ALUMINUM COMPOUNDS FOR USE IN THERAPEUTICS AND VACCINES

Abstract

The invention relates to means and methods for preparing aqueous composition comprising aluminium and a protein said composition comprising less than 700 ppm heavy metal on the basis of weight with respect to the aluminium content. The invention further relates to aqueous compositions comprising a protein and an aluminium-salt, said composition comprising less than 350 ppb heavy metal based on the weight of the aqueous composition.

Claims

1.-3. (canceled)

4. A method for preparing a clinical grade aluminum salt precipitate for incorporation into a medicament and/or vaccine, said method comprising preparing an aqueous solution of aluminum ions and precipitating said aluminum-ions from said solution to form an aluminum salt precipitate, and determining the level of copper (Cu) in the solution and/or the aluminum salt precipitate, wherein the precipitate is selected that is able to provide an aqueous composition comprising less than 3 ppb Cu based on the weight of the aqueous composition.

5.-26. (canceled)

27. The method of claim 4, wherein the precipitate is selected that is able to provide an aqueous composition comprising less than 2.5 ppb Cu based on the weight of the aqueous composition.

28. The method of claim 27, wherein the precipitate is selected that is able to provide an aqueous composition comprising less than 1.25 ppb Cu based on the weight of the aqueous composition.

28. The method of claim 27, further comprising determining the level of iron (Fe) in the solution and/or the aluminum salt precipitate, wherein the precipitate is selected that is able to provide an aqueous composition comprising between 5 ppb and 250 ppb Fe based on the weight of the aqueous composition.

29. The method of claim 27, further comprising determining the level of nickel (Ni) in the solution and/or the aluminum salt precipitate, wherein the precipitate is selected that is able to provide an aqueous composition comprising less than 40 ppb Ni based on the weight of the aqueous composition.

30. A method for manufacture of a medicine, comprising selecting an aluminium hydroxide concentrate that comprises less than 3.0 ppb Cu based on the weight of the medicine; and preparing the medicine using the aluminium hydroxide concentrate.

31. The method of claim 30, wherein the aluminium hydroxide concentrate comprises less than 2.5 ppb Cu based on the weight of the medicine.

32. The method of claim 31, wherein the aluminium hydroxide concentrate comprises less than 1.25 ppb Cu based on the weight of the medicine.

33. The method of claim 30, wherein the medicine is a vaccine.

34. The method of claim 30, wherein the aluminium hydroxide concentrate comprises 10 mg/ml of aluminium hydroxide.

35. The method of claim 30, wherein the aluminium hydroxide concentrate comprises between 5 ppb and 250 ppb iron (Fe) based on the weight of the medicine.

36. The method of claim 30, wherein the aluminium hydroxide concentrate comprises less than 40 ppb nickel (Ni) based on the weight of the medicine.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0087] FIG. 1: RP-HPLC elution profiles of chlorobutyl stopper extract (1:4 diluted) and JEV09L37 SN.

[0088] FIG. 2: SEC HPLC elution profiles of PS (2 mg/mL) before and after trypsin cleavage.

[0089] FIG. 3: SEC HPLC elution profiles of Trypsin treated PS and degraded PS as present in NIV11A74. Note that elution profiles were normalized to similar peak height to allow better comparison.

[0090] FIG. 4: DOE evaluation of the ratio monoclonal/polyclonal ELISA by Pareto chart analysis and main effects plots (4 weeks at 22° C.).

[0091] FIG. 5: DOE evaluation of the ratio monoclonal/polyclonal ELISA by Pareto chart analysis and main effects plots (8 weeks at 22° C.).

[0092] FIG. 6: Contour plot of the estimated response

[0093] FIG. 7: Residual plot of the estimated response

[0094] FIG. 8: ELISA ratio (monoclonal/polyclonal) for JEV formulations at pH 7 in presence of Ni, Cu and Cr. Samples were stored at 22° C. for 5 weeks

[0095] FIG. 9: Summary of results obtained after 7 weeks at 22° C. Shown are the raw data of ratio as a function of pH and metal ion type and combined results for each parameter.

[0096] FIG. 10: Mean ratio of DP formulations prepared with different Alum lots. Samples were stored for 6 weeks at 22° C. Error bars represent the 95% confidence interval calculated based on pooled standard deviation. Samples from left to right: Alum 3877, Alum 4074, Alum 4230 nonGI, Alum 4230 GI, Alum 4470, Alum 4563, Alum 4621, Alum Mix 4074_4230.

[0097] FIG. 11: Particle size distribution of ALHYDROGEL® samples.

[0098] FIG. 12: ALHYDROGEL® titration curves in PBS. .square-solid. non-irradiated AlOH (RQCS0890), .box-tangle-solidup. GI AlOH (RQCS1200), .diamond-solid. GI AlOH (RQCS1342), +GI AlOH (RQCS0448)

[0099] FIG. 13: Overview of tested ALHYDROGEL® batches. Total concentration of contaminating metal ions in ng/mL and share of the main metal ions Fe, Cr and Ni are shown.

[0100] FIG. 14: Amino acid sequence of Ala-(His)6-OprF190-342-OprI21-83 (SEQ ID NO: 1)—herein also referred as “protein A”.

EXAMPLES

Example 1

[0101] Aluminium hydroxide (Alum) Lot 4230 was previously identified to contribute significantly to antigen degradation in FVL09L37. In this particular Alum lot much higher residual metal ion content was observed compared to other Alum lots used for formulation of the inactivated JEV antigen. This Example demonstrates additional studies carried out to further identify the underlying root-cause mechanism and influence of metal ions on degradation pathway of JEV. Design of experiments (DOE) was performed to work out the influence of individual parameters on antigen stability.

[0102] Parameters tested in a 25 full factorial DOE were [0103] Aluminium hydroxide Lot 4230 vs. Aluminium hydroxide Lot 4074 [0104] Presence of excess Protamine sulfate fragments [0105] Presence of leachables from chlorobutyl rubber stopper [0106] pH range 7 to 8 [0107] Residual formaldehyde content

[0108] Alum lot 4230 contains much higher levels of residual metal ion impurities compared to other Alum lots used for formulation of JEV. A “Design-of-Experiment” (DOE) was selected to further investigate the potential root cause mechanism and interaction of parameters that finally could lead to product degradation. In factorial designs, multiple factors are investigated simultaneously during the test. As in one factor designs, qualitative and/or quantitative factors can be considered. The objective of these designs is to identify the factors that have a significant effect on the response, as well as investigate the effect of interactions (depending on the experiment design used). Predictions can also be performed when quantitative factors are present, but care must be taken since certain designs are very limited in the choice of the predictive model. For information about DOE in general see (Siebertz, Karl; van Bebber, David, Hochkirchen, Thomas: Statistische Versuchsplanung: Design of Experiments (DoE). Publisher: Springer Berlin Heidelberg; 1st Edition (2010), ISBN-10: 3642054927).

[0109] 1.1 DOE Study Design

[0110] 1.1.1 Definition of Parameters and Levels for DOE Design

[0111] The following parameters and levels were taken into consideration for designing an appropriate DOE experiment: [0112] Residual metal ion content of Alum: Aluminium hydroxide Lot 4230 and Lot 4074 were selected as representative of the two extremes quality with regard to residual metal ions content of Aluminium hydroxide. The center point was a mixture of 50/50% of both Alum lots. Initial analysis for remaining metal ion impurities in 2% Aluminium hydroxide stock solution by ICP-MS showed significant differences in Cr, Fe, Ni and Cu ion content between these two lots (see Table 1). [0113] Protamine Sulphate fragments: Protamine sulfate (PS) fragments are present at low quantity (<5 μg/mL) in the final vaccine lot. It was tested if PS fragments could contribute to virus surface modification (e.g. interaction/covalent linkage to the virus surface proteins) in combination with Alum and other factors used in this study. Therefore a stock solution of PS fragments was prepared by digestion with Trypsin followed by heat inactivation and ultrafiltration using a 5 kDa membrane for protease inactivation and removal of the enzyme. This stock solution was used for spiking additional PS fragments into the respective formulations at the high level of 50 μg/mL. In low level samples no additional PS fragments were spiked and the actual level in formulations was <5 μg/mL according to HPLC analysis. [0114] pH: Lower and upper level of pH in formulations was 7 and 8 with the center point at pH 7.5. [0115] Leachables/Extractables from syringe plunger: Syringe plungers (made of chlorobutyl PH701/50 black) that are currently used in the container closure system. It was tested if leachables from the chlorobutyl rubber in the formulation could contribute to antigen modification. Therefore a stock solution of leachables was prepared and used for spiking experiments. The high level of spiked leachables in formulation was estimated to be on average 1.4× higher compared to commercial Final Vaccine Lot (FVL). Due to the harsh extraction conditions additional peaks were detected not present in FVL samples. Therefore the spiked formulations represent a “worst case” with regard to leachables and extractables. Formulations at the low level did not contain any leachables from chlorobutyl rubber. [0116] Residual formaldehyde: For low level formulations no additional formaldehyde was spiked into the formulation samples. The lower level was the residual formaldehyde that was still present in diluted NIV sample after inactivation/neutralization and 2-fold dilution was in the range of approx. 37 ppm (recalculated from commercial DS release GMP analytical certificate). For the high level additional 40 ppm formaldehyde were spiked into the corresponding formulation (total final content approx. 77 ppm). It was tested if residual formaldehyde in combination with higher level of metal ions present in Alum 4230 and possible other factors could further react with the virus leading to hyper-cross linking of surface proteins and loss of relevant epitopes.

[0117] Determination of other residual process related impurities:

[0118] Residual formaldehyde, sulphite and sucrose in final formulations were estimated based on GMP certificates for commercial drug substance JEV11A74. Results were recalculated by the actual 2-fold dilution of NIV to DS used within the DOE experiments.

[0119] Residual Sulphite: The concentration of residual sulphite was constant in all formulations with approx. 93 ppm.

[0120] Residual sucrose: The concentration of residual sucrose was constant in all formulations with approx. 1% v/w.

[0121] 1.1.2 DOE Design

[0122] These 5 factors were combined in a 25 DOE plan resulting in a total number of 34 experiments including 2 center points with the following base design. DOE planning and evaluation were carried out with an appropriate software

[0123] (Statgraphic Plus 3.0)

[0124] Base Design: Factorial 25

[0125] Number of experimental factors: 5

[0126] Number of blocks: 1

[0127] Number of responses: 1

[0128] Number of centerpoints per block: 2

[0129] Number of runs: 34

[0130] Error degrees of freedom: 18

[0131] Randomized: Yes

TABLE-US-00002 Low High Units Continuous.sup.1) Factors pH 7.0 8.0 Yes Alum.sup.2) 0.0 100.0 Alum 4230% Yes Spiked PS Fragments.sup.3 0 50 μg/mL No Spiked leachables.sup.4) 0 1.4 relative content No compared to FVL Spiked Formaldehyde.sup.5) 0 40 ppm No Responses ELISA (monoclonal, polyclonal) of AU/mL desorbed antigen .sup.1)Continuous means that a center point (mean value from high and low level) is present in the study design. No Center point means that only low and high levels are present in the study design. .sup.2)Low level (0%) means that formulation was prepared with Alum Lot 4074. High level (100%) means that formulation was prepared with Alum lot 4230. For center point formulations an equal mixture (50/50%) of both Alum lots was used. .sup.3)Since PS is present in NIV used for preparation of drug product samples, actual PS concentration in non-spiked formulations was <5 μg/mL and ~50-55 μg/mL for PS-spiked formulations. .sup.4)No leachables were assumed in non-spiked formulations since samples were prepared/stored in low-bind Eppendorf tubes. Total content of leachables from chlorobutyl rubber syringe plunger in spiked samples was approx. 1.4 times higher compared to FVL. .sup.5)Actual formaldehyde concentration in non-spiked DP samples was approx. 37 ppm, total formaldehyde concentration in spiked samples was approx.77 ppm.

2 Definitions & Abbreviations

[0132] AcCN Acetonitrile

[0133] DOE Design of experiments

[0134] DS Drug substance

[0135] FVL Final Vaccine lot

[0136] HPLC High performance liquid chromatography

[0137] ICP-MS Inductively-coupled-plasma mass spectrometry

[0138] PS Protamine sulphate

[0139] RP Reversed Phase

[0140] SEC Size exclusion chromatography

[0141] SN Supernatant

[0142] TFA Trifluoroacetic acid

[0143] w/o without

3 Materials and Methods

[0144] 3.1 DOE Studies

[0145] 3.1.1 Materials [0146] Syringe plunger stoppers: PH701/50/C black Sil6 7002-1051 (obtained from West, Order No. 2116) [0147] 100 mL Glass bottle (Schott)+Teflon coated screw cap [0148] Aluminium foil [0149] HQ water [0150] Electrical water bath (IKA, HBR 4 digital) [0151] LoBind Eppis 2 ml (Eppendorf, Cat. No. 0030 108.132) [0152] Speed Vac (Christ, RVC-2-25) [0153] HPLC vials, clear glass, 900 μL, Chromacol (VWR, Cat. no.: 548-1124) [0154] HPLC vials, PP, 900 μL, (Agilent, Item no. 5182-0567) [0155] Caps for HPLC vials, pre-cut (VWR, Cat. no.: 548-1260) [0156] 15 ml Falcon tubes (Greiner, Cat. No. 188724) [0157] Alum batch 4470 (RQCS 1342); Alum Lot 4230 (RQCS 1200) [0158] 10×PBS (Gibco, Order No. 14200-091) [0159] Parafilm [0160] Waters Atlantis T3 column; 3 μm particle diameter; column diameter/length 2.1×100 mm (Order No. 186003718; Lot 0107372331) [0161] Acetonitril (Merck, Cat No. 1.13358.2500) [0162] TFA (Sigma, Order No. 302031 [0163] HPLC system Dionex 3000 [0164] Solvent Rack SR-3000 [0165] Pump UltiMate-3000, analytical low pressure gradient pump [0166] Autosampler WPS-3000 TSL, analytical autosampler—temperature controlled [0167] Column compartment TCC-3200, temperature controlled [0168] PDA-Detector PDA-3000 [0169] Formaldehyde solution 37% (Merck, Cat No. 1.040031000) [0170] Protamine sulphate (Intercell Biomedical Ltd, Batch no. 086056) [0171] Ultrafilatrion device (Amicon® Ultra 3 kDa) (Millipore, Cat No. UFC900324) [0172] Incubator Infors HT Incubator Multitron Standard (InforsAG) [0173] Trypsin (Sigma, Order No: T0303)

[0174] 3.1.2 Procedure for Preparation of Extractables from Syringe Plunger

[0175] Syringe plunger (made of chlorobutyl PH701/50 black) that are currently used in the FVL container closure system were obtained from West (Germany). Therefore a stock solution of leachables was prepared by heat treatment of syringe plunger in water (90° C./2 h) followed by concentration in a speed-vac. The relative content of leachables in this stock solution was estimated by RP-HPLC using a C18 column (Atlantis T3 column) and compared to FVL JEV09L37 supernatant.

[0176] Extraction Method

[0177] A 100 mL Schott glass bottle with a Teflon coated screw top and a piece of aluminum foil were washed with hot water and thoroughly rinsed with HQ water. 30 stoppers were filled in the bottle and 30 ml of HQ-water were added. The bottle was closed with the aluminum foil fitted between bottle and screw and sealed additionally with Parafilm. The bottle was heated in the water bath to 90° C. for 2 hours and allowed to cool to room temperature. The extract was transferred to 14 low-bind Eppendorf tubes (a total of 28 mL extract was recovered). Twelve vials (total of 24 mL) were concentrated in a Speed Vac for approximately 44 hours and pooled into a falcon tube to obtain 6 ml of 4× concentrated stopper extract. A control sample containing 30 mL HQ water w/o stopper was prepared in the same way to evaluate any possible contamination.

