+ All Categories
Home > Documents > PPB '05 abstracts

PPB '05 abstracts

Date post: 31-Jan-2017
Category:
Upload: tranthien
View: 223 times
Download: 0 times
Share this document with a friend
48
GE Healthcare Downstream PPB ’05 abstracts Extended Reports from the Fourth International Plasma Product Biotechnology Meeting Crete, Greece, May 9–12, 2005
Transcript

GE Healthcare

Downstream

PPB ’05 abstractsExtended Reports from the Fourth International Plasma Product Biotechnology Meeting

Crete, Greece, May 9–12, 2005

Downstream – PPB ’05 abstracts 3

5 From the chairman

Oral presentations

6 Purification of plasma derived mannose binding lectin (MBL)

8 A new, high yielding, affinity cascade for sequential isolation of plasma proteins of therapeutic value

10 New adsorbents for selective capture of plasma proteins from recovered and source plasma

12 The application of CaptureSelect affinity ligands for purification of plasma products

14 Identification and production of recombinant human polyclonal antibody drugs reflecting the natural human antibody diversity

16 Transgenic goats and cows for the production of human plasma proteins

19 Development of recombinant human thrombin as an aid to hemostasis in subjects undergoing surgery

22 A comparative study of Cohn and chromatographic fractionation using a novel affinity “Cascade Process”

25 Production of a virally safe despeciated equine botulinum antitoxin product

27 A high-yield IVIG process with efficient viral clearance

29 Emerging technology in hyperimmune therapeutic manufacturing

32 Development of an in vitro TSE infectivity assay: application to validation of manufacturing processes

34 Biochemical and pre-clinical characterization of a new 10% liquid triple virally reduced human intravenous immune globulin (IGIV, 10% TVR)

37 Efficacy and safety of subcutaneous immunologlobulin replacement therapy at home in patients with primary immunodeficiency diseases: combined analysis of two clinical studies, one in North America and one in Europe

40 Alpha-1 acid glycoprotein and apotransferrin effectively protect against ischemia-reperfusion injury

44 List of posters presented

46 Author index

In this issue

4 Downstream – PPB ’05 abstracts

Downstream – PPB ’05 abstracts 5

From the chairman

The highly successful PPB conference series held its fourth international meeting on the island of Crete. The meeting is now a well-established forum for open discussion and dialogue that brings together all working within the different aspects the plasma industry.

In this special edition of Downstream you will find some extended reports from the many excellent oral presentations given at this meeting. I hope that after reading these you will be strongly encouraged to join us next time, with a presentation, a poster or as a curious participant eager to learn of the latest trends and developments within our industry.

I would also thank our co-sponsors, CSL Bioplasma, Melbourne, Australia, for helping to make this once again a successful event. And I look forward to meeting you at PPB 07 on the island of Elba, off the coast of northern Italy. Watch our website for details. www.bo-conf.com/ppb07

Jan Berglöf Conference chairman, PPB ’05

6 Downstream – PPB ’05 abstracts

Purification of plasma derived mannose binding lectin (MBL)Antje Daehler*1,4, Teresa Martinelli1,4, Robert Pike3,4, and Robyn Minchinton2,4

1 CSL Bioplasma.2 Australian Red Cross Blood Service.3 Department of Biochemistry & Molecular Biology, Monash University.4 Cooperative Research Centre for Vaccine Technology.

e-mail: [email protected]

Mannose binding lectin (MBL) is a key component of the innate immune system. This plasma protein is considered an important factor in the first line host defence against various pathogens including bacteria, fungi, viruses, and parasites.

MBL is synthesized in the human liver and consists of polypeptide subunits with an apparent molecular weight of 32 kDa. Three of these subunits form structural units, or monomers, which in turn are assembled to oligomers. The MBL oligomers form complexes with MBL associated serine proteases (MASP-1, -2 and -3). MBL/MASP complexes bind to specific carbohydrate structures that are found on the surface of a variety of pathogens, but not on mammalian cells. This provides a mechanism to distinguish self from non-self.

Figure 1 shows the function of the MBL/MASP complexes. Upon binding to a target, a conformational change in MBL translates along the collagen domain to initiate activation of MASP. The activated MASP-2 is able to cleave and deposit C2 and C4, thus causing complement activation, which

can ultimately lead to lysis of the pathogen via a membrane attack complex.

The MBL pathway of complement activation coexists with the classical (antigen-antibody) pathway of the adaptive immune system and the alternative pathway. MBL as part of the innate immune system is particularly important when the adaptive immune system is compromised or immature.

MBL deficiency has been linked to higher susceptibility and/or less favorable outcomes for a number of conditions, including cystic fibrosis, infections, sepsis, common variable immune deficiency, and systemic lupus erythematosus.

Current research indicates that ~24% of the general population is affected by functional MBL deficiency due to genetic mutations and carriage of the low promoter polymorphisms and that MBL deficiency has been associated with increased susceptibility to infections. Thus MBL has potential as a therapeutic for a defined group of immunocompromised individuals.

Fig 1. Function of the MBL/MASP complex.

