WO2000040703A1

WO2000040703A1 - Virus inactivation process

Info

Publication number: WO2000040703A1
Application number: PCT/GB1999/004420
Authority: WO
Inventors: Peter Lewis Roberts; Jacqueline Rott; David Roger Lewin; John Edward More
Original assignee: National Blood Authority
Priority date: 1999-01-08
Filing date: 1999-12-23
Publication date: 2000-07-13
Also published as: GB9900428D0; AU1988600A

Abstract

A method of inactivating a virus in an aqueous albumin preparation comprising maintaining the albumin preparation at a pH of between about 8.5 and about 10.5 under conditions effective to result in substantial virus inactivation without adversely affecting the biological function of albumin. The method may be used to inactivate both lipid enveloped and non-lipid-enveloped viruses.

Description

Virus Inactivation Process

The present invention relates to a process for inactivating viruses m an aqueous albumin preparation. In particular, it relates to a method for inactivating viruses by means of incubation at a moderately alkaline pH.

Albumin is a therapeutic protein which may be purified m large quantity from human plasma. It is used for a variety of clinical purposes, including treatment of burns and of hypovolaemic shock (i.e. low blood volume) , in kidney dialysis and in plasma exchange .

As with all therapeutic proteins of human origin, there is a need to ensure safety of the product in terms of eliminating the risk of virus transmission. Viruses which give cause for concern m this regard include the hepatitis viruses and human immunodeficiency virus (HIV) . These may be present m biological fluids such as blood from which the plasma source of albumin is derived and may be carried through the protein preparation steps into the final product. If not inactivated, there is a risk that a subiect treated with albumin of human origin could receive a contaminated preparation and become infected with one or more of these life threatening viruses.

Albumin for clinical use is routinely sub3ected to pasteurisation, i.e. incubation of an aqueous preparation at 60°C for 10 hours. Although considered for a long time to be an effective method of inactivating transmissible viruses, there is nevertheless a theoretical risk of viral contamination in pasteurised albumin preparations, giving cause for concern about the efficacy of this treatment alone in eliminating infectious virus from preparations intended for human use. In order to minimise or eliminate completely the potential risk of transmission of viruses, there is thus a need for new method of inactivating blood borne viruses m albumin preparations. Such a method is desirably effective against both lipid enveloped viruses for example HIV, Hepatitis B and Hepatitis C, and non-lipid enveloped viruses, which are more difficult to eliminate, such as for example Hepatitis A and human parvovirus.

In any such method, there will inevitably be a balance between, on the one hand, achieving effective viral safety, whilst, on the other hand, maintaining activity and stability of the plasma derived albumin. There are various established methods for inactivating viruses m preparations of certain proteins derived from plasma, including solvent/detergent treatment and virus filtration. These methods whilst being effective for certain proteins, are not necessarily appropriate for albumin.

We have now surprisingly found that exposure of aqueous albumin preparations to moderately alkaline pH results m mactivation of both lipid enveloped and non- lipid enveloped viruses, the latter class m particular being traditionally more difficult to eradicate, without detriment to the stability and biological function of the albumin preparation.

Thus, viewed from one aspect, the present invention provides a method of inactivating a virus m an aqueous albumin preparation comprising maintaining the albumin preparation at a pH of between about 8.5 and about 10.5 under conditions effective to result in substantial virus mactivation without adversely affecting the biological function of albumin.

We have found this method is effective in inactivating all transmissible viruses, including non- lipid enveloped viruses, without sacrificing the biological properties of the albumin preparation It is also effective against those viruses which are known to be heat resistant, such as parvovirus, and which may not be completely inactivated by conventional pasteurisation .

The method is advantageous because of its elegant simplicity; it may be accomplished simply by the addition of an agent capable of elevating the pH to the range of the invention.

Furthermore, the method can readily be integrated into existing albumin purification processes, either as a stand alone mactivation step, or conjunction with conventional pasteurisation.