[0178] C18 RP-HPLC Method

[0179] Leachables were separated by RP-HPLC C18 column (Atlantis T3) operated at 40° C. and 0.25 mL/min. Solvent A was 0.1% TFA in H2O, solvent B was 0.1% TFA in AcCN. Separation was performed by linear gradient ranging from 0 to 95% B in 30 min. Detection was done at 214 nm, 254 nm and 280 nm. The total relative concentration of concentrated stopper extract was estimated to be 80 fold higher compared to peaks detected in Final Vaccine Lot supernatant (FVL SN; obtained by removal of Alum particles by centrifugation at 5000 g/5 min) as detected at 254 nm. Therefore a total relative content of 80 U/mL (arbitrary Units U) were assigned for the stock solution, whereas the total relative concentration of leachables in FVL SN was set to 1 U/mL. For DOE studies, the stock solution was diluted 16-fold into the respective formulations yielding approx. 5 U/mL of total extractables.

[0180] 3.1.3 Preparation of Protamine Sulfate Fragments

[0181] A stock solution of PS fragments was prepared by digesting a PS solution (2 mg/mL in PBS) with Trypsin (200 ng/mL for 60 min at 37° C.). The enzyme was subsequently inactivated by heat (90° C. for 10 min) followed by ultrafiltration using a 3 kDa membrane (Amicon® Ultra centrifugal filter). Due to the cut-off of the membrane Trypsin remained in the retentate, whereas the PS fragments were present in the permeate. Complete inactivation of the enzyme was evaluated by spiking 500 μg/mL of full length PS into an aliquot of the obtained PS fragment followed by incubation at 37° C. for 18 h. No degradation of full length PS was observed indication complete inactivation/removal of Trypsin. Degradation was monitored by PS-SEC HPLC.

[0182] 3.1.4 DOE Plan

[0183] Samples were prepared according to the pipetting scheme as shown in Table 2. NIV Batch JEV11A74 obtained from a commercial production run was used as starting sample. NIV was diluted 2-fold to DS using PBS buffer followed by pH adjustment. 5 mL aliquots were removed and adjuvanted with the corresponding Alum lot 4230, 4074 or a 50/50% mixture of both. The final amount of Alum stock (2% Al2O3) added was 500 μg/mL Aluminium (0.1% Al2O3). Each formulation (5 mL) was split into two parts (2×2.5 mL) using Lo-bind Eppendorf tubes. One aliquot was stored at 2-8° C., another aliquot stored at 22±1° C. (Infors HT Incubator) under gentle shaking (20 rpm).

[0184] 3.2 Inactivated JEV ELISA (Polyclonal Based)

[0185] Desorption of the antigen from Alum and ELISA analysis was carried out using polyclonal sheep anti JEV antibodies for coating the 96 well ELISA plates as described in Example 4.

[0186] 3.3 Inactivated JEV ELISA (Monoclonal Based)

[0187] A monoclonal (mAb) based JEV ELISA was developed. The assay is primarily based on the “polyclonal JEV ELISA” assay format, only a monoclonal anti-JEV antibody (clone 52-2-5) is used for coating. The employed mab 52-2-5 was shown to be specific for JEV and to recognize a neutralizing epitope. Mab clone 52-2-5 was obtained by subcutaneously immunizing BALB/c mice with commercially available vaccine lot JEV08J14B. Spleen cells of the mice were fused to myeloma cells. From resulting hybridoma cells single clones were selected and sub-cloned. The clones were negatively screened against Bovine Serum Albumin, Protamine sulphate and an extract of the production cell line of the JE-vaccine (Vero cells). A positive screen was done against Neutralized Inactivated Virus (NIV) of vaccine lot JEV08M20. For screening, microtiter plates were coated with the relevant antigen and reacted with supernatant of cultures of the selected clones. For detection a goat anti mouse polyclonal antibody conjugated with alkaline phosphatase was used. Mab clone 52-2-5 was shown to recognize a neutralizing epitope on domain III of the envelope (E) protein of JEV containing Ser331 and Asp332 (Lin C.-W. and Wu W.-C. J Virol. 2003; 77(4):2600-6). Binding of the mab to the indicated neutralizing epitope is for instance determined as described in Lin and Wu (2003) by site-directed mutagenesis of the domain III at position 331 (for instance: S.fwdarw.R), and/or by alanine mutations at or near position 331 of domain III, for instance of residues Ser 331 and Asp332, followed by immunoblots to determine binding of the mab to the mutated proteins. Negative binding results indicate that the epitope of the mab is the neutralizing epitope identified by Lin and Wu (2003). The neutralizing characteristic of the epitope gives rise to the assumption that the epitope might be of importance for the antigen to elicit a protective immune response.

[0188] JEV samples were analyzed by both ELISA assays, polyclonal and monoclonal. The relative specific epitope content can be expressed as the ratio of the total antigen content determined by “monoclonal ELISA” (clone 52-2-5) divided by total antigen content determined by “polyclonal ELISA”. Any differences in the ratio may indicate differences in specific epitope content 52-2-5. Results close to 1 would correspond to high epitope contents, and results close to 0 correspond to low relative epitope content. A low ratio indicates presence of structural changes of the neutralizing epitope.

[0189] In the course of development of this “mAb ELISA”, differences between vaccines lots were detected, which could be correlated with potency results of these lots.

[0190] 3.4 Protamine Sulfate SEC-HPLC

[0191] PS (full length) and its fragments were analyzed by size-exclusion HPLC (SEC-HPLC) using a Superdex Peptide 10/300 GL, 10×300 mm, 13 μm (GE Healthcare) using 0.1% (v/v) Trifluoroacetic acid (TFA) in 30% acetinitrile (CAN) as mobile phase at a flow rate of 0.6 mL/min. PS containing samples were prepared in duplicated, i.e. diluted with mobile phase before injection.

4 Results

[0192] 4.1 Analysis of Stopper Leachables Used for Spiking Experiments

[0193] RP-HPLC elution profiles of concentrated stock solution obtained after extraction of stoppers under heat compared to FVL SN is shown in FIG. 1. Similar peak pattern as observed for both samples. Due to the harsh extraction conditions additional peaks were detected in the concentrate that were not present in FVL samples or present only at a much lower relative content Therefore the spiked formulations represent a “worst case” with regard to leachables and extractables. The total relative content of individual peaks in extract concentrate and spiked formulation in comparison to FVL SN is summarized in Table 3. The total amount of leachables in the stock solution was calculated as the sum of all peaks detected and expressed in arbitrary units as 67 U/mL. Since the stock solution was diluted 16 times into the respective formulation, the resulting total content of leachables was estimated as 4.2 U/mL. This corresponds on average 1.4 fold increase compared to FVL JEV09L37 supernatant (3.0 U/mL).

[0194] 4.2 Analysis of Protamine Sulphate Fragments

[0195] PS fragments obtained after cleavage of full length PS by Trypsin are shown in FIG. 2. Similar peak profiles of Trypsin treated PS and already degraded PS present in NIV11A74 were obtained by HPLC (see FIG. 3).

[0196] 4.3 DOE Evaluation

[0197] Formulations prepared for this DOE were analyzed after 4 weeks and 8 weeks of incubation at accelerated conditions (22° C.). It was considered that any degradation reaction would be accelerated when stored at higher temperature compared to normal storage conditions (2-8° C.). However, samples are still stored at 2-8° C. and will be analyzed on a later time point (˜4-6 month). First analysis of samples stored at 22° C. for 4 and 8 weeks are shown in the Table 4.

[0198] 4.3.1 Doe Evaluation after 4 Weeks at 22° C.

[0199] Statistical evaluation of the DOE matrix results obtained after 4 weeks at 22° C. showed that the specific epitope content 52-2-5 (expressed as the ratio of desorbed antigen analyzed by monoclonal/polyclonal ELSIA) was statistically significant influenced (95% confidence level, see Table 5) by the following factors: [0200] lower specific epitope content 52-2-5 in presence of Alum lot 4230 [0201] lower specific epitope content 52-2-5 at lower pH 7 [0202] Higher specific epitope content 52-2-5 at increased concentration of formaldehyde

[0203] Presence of higher concentration of PS fragments and chlorobutyl rubber leachables did not show any influence on specific epitope content. No 2nd or higher order interactions between individual parameters were detected.

[0204] The ANOVA table partitions the variability in “Ratio 4 weeks” into separate pieces for each of the effects. It then tests the statistical significance of each effect by comparing the mean square against an estimate of the experimental error. In this case, 3 effects (Alum, pH, Formaldehyde) have P-values less than 0.05, indicating that they are significantly different from zero at the 95.0% confidence level. The R-Squared statistic indicates that the model as fitted explains 74.85% of the variability in Ratio 4 weeks. The adjusted R-squared statistic, which is more suitable for comparing models with different numbers of independent variables, is 51.27%. The standard error of the estimate shows the standard deviation of the residuals to be 0.063. The mean absolute error (MAE) of 0.0353 is the average value of the residuals. The Durbin-Watson (DW) statistic tests the residuals to determine if there is any significant correlation based on the order in which they occur in your data file. Since the DW value is greater than 1.4, there is probably not any serious autocorrelation in the residuals. Effects are also displayed by standardized Pareto chart and main effect plots as shown in FIG. 4.

[0205] 4.3.2 DOE Evaluation after 8 Weeks at 22° C.

[0206] Statistical evaluation of the DOE matrix results obtained after 8 weeks at 22° C. were similar to results obtained after 4 weeks. Evaluation shows that the specific epitope content 52-2-5 (expressed as the ratio of desorbed antigen analyzed by monoclonal/polyclonal ELSIA) was statistically significant influenced (95% confidence level, see Table 6) by the following factors: [0207] lower specific epitope content 52-2-5 in presence of Alum lot 4230 [0208] lower specific epitope content 52-2-5 at lower pH 7

[0209] Presence of higher concentration of PS fragments, chlorobutyl rubber leachables and formaldehyde did not show any influence on specific epitope content. Note that P-value for formaldehyde (P=0.08) is quite close to be significant. No 2nd or higher order interactions between individual parameters were detected.

[0210] The ANOVA table partitions the variability in “Ratio 8 weeks” into separate pieces for each of the effects. It then tests the statistical significance of each effect by comparing the mean square against an estimate of the experimental error. In this case, 2 effects (Alum and pH) have P-values less than 0.05, indicating that they are significantly different from zero at the 95.0% confidence level. The R-Squared statistic indicates that the model as fitted explains 75.9% of the variability in “Ratio 8 weeks”. The adjusted R-squared statistic, which is more suitable for comparing models with different numbers of independent variables, is 53.3%. The standard error of the estimate shows the standard deviation of the residuals to be 0.095. The mean absolute error (MAE) of 0.057 is the average value of the residuals. Effects are also displayed by standardized Pareto chart and main effect plots as shown in FIG. 5. Regression analysis was also performed to the fitted data and calculated regression coefficients are shown in Table 7. The regression equation is displayed below which has been fitted to the data including pH, Alum and Formaldehyde. The equation of the fitted model is

“Ratio 8 weeks”=0.0228125+0.113125*pH−0.00185625*Alum+0.0315625*Formaldehyde

where the values of the variables are specified in their original units, except for the categorical factors which take the values −1 for the low level and +1 for the high level. The contour of the estimated response and residual plot is shown in FIG. 6 and FIG. 7. The ratio increases when relative Alum content Lot 4230 decreases and pH increases.

[0211] Table 8 contains information about values of “Ratio 8 weeks” generated using the fitted model. The table includes: [0212] (1) the observed value of “Ratio 8 weeks” [0213] (2) the predicted value of “Ratio 8 weeks” using the fitted model [0214] (3) 95.0% confidence limits for the mean response

[0215] As shown the experimental results are well predicted by the regression model.

5 Summary

[0216] Out of the parameters tested, Alum lot 4230 was shown to contribute significantly to antigen degradation as analyzed by monoclonal/polyclonal ELISA under accelerated conditions (22° C., testing time points 4 and 8 weeks). DOE results obtained after 4 and 8 weeks at 22° C. show that Alum 4230 is the most significant factor with regard to antigen degradation as detected by the ratio of monoclonal/polyclonal ELISA. Formulations made with Alum 4074 (much higher purity with regard to residual metal ions) show in general much higher specific epitope content.

[0217] Formalaldehyde and pH also contributed to antigen stability, but to a lower extent. The effect of increased antigen stability in samples formulated with Alum 4230 at higher formaldehyde level was well demonstrated (e.g. samples #19 and 29). However, influence of formaldehyde was less pronounced after extended storage period (8 weeks at 22° C.).

[0218] Better stability of the antigen was observed at pH 8 compared to pH 7. Protamine sulphate and leachables from chlorobutyl rubber stopper did not contribute to the antigen degradation.

Example 2

[0219] In previous studies (see Example 1) Aluminium hydroxide Lot 4230 was identified a significant contributing factor to the observed antigen degradation in FVL09L37. In this particular lot of Aluminium hydroxide (Alum), a much higher residual metal ion content was observed compared to other Alum lots used for formulation of the inactivated JEV antigen. This Example summarizes additional studies carried out to evaluate the influence of metal ions on stability of inactivated JEV. Spiking studies were conducted with the antigen either present in inactivated neutralized virus (NIV) solution or in drug product (DP) suspension following formulation of the antigen with Aluminium hydroxide.

1 Study Description

[0220] It was previously shown that Alum lot 4230 contains much higher levels of residual metal ion impurities compared to other Alum lots used for formulation of JEV (see also Example 3). Additional studies were performed to evaluate the influence of metal ions on the stability and on a potential surface modification of JEV. The inactivated antigen was either present in neutralized inactivated virus (NW) solution or in drug product (DP) suspension following further dilution of NIV and formulation with Aluminium hydroxide. In another set of experiments, different Alum lots covering a broad content range of residual metal ions were used and formulated with a single defined NIV lot. All of these formulations still contained residual formaldehyde and bisulphite at representative concentrations compared to commercial product. Stock solution of metal salts were dissolved in water and spiked to the samples to the desired final concentration.