C4b

C4

Plasma of individual infected with bacteria

MBL binds to mannose sugarson bacterium

Conformational change in MBLtranslates along collagen domainto initiate activation of MASP

C2

C2a

Downstream – PPB ’05 abstracts 7

Extraction & depth filtration

SD treatment & filtration

Chromatographic capture step

Chromatographic polishing step

Ultrafiltration/Diafiltration

Virus reduction filtration

Formulation

plasma fractionation side fraction

MBL final product

Mannan Affinity Chromatography activation of MASP

MASP removal

The ability of MBL to cleave and deposit C4 can be measured as described by Minchinton et al. Figure 3 shows the C4 deposition ability of 3 large laboratory-scale batches of MASP depleted MBL compared to the average C4 deposition in a study of 236 Australian blood donors. The values are comparable, which demonstrates that the purified MBL is fully functional.

This work was carried out with the financial support of the Australian Government’s Cooperative Research Centres Program, as part of the Cooperative Research Centre for Vaccine Technology (CRC VT).

Fig 2. Purification of MASP depleted MBL.

Fig 3. C4 deposition ability of purified MBL. *236 Australian blood donors. Minchinton et al., 2002, Scandinavian Journal of Immunology 56: 630–641.

MBL/MASP complex can be purified from side fractions of existing plasma fractionation processes by affinity chromatography with carbohydrate resins. As the binding mechanism resembles the binding of MBL/MASP complexes to pathogens, the MASP can become activated during the purification process. These activated MASP could be capable of initiating an immune response independent of the presence of pathogens. Furthermore, inhibitors could potentially bind to the activated MASP before the MBL/MASP complex is bound to a pathogen and thus impair its ability to deposit C4 and cause lysis of the pathogen.

To enhance the safety and efficacy of the MBL product, a method has been developed to separate MASP from MBL. The fully scalable purification process includes two virus inactivation steps and is capable of delivering a fully functional, MASP-depleted MBL (Fig 2).

1.0

0.5

00

2

MBL concentration (µg/ml)

batch 1batch 2batch 3plasmaaverage*

C4

depo

sitio

n (U

/µl)

4 6

8 Downstream – PPB ’05 abstracts

A new, high yielding, affinity cascade for sequential isolation of plasma proteins of therapeutic valueChristopher Bryant1, Dev Baines1, Ruben Carbonell2, Tom Chen3, John Curling1, Timothy Hayes3, Steve Burton1, and David Hammond3

1 ProMetic BioSciences Ltd., Cambridge, UK.2 North Carolina State University College of Engineering, Raleigh NC, USA.3 American Red Cross, Plasma Derivatives, Gaithersburg, MD, USA.

e-mail: [email protected]

IntroductionA unique collaboration between ProMetic BioSciences and the American Red Cross, drawing on competences ranging from computational/combinatorial chemistry and specific affinity adsorbent design, to plasma protein separation, analysis and pathogen elimination has resulted in a new Cascade Process to recover the therapeutically valuable plasma proteins. Engineering and modelling resources from North Carolina State University and Alfa Laval Biokinetics have also been involved.

Disruptive technologyThe main research and process development has been carried out during a period in which Waeger1 (1) correctly projected industry consolidation and reduction of fractionation capacity at the same time that many haemophiliacs, immune-compromised patients, and emphysema sufferers still remain undiagnosed and untreated in the most populous areas of the world. Classic industrial doctrine dictates the assessment of plasma fractionation as a mature industry in need of rationalisation of R & D investment, implementation of life cycle management, and decrease in investment in innovation. This presents the true innovator’s dilemma and in Christensen’s terminology it is the time to adopt “disruptive technologies” (2). The highest performing companies, such as the major fractionators, find it difficult to invest adequate resources in disruptive technologies that current customers do not want, until they do need them – and then it is too late. On the other hand, neglected or under-served markets are generally more prone to investment in what will become the next generation of established technology.

The Cascade processThe Cascade Process, or Plasma Protein Purification System (PPPS), is an example of disruptive technology that focuses on the unmet needs of markets that are marginalized by the established industry which finds it difficult to operate outside the economic rules of the pharma and plasma enterprises. The PPPS is a set of affinity chromatography steps that isolate seven target proteins by adsorption to specifically designed synthetic affinity adsorbents (3). The full sequence is: Factor VIII/von Willebrand Factor, plasminogen, fibrinogen, immunoglobulin G, albumin and alpha1-proteinase inhibitor. Each adsorption step can be compared with a Cohn intermediate and abbreviated cascades can be designed if, for example, only Factor VIII and IgG are required. Each protein intermediate/eluate from the affinity capture step is subject to a downstream purification process incorporating viral inactivation steps and pathogen reduction using nanofiltration, etc. The Cascade backbone is compared to the classical Cohn pathway in Figure 1.

Fig 1. The Cascade and the Cohn processes follow similar backbone schemes. In the Cascade the design criteria were for initial capture of vWF/Factor VIII as a complex, followed by the removal of fibrinogen allowing unhindered capture of IgG. Alpha1-proteinase inhibitor is currently purified after the removal of albumin. If captured, plasminogen is isolated after the Factor VIII step. Adsorbents for vWF/Factor VIII, plasminogen, fibrinogen were specifically designed for the process. Commercially available MAbsorbent® A2P and Mimetic® Blue SA for IgG and albumin capture respectively, were standard products from ProMetic BioSciences Ltd., Cambridge, UK.

Plasma Intermediate Product Intermediate Plasma

Eluate 1 vWF/Factor VIII Cryosupernatant Cryo ppt.