In most countries it is now recommended that blood and plasma derived products such as albumin are subjected to two separate virus mactivation processes. The virus mactivation method of the invention is thus advantageous that it can be readily integrated into large scale manufacturing processes which include a pasteurisation step and used m conjunction with pasteurisation, in which case the two steps may be carried out simultaneously or sequentially.

In one embodiment, the pH of the aqueous albumin containing preparation is simply adjusted by means of the addition of an alkaline material, or a buffer which can be added solid or solution form. The material used should however be compatible with albumin, i.e. should not have any detrimental effect on its structure, function or stability. Suitable materials include alkaline materials, such as hydroxides of alkaline metals and alkaline earth metals, for example sodium hydroxide, preferably at a concentration of about O.OOIM to 5M, preferably from 0.5M to 1.0M such as 1.0M, however other agents for example buffers or methods known m the art such as diaflltration against high pH buffer may also be used. Suitable buffers are known m the art and examples which may be used include glycme- NaOH, tris base and sodium bicarbonate. Preferably however, since the treated albumin is for human administration, the pH adjusting material is physiologically acceptable. This has the advantage that no additional steps are required to remove the agent used to adjust the pH.

We have found that within the pH range at which the virus mactivation method of the invention is carried out, of from about 8.5 to about 10.5, all viruses tested, including hepatitis A and parvovirus, a non- lipid enveloped virus which is heat resistant and notoriously difficult to inactivate, are inactivated, whilst the structure, biological activity and stability of albumin are maintained. Furthermore, the albumin product treated in accordance with the method of the invention not only retains its stability immediately after alkaline treatment and after adjustment to a physiological pH, but this stability is maintained following subsequent treatment steps such as virus mactivation for example by pasteurisation and during subsequent storage of the product. Long term stability is important for products such as albumin extracted in large quantity from human plasma.

As used herein, structure of albumin refers to the molecular size distribution, in terms of the aggregate and monomer content, the proportion of monomer being important for the ability of albumin to maintain colloid osmostic pressure (or oncotic pressure) in the intravascular circulation. Biological activity refers to the ability of albumin to bind ligands, such as for example fatty acids, steroids, bilirubm, drugs, and small molecules and ions, as well as maintaining colloid osmotic pressure. Also, as used herein, moderately alkaline pH refers to a pH the range of from about 8.5 to 10.5, of the method of the invention.

Preferably, the pH of the method of the invention is between about 8.5 to about 10.0, preferably between about 9.0 to about 10.0, particularly preferably between about 9.2 and 9.8, such as about pH 9.5.

The method of the invention is carried out under conditions so that, at moderately alkaline pH, viruses are inactivated without detriment to the stability of the albumin. Thus the mactivation may be carried out at a temperature of up to about 70°C such as up to about 65°C, such as between about 20°C to about 70°C, for example between about 20°C and about 65°C, such as between about 30°C and about 65 °C preferably from about 37°C to about 65°C, for example at about 60°C, the latter being particularly advantageous because it permits the elevated pH virus mactivation method of the invention to be carried out simultaneously with pasteurisation. Generally, the higher the temperature used, the faster the virus mactivation takes place.

The combination of mactivation parameters, pH, temperature and incubation time may be chosen to achieve the desired mactivation, depending upon the users requirements. Whilst some viruses will be inactivated almost instantaneously the pH of the albumin is adjusted in accordance with the invention, the albumin preparation may be treated at the moderate pH of the invention for long periods of time, up to several weeks Generally, the albumin will be treated m accordance with the invention for up to three days, preferably up to 24 hours, more preferably between about 5 minutes and 20 hours, such as between about 1 and 15 hours, for example 10 hours.

As mentioned above, the rate of virus mactivation depends at least in part on the temperature. Thus towards 70°C such as for example 65°C, the upper end of the operating range of the method of the invention, viruses may be rapidly inactivated, often within minutes, so that periods of up to about 10 minutes may be used. We have however found no deleterious effect on the structure of the albumin preparation when subjected to temperatures towards the upper end of the operating range for extended periods.

The precise combination of operating parameters must of course be effective for virus kill, which can be determined according to conventional methods for measuring virus mfectivity, such as plaque assay. Periods of the order of ten hours are particularly convenient because this enables the virus mactivation of the invention to be carried out simultaneously with pasteurisation.