2 Definitions & Abbreviations

[0221] AcCN Acetonitrile

[0222] ANOVA Analysis of variance

[0223] DOE Design of experiments

[0224] DP Drug product

[0225] DS Drug substance

[0226] FBV Final bulk vaccine

[0227] FVL Final Vaccine lot

[0228] GI Gamma irradiated

[0229] HPLC High performance liquid chromatography

[0230] LSD Fisher's least significant difference

[0231] mAb Monoclonal antibody

[0232] NIV Neutralized inactivated virus

[0233] PS Protamine sulphate

[0234] RP Reversed Phase

[0235] SEC Size exclusion chromatography

[0236] SN Supernatant

[0237] TFA Trifluoroacetic acid

[0238] w/o without

3 Materials and Methods

[0239] 3.1 Materials [0240] Iron(II)chloride tetrahydrate (Sigma, Order no. 44939) [0241] Iron(III)chloride hexahydrate (Sigma, Order no. 31232) [0242] Nickle(II)sulphate hexahydrate (Sigma, Order no. N4882) [0243] Cobalt(II)chloride hexahydrate (Sigma, Order no. 31277) [0244] Copper(II)chloride dehydrate (Sigma, Order no. 807483) [0245] Zinksulphate heptahydrate (Sigma, Order no. 24750) [0246] Crom(III)chloride hexahydrate (AlfaAesar, Order no. 42114) [0247] Ethylenediaminetetraacetic acid disodium salt dehydrate (EDTA) (Sigma, E5134) [0248] Aqua bidest. (Fresenius Kabi, Art no. 0712221/01 A) [0249] 10×PBS (Gibco, Order No. 14200-091) [0250] Formaldehyde solution 37% (Merck, Cat No. 1.040031000) [0251] Protamine sulphate (Intercell Biomedical Ltd, Batch no. 086056) [0252] LoBind Eppis 2 ml (Eppendorf, Cat. No. 0030 108.132) [0253] 15 ml Falcon tubes (Greiner, Cat. No. 188724) [0254] Incubator Infors HT Incubator Multitron Standard (InforsAG) [0255] 0.2 μm filter Mini Kleenpak 25 mm (Pall) [0256] NIV11A74 and Final bulk vaccine (FBV, formulated with Alum lot 4539) JEV 11D87 from commercial production runs was obtained from Intercell Biomedical (Livingston, UK) and stored at 2-8° C. until further processing [0257] Stock solutions of metal salts in water (final concentration 1 mM) used for spiking experiments were prepared and stored at 2-8° C. until usage [0258] Aluminium hydroxide samples (2% Al2O3, Brenntag Biosector) were either retain samples obtained from Intercell Biomedical or purchased directly from Brenntag. Alum samples were stored at 2-8° C. The following Alum lots were used in this study: 4470, 4563, 4621, 3877, 4230 (non-gamma irradiated and gamma irradiated)

[0259] 3.2 Preparation of Metal Stock Solutions

[0260] 3.2.1 Iron(II) Stock Solution

[0261] 20 mM Iron(II) stock solution was prepared by dissolving 397 mg of

[0262] Iron(II)chloride tetrahydrate in 100 mL aqua bidest.

[0263] 3.2.2 Iron(III) Stock Solution

[0264] 20 mM Iron(III) stock solution was prepared by dissolving 540 mg of Iron(III)chloride hexahydrate in 100 mL of aqua bidest.

[0265] 3.2.3 Nickle(II) Stock Solution

[0266] 20 mM Nickle(II) stock solution was prepared by dissolving 525 mg of Nickle(II)sulphate hexahydrate in 100 mL of aqua bidest.

[0267] 3.2.4 Cobalt(II) Stock Solution

[0268] 20 mM Cobalt(II) stock solution was prepared by dissolving 476 mg of Cobalt(II)chloride hexahydrate in 100 mL of aqua bidest.

[0269] 3.2.5 Copper(II) Stock Solution

[0270] 20 mM Copper(II) stock solution was prepared by dissolving 341 mg of Copper(II)chloride dihydrate in 100 mL of aqua bidest.

[0271] 3.2.6 Zink Stock Solution

[0272] 20 mM Zink stock solution was prepared by dissolving 575 mg of Zinksulphate heptahydrate in 100 mL of aqua bidest.

[0273] 3.2.7 Crom(III) Stock Solution

[0274] 20 mM Crom(III) stock solution was prepared by dissolving 533 mg of Crom(III)chloride hexahydrate in 100 mL of aqua bidest.

[0275] 3.3 Preparation of Working Solutions

[0276] Working solutions of metal ions (1 mM final concentration if not otherwise stated) were prepared by dilution of metal ion stock solutions with aqua bidest and sterile filtration via 0.2 μm syringe filter.

[0277] 3.4 Preparation of Formulation

[0278] All formulations were prepared under sterile conditions. NIV and FBV obtained from commercial production runs were adjusted to the desired pH and spiked with aliquots of metal stock solution. All samples were stored in plastic tubes if not otherwise stated. In all formulations using Alum, the final Al content was 500 μg/mL, corresponding to 0.1% Al2O3. It has to be noted that metal ions, especially iron (II), iron (III) and to a certain extent Cr (III), formed a precipitate with the phosphate ions present in the buffer resulting in partial co-precipitation of the inactivated virus represented by the low recovery determined by size-exclusion HPLC (SEC-HPLC).

[0279] 3.4.1 Experiment 20110913(NIV): NIV Formulation at Different Metal Ion Concentration of Ni(II), Cu(II), Cr(III) with or w/o Presence of PS Fragments

[0280] NIV 11A74 was adjusted to pH 7 and pH 8 followed by spiking of metal ions (Ni(II), Cu(II), Cr(III)) at 100/500/1000 ng/mL final concentration. All formulations were stored in low-bind Eppendorf tubes at 22° C. Aliquots of all formulation were also prepared in presence of protamine sulphate fragments (50 μg/mL). This was done to evaluate for any effect of PS fragments on JEV stability in presence of metals. The preparation of Protamine Sulphate (PS) fragments is described in Example 1. Samples were prepared on the same day (see Table 9) and analyzed three weeks later. All samples were analyzed by SEC-HPLC, but only samples at pH 8 (#21-40) were analyzed by ELISA.

[0281] 3.4.2 Experiment 20110913(DP): DP Formulation at Different Metal Ion Concentration of Ni(II), Cu(II), Cr(III)

[0282] FBV 11D87 (formulated with Alum Lot 4539) was used in this study. FBV was adjusted to pH 7 and pH 8 and spiked with Ni(II)/Cu(II)/Cr(III) at 100, 500 and 1000 ng/mL to evaluate any metal ion concentration/pH depended effect. Table 10 shows the experimental design of this experiment. All formulations were stored in Falcon tubes at 2-8° C. and 22° C. Samples stored at 22° C. were analyzed by SEC-HPLC and ELISA after 5 weeks.

[0283] 3.4.3 Experiment 20110812-Metal Spiked DP

[0284] Final Bulk Vaccine 11D87 (formulated with Alum Lot 4539) was obtained from a commercial production run and used in this study. Residual formalin in DS was analyzed as 28.1 ppm, residual sulphites was 92.2 ppm. Actual content in DP can be considered to be in the same range. FBV JEV11D87 was adjusted to pH 7.0/7.4/7.8 and spiked with 500 ng/mL (final concentration) of Fe(II), Fe(III), Ni(II), Co(II), Cu(II), Zn(II). A metal ion mix formulation was also prepared containing all of the individual metal ions together in solution. Formulations with Cr(III) were prepared later on and Cr(III) was not included in metal ion mix. Control formulations were only adjusted to the desired pH, but not spiked with metals. All formulations (#1-24) were prepared on the same day and stored in Falcon tubes at 2-8° C. and 22° C.

[0285] Additional Cr(III) spiked samples (#25-27) were prepared by taking aliquots of the control samples stored at 2-8° C. and spiked with Cr(III) to a final concentration of 500 ng/mL. Formulations were stored at accelerated conditions (22° C.) only. Table 11 shows the experimental set-up of this experiment. All samples stored at 22° C. were analyzed by ELISA (monoclonal and polyclonal) after 4 weeks and 7 weeks.

[0286] 3.4.4 Experiment 20110819: DP Formulation Using Various Alum Lots

[0287] Spiking studies as described above can give first evidence of possible instability of the formulated antigen in presence of certain metals, but might not be completely representative of the real conditions where metals present in Aluminium hydroxide are incorporated in the three-dimensional structure of the gel resulting in different local concentration and orientation/accessibility. To overcome these limitations an initial study was started to simulate the real conditions. A single NIV batch (11A74) obtained from a commercial production run was formulated with various Alum lots produced by Brenntag covering a broad range of residual metals. 4.75 mL of NIV was mixed with 0.25 mL Alum (2%) in Falcon tubes. The final Aluminium hydroxide concentration was 500 μg/mL (=0.1% Al2O3). Formulated vaccine samples were stored at 2-8° C. and under accelerated conditions at 22° C. All of these Alum lots contained residual metal ions at different concentrations. Alum lot 4230 has the highest level for Fe, Cu, Ni and V (see Example 3). Note that metal ion valences cannot be specified by ICP-MS. A mixed Alum sample containing equal amounts of 4230 and 4074 was also prepared to get an “intermediate” level for Ni(II) and Cu(II). Samples were analyzed after 6 weeks of storage at 22° C. The residual amount of formaldehyde and sulphite estimated by recalculation from available DS analysis results corrected by dilution factor of NIV to DS was 76 ppm formaldehyde and 192 ppm sulphite respectively.

[0288] 3.1 Antigen Desorption from Aluminium Hydroxide for SEC-MALLS Analysis

[0289] Viral particles were desorbed from Aluminium hydroxide. ˜625 μL of DP was spun down (8° C., 5 min, 3300×g) and the supernatant was either discarded if not otherwise stated or analyzed by JEV-SEC-MALLS to detect the unbound antigen concentration. Viral particles were desorbed by suspending the Aluminium hydroxide particles with 62.5 μL 0.8 M potassium phosphate buffer (pH 8) containing BSA (50 μg/mL). BSA was added to the desorption buffer for SEC-MALLS analysis to minimize losses caused by unspecific adsorption of the antigen. After shaking (500 rpm) the Aluminium hydroxide particles for 10 min at room temperature, particles were removed by centrifugation and the supernatant was collected into a LoBind Eppendorf tube and the desorption procedure repeated on remaining sample. The pooled desorbed antigen (˜5× concentrated sample; final volume 125 μL; starting volume ˜625 μL) was then further analyzed by SEC-MALLS.

[0290] 3.2 SEC-MALLS HPLC Method

[0291] Desorbed antigen was analyzed by SEC-MALLS. In brief, following desorption of the antigen from Aluminium hydroxide 100 μL of the pooled desorbed material (˜5× concentrated) were subsequently loaded onto a Superose 6 10/300 GL SEC column. 1×PBS+250 mM NaCl was used as mobile phase. Ultraviolet (UV) 214 nm and MALLS signals of viral particles were recorded and analysed using Chromeleon and ASTRA software packages.

[0292] 3.3 Inactivated JEV ELISA (Polyclonal Based)

[0293] Desorption of the antigen from Alum and ELISA analysis was carried out using polyclonal sheep anti JEV antibodies for coating the 96 well ELISA plates as described in Example 4.

[0294] 3.4 Inactivated JEV ELISA (Monoclonal Based)

[0295] During course of this investigational testing, a monoclonal (mAb) based JEV ELISA was developed. The assay is primarily based on the “polyclonal JEV ELISA” assay format, only a monoclonal anti-JEV antibody (clone 52-2-5) is used for coating and the current polyclonal antibody for detection. The employed mab 52-2-5 was shown to be specific for JEV and to recognize a neutralizing epitope. Mab clone 52-2-5 was obtained by subcutaneously immunizing BALB/c mice with commercially available vaccine lot JEV08J14B. Spleen cells of the mice were fused to myeloma cells. From resulting hybridoma cells single clones were selected and sub-cloned. The clones were negatively screened against Bovine Serum Albumin, Protamine sulphate and an extract of the production cell line of the JE-vaccine (Vero cells). A positive screen was done against Neutralized Inactivated Virus (NW) of vaccine lot JEV08M20. For screening, microtiter plates were coated with the relevant antigen and reacted with supernatant of cultures of the selected clones. For detection a goat anti mouse polyclonal antibody conjugated with alkaline phosphatase was used. Mab clone 52-2-5 was shown to recognize a neutralizing epitope on domain III of the envelope (E) protein of JEV containing Ser331 and Asp332 (Lin C.-W. and Wu W.-C. J Virol. 2003; 77(4):2600-6). Binding of the mab to the indicated neutralizing epitope is for instance determined as described in Lin and Wu (2003) by site-directed mutagenesis of the domain III at position 331 (for instance: S.fwdarw.R), and/or by alanine mutations at or near position 331 of domain III, for instance of residues Ser 331 and Asp332, followed by immunoblots to determine binding of the mab to the mutated proteins. Negative binding results indicate that the epitope of the mab is the neutralizing epitope identified by Lin and Wu (2003). The neutralizing characteristic of the epitope gives rise to the assumption that the epitope might be of importance for the antigen to elicit a protective immune response.

[0296] JEV samples were analyzed by both ELISA assays, polyclonal and monoclonal. The relative specific epitope content can be expressed as the ratio of the total antigen content determined by “monoclonal ELISA” (clone 52-2-5) divided by total antigen content determined by “polyclonal ELISA”. Any differences in the ratio may indicate differences in specific epitope content 52-2-5. Results close to 1 would correspond to high epitope contents, and results close to 0 correspond to low relative epitope content. A low ratio indicates presence of structural changes of the neutralizing epitope.

[0297] In the course of development of this “mAb ELISA”, differences between vaccines lots were detected, which could be correlated with potency results of these lots.

[0298] 3.5 Statistical Evaluation

[0299] Statistical evaluation was done with Statgraphic Plus 3.0.

4 Results

[0300] 4.1 Experiment 20110913(NIV): NIV Formulation at Different Metal Ion Concentration of Ni(II), Cu(II), Cr(III) with or w/o Presence of PS Fragments

[0301] SEC-HPLC results of NIV formulations (pH 7 and pH 8) containing metal ions [Ni(II), Cu(II), Cr(III)] w/and w/o PS fragments are summarized in Table 12. SEC-HPLC results show that antigen recoveries of most of the samples was >80%. Some samples (#7, #36, #38) showed slightly reduced recoveries in the range of 70-80%. It has to be noted that the actual virus content is quite low and precision of HPLC results can be estimated as approx.±20%. Since for samples #36 and #38 the recoveries for following formulations (#37, #39) at next level of individual metal ion content were higher again, these differences might be caused by assay variability and were not considered as significant. Based on the results obtained it was not possible to clarify the influence of metal ions with respect to the recovery of soluble inactivated JEV. However, SEC-HPLC only gives information about content of soluble virus, but no information about any potential surface modification. Only formulations prepared at pH 8 were also analyzed by ELISA (duplicate analysis). The ratio of monoclonal/polyclonal ELISA was calculated and can be used for comparison purpose of results. Analysis of samples by ELISA (see Table 13) do not show any significant influence of tested metals on degradation of inactivated JEV at pH 8 after three weeks at 22° C. There might be a trend of decreasing ratio in presence of Cu(II), but overall it appears that an incubation time of three weeks at 22° C. seems not be sufficient to detect any significant degradation. As also shown in DOE experiment (Example 1) inactivated JEV appears to have higher stability at pH 8 when stored at accelerated conditions at 22° C. and this would also contribute that significant effects were not observed. In this experiment it was also shown that PS fragments do not have any influence on JEV stability. This is also well in agreement with DOE results. NIV samples 1-20 formulated at pH 7 showed significant reduction in monoclonal epitope content in presence of Cu(II). At the highest tested concentration (1000 ng/mL) the ratio was close to zero indication significant structural changes of the antigen.

[0302] 4.2 Experiment 20110913(DP): DP Formulation with Different Metal Ion Concentration of Ni(II), Cu(II), Cr(III)

[0303] Analysis of desorbed JEV antigen is summarized in Table 14 (SEC-HPLC) and Table 15 (ELISA). Antigen recoveries for all samples as determined by SEC-HPLC was >80% after 5 weeks at 22° C. indicating no significant influence of tested metal ions on desorption recovery. As shown in FIG. 8, there is a trend of decreasing ratio as analyzed by ELISA in presence of Cu(II) and Cr(III) at pH 7. Formulations at pH 8 appear to be more stable.