Eluate 2 Fibrinogen Fr. I suspended Fraction I

Eluate 3 Immunoglobulin G Fr. II+III suspended Fraction II+III

Eluate 4 Alpha1-PI Fr. IV suspended Fraction IV

Eluate 5 Albumin Fr. V suspended Fraction V

CohnCascade

Downstream – PPB ’05 abstracts 9

Table 1. Yield and purity data for the Cohn intermediates are from literature and industry sources. Data for the Cascade (PPPS) Process are from engineering and consistency runs at the 4 l scale and have been confirmed by scale up to the 30 l plasma batch size at Hemosol, Inc., Mississauga, Canada.

Purity YieldSpecific activity/SDS-PAGE % Plasma input

Target protein Cohn intermediate PPPS Capture eluate Cohn intermediate PPPS Capture eluate

vWF/FVIII 2 x 10-5 Cryo precipitate

0.25 45 60

Plasminogen – Fr. III precipitate

0.79 35 77

Fibrinogen 0.65 Cryo precipitate

0.74 40 80

IgG 0.53 Fr. II + III precipitate

0.50 70 87

Albumin 0.95 Fr V precipitate

0.94 85 85

A1PI 0.04 Fr. IV precipitate

0.25 23 90

study was undertaken (and reported separately – see Curling et al.) with Alfa Laval Biokinetics. Although processing costs are roughly equivalent there are 2-fold revenue advantages contributed by vWF/Factor VIII, IgG, albumin and A1PI if all these proteins are isolated and sold as products.

Time for new technologyAs Manucci (4) comments: “Progress is warranted in the quality of plasma fractionation technologies. The yield of Factor VIII from plasma is still only 5–10%, a loss that is difficult to accept in an era of high technology”. Such progress can only be achieved through disruptive technology breakthroughs that, although they can only be used initially in small markets remote from the mainstream, are disruptive because they subsequently can become fully performance-competitive within the mainstream market.

References1. Waeger, R. The future of plasma protein therapies. PPTA Global

Forum, Washington, DC, 11–13 June (2003).

2. Christensen, C.M. The Innovator’s Dilemma. Harper Business, Boston, MA (1997).

3. Curling, J.M, Affinity chromatography. Intl. BioPharm 17, 34–42 (2003) and 17, 60–66 (2004).

4. Manucci, P.M. Hemophilia: treatment options in the twenty-first century. J. Thromb. Haemost., 1349–1355 (2003).

Yield and purityIn contrast to the 5-variable, Cohn cold ethanol precipitation backbone, the Cascade Process operates at ambient temperature with variations of buffer composition as well as ionic strength (conductivity) and pH. Proteins are selectively adsorbed and selectively desorbed from the affinity adsorbents and not subject to precipitation or non-specific adsorption/filter aid steps commonly used in the Cohn backbone or side-fraction work ups. This provides the Cascade with a significant yield advantage and allows closed system and highly automated operation. The process yields and intermediate purities are compared in Table 1.

The improved yields of immunoglobulin G and alpha1-proteinase inhibitor are particularly noteworthy. In the case of IgG, the intermediate purities are similar in both processes but the Cascade provides 6 g of IgG compared to 5 g of IgG out of source plasma with an average IgG content of 7 g/l. For A1PI the intermediate purity is improved 6-fold with a concomitant increase in yield from 23% to 90%.

It was expected that higher in-process yields would lead to higher final product yields and that higher intermediate product purity would reduce the number of downstream processing steps and, therefore, also contribute to improved yield. Potential savings in operating costs could be expected and less plasma would be used to produce an equivalent amount of product. To elucidate the hypothesis, a

10 Downstream – PPB ’05 abstracts

New adsorbents for selective capture of plasma proteins from recovered and source plasmaDev Baines1, Jason Betley1, Ben Beacom1, Tom Chen2, Timothy Hayes2, Jim Pearson1, and Pilar Vazquez1

1 ProMetic BioSciences Ltd., Cambridge, UK. 2 American Red Cross, Plasma Derivatives, Gaithersburg, MD, USA.

e-mail: [email protected]

IntroductionThe Plasma Protein Purification Scheme ‘Affinity Cascade Process’ for the purification of therapeutically useful proteins from human plasma required the discovery of a number of new specific affinity ligands and the development of affinity adsorbents. An overview of the process used to generate three of five new and specific adsorbents for individual plasma proteins is presented. The performance of these affinity adsorbents in the context of the ‘Affinity Cascade Process’ is also discussed.

Ligand designWhen designing Mimetic™ ligands based on ProMetic’s proprietary triazine chemical scaffold, a number of approaches may be employed. If detailed structural information is available for the Target Protein and/or a previously identified ligand, it is possible to use computer assisted rational design to instruct the combinatorial synthesis of candidate ligands on our Purabead™ matrix. In addition, sets of libraries spanning the entire diversity space of chemical structure are routinely screened. If no detailed structural information is available, only the latter approach is utilised in the first instance. Both approaches were utilized in the screens described.

Library synthesis and screening methodsLibraries are synthesised robotically in 96-well plate format, in fritted blocks, generating 8 × 8 arrays of candidate adsorbents occupying Rows A–H and Columns 1–8. The quality of each individual synthesis is monitored by appropriate in-process assays. Libraries are screened with extensive use of robotic liquid handling techniques. A schematic diagram of

the library synthesis process is shown in Figure 1. Each individual library member is treated as a miniaturized chromatography column with a packed bed volume of 0.25 ml; after equilibration of each well and application of feedstock, non-bound, wash, elution, and sanitization fractions are collected in fraction blocks. All of these fractions are then assayed using various high-throughput 96-well plate assays for recovery of Target Protein and Total Protein. Relevant triplicate standard curves are generated for each fraction and each assay, utilizing columns 10–12. Examples of assays used for these targets are Bradford Assay for Total Protein, ELISA assays, and 96-well SDS PAGE (E-PAGE™ from Invitrogen).