The method of the invention may be applied to a variety of albumin containing starting materials, of varying levels of purity, and at albumin concentrations of from about lOg/L to about 350g/L, for example lOg/1 to 250 g/1, preferably between about 40g/L and 250g/L, such as for example 45g/L and 200g/L. The method may thus be used on plasma fractions such as Fraction V, and on purified albumin preparations, for example chromatographically purified albumin from human or animal or synthetic sources e.g. recombmant human albumin, or albumin purified by heat-ethanol fractionation.

The virus mactivation according to the invention may be carried out in the presence of one or more stabilising agents which act to protect albumin during the elevated pH treatment. Examples of agents which may be used to stabilise albumin are known m the art and include sugars, such as maltose, sugar alcohols, polyols, organic acids and/or salts thereof such as octanoate (caprylate) , and ammo acids such as N-acetyl tryptophanate . Caprylate is the conventional stabiliser used for albumin pasteurisation. When used with or without any other stabiliser this is generally m the range of from llg/Kg protein to 44 g/Kg protein, and is the preferred stabilising agent for the virus mactivation method of the invention because these levels are physiologically acceptable and thus the product does not require any further treatment to remove the stabiliser, such as diaflltration which would involve a risk of subsequent virus contamination.

Moreover, the process of the invention is readily integrated into conventional albumin preparative methods which involve addition of caprylate prior to pasteurisation. The same bulk preparation to which caprylate has been added may thus be used for the method of the invention. No additional stabiliser needs to be added, and thus no stabiliser needs to be removed.

The method of the invention may be carried out before, after or simultaneously with conventional pasteurisation, the latter being preferred because of the economies of time, and energy, and the savings associated cost .

Among other advantages, the method of the invention is more cost effective than other virus mactivation methods, such as solvent/detergent. As such, the method of the invention is particularly advantageous for large scale manufacturing typically up to lOOKg albumin and volumes up to 2500L; these are quantities for which other virus mactivation methods would prove prohibitively expensive.

After the virus mactivation has been carried out m accordance with the invention, the pH of the albumin preparation may be adjusted to a value which is physiologically acceptable prior to administering the protein. The pH may thus be adjusted to from about 8.0 to about 6.0, preferably from about 7.3 to about 6.7 such as about 7.0. This may be done by means of addition of any physiologically acceptable acid, such as hydrochloric acid or carboxylic acids such as acetic acid, formic, citric, propanoic or octanoic acid the range of 0.01M to 5M, preferably 0.5M to 2M, such as at 1.0M. Preferably, the material used is sterile, and is introduced into the albumin preparation using sterile techniques, thereby avoiding any risk of subsequent viral contamination. According to a further aspect, the present invention provides a virus inactivated albumin preparation produced by the method of the invention.

The invention will now be described with reference to the following non- limiting examples and the drawings in which:

Figure 1 is a photograph of an agarose gel on which albumin samples heat treated at 60°C at various pHs between 5 and 10 are analysed;

Figure 2 is a graph showing the extent of inactivation of canine parvovirus (CPV) in albumin at various pHs between 7.0 and 10.4 at 60°C;

Figures 3A and B show the effect of alkaline treatment on the subsequent stability of 4.5% albumin upon storage. Figure 3A shows the effect of treatment pH on monomer and dimer composition of albumin and Figure 3B shows the effect of storage time on monomer and dimer composition of albumin;

Figures 4A and B show the effect of alkaline treatment on the subsequent stability of 20% albumin upon storage. Figure 4A shows the effect of treatment pH on monomer and dimer composition of albumin and Figure 4B shows the effect of storage time on monomer and dimer composition of albumin;

Figures 5A and B show the characteristic absorbance pattern of phenol red (Figure 5A) and the effect of albumin addition on the absorbance pattern of phenol red (Figure 5B) .