[0304] 4.3 Experiment 20110812(DP): Metal Ion Spiked DP

[0305] In this experiment FBV11D87 was used as starting material. The formulation pH value was adjusted in a more narrow range (pH 7.0, 7.4, 7.8) and additional metal ions were used for spiking, each at 500 ng/mL (final concentration). The metal ion mix contained all individual metal ions with the exception of Cr(III) in single formulations (each metal at 500 ng/mL). ELISA results obtained after 4 weeks and 7 weeks at 22° C. are summarized in Table 16. Results are also displayed as graphs in FIG. 9.

[0306] Statistical evaluation of stability samples stored at 22° C. for 7 weeks was performed. ANOVA (analysis of variance) showed significant effects of parameters (pH and metal type) on antigen stability, that is expressed as the ratio of monoclonal/polyclonal ELISA (see Table 17). The ANOVA table decomposes the variability of Ratio into contributions due to various factors. Since Type III sums of squares have been chosen, the contribution of each factor is measured having removed the effects of all other factors. The P-values test the statistical significance of each of the factors. Since the P-values for pH and metal ion type are less than 0.05, these factors have a statistically significant effect on Ratio at the 95.0% confidence level.

[0307] In Table 18 a multiple comparison procedure is applied to determine the significance of the differences observed with respect to the means. Significant effect on ratio was shown for Cu(II) and the metal mix compared to the non-spiked control formulations. The bottom half of the output shows the estimated difference between each pair of means. An asterisk has been placed next to 7 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. At the top of the page, 3 homogenous groups are identified using columns of X's. Within each column, the levels containing X's form a group of means within which there are no statistically significant differences. The method currently being used to discriminate among the means is Fisher's least significant difference (LSD) procedure. With this method, there is a 5.0% risk of calling each pair of means significantly different when the actual difference equals 0.

[0308] Significant effect on ratio was shown for Cu(II) and the metal ion mix. The influence of other metal ions might become significant at longer storage period. The metal ion mix contained the highest total concentration and might represent a worst case. However, it was concluded that several metal ions present in Alum might contribute to degradation, each to different extent. These results further support the proposed root cause of metal ion-catalysed antigen degradation. It has to be noted that spiking experiment might not fully simulate the real conditions of residual metal ion impurities present in Alum 4230. Metal ions are incorporated in the Alum structure and the local concentration and orientation might be different from metal ions used in spiking experiments. It is also known that metal ions (e.g. Fe) have low solubility in presence of phosphate ion (PO43-). Therefore the actual concentration of soluble metal ions and contribution of metals present as metal-phosphate complex on JEV degradation is unknown.

[0309] 4.4 Experiment 20110819: Preparation of DP Samples with Different Alum Lots

[0310] Spiking studies as described earlier can give first evidence of possible instability of the formulated antigen in the presence of some metal ions, but might not be completely representative of the real conditions where these metal ions present in Aluminium hydroxide are expected to be incorporated in the three-dimensional structure of the Aluminium-hydroxide gel resulting in different local concentration and orientation/accessibility. To overcome these limitations an initial study was started to simulate the real conditions. A single NIV (11A74) was formulated with various Alum lots obtained by Brenntag covering a broad range of residual metals. Formulated vaccine samples were stored at 2-8° C. and under accelerated conditions at 22° C. All of these Alum lots contained residual metal ions at different levels. Lot 4230 had the highest level for Fe, Cu, Ni and V (see Table 19). Note that the actual content of metal ions present in the formulated product is only 1/20 of the concentration in Alum (2%) stock solution. Note that metal ion valences cannot be specified by ICP-MS. Analysis of desorbed antigen by ELISA of samples stored at 22° C. for 6 weeks is shown in Table 20.

[0311] Pooled standard deviation was calculated from all samples (spooled ˜0.075) as a measurement of experimental uncertainty. Mean values for ratio and 95% confidence intervals (calculated based on spooled) were plotted against the individual formulations (see FIG. 10). Samples formulated with Alum 4230 showed a trend to lower ratio compared to the other samples. However, differences were not as large to show statistical significant differences between the various formulations.

5 Summary

[0312] It was shown that metal ions contribute to the degradation of the inactivated JEV under accelerated storage conditions (22°). In spiking studies higher concentration of residual metal ions (range 100-1000 ng/mL) were used than present in FVL formulated with Alum lot 4230 (e.g. Fe˜310 ng/mL; Cr˜64 ng/mL; Ni˜52 ng/mL). Cu content in FVL can be only estimated as 3 ng/mL based on ICP-MS data of 2% Alum stock solution since LOD is 25 ng/mL. Higher metal concentration and storage temperature were chosen to increase the rate of any potential degradation reaction. In fact, for FVL JEV09L37, the potency loss occurred after 11 month stored at 2-8° C. It was also shown that metals can form insoluble complexes with phosphate ions making estimation of actual levels of metals present difficult.

[0313] In spiking experiments ELISA results showed statistically significant structural changes of virus surface in as little as 4 weeks at 22° C. in presence of metal ions. It was shown that the ELISA ratios for formulations containing Cu(II) and metal ion mix (containing Fe(II), Fe(III), Co(II), Cu(II) and Zn(II)) were statistically significant lower compared to the non-spiked control formulation. Cu(II) was also found in Alum lot 4230 (2% stock solution) at 64 ng/mL, corresponding to ˜3 ng/mL in FVL. In all other Alum (2%) lots Cu(II) content was <25 ng/mL (below limit of detection).

[0314] For formulation experiments of the antigen using different Alum lots, longer storage time (>6 weeks at 22° C.) at accelerated conditions is required. There is a trend that formulations prepared with Alum 4230 showed lower ELISA ratios compared to other lots. Slower degradation rate compared to spiked formulation might be contributing to lower metal ion content in commercial Alum lots. It was also observed that the antigen shows higher stability at pH 7.5-8 compared to pH 7 and that PS fragments do not contribute to any degradation reaction. These results are in good agreement with DOE results described in Example 1.

Example 3

[0315] As part of the out-of-specification investigation concerning FVL JEV09L37, the used Aluminum hydroxide lot (lot 4230) was determined to be the most probable root cause for the observed loss of potency. Aluminum hydroxide (referred to as Alum during the manufacturing process of JE-PIV) is purchased from Brenntag Biosector as autoclaved suspension termed “ALHYDROGEL® Aluminium Hydroxide Gel Adjuvant”. Each batch is sterilized by radiation prior to use in the JEV production process.

[0316] A number of different ALHYDROGEL® batches were analyzed for appearance, metal ion content and physical properties.

1 Introduction

[0317] 1.1 Aluminum Hydroxide

[0318] Brenntag Biosector's ALHYDROGEL® has a specified Aluminum content of 10 mg/mL which translates to 2% Al2O3 and 3% Al(OH)3, respectively. Further specifications are Nitrogen (max 0.005%), free Sulphate (max 0.05%), total Sulphate (max 0.1%) and pH (6.5±0.5). it has a shelf life of 26 months when stored at room temperature.

[0319] 1.2 Generation of Aluminum Hydroxide

[0320] ALHYDROGEL® 2% (referred to as Aluminum hydroxide) is manufactured by Brenntag (CAS no. 21645-51-2).

[0321] 1.3 Use of Aluminum Hydroxide Lots in JEV Manufacturing

[0322] For the production of commercial JEV vaccine batches a total of 5 different Brenntag ALHYDROGEL® 2% lots have been used so far.

2 Definitions & Abbreviations

[0323] ALHYDROGEL® 2% Aluminum hydroxide solution (also referred to as alum)

[0324] DS/DP Drug Substance/Drug Product

[0325] ESG Environmental Scientifics Group

[0326] F-AAS Flame Atomic Absorption Spectrometry

[0327] FVL Final Vaccine Lot

[0328] GF-AAS Graphite Furnace Atomic Absorption Spectrometry

[0329] ICP-MS Inductively-Coupled-Plasma Mass Spectrometry

[0330] JEV Japan Encephalitis Virus

[0331] JE-PIV Japan Encephalitis Purified Inactivated Virus

[0332] LOQ Limit Of Quantification

[0333] P&TD Patch & Technical Development

[0334] PSD Particle Size Distribution

[0335] PZC Point of Zero Charge

[0336] QCI Quality Control Immunology

3 Materials and Methods

[0337] 3.1 ALHYDROGEL® batches

[0338] ALHYDROGEL® 2% lots: 3877, 4074, 4187, 4230, 4414, 4470, 4539, 4563, 4587, 4621 (not all Alum lots listed were used in the formulation of JE-PIV) ALHYDROGEL® 2% 7× washed lots: 4577, 4580, 4596 (sourced from Brenntag, not typical of the 2% Alum received for formulation)

[0339] 3.2 ALHYDROGEL® PSD Measurements

[0340] Aluminum hydroxide particle size distribution (PSD) was analyzed on a Malvern Mastersizer 2000 μP system with a 20 mL sample cell.

[0341] ALHYDROGEL® 2% bulk substance was diluted 1:20 in water and 1 mL was added to the sample cell. Final dilution of sample in sample cell was therefore 400 fold (0.005% Aluminum hydroxide).

[0342] 3.3 ALHYDROGEL® Zeta-Potential Measurements

[0343] Zetapotential and point of zero charge (PZC) was measured on a Malvern Zetasizer ZS system equipped with a MPT-2 autotitrator. ALHYDROGEL® 2% bulk substance was diluted 1:20 in PBS and equilibrated over night at room temperature. For recording of the charge titration curve the pH was adjusted using 100 mM HCl and 100 mM NaOH solutions. PZC was determined by extrapolation of the zero charge in the titration plot (intersection of titration curve and x-axis). Point of zero charge corresponds to the pH value where the surface of the sample has no net charge.

[0344] 3.4 Analysis of Metal Ions in Aluminum Hydroxide

[0345] Selected metal ions were analyzed either by inductively-coupled-plasma mass spectrometry (ICP-MS), flame atomic absorption spectrometry (F-AAS) and graphite furnace atomic absorption spectrometry (GF-AAS) at the Medical Laboratory Bremen (Germany). In short, samples containing Aluminum hydroxide were treated with conc. HNO3 under heat until a clear solution is obtained. The clear solution is then further diluted and analyzed. The presence and content of following metal ions were determined: Pb, Cd, Cr, Co, Fe, Cu, Ni, Ag, W and Al. Depending on the sample dilution, the limit of quantification (LOQ) was 5 to 25 ng/mL.

[0346] In addition a semi-quantitative 70-Element scan was performed by ESG (UK) using a combination of ICP-MS (Agilent 7500ce) and ICP-AES (Perkin Elmer Optima 4300DV), which were calibrated using certified standards. The element scan is a screening method and not as sensitive as trace metal analysis for selected metals as performed by Medical Laboratory Bremen. However, such a screening gives a good overview about the presence and levels of certain metals.

4 Results

[0347] 4.1 Determination of ALHYDROGEL® Particle Size Distribution

[0348] Table 21 summarizes PSD data of two sublots each of ALHYDROGEL® lots 4230 and 4740. Distribution results are shown in FIG. 11. Mean particle was ˜2-4 μm with populations of smaller (<1 μm) and larger (>20 μm) particles being present in all four samples. The four tested ALHYDROGEL® samples showed no significant difference in mean particle size distribution.

[0349] 4.2 Zeta-Potential Measurements

[0350] Two sublots each of ALHYDROGEL® lots 4230 and 4740 (2% stock solution diluted 20 fold in PBS and equilibrated overnight at RT before analysis) were analyzed for point of zero charge. Table 22 summarizes results of PZC for the four samples showing very similar PZC in PBS buffer. Titration curves are shown in FIG. 12. No difference in the titration curves and PZC could be observed between the four analyzed samples.

[0351] 4.3 Determination of Residual Metal Ion Content in ALHYDROGEL® Batches

[0352] The current limits for Fe in 2% Aluminum hydroxide solutions according to the Ph. Eur. are 15 ppm (=15 μg/mL) and a total maximum of 20 ppm (=20 μg/mL) for other heavy metals (such as Pb). However, a concentration of 15 ppm Fe would correspond to 0.27 mM Fe in solution. Taking into consideration that even trace amounts of residual metal ions can catalyze a variety of degradation reactions for proteins (e.g. oxidation, activation of proteases etc.) and that metals remain stable in solution, differences in metal ion content between Aluminum hydroxide lots might cause differences in antigen stability over time.

[0353] The concentrations of a number of metal ions in commercially available aluminum hydroxide lots were analyzed using ICP-MS. The results of these analyses are summarized in Table 23. Lots 4074, 4230, 4470, 4414 and 4539 were used in the production of commercial JEV batches. As a 2% ALHYDROGEL® stock solution equals an Al concentration of 10 mg/mL the quantification of Al content in the different samples can be used as reference for the results obtained for the other metal ions. Indeed an average Aluminum content of 10.3 mg/mL could be measured showing the accuracy and reproducibility of the method.

[0354] When comparing the different ALHYDROGEL® lots large variations in the amount of contaminating metal ions could be observed. Most notable contaminating metal ions are Fe, Cr and Ni which were detected in all batches. In addition lot 4230 contained detectable amounts of Cu which was below LOQ in all other batches.

[0355] However, it has to be noted that none of these metals were detected in quantities near the specifications of ALHYDROGEL® mentioned above. For example the highest concentration of iron found in lot 4230 was 5.6 μg/mL or roughly 40% of the permitted concentration.

[0356] An “improved” ALHYDROGEL® is washed 7 times with water during the purification step instead of only 4 times for standard ALHYDROGEL®. To test if this additional washing steps would result in reduced metal ion contamination three different lots (4580, 4596 and 4577) were analyzed. Results are included in Table 23. No difference in metal ions comparing to the standard grade ALHYDROGEL® could be observed suggesting that the metal ions are either strongly bound to the surface of the Aluminum hydroxide particles or actually co-precipitate during the production process.

[0357] FIG. 13 shows a comparison of the different ALHYDROGEL® lots analyzed. The total contaminating metal ion content for all tested contaminating elements are shown with the absolute shares for the three major metals Fe, Cr and Ni depicted in different colors. As can be seen lot 4074 has very little contaminating metal ions compared to the majority of the other analyzed batches. Only lot 3877 showed a similar lot contamination whereas lot 4230 shows by far the highest contamination of all batches analyzed during this investigation.

[0358] During the investigational testing a large variation in the metal ion content was observed between different batches of ALHYDROGEL® (see FIG. 13). To test if these contaminating metal ions are located in the Aluminum hydroxide fraction or in the supernatant Lot 4230 was separated into a supernatant and a sediment fraction (see Table 24). As can be seen less than 2% of the metal ions could be detected in the supernatant indicating that all contaminating metal ions are either bound to the Aluminum hydroxide particle surface or inside the particle structures. The local metal ion concentration can therefore be estimated to be at least 50-100 times higher since the solid volume fraction (volume of Alum-pellet after centrifugation) of 0.1% Al2O3 (corresponding 0.5 mg/mL Al) used in JEV vaccine formulation is approx. 10-20 μL per 1000 μL of FVL.

5 Summary

[0359] ALHYDROGEL® is used in a 0.1% final concentration as adjuvant in the current JEV vaccine formulation. During the investigation of an out-of-specification (OOS) potency result for production FVL JEV09L37 an evaluation of the ALHYDROGEL® production process was initiated. A total of 13 different ALHYDROGEL® lots were analyzed for the presence of contaminating metal ions that could reduce protein stability.

[0360] Large variations in the concentration of a number of metal ions were observed for different ALHYDROGEL® lots. When analyzing the raw materials it was shown that these contaminations were present at the same concentration as found within the ALHYDROGEL®.