Specific screens for human plasma proteinsThree separate screens for candidate ligands specifically binding and eluting Fibrinogen, von Willebrand’s Factor/Factor VIII Complex, and Plasminogen were carried out, using undepleted human plasma as feedstock. For each assay, high-throughput primary screens were used to identify a small number (<100) of candidate adsorbents

Fig 1. Schematic diagram of ligand library synthesis: Amines R1–R8 are added to activated triazine and arrayed in columns 1–8. Amines A–H are then added in rows A–H to generate a 64 component combinatorial library.

Downstream – PPB ’05 abstracts 11

displaying the desired specificity of binding and elution for the Target Protein. A secondary screen was then employed using further analytical techniques to permit the recommendation of a number (4–6) of candidate adsorbents to scale up to the 50 ml scale. The analytical techniques used for the three screens are shown in Table 1. An example of an E-PAGE gel generated during primary screening for a plasminogen ligand is also shown in Figure 2.

Adsorbent developmentOnce a small number of candidate ligands (4–6) were synthesised for each Target Protein, their performance was tested chromatographically, and the most favorable candidate was taken forward for full adsorbent development. Desirable properties for these adsorbents included specificity, binding capacity, purity of eluted protein, step yield, cost-of-goods, and ease of synthesis. After the selection of a single candidate, ligand loading and spacer arm combinations were explored, and the combination permitting the best combination of yield, binding capacity, elution capacity, and purity for the Target Protein was progressed through to technical transfer and manufacturing. The chromatographic performance of the best candidates identified during the three screening projects is shown in Table 2.

Fig 2. E-PAGE 96 Sample Gel Electrophoresis on elution fractions from screening of library 254 for adsorbents specific for human plasminogen from human plasma feedstock. A1–H8: Elution samples from adsorbents A1–H8; Columns 9, 10: Blank; Columns 11, 12: Plasminogen Standard (D11, D12 highlighted in blue); AM, CM, EM, GM: Molecular Weight Markers. Promising candidates in row H are highlighted in red.

ConclusionsLibraries of specifically-designed and/or diverse Mimetic ligands were synthesized, and screened for their ability to purify specific proteins from human plasma. The screen combined specific and ‘Total Protein’ assays in a high throughput fashion to generate leads for each protein. After secondary screening, verification chromatography, and adsorbent development, single candidates were progressed through to manufacturing. The projects were carried out concurrently, and progressed from inception to manufacturing in less than 12 months. All ligands discovered were robust and base stable. All three adsorbents described have been incorporated into the cascade, and display good levels of recovery and purity, and relatively low levels of non-specific binding of non-Target Protein.

Table 1. Assays used in primary and secondary screening.

Screen Assays used in:Primary screen Secondary screen

Fibrinogen Bradford Total Protein Fibrinogen ELISA

Bradford Total Protein Fibrinogen ELISA

SDS PAGE Nephelometry

vWF/FVIII Bradford Total Protein Fibrinogen ELISA

Bradford Total Protein Fibrinogen ELISA

SDS PAGE

Plasminogen E-PAGE SDS PAGE Nephelometry Table 2. Performance parameters for lead candidate adsorbents

after adsorbent development. Undepleted human plasma was used as the feedstock material.

Preformance Fibrinogen vWF/FVIII Plasminogen

Binding capacity (g/l at 10% Breakthrough)

15 – 17

Step recovery (%) > 90 65 > 90

Purity (%) > 85 – > 90

l Plasma/l of adsorbent

7 10 140

Specific activity of eluate (IU/mg)

– 0.25 –

Purification factor 15 40 340

12 Downstream – PPB ’05 abstracts

The application of CaptureSelect affinity ligands for purification of plasma productsLaurens N. Sierkstra*, Mark ten Haaft, and Pim Hermans

B.A.C. BV, Naarden, The Netherlands.www.bac.nlwww.captureselect.com

e-mail: [email protected]

Affinity chromatography is a well-established technology for the purification of molecules from complex source materials. We have developed a technology for the generation of affinity ligands, using small 12 kD fragments derived from single domain antibodies (CaptureSelect™) that can be used for purification of products as well as scavenging of impurities. The advantages of these ligands as compared with other affinity technologies is the fact that our technology combines specificity, affinity, stability, short development times, and ease of non-animal derived production. Moreover, these affinity ligands can be very easily immobilized on different solid support. The specific ligands are generated from immunized antibody fragment libraries, isolated, and recloned into yeast. The requirements for a ligand for a particular application are incorporated in the screening; this could be specific binding and elution conditions, and/or stability of the affinity ligand against caustic cleaning agents. Subsequently, ligands are selected and tested for affinity, selectivity, and stability in chromatography experiments. Finally the ligands are recombinantly produced at any scale in Saccharomyces cerevisiae (Baker’s yeast).