EXAMPLE 1

Stability of albumin at varying pH

These experiments were carried out on chromatographically purified Cohn Fraction V solution (Kistler & Nitschmann 1962, Vox Sang, 7 414-424) . Methods

Albumin samples at 6-7% w/v were formulated with sodium octanoate in the molar ratio of 10:1 octanoate to albumin (23g octanoate per kg albumin) . Aliquots of the formulated albumin were titrated to pH 5, 6, 7, 8, 9, 10, 11 with 1M HC1/1M NaOH as necessary. All pH measurements were done on a 1 m 10 dilution of the sample m saline. The samples were then heated at 60°C for 10 hours.

All samples were analysed by size exclusion chromatography by Pharmacia FPLC on Superose 12 and Agarose Electrophoresis .

Results of FPLC Superose 12 chromatography are presented m Table 1. The results of agarose electrophoresis are shown m Fig. 1.

Results

Table 1 - Size Exclusion Chromatography of pH Adjusted Heated Albumin

ND: not detectable Discussion of Results

These experiments were carried out at 60°C, the temperature used for conventional pasteurisation, at which albumin is known to be stable.

Analysis of treated albumin by size exclusion chromatography showed monomer to be greater than 93% m samples m the range pH 5 to 10. Albumin at pH 4 solidified after heating at 60°C for 10 hours; albumin at pH 11 showed greater than 90% aggregation after heating .

Agarose electrophoresis of albumin showed little difference between un-heated albumin at pH 7 and albumin heat-treated m the range pH 5-10, there being a single major band detectable m all samples, with one minor band visible (see Figure 1) . The pH 11 samples showed some deterioration.

The gross physical and molecular integrity of the albumin does not appear to be affected by incubation at pHs m the range of 5 to 10.

EXAMPLE 2

Virus Inactivat on by Treatment at Alkaline pH at 60 °C

This investigates the mactivation of several non- enveloped viruses at pH m the range 9.0 to 10.5 and at a temperature of 60°C.

Hepatitis A virus (HAV) and Canme parvovirus (CPV) were used as they have been shown to be highly heat resistant. Simian Vιrus-40 (SV-40) was included as it has also been reported to be a relatively resistant virus. These are all non-lipid enveloped viruses. Materials and Methods

Formulated albumin solution (20%) was adjusted to pHs m the range of about pH 9.0 to 10.5 with sodium hydroxide. Samples were then heated at 60.0°C and HAV, SV-40 or CPV added at a titre of approximately 5 to 6 logs PFU/ml

(plaque forming units/ml) and at a dilution of 1 in 20. Samples were taken after various times, immediately diluted 1 10 m cell culture medium and adjusted to neutral pH where necessary. Controls consisting of virus m cell culture medium or albumin at standard pH

(pH7) , were also prepared. Virus mfectivity was determined by plaque assay on suitable indicator cells, i.e. Vero for SV-40, BSC-1 for HAV or A-72 cells for CPV. The log virus mactivation value was determined by subtracting the log virus titre after treatment from the log virus titre of the albumin pH7 control.

Results and Discussion

Virus mactivation

Hepatitis A

The mactivation of HAV at elevated pH during the heat- treatment of albumin at 60°C was investigated. After 5 hr, 4.8 log mactivation was seen with standard albumin i.e. approx pH 7.0 (Table 2) . However at pH 9.5 or approx 10.0, the mactivation of >5.1->5.4 log was found m 10mm. There was no significant difference between pH 9.5 or approx 10.0.

Simian Virus - 40

SV-40 was readily inactivated in albumin at standard pH (approx pH 7.0) i.e. >6.4 log in 8 mm (Table 3) . Increasing the pH to approx 9.5 or approx 10.0 led to virus being usually undetectable by 2min i.e. 6.1->6.4 log inactivation.

Canine Parvovirus

The results for CPV are shown in Table 4 and Fig. 2. This data shows that CPV is highly resistant to inactivation, under neutral conditions i.e. approx pH 7.0, with only about 2.7 log being inactivated within 10 hr at 60 °C. However at the higher pH of 9.5 and 10.0, virus inactivation after 10 hr at 60°C increased substantially i.e. to 5.0 and >5.7 log respectively.