[0361] Higher levels of Fe, Ni and Cu ions were noted in ALHYDROGEL® lot 4230 when compared to the other investigated lots. Lot 4230 was the only one where residual Cu ions were detected. This lot 4230 was used for the formulation of FVL JEV09L37.

[0362] When analyzing supernatant and insoluble fraction of an ALHYDROGEL® batch these contaminating metal ions could only be found in the precipitate indicating that these ions are either bound to the Aluminum hydroxide particle surface or actually part of the particle.

[0363] Although macroscopic and in composition different from other ALHYDROGEL® lots used for JEV production, lot 4230 fulfilled all requirements detailed by the Ph. Eur. Also physical characterization (particle size distribution and point of zero charge) showed no differences between lot 4230 and other ALHYDROGEL® lots not showing these high metal ion contaminations.

Example 4

[0364] 1.1. Materials, Equipment and Methods

[0365] 1.2. Equipment

[0366] Analytical balance (readability of 0.1 mg; e.g. Mettler Toledo XP205DR/M)

[0367] Precision balance (readability of 0.1 g; e.g. Mettler Toledo, Model No XS6002S Delta Range)

[0368] Filter Units 0.22 μm (e.g. Stericup Cat No SCGPV01RE) or 0.2 μm filter system (e.g. 50 mL Millipore Steriflip)

[0369] Freezer (−20° C.) and Ultra-Freezer (−80° C.)

[0370] Fridge (+2 to 8° C.)

[0371] Magnetic stirrer (e.g. KIKA Labortechnik RCT basic) and magnetic stir bars

[0372] Microplate Washer: e.g. BioTek ELx405

[0373] Microplate Reader: e.g. BioTek Synergy 2 and Gen5 Secure software

[0374] Microplate Incubator (37° C.)

[0375] Microtiter Sealing tape (e.g. Thermo Electron 9503130)

[0376] Multichannel pipettes and tips (e.g. Eppendorf Research Pro 50-1200 μL, Eppendorf Research, 10-100 μL)

[0377] pH meter (e.g. WTW Series ino Lab, Terminal 740 and pH/Cond. 740)

[0378] Pipettes and tips (e.g. Eppendorf Research, 0.5-10 μL, 2-20 μL, 20-200 μL, 100-1000 μL, 500-5000 μL)

[0379] Pipettor (e.g. IBS Biosciences Pipetboy)

[0380] PP Tubes 15 mL (e.g. Sarstedt 62.515.006) or PP tubes 50 mL (e.g. Greiner 227261)

[0381] Reagent Reservoir 50 mL (e.g. Corning Incorporated 4870)

[0382] Serological pipettes (e.g. Falcon, 2 mL, 5 mL, 10 mL, 25 mL, 50 mL)

[0383] Titertube Micro Tubes—Bulk (BioRad 223-9391)

[0384] Vortex mixer (e.g. VWR Analog Vortex Mixer, Model No 945304)

[0385] 1.5 mL or 2.0 mL Eppendorf LoBind tubes (Cat No 0030 108.116, CatNo 0030 108.132, respectively)

[0386] 96 well Microplate (F96 Cert. Maxisorp Nunc-Immunoplates)

[0387] For Analysis of DP Samples in Addition:

[0388] Bench top centrifuge (e.g. Beckman coulter, Microfuge 16 Centrifuge, Cat No A46473)

[0389] Orbital shaker (e.g. Eppendorfer Thermomixer compact)

[0390] 50 mL PP tubes (e.g. Greiner 227261)

[0391] 1.3. Reagents

[0392] PBS 10× (e.g. Gibco, Cat. No 14200-083)

[0393] Tween 20 (e.g. Sigma Cat. No P7949)

[0394] 2M Sulphuric Acid (Volumetric solution, e.g. Fisher, Cat. No. J/8410/17)

[0395] De-ionised water, e.g. (Milli-Q, 18.2Ω)

[0396] Sodium carbonate—bicarbonate capsules (e.g. Sigma, Cat. No. C3041)

[0397] Hydrochloric acid (HCl) 1 mol/L (e.g. Merck, Cat. No 1.09057.1000)

[0398] Sodium hydroxide (NaOH) 1 mol/L (e.g. Merck, Cat. No 1.09132.1000)

[0399] Glycerol (e.g. Sigma)

[0400] For Analysis of DP Samples in Addition:

[0401] Di-potassium hydrogen phosphate trihydrate (e.g. Sigma, Cat No. P5504)

[0402] Potassium di-hydrogen phosphate (e.g. VWR, AnalaR Normapur, Cat No. 26936.260)

[0403] Albumin, Bovine Serum (BSA), ELISA grade (e.g. Sigma, Cat. No. A3059)

[0404] TMB Substrate (e.g. BioFX, TMBW-1000-01)

[0405] Donkey anti rabbit IgG HRP Conjugate (Jackson Immuno Research, Cat No 711-035-152)

[0406] Reconstitution:

[0407] The content of 1 vial (0.4 mg) is reconstituted in 0.5 mL of de-ionised water and thoroughly mixed until total dissolution. Add 0.5 mL of Glycerol and mix it further until homogeneity. Aliquots are stored at −20° C. until use.

[0408] Inactivated JEV Reference Standard (Intercell Biomedical Ltd.)

[0409] Purified sheep anti-JEV (Intercell Biomedical Ltd.)

[0410] Purified rabbit anti-JEV (Intercell Biomedical Ltd.)

[0411] 1.4. Solutions

[0412] a) 0.05M carbonate buffer at pH 9.6 (used for coating of ELISA plates) For 100 mL buffer, dissolve one bi-carbonate/carbonate buffer capsule in 100 mL de-ionised water. Check the pH and adjust to 9.6±0.1 with HCl or NaOH if required. Use on the day of preparation only. Keep ELISA coating buffer at RT during the day of use, then discard.

[0413] b) ELISA wash buffer and part of block/sample diluent (PBS-T) Prepare approximately 1 litre for every plate used. Dilute 10×PBS stock 1+9 in de-ionised water, mix well and check pH (7.4+/−0.1), adjust with 1M HCl or 1M NaOH as required. Add 0.05% (v/v) TWEEN® 20, mix well.

[0414] e.g. ELISA wash buffer (PBS-T) [1L]:

[0415] 100 mL 10×PBS

[0416] 900 mL de-ionised water

[0417] Mix well, check/adjust pH (7.4+/−0.1).

[0418] 0.5 mL TWEEN® 20

[0419] Mix well.

[0420] Use on the day of preparation only; keep ELISA wash buffer at RT during the day of use, then discard.

[0421] c) Blocking solution: 5% BSA in PBS-T

[0422] Prepare approximately 25 mL for every plate. Measure required quantity of PBS-T into a clean glass bottle using a serological pipette. Add a clean magnetic stir bar. Weigh the required amount of BSA, add to the surface of the PBS-T and mix gently on a magnetic stirrer until all the BSA has gone into solution. Filter solution using a 0.2 μm filter (either Steriflip filter system or syringe filter).

[0423] e.g. Blocking solution [100 mL]

[0424] 5 g BSA

[0425] 100 mL PBS-T

[0426] Use on the day of preparation only; keep blocking solution at RT during the day of use then discard.

[0427] d) Sample diluent: 1% BSA in PBS-T

[0428] Prepare as above but using 1 g of BSA per 100 mL PBS-T, approximately 25 mL is required per plate.

[0429] e.g. Sample diluent [100 mL]

[0430] 1 g BSA

[0431] 100 mL PBS-T

[0432] Use on the day of preparation only; keep sample diluent at RT during the day of use, then discard.

[0433] For Analysis of DP Samples in Addition:

[0434] e) 1×PBS

[0435] Prepare 1 part 10×PBS with 9 parts de-ionised water

[0436] e.g. 1×PBS [100 mL]

[0437] 10 mL 10×PBS

[0438] 90 mL de-ionised water

[0439] Use on the day of preparation only; keep 1×PBS at RT during the day of use, then discard.

[0440] f) 20×ELISA buffer

[0441] Weigh an appropriate amount of BSA into a suitable container to make a 20× solution. Add the appropriate volume of 1×PBS. Add TWEEN® 20 to a final concentration of 0.05% Mix on the magnetic stirrer until the BSA is fully dissolved. Filter the solution through a 0.2 μm filter (using either STERIFLIP filter system or syringe filter) into a sterile container (and aliquoted as needed).

[0442] e.g. 20×ELISA Buffer [25 mL]

[0443] 5 g BSA

[0444] 25 mL 1×PBS

[0445] 12.5 μL TWEEN® 20

[0446] The solution can be stored at +2-8° C. for 1 week.

[0447] g) 2×ELISA buffer

[0448] It is prepared by dilution of the 20×ELISA Buffer with 1×PBS (1 part 20× ELISA Buffer and 9 part 1×PBS).

[0449] e.g. 2×ELISA Buffer [20 mL]

[0450] 2 mL 20×ELISA Buffer

[0451] 18 mL 1×PBS

[0452] Use on the day of preparation only; keep 2×ELISA buffer at RT during the day of use, then discard.

[0453] h) Desorption buffer [0454] Potassium phosphate stock solution: Make a 3× stock solution of potassium phosphate (2.4M) by dissolving the appropriate volume of di-potassium phosphate trihydrate and of potassium dihydrogen phosphate in de-ionised water. Place on a magnetic stirrer and once dissolved make up the required volume, check that the pH of the solution is 8.0+/−0.1. Filter through a 0.2 μm filter. [0455] e.g. 3× stock solution of Potassium phosphate (2.4M) [50 mL] [0456] 23.963 g Di-potassium phosphate trihydrate [0457] 2.041 g Potassium dihydrogen phosphate [0458] Make up to 50 mL De-ionised water [0459] Store at +2°−8° C. for up to 1 month. [0460] Make working strength desorption buffer (0.8M potassium phosphate buffer containing 1% BSA and 0.05% TWEEN® 20) by adding the appropriate volume of potassium phosphate stock (2.4M), of TWEEN® 20 and of BSA to the required volume of de-ionised water. Mix thoroughly and use on the day of preparation. [0461] e.g. working strength desorption buffer [15 mL] [0462] 5 mL Potassium Phosphate (2.4M) [0463] 7.5 μL TWEEN® 20 [0464] 0.15 g BSA [0465] 10 mL De-ionised water [0466] Keep working strength desorption buffer at RT during the day of use.

[0467] 1.5. Test Samples and Antibodies

[0468] Test samples: [0469] Drug Substance and/or NIV (various batches) [0470] JEV Vaccine samples (final bulk vaccine and final vaccine lot) Inactivated JEV Reference Standard (Neutralised Inactivated Virus—NIV) (Intercell Biomedical Ltd.)

[0471] Polyclonal Antibodies: [0472] Coating antibody: Purified Sheep anti-JEV (Intercell Biomedical Ltd.) [0473] Primary detection antibody: Purified Rabbit anti-JEV (Intercell Biomedical Ltd.)

[0474] Secondary conjugated antibody: Donkey anti-Rabbit HRP Conjugate (Jackson Immuno Research Cat. No 711-035-152)

2 Procedure

[0475] 2.1. Plate Coating [0476] Label the plate with plate number, date and analyst. [0477] Prepare fresh 0.05M carbonate buffer (pH 9.6) on the day of plate coating. Allow approximately 12 mL for each plate coated. [0478] Remove the required number of aliquots of the coating antibody from the freezer and allow thawing at RT. Prepare a dilution of Purified Sheep anti-JEV in carbonate buffer. Mix well by inversion of the tube. [0479] Using the multichannel pipette, apply 100 μL/well to a 96-well Maxisorp plate within 15 min of preparation of the antibody dilution. [0480] Cover with microtiter sealing tape and incubate 17 to 72 hrs at +2-8° C. 2.2. Washing [0481] Remove plate from the refrigerator and allow warming to room temperature. [0482] Wash the plate/s with the Microtiter plate washer 3 times using the respective wash program (300 μL per well, three times, final dispense). After that: remove any remaining wash buffer by decanting. Invert the plate and blot it against a clean paper towel. Do not allow microtiter plate to dry between wash steps and reagent addition.

[0483] 2.3. Blocking [0484] Prepare a Blocking Solution 5% (w/v) of BSA in PBS-T as above. [0485] Apply 200 μL Blocking Solution per well, cover the plate(s) with a cover plate and incubate at 37° C. for 1 hour+/−10 min.

[0486] 2.4. Preparation of Standard Curve Dilutions [0487] Remove the NIV reference standard from the freezer, allow thawing at RT, mix well. Prepare a 1 AU/mL stock dilution of the current reference standard; use at least 20 μL of NIV reference standard for dilution.

[0488] e.g. NIV reference standard Pre-dilution:

[0489] Concentration: 235 AU/mL (lot No 03/2009)

[0490] To prepare a 1 AU/mL working standard solution dilute it 1 to 235 in sample diluent:

[0491] 4680 μL sample diluent

[0492] 20 μL NIV reference standard [0493] Prepare then the following working standard solutions from the 1 AU/mL pre-dilution:

[0494] 0.8 AU/mL, 0.6 AU/mL, 0.4 AU/mL, 0.2 AU/mL, 0.1 AU/mL and 0.05 AU/mL in sample diluent.

[0495] 2.5. Quality Control Samples

[0496] a) Quality control (QC) samples (for example at 0.75, 0.30 and 0.18 AU/mL) should be made from the NIV reference standard pre-dilution freshly at the time of the assay then discarded once used.

[0497] b) These controls are part of the system suitability criteria and allow the performance of the assay to be monitored over time.

[0498] 2.6. Preparation of Test Samples

[0499] Drug Substance Preparation

[0500] Drug substance test samples are received for testing at unknown concentrations. These will be tested at six dilutions in triplicate. The dilutions will be made independently into the range of the standard curve, e.g. pre-dilution of 1 in 15 or other suitable dilution then six dilutions with sample buffer.

[0501] NIV Sample Preparation

[0502] NIV samples will be received for testing at unknown concentrations and pre-diluted in the range of the standard curve (e.g. 1 in 30 or other suitable dilution) then diluted six times in the same manner as the DS samples.

[0503] Drug Product Supernatant Preparation

[0504] a) For the analysis of Bulk-DP samples mix well sample by vortexing. Transfer exactly 1 mL into a 1.5 mL LoBind Eppendorf tube.

[0505] For the analysis of final product container samples transfer the content of 2 syringes (0.6 mL per syringe) of the same lot into a 1.5 mL LoBind Eppendorf Tube. Mix content of tube thoroughly by inversion to ensure homogeneity of the DP and transfer exactly 1 mL into fresh 1.5 mL LoBind Eppendorf tube.

[0506] b) Centrifuge tubes containing 1 mL DP each at 3300×g for 5 minutes.

[0507] c) For each sample, pipette 25 μL of 20×ELISA buffer into fresh LoBind Eppendorf tube.

[0508] d) Carefully remove 475 μL of the supernatant without disturbing the alum pellet and transfer into the tube containing the 20×ELISA buffer. Mix gently by inversion. Re-spin 2 min at 16,000×g. Store sample at +2-8° C. prior to analysis.

[0509] NOTE: DP supernatant samples prepared in this way should be measured neat in triplicate in the inactivated JEV ELISA.

[0510] e) Carefully remove as much of the residual supernatant from the centrifuged tube using a 10-200 μL pipette without disturbing the alum pellet and discard the supernatant.

[0511] f) The pellet obtained is subjected to the Desorption procedure as described below.