A llama was immunized with a human Fc fragment and the resulting affinity ligands were carefully screened for their ability to bind to all subclasses of IgG (to IgG1–4). By incorporating stability towards causticity in the screening phase we were able to select a human IgG affinity ligand that binds to all IgG subclasses and, additionally, is stable towards caustic cleaning agents. The affinity ligand does not cross-react with IgG’s from other species.

Subsequently the affinity ligand was immobilized on NHS Sepharose™ and we demonstrated repeated cycling with intermediate caustic cleaning for over 120 cycles when using 0.1 M NaOH. This immobilized affinity ligand was then used to purify IVIG from fribrogen free plasma (see Table 1). These data show the excellent yield of the purification step (87.6%) as well as the extremely good reduction in contaminants like IgA and Human Serum Albumin. After the 2nd wash step levels of these contaminants dropped below detection level. For the functional recovery of IVIG it is extremely important that the purified material should have a matching sub-class distribution with respect to the starting material. Table 2 shows that there is a matching subclass distribution for two different pools of IVIG.

Table 1. Purification of IVIG from plasma.NHS Sepharose immobilized human IgG ligandsource material: fibrinogen free plasmaprocess: 25 mg/ml load, 100 cm/h, wash 5CV, eluate I, 5CV, eluate 2, 5CVEq buffer PBS pH 7.4, elution buffer 1, 0.1M HCl/Glyc pH 3.0, elution buffer 2, 0.1 M HCl/Glyc pH=2.0

Albuminmg/ml

IgGmg/ml

IgAmg/ml

Albuminrecovery (%)

IgGrecovery (%)

IgArecovery (%)

Load 15.35 2.58 1.24 100 100 100

Total FT 14.65 0.19 1.13 95.4 7.4 91.1

Wash 1 2.85 0.18 0.2 9.7 3.6 8.4

Wash 2 <0.01 0.08 <0.01 0 1.6 <0.4

Elution pH 3 <0.01 4.15 <0.01 0 87.6 0.2

Downstream – PPB ’05 abstracts 13

Table 2. Subclass distribution of purified IVIG as compared with starting material.

Table 3. Purification of IgG4 from serum.Elution buffer: Glycine/citric acid/HCl pH 3NHS Sepharose

As a second alternative approach for the plasma industry we wanted to demonstrate that besides broad binding (to all IgG subclasses) this technology has the ability to purify single subclass IgG’s as well. Consequently, we immunized a llama with IgG4 only, and carefully screened for affinity ligands that would bind to human IgG4 only and not to other IgG subclasses and species. Table 3 shows the excellent purification of IgG4 from serum. Besides the high recovery levels of IgG4, the level of the other IgG subclasses were signifcantly lowered.

These two examples demonstrate the feasibility of using single domain antibodies as affinity ligands for the purification of plasma products. The use of our technology could greatly enhance yields and purity of these very important products and could also lead to new products isolated from this valuable feed-stock.

Cryo richplasma (1)

Eluate pool(1)

Cryo richplasma (2)

Eluate pool(2)

IgG 1 41.1% 42.7% 37.4% 42.3%

IgG 2 49.8% 52.2% 56.4% 50.9%

IgG 3 3.1% 2.1% 2.1% 2.1%

IgG 4 5.9% 3.1% 4.2% 4.8%

IgG subclass (%) starting material

IgG subclass (%) in elution fraction

IgG 1 57.1 1.7

IgG 2 34.2 0

IgG 3 4.0 0.6

IgG 4 4.7 97.7

14 Downstream – PPB ’05 abstracts

Identification and production of recombinant human polyclonal antibody drugs reflecting the natural human antibody diversitySøren Bregenholt and Allan Jensen

Symphogen A/S, Elektrovej 375, DK-2800 Lyngby, Denmark.

e-mail: [email protected]

Antibodies constitute the so-called humoral arm of the adaptive mammalian immune system protective against infection with pathogenic microorganisms. Antibodies exert their function by binding to antigenic structures, e.g. proteins or carbohydrates, on the surface of the microorganism and mediating the recruitment of a range of effector mechanisms. In natural infections a polyclonal antibody response is elicited, comprising multiple antibodies against the existing antigenic determinants of the pathogen, hence increasing the likelihood of obtaining sufficient antibody binding to the pathogen to confer immunity. The therapeutic principles of antibody-based drugs have long been recognized and so-called passive antibody therapy has been used for more than a century. The development of recombinant technologies has allowed cloning and production of fully human recombinant monoclonal antibodies (MAb), which has significantly expanded the use of therapeutic antibodies. However, MAb are intrinsically mono-specific and might not offer sufficient antibody binding to mediate protective immunity to more complex structures such as pathogens. This is underscored by the widespread use of plasma-derived polyclonal immunoglobulin products against Rhesus D (RhD), hepatitis B virus, rabies virus, cytomegalovirus, vaccinia virus, etc. Although plasma-derived immunoglobulin products have advantageous pharmacological characteristics, their use is limited by a number of disadvantages including high production costs, supply constraints, low specific activity, and risk of pathogen transmission.

We believe that recombinant polyclonal antibodies could overcome these problems associated with plasma-derived immunoglobulin products while retaining the efficacy associated with a polyclonal antibody drug. In order to develop a drug lead with pharmaceutical properties similar to plasma-derived immunoglobulin products, cloning methods able to

capture large proportions of the natural antibody diversity must be employed. Also, methods allowing the cost efficient and reproducible large-scale manufacturing of complex antibody compositions are required.

Fig 1. Overview of the individual steps of the Symplex technology.