Conclusion

This investigates the inactivation of several non- enveloped viruses under conditions of elevated pH in the range approx 9 to 10.5. At elevated pH, the rate of HAV inactivation was greatly increased compared to that which occurs under standard conditions i.e. approx pH 7.0. In addition, increasing the temperature from 37°C (see Example 3) to 60°C increased the rate of virus inactivation at approx pH 9.5/10.0. For instance .5 log was inactivated after 2-3 days at 37°C but the same level of inactivation was reached after lOmin at 60°C and approx pH 9.5/10.0. For CPV, which is highly resistant to inactivation at pH 7.0 at 60°C, increasing the pH to 9.5- 10.0 resulted in increased inactivation. SV40 was rapidly inactivated at 60°C without the need to raise the pH.

In conclusion, virus inactivation at elevated pH in the range about 9 to 10.5 has been shown to be effective at inactivating a range of non-enveloped viruses. These include hepatitis A and parvovirus which are viruses that may be transmitted by plasma products. Table 2 - Summary of HAV Inactivation n Albumin at Various pH During Heating at 60°C

Table 3 - Summary of SV-40 Inactivation in albumin at Various pH During Heating at 60°C

Table 4 - Summary of CPV Inactivation m Albumin at Various pH During Heating at 60°C

EXAMPLE 3

Virus activation m albumin by treatment at alkali pH and 37°C

This investigates mactivation of two non-enveloped viruses, HAV and bovme parvovirus iBPV) , m albumin, at alkaline pH and at 37°C.

Materials and Methods

Albumin solution (10%) , collected prior to final formulation, was adjusted to about pH 9.5 and about 10.0 with sodium hydroxide; samples were then heated at 37°C and BPV or HAV added at a titre of approximately 5 to 6 logs PFU/ml and at a dilution of 1 m 20. Samples were taken after various times, immediately diluted 1 10 in cell culture medium and adjusted to neutral pH where necessary. Controls consisting of virus m cell culture medium or albumin at standard pH (pH7) were also prepared. Virus mfectivity was determined by plaque assay on suitable indicator cells i.e. MDBK for BPV and BSC-1 for HAV. The log virus mactivation value was determined by subtracting the log virus titre after treatment from the log virus titre of the albumin pH7 control .

Results and Discussion

The results are presented in Table 5.

Hepatitis A

Table 5 shows that activation of HAV was only about 0.8 log after incubation for 5 hr at approx pH 9.5 or approx 10.0. When incubation was extended to 3 days, the mactivation of 5.1 log of HAV occurred at approximately pH 9.5 after 3 days and >4.9 logs after 2 days at approximately pH 10.0.

Bovme parvovirus

Virus mactivation studies were also conducted with BPV. This proved to be more susceptible to mactivation by high pH than HAV, with 1.6-4.2 log being inactivated after 5 hr at approximately pH 9.5 and >6.1->6.3 log at approximately pH 10.0. Extending the incubation time increased virus mactivation at approximately pH 9.5 to >6.3 log after 1 day (Table 5).

Conclusion

At alkaline pH, incubation at 37°C for 1-2 days was required to inactivate about 4 log of HAV. At a pH of approximately 9.5, 5.1 log of HAV was inactivated after 3 days. Increasing the pH to approximately 10.0 increased the level of HAV mactivation to >4.9 log after 2 days. However, BPV was more susceptible to inactivation with 1.6-4.2 log being inactivated after 5 hr and >6.3 log after 1 day at approximately 9.5. Increasing the pH to approximately 10.0 substantially increased the level of BPV inactivation to >6.3 log after 5 hours.

Table 5. Virus Inactivation in Human Albumin Intermediate at high pH and 37°C: Summary

nd: not done

EXAMPLE 4

Stability of 4.5% albumin following treatment at alkaline pH

Introduction

This study was to show that incubation at alkaline pH can be used without affecting the long term stability of albumin. Materials

Un-heated 4.5% Albumin solution, formulated with sodium octanoate (23g/kg albumin) .