[0512] Drug Product Desorption Procedure

[0513] a) Add 158 μL of working strength desorption buffer to each pellet left in the LoBind tube.

[0514] b) Resuspend the pellet by pipetting up and down several times to ensure complete re-suspension of the pellet and homogenisation of the sample.

[0515] c) Incubate samples for 10 min at RT on an orbital shaker at 500 rpm.

[0516] d) After incubation centrifuge samples at 3300×g for 5 minutes.

[0517] e) For each sample pipette 250 μL of 2×ELISA buffer into a fresh LoBind Eppendorf tube.

[0518] f) Carefully remove 83.3 μL from the upper part of the supernatant containing the desorbed product without disturbing the pellet and transfer into the tube containing the 2×ELISA buffer. Remove remaining supernatant using a 20-200 μL pipette without disturbing the pellet and discard the supernatant.

[0519] g) Add another 158 μL of working strength desorption buffer to each pellet.

[0520] h) Carry out 2 more desorption cycles (3 in total) pooling the 3×83.3 μL of the desorbed material +250 μL ELISA buffer into the appropriate tube. After the last step the remaining pellet can be discarded.

[0521] i) The final concentration of the viral antigen in the desorbed pool(s) should now be the same as the original 1 mL of DP from which it was desorbed. Therefore, the concentration of inactivated JEV antigen content measured in the desorbed pool can be directly related to the original DP.

[0522] Note: Analyse the desorbed samples by ELISA on the day of desorption.

[0523] j) Dilution of desorbed DP samples:

[0524] These will be tested at six dilutions in triplicate. An appropriate pre-dilution will be performed in the range of the standard curve e.g. 1 in 15 (100 μL to 1400 μL diluent) or other suitable dilution, then six dilutions of 1 in 15 pre-dilution will be made independently using sample diluent.

[0525] 2.7. Sample Loading and Plate Plan

[0526] Prepare samples and standards before analysis.

[0527] After blocking wash the plate using the plate washer employing the JEV ELISA program. After that, remove any remaining wash buffer by decanting. Invert the plate and blot it against a clean paper towel. Do not allow microtiter plate to dry between wash steps and reagent addition.

[0528] Add 100 μL/well of standards/controls/samples and cover with cover plate and incubate for 1 hour+/−10 min at 37° C.

[0529] Add 100 μL of sample diluent to all wells not required for testing.

[0530] 2.8. Preparation of Primary Antibody

[0531] Remove the required number of aliquots of the primary antibody from the freezer and allow to thaw at RT. Prepare max 15 min before use Rabbit anti-JEV in sample diluent at a suitable dilution. Following sample incubation, wash the plate using the plate washer employing the JEV ELISA program. After that, remove any remaining wash buffer by decanting. Invert the plate and blot it against a clean paper towel. Do not allow microtiter plate to dry between wash steps and reagent addition.

[0532] Add 100 μL/well of diluted primary antibody, cover with cover plate and incubate for 1 hour+/−10 min at 37° C.

[0533] 2.9. Preparation of Secondary Antibody Conjugate

[0534] Remove the required number of aliquots of the secondary antibody conjugate from the freezer and allow to thaw at RT. Prepare max 15 min before use a dilution of Donkey anti-Rabbit-HRP in sample diluent; e.g. for a 1 in 10,000 dilution for make a 1 in 100 pre-dilution then make a second dilution of 1 in 100.

[0535] Following primary antibody incubation, wash the plate using the plate washer employing the JEV ELISA program. After that, remove any remaining wash buffer by decanting. Invert the plate and blot it against a clean paper towel. Do not allow microtiter plate to dry between wash steps and reagent addition. Add 100 μL/well of diluted secondary antibody conjugate, cover with cover plate and incubate for 1 hour+/−10 min at 37° C.

[0536] 2.10. Substrate Incubation

[0537] When the conjugate has been added, remove TMB from the 2-8° C. refrigerator. Pipette the required volume (12 mL of TMB per plate) into a 50 mL centrifuge tube, using a serological pipette. Allow the TMB to reach room temperature in the dark. Following conjugate incubation wash the plate 3 times with the plate washer employing the JEV ELISA program. After that: remove any remaining wash buffer by decanting. Invert the plate and blot it against a clean paper towel. Do not allow microtiter plate to dry between wash steps and reagent addition. Add 100 μL/well of TMB and develop the plate in the dark at Room Temperature for 10 minutes.

[0538] 2.11. Stopping and Reading

[0539] After 10 minutes of TMB incubation, stop the development by adding 100 μL/well 2M sulphuric acid. Read the plate at 450 nm (reference filter 630 nm) within 10 minutes of stopping using the BioTek reader and Gen5 Secure software.

[0540] 2.12. Data Analysis

[0541] NIV/DS Data Analysis:

[0542] Gen5Secure software will be used to calculate the % Recovery of the QCs, concentration×dilutions, mean concentration of the samples corrected for dilutions and this value multiplied by 1.05 to correct for the addition of ELISA buffer.

[0543] DP Data Analysis:

[0544] Gen5Secure software will be used to calculate the % Recovery of the QCs, concentration×dilutions and mean concentration of the dilutions for the samples.

[0545] For DP Supernatant Samples:

[0546] If the concentration of the supernatant sample is below the LLOQ of the assay (i.e. 0.05 AU/ml), then the supernatant sample should be recorded as <0.05 AU/ml

[0547] If the concentration of the supernatant sample is within the LOQs of the assay (i.e. 0.05 AU/ml to 1.25 AU/ml), the concentration value is recorded for the supernatant sample.

[0548] If the concentration of the supernatant sample is above the ULOQ of the assay (i.e. 1.25 AU/ml), the preparation of the drug product supernatant should be repeated. The supernatant sample will be re-tested by performing a suitable pre-dilution into the range of the standard curve followed by 6 sample dilutions. (The desorbed Drug Product sample does not need to be repeated.) The mean concentration for the dilutions that are within the LOQs of the assay (LLOQ 0.05 AU/ml to 1.25 AU/ml) will be the recorded concentration value for the supernatant sample, provided that the system suitability are met and at least 4 out of the 6 sample dilutions are within the LOQs.

[0549] 2.13. Assay Acceptance Criteria

[0550] a) The correlation coefficient for the calibration curve must be >0.980.

[0551] b) % CVs≦15% for standards and samples (except DP supernatant) for the four highest concentrations of the dilutions, % CV≦15% for controls

[0552] c) Individual blank ODs must be ≦0.2.

[0553] d) Assay controls must be within specified defined limits (for freshly prepared controls 2 out of 3 QCs should have observed concentrations within ±30% of the nominal values; OR the levels set during QC qualification) for the assay to pass.

[0554] e) Assay validity will be recorded on the Gen5-print-out. If the plate fails to meet the defined acceptance criteria the assay is deemed invalid.

[0555] 2.14. Reporting of Data NIV

[0556] a) Antigen Content

[0557] The reported value for inactivated AU/mL is the mean of the concentrations (which are within the LOQs of 0.04 to 1.25 AU/mL) calculated for the single sample dilutions corrected by the respective dilution factors, and the mean multiplied by a correction factor of 1.05 to account for the 5% volume of 20× ELISA buffer that was added to each sample when it was taken. Antigen content will be recorded on Gen5 print-out.

[0558] DS:

[0559] a) Identity

[0560] If the absorbances of the samples at lowest dilution (highest concentration) are higher than 3 standard deviations above the mean value of the blank the result will be reported as positive

[0561] b) Antigen Content

[0562] The reported value for inactivated AU/mL is the mean of the concentrations (which are within the LOQs of 0.04 to 1.25 AU/mL) calculated for the single sample dilutions corrected by the respective dilution factors, and the mean multiplied by a correction factor of 1.05 to account for the 5% volume of 20× ELISA buffer that was added to each sample when it was taken. Antigen content will be recorded on Gen5 print-out.

[0563] Desorbed DP:

[0564] a) Identity:

[0565] If the absorbances of the samples at lowest dilution (highest concentration) are higher than 3 standard deviations above the mean value of the blank the result will be reported as positive.

[0566] b) Antigen Content

[0567] The reported value for inactivated AU/mL is the mean of the concentrations (which are within the LOQs of 0.05 to 0.8 AU/mL) calculated for the single sample dilutions corrected by the respective dilution factors. Antigen content will be recorded on Gen5 print-out.

[0568] DP supernatant (degree of adsorption/degree of non-adsorption):

[0569] Degree of adsorption is reported in relationship to aluminium hydroxide formulated drug substance post filtration.

[0570] a) For calculation of the reported value, the reported antigen content (AU/mL) for the respective DS sample (post filtration) corrected for the dilution with aluminium hydroxide (5%) will be set at 100% and the percentage of the concentration measured in the supernatant (corrected for the addition of 5% ELISA buffer) calculated in relation to that. The reportable value will be the difference between 100% and the percentage calculated for the supernatant. Results will be reported to 2 decimal places.

[00001]$\mathrm{Degree}.Math..Math.\mathrm{of}.Math..Math.\mathrm{adsorption}.Math..Math.\left(\%\right)=100.Math.\%-\frac{\mathrm{DP}.Math..Math.\mathrm{supernatant}.Math..Math.\left(\mathrm{AU}.Math.\text{/}.Math.\mathrm{mL}\right)\star 1.05}{\mathrm{DS}.Math..Math.\left(\mathrm{AU}.Math.\text{/}.Math.\mathrm{mL}\right)\star 0.95}\star 100.Math.\%$

[0571] The degree of non-adsorption will be calculated as detailed below and results will be reported to 2 decimal places:

[00002]$\mathrm{Degree}.Math..Math.\mathrm{of}.Math..Math.\mathrm{non}.Math.\text{-}.Math.\mathrm{adsorption}.Math..Math.\left(\%\right)=\frac{\mathrm{DP}.Math..Math.\mathrm{supernatant}.Math..Math.\left(\mathrm{AU}.Math.\text{/}.Math.\mathrm{mL}\right)\star 1.05}{\mathrm{DS}.Math..Math.\left(\mathrm{AU}.Math.\text{/}.Math.\mathrm{mL}\right)\star 0.95}\star 100.Math.\%$

[0572] b) In case the neat supernatant does not contain any measurable antigen (ie. observed supernatant concentration less than LLOQ, where LLOQ=0.05 AU/mL), the LLOQ will be used for the calculation of the result. The result in this case is reported as “greater than x %”. For example if the DS sample is measured as 12.00 AU/mL and no signal was measured in the supernatant; with the LLOQ of 0.05 AU/mL then amount in supernatant is <0.05*1.05=<0.0525 AU/mL. The amount of DS after buffer correction is 12.00*0.95=11.40 AU/mL, and the reported result for degree of adsorption is <100-0.0525/11.4*100=>99.54%. The degree of non-adsorption will also be reported (i.e. 100—the % degree of adsorption).

Example 5

[0573] Introduction:

[0574] In order to further investigate the mode of action that leads to product instability/potency loss of the JEV vaccine, Ala-(His)6-OprF190-342-OprI21-83 (SEQ ID NO: 1, FIG. 14)—herein also referred to as “protein A” was used in a preliminary screening assay incorporating ALHYDROGEL® lots with different metal content and spiking with copper ions and sulfite.

[0575] Material: [0576] Copper(II)chloride dihydrate (Sigma, Order no. 807483) [0577] 10×PBS (Gibco, Order No. 14200-091) [0578] 15 ml Falcon tubes (Greiner, Cat. No. 188724) [0579] Incubator Infors HT Incubator Multitron Standard (InforsAG) [0580] Aqua bidest. (Fresenius Kabi, Art no. 0712221/01 A)

[0581] Preparation of Stock Solutions: [0582] Copper(II) stock solution [0583] 20 mM Copper(II) stock solution was prepared by dissolving 341 mg of Copper(II)chloride dihydrate in 100 mL of aqua bidest. [0584] Sodiummetabisulfite stock solution [0585] 200 mM Sodiummetabisulfite stock solution was prepared by dissolving [0586] 1.52 g of Sodiummetabisulfite in 35 mL PBS. This solution was adjusted the [0587] pH to 7.3 with NaOH and filled up to a volume of 40 mL with PBS. The solution was them filtered via 0.2μ syringe filter.

[0588] Preparation of Working Solutions [0589] Working solutions were prepared by dilution of metal stock solutions with aqua bidest. and sterile filtration via 0.2μ syringe filter. (Mini Kleenpak 25 mm-Pall)

[0590] Preparation of Buffer Solutions [0591] ⅓ PBS+0.9% NaCl [0592] 1×PBS buffer solution was prepared by 1:10 dilution of 10×PBS with aqua bidest. The pH of this buffer solution was 7.5. PBS buffer solutions adjusted to pH 7.3 and 8.0 were prepared by adjusting the pH with HCl or NaOH respectively. [0593] 9 g of NaCl were dissolved in 333 mL of either pH 7.3 of pH 8.0 buffer solution and then brought to 1000 mL with aqua bidest. followed by filtration via 0.2μ bottletop filter.

[0594] Sample Preparation:

[0595] Formulations of Protein A and different lots of ALHYDROGEL® (Lot 4230 & Lot 4074) were prepared in ⅓ PBS+0.9% NaCl at two different pH values and were spiked with sulfite according to the following scheme:

TABLE-US-00003 Sample Spike Alum Cu(II) Sulfit No. Name pH batch [ng/mL] [mM] 1 17112011_PROTEIN A_4074_ref_4°_pH7.3 7.3 4074 2 17112011_PROTEIN A_4074_ref_37°_pH7.3 7.3 4074 3 17112011_PROTEIN 7.3 4074 1 A_4074_Sulfit_37°_pH7.3 4 17112011_PROTEIN A_4230_ref_4°_pH7.3 7.3 4230 5 17112011_PROTEIN A_4230_ref_37°_pH7.3 7.3 4230 6 17112011_PROTEIN 7.3 4230 1 A_4230_Sulfit_37°_pH7.3 7 17112011_PROTEIN A_4074_ref_4°_pH8 8 4074 8 17112011_PROTEIN A_4074_ref_37°_pH8 8 4074 9 17112011_PROTEIN A_4074_Sulfit_37°_pH8 8 4074 1 10 17112011_PROTEIN A_4230_ref_4°_pH8 8 4230 11 17112011_PROTEIN A_4230_ref_37°_pH8 8 4230 12 17112011_PROTEIN A_4230_Sulfit_37°_pH8 8 4230 1

[0596] Samples 1, 6, 11 and 16 were stored at 4° C. (reference samples). All other samples were incubated at 37° C. for 96 hours.

[0597] After 96 hours all samples were subjected to a desorption procedure to separate the antigen from ALHYDROGEL®. The desorbed antigen was analyzed by RPC.

[0598] Results:

[0599] Results showed severe degradation of the antigen Protein A in the presence of sulfite. The degradation was more pronounced in the samples formulated with ALHYDROGEL® of higher metal impurity content.