Symphogen’s proprietary technology platform, Symplex™, allows the direct cloning and identification of high-affinity antibodies. The technology platform employs a multiplex overlap PCR of variable heavy and light chain antibody genes from human immunoglobulin producing cells, followed by high through-put screening of antigen specific cognate variable heavy and light chain gene pairs (Fig 1). Employing the Symplex technology on peripheral blood mononuclear cells from volunteers vaccinated with a commercial tetanus toxoid vaccine resulted in the isolation of more than 30 unique tetanus toxoid-specific antibodies from each of the processed donors. Diverse repertoires of unique antigen-specific antibodies have also been isolated from donors either vaccinated with an influenza vaccine or infected with vaccinia virus used for smallpox vaccination. Analysis of the vaccinia virus-specific antibody repertoires demonstrated a strong correlation between the immune status of the donors and the

Immune individual- vaccination- natural immunity

Blood donation

Lymphocyteisolation

Symplex™ PCR

High throughput screening e.g. byELISA for antigen-specificity

Symplex™ TechnologyMirroring the naturalimmune response

Cognateantigen-specificVH-VL pairs

CD19/CD38/CD45 positive plasma blasts Single cell-sorted by FACS

Discarded cells

Linked cognate pairs of VH-VL antibodygenes pooled & expressed

Downstream – PPB ’05 abstracts 15

antigen specificity, number and diversity of high-affinity antibody leads isolated from that individual. In addition, these analyses revealed that antibodies from all of the VH families can be isolated by the Symplex technology. Thus, the antibody diversity captured by Symplex is significantly broader than observed with other technologies, thus facilitating the isolation and compilation of polyclonal drug leads that reflect the natural human repertoire against a given antigenic structure.

Our proprietary mammalian expression platform, Sympress™, allows the reproducible expression of complex polyclonal antibody drug leads in a “one-pot” system. The Sympress technology employs site-specific integration securing the insertion of the expression constructs in a predetermined site in the genome of Chinese Hamster ovary cells, which circumvent the potential problem of inconsistent productivity and antibody composition due to genomic positioning effects and potential antibody chain shuffling. The nature of the expression system is so that the individual antibodies comprised in a given drug lead are expressed with identical constant regions. Moreover, the system allows for a free choice of constant region compatible with the desired pharmacological properties of the drug lead. The manufacturing of recombinant human polyclonal antibodies involves conventional cell banking, as well as upstream and downstream processing technologies used for manufacturing of MAb drug leads. The test battery for CMC assessment of the antibodies is also based on conventional methods. Only the assay systems employed to analyze the compositional diversity differ from the conventional CMC test battery. The structural protein characterization occurs at three levels: characterization of structural integrity; detailed description of antibody constant regions; and assessment of polyclonality, the latter combining various techniques to demonstrate that the product is indeed polyclonal and allows batch-to-batch comparison. The Sympress technology has been transferred to an industrial setting for clinical trials manufacturing of Sym001, a recombinant polyclonal anti-RhD antibody composition comprising 25 unique RhD-specific IgG1 antibodies (20 kappa and five lambda antibodies). Sym001 is being developed for treatment of Idiopathic thrombocytopenic purpura (ITP) and prevention of Hemolytic disease of the newborn (HDN). These experiments demonstrated a high level of consistency between different

laboratory-scale batches with respect to productivity (Fig 2), compositional diversity, structural integrity, glycosylation, antigen binding potency, and in vitro pharmacology; erythrocyte phagocytosis and antibody-dependent cellular cytotoxicity (ADCC) (Fig 3). A similar consistency was observed when the process was run scaled up for clinical trial manufacturing.

Fig 2. Reproducible laboratory scale production of Sym001 using Sympress. Antibody titer, from seven individual fed batch runs at laboratory scale, as determined using kappa- and lambda-specific ELISA.

Fig 3. Similar in vitro pharmacology of Sym001 from two individual production runs at laboratory scale. The ability of Sym001, 25 ng/ml to mediate erythrocyte elimination by phagocytosis and ADCC was determined in vitro. Data represent the mean of triplicate cultures, SD values are indicated.

The ability to isolate a diverse natural human antibody repertoire against complex antigenic structures offered by the Symplex technology, combined with the consistent manufacturing of complex polyclonal antibody compositions using the Sympress technology make therapeutic recombinant human polyclonal antibodies a future alternative to MAb and plasma-derived immunoglobulin products for treatment of human diseases.

0

01

02

03

04

05

06

07

08

09

ADCCPhagocytosis

phag

ocyt

osis

/ ly

sis

(%)

16 Downstream – PPB ’05 abstracts

Transgenic goats and cows for the production of human plasma proteinsYann Echelard* and Harry M. Meade

GTC Biotherapeutics, Inc., 5 the Mountain Rd, Framingham, MA 01701, USA.

e-mail: [email protected]

The use of recombinant therapeutic proteins has steadily increased during the last two decades. Clinical applications often require large amounts of highly purified molecules, often for multiple or chronic treatments. The development of very efficient expression systems has been the key to the full exploitation of the recombinant technology. This is particularly true for some of the proteins derived from human plasma, where the combination of a complex structure and large therapeutic dosing have until now precluded the use of traditional bacterial and cell culture bioreactors for recombinant production. However, there is a need for alternative methods. For example, commercial recombinant production of complex molecules such as antithrombin, alpha-1-antitrypsin, or serum albumin, all used in high doses and currently extracted from human plasma, has not yet been achieved in microbial or mammalian cell derived bioreactors. In addition, since capital investments associated with production plants represent a significant portion of the development cost of new recombinant drugs, the inherent risk associated with the regulatory approval process is another stimulus for the development of flexible and inexpensive approaches for the manufacture of therapeutic proteins (Fig 1).

