. Novel formulations enhance the thermal stability of live- attenuated flavivirus vaccines. Vaccine. 2011; 29(43):7456-7462.

[0440] • Patent references

[0441] - WO9640933

[0442] - WO9837911

[0443] - WOO 160847

[0444] - W003101397

Claims

68CLAIMS1. A stabilizer for compositions comprising at least one or more live-attenuated flaviviruses, which comprises: a buffer; trehalose;- CaCh; a polyvinylpyrrolidone (PVP) having an average molecular weight from about 2 500 to about 40 000 Da;L-Proline; a poloxamer; and urea, wherein said stabilizer has a percentage of dry content which is above 5% and less than 20%.

2. The stabilizer of claim 1 wherein said buffer is a TRIS buffer, and / or wherein said PVP has an average molecular weight of about 10 000 Da, and / or wherein said poloxamer is P407.

3. The stabilizer of claim 1 or 2 which is devoid of sorbitol, and / or devoid of lysine, and / or devoid of albumin, and / or devoid of sorbitol and albumin, and / or devoid of sorbitol, lysine and albumin.

4. The stabilizer of any one of claims 1 to 3, which has a percentage of dry matter which is below 18%, or below 15%, or below 13%.

5. The stabilizer of any one of claims 1 to 4, wherein the CaCh is present at a concentration below 7 g / L and / or the PVP is present at a concentration below 20 g / L.

6. The stabilizer of any one of claims 1 to 5, wherein L-Proline is present at a concentration below 20 g / L and / or urea is present at a concentration below 15 g / L and / or the poloxamer is present at a concentration below 5 g / L.

697. The stabilizer of any one of claims 1 to 6, which comprises:TRIS buffer, at pH =8;Trehalose at a concentration from 50 to 150 g / L;CaCh at a concentration from 0.6 to 5 g / L;PVP10, at a concentration from 5 to 15 g / L;L-Proline, at a concentration from 5 to 15 g / L;P407 at a concentration from 0.02 to 1 g / L; and urea at a concentration from 3 to 10 g / L.

8. The stabilizer of any one of claims 1 to 7, wherein the buffer is TRIS buffer at a concentration from 0.8 to 2.5 g / L and / or wherein said stabilizer does not comprise sorbitol.

9. The stabilizer of any one of claims 1 to 8, which is devoid of protein.

10. The stabilizer of any one of claims 1 to 9, which is devoid of lactalbumin, human serum albumin, recombinant human serum albumin (rHA), bovine serum albumin, mammalian serum albumin, recombinant bovine serum albumin, other serum albumins, albumin gene family members, or modified albumin.

11. The stabilizer of any one of claims 1 to 10, which is devoid of sorbitol, Lysine, and albumin.

12. The stabilizer of any one of claims 1 to 11, which is devoid of sorbitol, Lysine, and protein.

13. The stabilizer of any one of claims 1 to 12, which is devoid of sucrose.

14. The stabilizer of any one of claims 1 to 13, which does not comprise more than 1 or 2 additional excipients (water excluded), in addition to the buffer solution, trehalose, CaC12, PVP, L-Proline, poloxamer and urea, and none of these potential additional excipients is sorbitol, protein or lysine.7015. The stabilizer of any one of claims 1 to 14, which comprises one or more buffers or buffering media with a pH from about 7.0 to about 9.0, from about 7.3 to about 8.7, from about 7.5 to about 8.5, or about 8.0.

16. The stabilizer of any one of claims 1 to 15, which comprises one or more buffers or buffering media having concentrations of from about 4.0 to about 50.0 mM, from about 5.0 to about 30.0mM, from about 6.0 to about 25.0 mM, or about 8.3 mM.

17. The stabilizer of any one of claims 1 to 16, wherein trehalose is present at a concentration of about 100 g / L or about 120 g / L.

18. The stabilizer of any one of claims 1 to 17, wherein CaCh is present at a concentration of about 1 g / L, about 1.3 g / L, or 1.5 g / L.

19. The stabilizer of any one of claims 1 to 18, wherein urea is present at a concentration of about 5 g / L.

20. The stabilizer of any one of claims 1 to 19, wherein proline is present at a concentration of about 10 g / L.

21. The stabilizer of any one of claims 1 to 20, wherein the PVP is present at a concentration of about 7 g / L, optionally wherein the PVP is PVP10.

22. The stabilizer of any one of claims 1 to 21, wherein the poloxamer is present at a concentration of about 0.05 g / L, optionally wherein the poloxamer is P407.