Method

Aliquots of the albumin solution were pH adjusted to pH 7 and about 9.5-10.5 by means of 1M NaOH. The pH adjusted aliquots were then heated at 60°C for 10 hours. The aliquots were pH neutralised with 1M HC1 and subjected to pasteurisation step at 60°C for a further 10 hours.

Aliquots of all samples were then incubated at 37°C to provide accelerated stability data to support the shelf life of albumin at room temperature (about 25°C) . The composition of the samples was analysed by HPLC Size exclusion chromatography on media TSK3000 at suitable time intervals to determine the monomer and dimer content .

Results

Results are presented m Figure 3 A and B.

Discussion of Results

Analysis of samples treated at alkaline pH and terminally pasteurised showed a very marginal downward shift m monomer composition during storage at 37°C.

Conclusions

High levels of monomer + dimer are maintained m 4.5% Albumin which has been treated at up to about pH 10.0, at 60°C, followed by pasteurisation at neutral pH. Furthermore, the molecular weight composition of these sample remains at acceptable levels for at least 26 weeks when stored at 37°C.

EXAMPLE 5

Stability of 20% albumin following treatment at elevated pH

Introduction

This example shows that the heating at elevated pH in accordance with the present invention can be used without affecting long term stability of albumin.

Materials

Unheated 20% albumin solution (formulated with sodium octanoate (23g/kg albumin) .

Method

Aliquots of the albumin solution were pH adjusted to pH 7 and in the range between 8-11 by means of 1M NaOH. The pH adjusted aliquots were then heated at 60°C for 10 hours. The aliquots were pH neutralised with 1M HCl and subjected to pasteurisation at 60°C for a further 10 hours .

Aliquots of all samples were then incubated at 37°C to provide accelerated stability study data. The composition of the samples was analysed by HPLC size exclusion chromatography using TSK3000 media at suitable time intervals to determine monomer and dimer content. Results

Results are presented in Figure 4 A and B.

Discussion of Results

The results show that following treatment at alkaline pHs, a decrease in monomer composition occurs at pHs greater than 10 in 20% albumin preparations at 60°C. Following neutralisation and pasteurisation, samples that had been treated at pH greater than 10.4 solidified and were not able to be assayed. Subsequent analysis of samples stored at 37°C showed a general downward shift in monomer composition with increasing storage time. Although albumin heated at alkaline pHs started at a lower monomeric composition, the extent of monomeric reduction with time, appeared to be independent of pH at which it was treated.

Albumin treated at up to pH 9.8 and at 60°C remained within a 2-3% difference in monomer composition when compared to the pH 7 control sample.

Conclusions

A high level of monomer and dimer is maintained in 20% Albumin which has been treated at elevated pH up to pH 9.8 followed by conventional pasteurisation at neutral pH. Furthermore, the molecular weight composition of these samples remained at acceptable levels for at least 28 weeks when stored at 37°C.

Examples 4 and 5 show that alkaline treatment of albumin at the two concentrations in clinical use, 4.5% (the normal plasma concentration) and 20%, is not detrimental to stability. EXAMPLE 6

Inactivation of Canine Parvovirus in Albumin by Treatment of Alkaline pH : Effect of Temperature

This study was designed to investigate the effect of temperature on the inactivation of the non enveloped virus, canine parvovirus (CPV) . A pH of 9.5 was used.

Materials and Methods

Formulated albumin solution (20%) was adjusted to pH 9.5 and infected with CPV and treated at various temperatures according to details given in Example 2.

Results and Discussion

The results are given in Table 6. Virus inactivation was greater and more rapid at higher temperatures and ranged from 1.1 log after 45°C/I0hr to >6.0 log after 65°C/2hr.

Conclusion

The use of a temperature of about 55-60°C resulted in substantial levels of inactivation i.e. 3.2-3.8 log. At a temperature of 65°C, CPV inactivation was even greater i.e. >6.0 log after 2hr.