TABLE-US-00004 TABLE 1 Metal ion content in Aluminium hydroxide lot 4230 and 4074 analyzed by ICP-MS. Alum Lot (2% solution) Al Cr Fe Co Ni Cu Ag Cd W Pb V Rb Mo μg/mL ng/mL ng/mL ng/mL ng/mL ng/mL ng/mL ng/mL ng/mL ng/mL ng/mL ng/mL ng/mL Alum Lot 4230 9570 1139 5640 7 816 64 <5 <5 <25 24 13 <5 11 RQCS 0890 Alum Lot 4074 9130 20 266 <5 15 <25 <5 <5 <25 19 <5 <5 <5 RQCS0013 Mix 50/50% of 9130 579 2952 <5.8 415 <45 <5 <5 <25 21.6 <9 <5 <8 both Lots* *Calculated residual metal content

[0600]

TABLE-US-00005 TABLE 3 Comparison of leachables from stopper extract and JEV09L37 SN. All peaks with an area of >0.1 mAU .Math. min are included in this table. Stopper extract Stopper concentrate FVL Relative Retention extract 1:16 diluted in L37 concentration time concentrate formulation SN compared to FVL (min) Peak area (mAU .Math. min) (%) 12.40 0.79 0.05 0.10 47 13.14 1.91 0.12 n.d. additional peak compared to FVL 13.48 0.81 0.05 n.d. additional peak compared to FVL 13.74 0.43 0.03 n.d. additional peak compared to FVL 14.10 0.75 0.05 n.d. additional peak compared to FVL 14.41 0.49 0.03 0.25 12 16.12 29.64 1.85 0.45 414  16.71 2.75 0.17 n.d. additional peak compared to FVL 17.15 14.68 0.92 0.11 815  18.68 0.90 0.06 0.57 10 20.08 0.70 0.04 0.25 18 20.75 0.42 0.03 n.d. additional peak compared to FVL 21.27 0.54 0.03 n.d. additional peak compared to FVL 22.13 2.84 0.18 0.15 122  22.74 0.55 0.03 0.17 20 23.74 1.74 0.11 0.27 41 24.88 0.98 0.06 0.12 51 27.63 0.84 0.05 0.16 32 29.46 0.72 0.04 n.d. additional peak compared to FVL 31.40 0.39 0.02 n.d. additional peak compared to FVL 35.23 4.70 0.29 0.41 72 SUM 67.59 4.2 3.0  140 

[0601]

TABLE-US-00006 TABLE 5 Analysis of Variance for ratio Monoclonal/Polyclonal ELISA after storage 4 weeks at 22° C. Analysis of Variance for Ratio 4 weeks Source Sum of Squares Df Mean Square F-Ratio P-Value A: pH 0.02205 1 0.02205 5.48 0.0326 B: Alum 0.117612 1 0.117612 29.20 0.0001 C: PS Fragments 0.0008 1 0.0008 0.20 0.6618 D: Extractables 0.0008 1 0.0008 0.20 0.6618 E: Formaline 0.0242 1 0.0242 6.01 0.0261 AB 0.0001125 1 0.0001125 0.03 0.8694 AC 0.00045 1 0.00045 0.11 0.7425 AD 0.01125 1 0.01125 2.79 0.1141 AE 0.00405 1 0.00405 1.01 0.3309 BC 0.0000125 1 0.0000125 0.00 0.9563 BD 0.0010125 1 0.0010125 0.25 0.6229 BE 0.0006125 1 0.0006125 0.15 0.7017 CD 0.0008 1 0.0008 0.20 0.6618 CE 0.0008 1 0.0008 0.20 0.6618 DE 0.0072 1 0.0072 1.79 0.1999 Total error 0.0644375 16 0.00402734 Total (corr.) 0.2562 31 R-squared = 74.8488 percent R-squared (adjusted for d.f.) = 51.2695 percent Standard Error of Est. = 0.0634614 Mean absolute error = 0.0352734 Durbin-Watson statistic = 1.47556

TABLE-US-00007 TABLE 6 Analysis of Variance for ratio Monoclonal/Polyclonal ELISA after storage at 22° C. for 8 weeks. Analysis of Variance for Ratio 8 weeks Source Sum of Squares Df Mean Square F-Ratio P-Value A: pH 0.102378 1 0.102378 11.19 0.0041 B: Alum 0.275653 1 0.275653 30.13 0.0000 C: PS Fragments 0.00300312 1 0.00300312 0.33 0.5747 D: Extractables 0.0108781 1 0.0108781 1.19 0.2917 E: Formaline 0.0318781 1 0.0318781 3.48 0.0804 AB 0.00137813 1 0.00137813 0.15 0.7031 AC 0.00382812 1 0.00382812 0.42 0.5269 AD 0.00137813 1 0.00137813 0.15 0.7031 AE 0.0166531 1 0.0166531 1.82 0.1961 BC 0.000253125 1 0.000253125 0.03 0.8700 BD 0.00382813 1 0.00382813 0.42 0.5269 BE 0.00195313 1 0.00195313 0.21 0.6503 CD 0.00195313 1 0.00195313 0.21 0.6503 CE 0.000153125 1 0.000153125 0.02 0.8987 DE 0.00525313 1 0.00525313 0.57 0.4596 Total error 0.1464 16 0.00915 Total (corr.) 0.606822 31 R-squared = 75.8743 percent R-squared (adjusted for d.f.) = 53.2565 percent Standard Error of Est. = 0.0956556 Mean absolute error = 0.0571484 Durbin-Watson statistic = 0.888586

TABLE-US-00008 TABLE 7 Regression analysis for “Ratio 8 weeks” including pH, Alum and Formaldehyde. Regression coeffs. for Ratio 8 weeks constant = 0.0228125 A: pH = 0.113125 B: Alum = −0.00185625 E: Formaline = 0.0315625

TABLE-US-00009 TABLE 8 Estimation of results “Ratio 8 weeks” generated using the fitted model. Estimation Results for Ratio 8 weeks Observed Fitted Lower 95.0% CL Upper 95.0% CL Row Value Value for Mean for Mean 1 0.58 0.5975 0.536766 0.658234 2 0.81 0.783125 0.722391 0.843859 3 0.9 0.84625 0.785516 0.906984 4 0.92 0.89625 0.835516 0.956984 6 0.99 0.84625 0.785516 0.906984 7 0.67 0.5975 0.536766 0.658234 8 0.99 0.89625 0.835516 0.956984 9 0.84 0.710625 0.649891 0.771359 10 0.78 0.710625 0.649891 0.771359 11 0.98 0.959375 0.898641 1.02011 12 0.59 0.5975 0.536766 0.658234 13 1.03 0.959375 0.898641 1.02011 14 0.78 0.660625 0.599891 0.721359 15 0.94 0.89625 0.835516 0.956984 16 0.81 0.77375 0.713016 0.834484 17 0.82 0.783125 0.722391 0.843859 18 0.88 0.84625 0.785516 0.906984 19 0.82 0.77375 0.713016 0.834484 20 0.82 0.959375 0.898641 1.02011 21 0.82 0.959375 0.898641 1.02011 22 0.63 0.710625 0.649891 0.771359 23 0.74 0.84625 0.785516 0.906984 24 0.66 0.660625 0.599891 0.721359 25 0.67 0.783125 0.722391 0.843859 26 0.76 0.783125 0.722391 0.843859 27 0.64 0.710625 0.649891 0.771359 28 0.74 0.77375 0.713016 0.834484 29 0.73 0.77375 0.713016 0.834484 30 0.44 0.5975 0.536766 0.658234 32 0.87 0.89625 0.835516 0.956984 33 0.67 0.660625 0.599891 0.721359 34 0.59 0.660625 0.599891 0.721359

[0602]

TABLE-US-00010 TABLE 14 Antigen recoveries after 5 weeks at 22° C. of desorbed JEV obtained by SEC-HPLC. Recoveries were based on non-spiked DP control samples stored at either pH 7 or pH 8 JEV area No Sample mAU .Math. min recovery 1 DP_unspiked_pH7 22° C. 3.710 100% 2 DP_Ni(II)_100_pH7 22° C. 3.656 99% 3 DP_Ni(II)_500_pH7 22° C. 3.705 100% 4 DP_Ni(II)_1000_pH7 22° C. 3.444 93% 5 DP_Cu(II)_100_pH7 22° C. 3.565 96% 6 DP_Cu(II)_500_pH7 22° C. 3.313 89% 7 DP_Cu(II)_1000_pH7 22° C. 3.367 91% 8 DP_Cr(III)_100_pH7 22° C. 3.562 96% 9 DP_Cr(III)_500_pH7 22° C. 3.422 92% 10 DP_Cr(III)_1000_pH7 3.148 85% 22° C. 11 DP_unspiked_pH8 22° C. 4.297 100% 12 DP_Ni(II)_100_pH8 22° C. 4.029 94% 13 DP_Ni(II)_500_pH8 22° C. 4.306 100% 14 DP_Ni(II)_1000_pH8 22° C. 4.065 95% 15 DP_Cu(II)_100_pH8 22° C. 3.751 87% 16 DP_Cu(II)_500_pH8 22° C. 3.698 86% 17 DP_Cu(II)_1000_pH8 22° C. 3.511 82% 18 DP_Cr(III)_100_pH8 22° C. 3.805 89% 19 DP_Cr(III)_500_pH8 22° C. 3.843 89% 20 DP_Cr(III)_1000_pH8 4.212 98% 22° C.

TABLE-US-00011 TABLE 15 ELISA results of desorbed JEV antigen after 5 weeks at 22° C. % Ratio compared to Poly. Mono. non-spiked No. Name pH ELISA ELISA Ratio Control 1 DP_unspiked_pH7 7 14.203 13.13 0.924 100 2 DP_Ni(II)_100_pH7 7 13.089 12.623 0.964 104 3 DP_Ni(II)_500_pH7 7 12.640 12.572 0.995 108 4 DP_Ni(II)_1000_pH7 7 15.051 11.757 0.781 84 5 DP_Cu(II)_100_pH7 7 13.420 10.792 0.804 87 6 DP_Cu(II)_500_pH7 7 13.247 10.079 0.761 82 7 DP_Cu(II)_1000_pH7 7 12.981 9.654 0.744 80 8 DP_Cr(III)_100_pH7 7 16.936 11.886 0.702 76 9 DP_Cr(III)_500_pH7 7 13.991 11.219 0.802 87 10 DP_Cr(III)_1000_pH7 7 13.061 10.438 0.799 86 11 DP_unspiked_pH8 8 12.647 11.287 0.892 100 12 DP_Ni(II)_100_pH8 8 12.308 10.689 0.868 97 13 DP_Ni(II)_500_pH8 8 14.300 12.623 0.883 99 14 DP_Ni(II)_1000_pH8 8 12.082 10.930 0.905 101 15 DP_Cu(II)_100_pH8 8 11.041 9.937 0.900 101 16 DP_Cu(II)_500_pH8 8 9.869 9.176 0.930 104 17 DP_Cu(II)_1000_pH8 8 9.379 8.802 0.938 105 18 DP_Cr(III)_100_pH8 8 10.164 9.545 0.939 105 19 DP_Cr(III)_500_pH8 8 11.241 10.057 0.895 100 20 DP_Cr(III)_1000_pH8 8 12.400 11.183 0.902 101

TABLE-US-00012 TABLE 16 ELISA results of desorbed JEV antigen after 4 weeks and 7 weeks stored at 22° C. Samples marked with “n.a” were not analyzed due to sample prioritization 4 weeks @ 22° C. 7 weeks @ 22° C. No pH poly mono ratio poly mono ratio m/p 1 DP Fe(II) pH7 7 14.285 11.213 0.785 14.181 11.378 0.802 2 DP Fe(III) pH7 7 14.879 11.552 0.776 14.323 11.765 0.821 3 DP Ni(II) pH7 7 16.572 11.862 0.716 14.231 11.666 0.820 4 DP Co(II) pH7 7 12.81 12.629 0.986 14.246 11.244 0.789 5 DP Cu(II) pH7 7 12.474 9.747 0.781 11.464 7.654 0.668 6 DP Zn(II) pH7 7 13.131 11.186 0.852 15.122 11.514 0.761 25 DP Cr(III) pH7 7 12.144 11.598 0.955 13.543 10.73 0.792 7 DP metalmix pH7 7 11.078 8.366 0.755 8.617 5.392 0.626 8 DP unspiked pH7 7 16.159 13.224 0.818 14.158 11.559 0.816 9 DP Fe(II) pH7.4 7.4 15.096 12.649 0.838 14.417 11.239 0.780 10 DP Fe(III) pH7.4 7.4 13.509 12.029 0.890 16.948 12.344 0.728 11 DP Ni(II) pH7.4 7.4 13.036 11.306 0.867 13.293 12.571 0.946 12 DP Co(II) pH7.4 7.4 13.715 11.714 0.854 13.079 11.96 0.914 13 DP Cu(II) pH7.4 7.4 14.235 11.748 0.825 10.843 9.276 0.855 14 DP Zn(II) pH7.4 7.4 13.815 12.882 0.932 13.28 12.619 0.950 26 DP Cr(III) pH7.4 7.4 12.324 11.659 0.946 13.312 10.804 0.812 15 DP_metalmix_pH7.4 7.4 n.a. 9.74 6.551 0.673 16 DP unspiked pH7.4 7.4 13.951 13.225 0.948 11.97 12.672 1.059 17 DP Fe(II) pH7.8 7.8 14.356 13.102 0.913 12.625 13.304 1.054 18 DP Fe(III) pH7.8 7.8 13.554 12.388 0.914 13.003 10.811 0.831 19 DP Ni(II) pH7.8 7.8 13.949 12.496 0.896 13.106 11.757 0.897 20 DP Co(II) pH7.8 7.8 12.826 11.931 0.930 13.333 11.059 0.829 21 DP Cu(II) pH7.8 7.8 12.593 11.268 0.895 12.538 9.872 0.787 22 DPZn(II) pH7.8 7.8 15.217 14.204 0.933 15.55 13.217 0.850 27 DP Cr(III) pH7.8 7.8 12.977 13.228 1.019 14.642 11.975 0.818 23 DP metalmix pH7.8 7.8 11.196 9.811 0.876 10.771 7.539 0.700 24 DP unspiked pH7.8 7.8 11.906 11.819 0.993 12.472 11.034 0.885

TABLE-US-00013 TABLE 17 ANOVA for stability samples stored at 22° C. for 7 weeks. Analysis of Variance for Ratio - Type III Sums of Squares Source Sum of Squares Df Mean Square F-Ratio P-Value MAIN EFFECTS A: Metal Type 0.139643 8 0.0174554 3.00 0.0293 B: pH 0.0462028 2 0.0231014 3.97 0.0398 RESIDUAL 0.0931046 16 0.00581904 TOTAL 0.27895 26 (CORRECTED) All F-ratios are based on the residual mean square error.

TABLE-US-00014 TABLE 18 Multiple range test for ratio by metal ion type. 1 = Fe(II); 2 = Fe(III); 3 = Ni(II); 4 = Co(II); 5 = Cu(II); 6 = Zn(II); 7 = Cr(III); 8 = Mix[1-6]; 9 = non-spiked control Multiple Range Tests for Ratio by Metal Type Method: 95.0 percent LSD Metal Type Count LS Mean Homogeneous Groups 8 3 0.666087 X 5 3 0.770168 XX 2 3 0.793725 XXX 7 3 0.807248 XX 4 3 0.844388 XX 6 3 0.853867 XX 1 3 0.878563 XX 3 3 0.887505 XX 9 3 0.919926 X Contrast Difference +/−Limits 1-2 0.084838 0.132037 1-3 −0.0089421 0.132037 1-4 0.0341754 0.132037 1-5 0.108395 0.132037 1-6 0.0246961 0.132037 1-7 0.0713155 0.132037 1-8 *0.212476 0.132037 1-9 −0.0413627 0.132037 2-3 −0.0937801 0.132037 2-4 −0.0506626 0.132037 2-5 0.0235569 0.132037 2-6 −0.0601419 0.132037 2-7 −0.0135225 0.132037 2-8 0.127638 0.132037 2-9 −0.126201 0.132037 3-4 0.0431175 0.132037 3-5 0.117337 0.132037 3-6 0.0336382 0.132037 3-7 0.0802576 0.132037 3-8 *0.221418 0.132037 3-9 −0.0324206 0.132037 4-5 0.0742195 0.132037 4-6 −0.00947935 0.132037 4-7 0.0371401 0.132037 4-8 *0.1783 0.132037 4-9 −0.0755381 0.132037 5-6 −0.0836988 0.132037 5-7 −0.0370794 0.132037 5-8 0.104081 0.132037 5-9 *−0.149758 0.132037 6-7 0.0466195 0.132037 6-8 *0.18778 0.132037 6-9 −0.0660588 0.132037 7-8 *0.14116 0.132037 7-9 −0.112678 0.132037 8-9 *−0.253838 0.132037 *denotes a statistically significant difference.