23. The stabilizer of any one of claims 1 to 22, which reduces the aqueous virus infectious titer loss of the at least one or more live-attenuated flaviviruses to less than 1.0 Logio (optionally in CCID50) for up to 2 hours, up to 3 hours, up to 4 hours, up to 6 hours, or up to 12 hours at about 37°C.7124. The stabilizer of any one of claims 1 to 23, which reduces the aqueous virus infectious titer loss of the at least one or more live-attenuated flaviviruses to less than 1.0 Logio (optionally in CCID50) for up tolO hours at about 25°C.

25. The stabilizer of any one of claims 1 to 24, which reduces the aqueous virus infectious titer loss of the at least one or more live-attenuated flaviviruses to less than 1.0 Logio (optionally in CCID50) for up to 1 day, up to 2 days, up to 3 days, or up to 5 days, at about 5°C.

26. The stabilizer of any one of claims 1 to 25, which reduces the infectious titer loss of the at least one or more live-attenuated flaviviruses to less than 1.0 Logio (optionally in CCID50) during freeze-drying.

27. The stabilizer of any one of claims 1 to 26, which reduces the infectious titer loss of the at least one or more live-attenuated flaviviruses in a dry composition to less than 1.0 Logio (optionally in CCID50) during 1 or 3 or 5 years at about 5°C, during 3 months at about 25° C, during one week or 14 days at about 37°C, and / or during 4 days at about 40°C.

28. A composition comprising at least one or more live-attenuated flaviviruses and the stabilizer according to any one of claims 1-27.

29. A vaccine composition comprising at least one or more live-attenuated flaviviruses and the stabilizer according to any one of claims 1-27.

30. A liquid vaccine composition comprising at least one or more live-attenuated flaviviruses and the stabilizer of in any one of claims 1 to 27.

31. A vaccine composition, which is a lyophilizate of the vaccine composition according to claim 30.

32. A vaccine composition, which is a freeze-dried form of the vaccine composition according to claim 30 and which is not in dry micropellets form.7233. A method for stabilizing at least one or more live-attenuated flaviviruses, comprising combining at least one or more live-attenuated flaviviruses with the stabilizer according to any one of claims 1 to 27.

34. The method according to claim 33, comprising combining a viral harvest comprising at least one or more live-attenuated flaviviruses with the stabilizer according to claims 1-27 in order to obtain a stabilized bulk aqueous composition comprising said one or more live-attenuated flaviviruses.

35. The method according to claim 34, wherein the viral harvest is a purified and / or concentrated viral harvest.

36. The method of any one of claims 33-35, further comprising drying or freeze-drying, the one or more live-attenuated flaviviruses combined with the stabilizer.

37. A method for preparing a vaccine composition comprising at least one or more live- attenuated flaviviruses, which comprises at least: a) culturing the one or more live-attenuated flaviviruses; b) harvesting the one or more cultured live-attenuated flaviviruses; and c) combining the harvested one or more live-attenuated flaviviruses with the stabilizer of any one of claims 1 to 27.

38. The method of claim 37, further comprising:- filling the vaccine composition and, drying or freeze-drying the filled vaccine composition ; or- drying or freeze-drying the vaccine composition, and filling the dry vaccine composition.

39. A vaccine composition obtained by the method of claim 37.

40. A dry vaccine composition obtained by the method of claim 38.7341. A vaccine kit comprising a first container containing the dry vaccine composition of claims 31-32 or 40 and a second container containing an aqueous solution for reconstituting the vaccine.

42. The stabilizer of any one of claims 1 to 27, the composition of claim 28, the vaccine composition of any one of claims 29-32 or 39-40, the method for stabilizing of any one of claims 33-36, the method for preparing of claim 37 or 38, and the vaccine kit of claim 41, wherein the at least one or more live-attenuated flaviviruses comprise live-attenuated dengue virus, West Nile virus, yellow fever virus, Japanese encephalitis virus, St. Louis encephalitis virus, tick-borne encephalitis virus, or zika virus.

43. The stabilizer of any one of claims 1 to 27, the composition of claim 28, the vaccine composition of any one of claims 29-32 or 39-40, the method for stabilizing of any one of claims 33-36, the method for preparing of claim 37 or 38, and the vaccine kit of claim 41, wherein the at least one or more live-attenuated flaviviruses comprise live-attenuated dengue virus, or yellow fever virus, optionally chimeric.

44. The stabilizer of any one of claims 1 to 27, the composition of claim 28, the vaccine composition of any one of claims 29-32 or 39-40, the method for stabilizing of any one of claims 33-36, the method for preparing of claim 37 or 38, and the vaccine kit of claim 41, wherein the at least one or more live-attenuated flaviviruses comprise live-attenuated yellow fever virus.

Images