Thanks to a careful integration of molecular biology, large animal embryology, and protein chemistry, transgenic milk production offers a cost-effective system for the manufacturing of large amounts of complex proteins. To achieve milk-specific recombinant protein production, an expression vector, comprising a gene encoding for the human or humanized target protein is fused with mammary gland-specific regulatory sequences and then is inserted into the germline of the selected production species. When integrated, the milk specific expression construct becomes a dominant genetic characteristic that is inherited by the progeny of the founder animal (Fig 2). This general strategy makes it possible to harness the ability of dairy animal mammary glands to produce large quantities of complex proteins. Using this approach, GTC Biotherapeutics and others have generated transgenic animal herds that yield large amounts of proteins as diverse as: human antithrombin (AT), alpha-1 antitrypsin (AAT), C1 esterase inhibitor, fibrinogen, albumin, or even monoclonal antibodies (Table 1). Furthermore, technologies that permit the clinical-grade purification of recombinant therapeutic proteins from the milk of transgenic dairy animals have been developed and implemented.

Fig 1. Schematic representation of the estimated impact of transgenic production on capital costs for the production and harvest of recombinant protein, as compared to cell culture.

Fig 2. Schematic representation of the transgenic production process. In this case, pronuclear microinjection is used to generate transgenic animals. MI, pronuclear microinjection.

Cell Culture

TransgenicGoats

200

100

250 500 750 1000Annual Scale kg

Cap

ital C

ost

$ M

M Increased expression rate Inducelactation

Measureprotein

expression

+

Goatbeta-caseinDNA

gene of interest

•1 mo.

•5 mo.

•5 mo.•6 mo.

transgene

Isolatefertilizedembryos

MItransgene

intoembryo

Transfer intorecipientfemale

Test offspringfor transgene

Select & matefounder

therapeutic protein production herd

milk containing therapeutic protein

• Milk • Milk • Milk • Milk

Downstream – PPB ’05 abstracts 17

Recombinant human antithrombin (rhAT, commercial name ATryn®) is the most advanced of the transgenic milk-derived compounds. After several years in clinical development it has recently completed pivotal trials. The recombinant production of AT presented numerous challenges. Antithrombin is a complex glycoprotein carrying 4 N-linked glycosylation sites and 3 disulfide bonds; these characteristics, crucial for the functions of AT, precluded the use of microbial bioreactors for its recombinant production. In addition, AT is generally used therapeutically in large amounts, often grams of purified protein per course of treatment. This ruled out the use of standard mammalian cell culture bioreactors since production costs with this approach would be prohibitive. Therefore, expression in the milk of transgenic dairy goats was employed. The promoter region of the goat beta-casein gene was linked to hAT cDNA. This transgene was introduced into the chromosomes

of goat embryos, which were then transferred to surrogate mothers. The resulting goats carrying this transgene produce the gene product, rhAT, in their milk. Transgenic offspring consistently express rhAT in their milk at approximately 2 g/l. Goats typically lactate 300 days per year, producing 2–4 liters of milk per day. The rhAT protein is isolated from the milk of the transgenic females and conventionally purified using tangential flow filtration, heparin affinity chromatography, nanofiltration, anion exchange chromatography, and hydrophobic interaction chromatography, with a yield of greater than 50% (Fig 3 and Fig 4). The human AT purified from transgenic goat’s milk is structurally indistinguishable from human plasma-derived AT (hpAT) with the exception of the carbohydrates. The main glycosylation differences observed for rhAT were the presence of fucose and GalNAc, a higher level of mannose, and a lower level of galactose and sialic acid. There was also substitution of 40–50% of the N-acetyl neuraminic acid with N-glycolyl-neuraminic acid. The terminal sialic acid in the rhAT contained the same 2-6 linkage found in hpAT. Several idependent laboratories have determined that differences in glycosylation of AT do not affect the intrinsic rate constant of the uncatalyzed or heparin catalyzed inhibition of thrombin, indicating that the carbohydrate chains solely affect heparin binding and not heparin activation or proteinase binding functions. Thus, glycosylation does not impact the major biological activity of AT which is thrombin inhibition, but does explain the differences in affinity for heparin and in pharmacokinetics.

Fig 3. Schematic representation of rhAT purification from the milk of transgenic animals.

Table 1. Transgenic production pipeline as of June 2005.