TABLE-US-00015 TABLE 19 ICP-MS results of residual metal ion impurities present in various Alum (2%) lots Residual metal content (ng/mL) Alum (2%)Lot Cr Fe Ni Cu V Co 4074 19.8 266 14.8 <25 <5 <5 4470 1637 1179 17.5 <25 <5 <5 4563 1874 2485 8.9 <25 <5 <5 4621 1333 1183 7.6 <25 <5 <5 3877 48.2 183 12.2 <25 <5 <5 4230 (nonGI**, 1139 5640 816 64 12.6 7 GI***) Mix* 4074/4230 579.4 2953 415.4 <44.5 <6 <6 *Calculated content of residual metals based on Alum lot 4230 and 4074 **nonGI: non gamma irradiated ***GI: gamma irradiated

TABLE-US-00016 TABLE 20 Summary of metal ion content and analysis of DP samples formulated with various Alum lots. Samples were analysed in duplicate by ELISA and the ratio of monoclonal/polyclonal ELISA is reported. Formulations were stored at 22° C. for 6 weeks. Mean Alum Lot Monoclonal Polyclonal Monoclonal Polyclonal Ratio Ratio ratio (2% Stock 1.sup.st 1.sup.st 2.sup.nd 2.sup.nd 1.sup.st 2.sup.nd (Mono/ Range* # solution) analysis analysis analysis analysis analysis analysis poly) Ratio 1 4470 23.584 26.397 21.666 22.948 0.893 0.944 0.918 0.025 2 4563 23.027 23.397 19.051 18.862 0.984 1.010 0.997 0.013 3 4621 22.758 24.056 19.041 19.196 0.946 0.991 0.968 0.023 4 3877 23.5 23.85 21.186 21.24 0.985 0.997 0.991 0.006 5 4230 21.85 25.155 20.682 23.792 0.868 0.869 0.869 0.000 (non-gamma irradiated) 6 4230 20.509 22.904 18.002 23.512 0.895 0.765 0.830 0.065 (gamma irradiated) 7 4074 23.022 24.047 16.695 19.866 0.957 0.840 0.898 0.058 8 Mixture 20.833 22.954 22.473 21.217 0.908 1.059 0.983 0.076 (50%/50%) of 4074 and 4230 *Range is the absolute difference between 1.sup.st and 2.sup.nd analysis.

TABLE-US-00017 TABLE 21 Results of PSD analysis of ALHYDROGEL ® (2%) stock solution in water. Obscuration No Sample Name d (0.1) d (0.5) d (0.9) (%) 1 Non-irradiated AlOH 0.70 2.13 46.53 1.51 RQCS0890 Lot 4230 2 GI AlOH RQCS1200 Lot 0.71 4.14 69.64 1.98 4230 3 GI AlOH RQCS1342 Lot 0.78 2.23 53.44 2.11 4740 4 GI AlOH RQCS0448 Lot 0.73 4.49 78.58 1.96 4074 d (0.1): 10% of all measured particles have a diameter below this value d (0.5): 50% of all measured particles have a diameter below this value d (0.9): 90% of all measured particles have a diameter below this value Obscuration: amount of laser light reduction by sample; corresponds to concentration of sample in measurement chamber

TABLE-US-00018 TABLE 22 Results of ALHYDROGEL ® titration curves for determination of POZ. Samples were analyzed in PBS (1:20 dilution). Sample ID PZC (pH) Non-irradiated AlOH RQCS0890 4.58 Lot 4230 GI AlOH RQCS1200 4.62 Lot 4230 GI AlOH RQCS1342 4.49 Lot 4740 GI AlOH RQCS0448 4.48 Lot 4074

[0603]

TABLE-US-00019 TABLE 24 Analysis of supernatant and Aluminum hydroxide (Lot 4230) gel fraction for contaminating metal ions shows metal ions are located in the gel, not the supernatant Sample Fe Ni Cu Co Cr Ag Cd W Pb V ng/mL Lot 4230 Supernatant 82 12 <25 <5 7 <5 <5 <25 70 <5 Lot 4230 Sediment 6200 920 <25 <5 1200 <5 <5 <25 45 13 % supernatant 1.3 1.3 n.a. n.a. 0.6 n.a. n.a. n.a. 155.6 n.a. compared to sediment

Preferred Aspects

[0604] Aspect 1. A method for preparing an aqueous composition comprising aluminium and a protein said method comprising [0605] combining an aluminium-salt, said protein and water to produce said aqueous composition and [0606] determining the level of a heavy metal in the aqueous composition and/or the aluminium-salt.

[0607] Aspect 2. A method for preparing an aqueous composition comprising aluminium and a protein said method comprising [0608] preparing or selecting an aluminium-salt that is able to provide an aqueous composition having less than 350 ppb heavy metal based on the weight of the aqueous composition and [0609] combining said aluminium salt, said protein and water to produce said aqueous composition.

[0610] Aspect 3. A method according to aspect 1-2, further comprising buffering said aqueous composition at a pH of between 6.5 and 8.5, preferably at a pH between 7.5 and 8.5.

[0611] Aspect 4. A method according to aspect 1-3, further comprising packaging aliquots of said aqueous composition having less than 350 ppb heavy metal based on the weight of the aqueous composition in separate air-tight storage containers.

[0612] Aspect 5. A method for preparing a clinical grade aluminium-salt precipitate for incorporation into a medicament and/or vaccine, said method comprising preparing an aqueous solution of aluminium ions and precipitating said aluminium-ions from said solution, and determining the level of a heavy metal in the solution and/or the aluminium-salt precipitate.

[0613] Aspect 6. A method according to aspect 5, wherein the precipitate is selected that is able to provide an aqueous composition comprising less than 350 ppb heavy metal based on the weight of the aqueous composition.

[0614] Aspect 7. An aqueous composition comprising a protein and an aluminium-salt, said composition comprising less than 350 ppb heavy metal based on the weight of the aqueous composition.

[0615] Aspect 8. An aqueous composition according to aspect 7, which has been stored at temperatures higher than 20° C. for at least 1 month.

[0616] Aspect 9. An aqueous composition according to aspect 7 having a shelf-life of at least 20 month.

[0617] Aspect 10. An aqueous composition according to aspect 7-9, wherein said heavy metal is selected from Cu, Ni, W, Co, Os, Ru, Cd, Ag, Fe, V, Cr, Pb, Rb and Mo.

[0618] Aspect 11. An aqueous composition according to aspect 7-10, wherein said heavy metal is selected from Cu, Ni, W, Co, Os, Ru, Cd, Ag, Fe, V.

[0619] Aspect 12. An aqueous composition according to aspect 7-11, wherein said heavy metal is selected from Cu or Ni.

[0620] Aspect 13. An aqueous composition according to aspect 7-12, wherein said heavy metal is present in ionic form.

[0621] Aspect 14. An aqueous composition according to aspect 7-13, wherein the aluminium-salt is aluminiumhydroxide (Al(OH)3) or aluminiumphosphate (AlPO4).

[0622] Aspect 15. An aqueous composition according to aspect 7-14, further comprising a reactive compound.

[0623] Aspect 16. An aqueous composition according to aspect 15, wherein the reactive compound is selected from the group consisting of a redox active compound, a radical building compound, a stabilizing compound and a combination of any thereof.

[0624] Aspect 17. An aqueous composition according to aspect 15-16, wherein the reactive compound is selected from the group consisting of formaldehyde, ethanol, chloroform, trichloroethylene, acetone, TRITON™ X-100 (Polyethylene glycol tert-octylphenyl ether), deoxycholate, diethylpyrocarbonate, sulphite, Na.sub.2S.sub.2O.sub.5, beta-proprio-lacton, polysorbate such as TWEEN® 20 (Polysorbate 20) or TWEEN® 80 (Polysorbate 80), O.sub.2, phenol, PLURONIC (poloxamer) type copolymers, and a combination of any thereof.

[0625] Aspect 18. An aqueous composition according to aspect 7-17, comprising between 5 μg/ml and 50 mg/ml aluminium.

[0626] Aspect 19. An aqueous composition according to aspect 7-18, comprising between 50 μg/ml and 5 mg/ml aluminium.

[0627] Aspect 20. An aqueous composition according to aspect 7-19, comprising between 5 ppb and 250 ppb Fe based on the weight of the aqueous composition.

[0628] Aspect 21. An aqueous composition according to aspect 7-20, comprising less than 3 ppb Cu based on the weight of the aqueous composition.

[0629] Aspect 22. An aqueous composition according to aspect 7-21, comprising less than 40 ppb Ni based on the weight of the aqueous composition.

[0630] Aspect 23. An aqueous composition according to aspect 7-22, wherein said protein is a therapeutic and/or a vaccine.

[0631] Aspect 24. An aqueous composition according to aspect 7-23, wherein said protein is a viral or bacterial protein.

[0632] Aspect 25. An aqueous composition according to aspect 7-24, wherein said viral protein is a protein of the Japanese encephalitis virus or a protein of the Pseudomonas aeruginosa bacterium.

[0633] Aspect 26. An aqueous composition according to aspect 7-25, wherein said protein is protein within a formaldehyde inactivated virus particles.

[0634] Aspect 27. An aqueous composition according to aspect 7-26, further comprising sulphite.

[0635] Aspect 28. An aqueous composition according to aspect 7-27, obtained by a method according to aspect 1-6.

[0636] Aspect 29. A vaccine comprising an aqueous composition according to aspect 7-28.

[0637] Aspect 30. An aluminium hydroxide concentrate that a) comprises 10 mg/ml of said aluminium hydroxide and b) less than 7 microgram heavy metal, for use in the manufacture of a medicine, preferably for use in the manufacture of a vaccine.

[0638] Aspect 31. An aluminium hydroxide concentrate that comprises 10 less than 7 microgram heavy metal, for use in the manufacture of a medicine, preferably for use in the manufacture of a vaccine.

[0639] Aspect 32. An aluminium salt concentrate that comprises less than 7 ppm heavy metal based on the weight of the concentrate, for use in the manufacture of a medicine, preferably for use in the manufacture of a vaccine.

[0640] Aspect 33. An aluminium salt concentrate that comprises less than 700 ppm heavy metal based on the weight of the aluminium salt, for use in the manufacture of a medicine, preferably for use in the manufacture of a vaccine.

[0641] Aspect 34. An aluminium hydroxide concentrate that comprises less than 700 ppm heavy metal based on the weight of the aluminium hydroxide, for use in the manufacture of a medicine, preferably for use in the manufacture of a vaccine.

[0642] Aspect 35. A method for preparing an aqueous composition comprising aluminium, a reactive compound and a protein said method comprising [0643] preparing or selecting an aluminium-salt that is able to provide an aqueous composition having less than 350 ppb heavy metal based on the weight of the aqueous composition and [0644] combining said aluminium salt, said protein, reactive compound and water to produce said aqueous composition.

[0645] Aspect 36. The method for preparing an aqueous composition according to aspect 35, wherein said heavy metal is selected from Cu, Ni, W, Co, Os, Ru, Cd, Ag, Fe, V, Cr, Pb, Rb and Mo.

[0646] Aspect 37. The method for preparing an aqueous composition according to aspect 35-36, wherein said heavy metal is selected from Cu, Ni, W, Co, Os, Ru, Cd, Ag, Fe, V.

[0647] Aspect 38. The method for preparing an aqueous composition according to aspect 35-37, wherein said heavy metal is selected from Cu or Ni.

[0648] Aspect 39. The method for preparing an aqueous composition according to aspect 35-38, wherein said heavy metal is present in ionic form.

[0649] Aspect 40. The method for preparing an aqueous composition according to aspect 35-39, wherein the aluminium-salt is aluminiumhydroxide (Al(OH)3) or aluminiumphosphate (AlPO4).

[0650] Aspect 41. The method for preparing an aqueous composition according to aspect 35-40, wherein the reactive compound is selected from the group consisting of a redox active compound, a radical building compound, a stabilizing compound and a combination of any thereof.

[0651] Aspect 42. The method for preparing an aqueous composition according to aspect 35-41, wherein the reactive compound is selected from the group consisting of formaldehyde, ethanol, chloroform, trichloroethylene, acetone, TRITON™ X-100 (Polyethylene glycol tert-octylphenyl ether), deoxycholate, diethylpyrocarbonate, sulphite, Na.sub.2S.sub.2O.sub.5, beta-proprio-lacton, polysorbate such as TWEEN® 20 (Polysorbate 20) or TWEEN® 80 (Polysorbate 80), O.sub.2, phenol, PLURONIC (poloxamer) type copolymers, and a combination of any thereof.

[0652] Aspect 43. The method for preparing an aqueous composition according to aspect 35-42, comprising between 5 μg/ml and 50 mg/ml aluminium.

[0653] Aspect 44. The method for preparing an aqueous composition according to aspect 35-43, comprising between 50 μg/ml and 5 mg/ml aluminium.

[0654] Aspect 45. The method for preparing an aqueous composition according to aspect 35-44, comprising between 5 ppb and 250 ppb Fe based on the weight of the aqueous composition.

[0655] Aspect 46. The method for preparing an aqueous composition according to aspect 35-45, comprising less than 3 ppb Cu based on the weight of the aqueous composition.

[0656] Aspect 47. The method for preparing an aqueous composition according to aspect 35-46, comprising less than 40 ppb Ni based on the weight of the aqueous composition.

[0657] Aspect 48. A method for preparing an aqueous composition according to aspect 35-47, further comprising buffering said aqueous composition at a pH of between 6.5 and 8.5, preferably 7.5 and 8.5.

[0658] Aspect 49. A method for preparing an aqueous composition according to aspect 35-48, further comprising packaging aliquots of said aqueous composition in separate air-tight storage containers.

[0659] Aspect 50. A method for prevention and/or treatment of a subject in need thereof that comprises the administration of an aqueous composition comprising an effective dose of an antigen, an aluminium compound, a reactive compound and less than 350 ppb heavy metal based on the weight of the aqueous composition.

[0660] Aspect 51. A method for prevention and/or treatment of a subject in need thereof that comprises the administration of an effective dose of a composition according to aspect 7-29.

[0661] Having thus described several aspects of at least one embodiment of this invention, it is to be appreciated various alterations, modifications, and improvements will readily occur to those skilled in the art. Such alterations, modifications, and improvements are intended to be part of this disclosure, and are intended to be within the spirit and scope of the invention. Accordingly, the foregoing description and drawings are by way of example only.

EQUIVALENTS

[0662] Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many equivalents to the specific embodiments of the invention described herein. Such equivalents are intended to be encompassed by the following claims.

[0663] All references disclosed herein are incorporated by reference in their entirety for the purposes cited herein.