Table 6. Effect of Temperature on the Inactivation of CPV in Albumin at alkaline pH

Example 7

Effect of Alkaline pH treatment on the Ligand Binding Properties of Albumin

Introduction

Ligand binding is an important physiological function of albumin. This Example shows that a biological property of albumin is unaffected by treatment at alkaline pH.

Phenol red is a compound which is bound by albumin. The phenol red binding site is believed to be the same site as for bilirubin, a known physiological ligand. Therefore, the use of phenol red binding can be used as a marker for albumin binding activity.

This assay relies upon the visible light absorption properties of phenol red. When a phenol red solution is scanned in the visible light region a characteristic two peak pattern is produced. These peaks are termed 'acidic' and 'basic' (Figure 5A) . On addition of an albumin solution, phenol red is bound and the absorbance of the basic peak is reduced whilst the acid peak absorbance remains relatively unchanged (Figure 5B) . This property can be used to quantify the 'ligand binding activity¹ of albumin.

Method

A 3.03 x 10 M phenol red working solution was made in a 33 mM sodium phosphate, pH 7.4 buffer.

Formulated, unheated, albumin samples of 4.5% and 20% were adjusted to pHs m the range 7.0 to 10.5 by means of 1M NaOH, and subjected to incubation at 60 °C then neutralised and pasteurised at 60°C for 10 hours then diluted accurately to 40 g/1 with 33 mM sodium phosphate, pH 7.4 buffer.

The assay mixture was made by mixing 0.25 ml diluted albumin sample, 0.25 ml working phenol red solution and 2.00 ml 33 mM sodium phosphate, pH 7.4 buffer and incubated at room temperature for 30 minutes.

A control sample was prepared using buffer m place of the albumin sample.

Following phenol red/sample incubation, the control and test samples were spectrophotometrically scanned between 650 and 350 nm.

Calculations

To determine the binding activity of albumin samples, relation to a 'reference' albumin sample the following procedure was used.

The λ~_ay of the acidic phenol red peak of the control sample was determined. The absorbance of the test sample and the reference albumin at the λ_raax of the phenol red control was measured. The values were entered in to equation below to give % binding activity.

A (control acid peak Λmax _l

- A {_Control acid peak Λmax i sampl e x 100^s (control acid peak _λmax i Con ol - A i_control acid peak λmax ι Ref ere ce

Results

Table 7 - Functional activity of albumin treated at alkaline pH

(a) 4.5% Albumin

(b) 20% Albumin

Conclusion

The data on 4.5% and 20% albumin shows that the functional activity of the albumin molecule as measured by phenol binding is largely unaffected up to at least pH 10.

Claims

1. A method of inactivating virus in an aqueous albumin preparation comprising maintaining the albumin preparation at a pH of between about 8.5 and about 10.5 under conditions effective to result in substantial virus inactivation without adversely affecting the biological function of albumin.

2. A method as claimed in claim 1 wherein the pH is between about 8.5 and about 10.0.

3. A method as claimed in claim 2 wherein the pH is between about 9.0 and about 10.0.

4. A method as claimed in claim 3 wherein the pH is between about 9.2 and about 9.8.

5. A method as claimed in claim 4 wherein the pH is about 9.5.

6. A method as claimed in any one of the preceding claims wherein the temperature is up to about 70°C.

7. A method as claimed in claim 6 wherein the temperature is from about 20°C to about 70°C.

8. A method as claimed in claim 7 wherein the temperature is from about 30°C to about 65 °C.

9. A method as claimed in claim 8 wherein the temperature is from about 37°C to about 65°C.

10. A method as claimed in claim 9 wherein the temperature is about 60°C.

11. A method as claimed in any one of the preceding claims wherein the method is carried out in the presence of one or more stabilising agents.

12. A method as claimed in any one of the preceding claims wherein following the virus inactivation, the pH is adjusted to neutral pH.

13. A method as claimed in any one of the preceding claims for inactivating lipid enveloped and/or non- lipid-enveloped viruses.

14. A virus inactivated albumin preparation.

15. A virus inactivated albumin preparation substantially free of non- lipid enveloped viruses.

16. A virus inactivated albumin preparation obtainable by the method of any one of claims 1 to 13.