Product Company Status

ATryn® (rhAT) GTC MAA in Review US, Phase III

C1 Esterase Inh. Pharming Phase II/III

MM-093 (hAFP) Merrimack/GTC Phase II

rhGH BioSidus Development

rhAlbumin Taurus (GTC/Fresenius)

Development

AAT-r GTC Development

Fibrinogen Pharming Development

MSP-1 Vaccine GTC/NIH Transgenic Goats

CD137 (4-1BB) MAb GTC Transgenic Goats

Step: 1 2 3 4 5 6Fill/lyo

kliM

kliM

kliM

rhAT milk300 l

TFF

Milkfeed

Heparincolumn

Anionexchange

column

Methylcolumn

rhAT300g

~55% Yield

TAhr

TAhr

TAhr

TAhr

TAhr

TAhr

81% 97% 80%

Viral Removal

Filter

Viralinactivation

step

Removed/inactivated

ateach step:

Casein micellesFat globules

BacteriaSomatic cells

Viruses

DNA Lactose

Mineral saltsVitamins

HormonesMilk proteins

Viruses

Viruses LactoferrinMilk proteins

DNAViruses

Goat ATMilk proteins

Viruses

>99.9% pure bycontaminatingprotein ELISA

Viruses

18 Downstream – PPB ’05 abstracts

SDS-PAGE gel

110 kd

84 kd

47 kd

33 kd

24 kd

16 kd

1 5432MW

markersMilk Heparin

eluateANX

eluateHIC

eluate

Silver stained gel

110 kd

84 kd

47 kd

33 kd

24 kd

16 kd

4321MW

markersrhAT rhAT

standardhpAT

Western blot

110 kd

84 kd

47 kd

33 kd

24 kd

16 kd

21 43MW

markersrhAT rhAT

standardhpAT

The manufacturing process for rhAT has been validated for its viral and prion removal capacity. The rhAT viral validation studies demonstrated that a significant virus reduction of ≥8.5 to ≥25.3 log10 was accomplished across the distinctly different modes of the rhAT process. In addition, all GTC goats are certified free of scrapie in the United States Department of Agriculture (USDA) Scrapie Flock Certification Program and various risk minimization measures have been instituted to reduce any potential risk from this TSE in this highly controlled closed donor goat population. The rhAT purification process has been validated for its ability to remove ≥11.3 log10 scrapie.

At this point, ten human clinical studies have been undertaken with rhAT. Two clinical indications were pursued:

• Heparin resistance in patients undergoing cardiac surgery involving CPB

• Prevention of deep vein thrombosis (DVT) in patients who have a hereditary deficiency of AT and who are in high risk situations such as delivery or surgery.

Fig 4. Purity profile for rhAT (process & product). MW, molecular weight markers; ANX, anion exchange column; HIC, methyl column; hpAT, human plasma-derived antithrombin.

In all the human studies completed to date, rhAT has proved safe and met the primary endpoints of the study. All studies demonstrated that rhAT was well tolerated in these patient populations.

A European regulatory filing was submitted in January 2004, for the use of rhAT in the prophylaxis of DVT in Hereditary AT deficient patients in a high-risk situation. If this application is approved, it will constitute the first approval of a transgenically produced biopharmaceutical. Indeed, this will constitute the first approval of a biologic manufactured in a new recombinant production system, since the approval of the first product manufactured in cell culture in the early nineties. The development of a recombinant option for antithrombin will provide a safe and reliable supply of this important factor and may facilitate the resumption of clinical trials aimed at acquired deficiencies of antithrombin such as cardiovascular surgery, severe burns, and severe sepsis. Furthermore, the emergence of the transgenic manufacturing platform will provide an attractive option for the recombinant production of other complex therapeutic proteins that are needed in large amounts.

Downstream – PPB ’05 abstracts 19

Development of recombinant human thrombin as an aid to hemostasis in subjects undergoing surgeryJan Ohrstrom*, John Forstrom, Linda Zuckerman, and Thomas Reynolds

ZymoGenetics, Inc.; Seattle, WA.

e-mail: [email protected]

Intraoperative blood loss is always a concern in surgical settings. Although advances in surgical techniques that result in more efficient surgeries have been made, minimizing bleeding during these procedures remains crucial to reducing peri- and post-operative complications and helps to minimize the overall costs associated with surgical procedures.

Due to their recognized ability to control bleeding, topical hemostatic agents have become standard and are widely used in surgical practice. These agents include various products with differing functions, from passive hemostats such as absorbable gelatin sponges, which provide a physical matrix in which clot formation occurs, to active hemostats, which contain biologic component(s) to facilitate hemostasis. Active hemostats include products such as purified thrombin, thrombin in combination with a gelatin sponge, and fibrin-containing sealants. The hemostatic effectiveness of these products varies across different types of bleeding and patient characteristics (e.g. underlying coagulopathy).

Thrombin, the terminal protease in the coagulation cascade, has many functions in blood coagulation, including conversion of fibrinogen to fibrin, activation of platelets, and formation of fibrin clots. Because of the direct role of thrombin in forming a fibrin clot, thrombin has long been recognized for its hemostatic ability. Today, thrombin is used as a topical hemostat to treat surgical bleeding in over 700,000 surgeries annually in the U.S., and over a million worldwide.

All currently marketed thrombin-containing products are derived from pooled human or bovine plasma. Due to the sources of thrombin used in these products, patients may be exposed to plasma-borne adventitious agents or plasma protein contaminants, which have been associated with detrimental clinical sequelae. ZymoGenetics, Inc. has developed a recombinant human thrombin (rhThrombin) that is efficiently produced from precursor recombinant prethrombin-1 derived from cell culture (Fig 1). The recombinant manufacturing process used to produce rhThrombin has been shown to provide a consistent and well-characterized product that does not contain contaminating human plasma proteins or plasma-borne adventitious agents. The purified rhThrombin product is substantially free of non-thrombin proteins, as assessed by polyacrylamide gel electrophoresis.

Fig 1. Production of Thrombin from Prethrombin-1.

F1 F2 A Bss

Kringle 2Kringle 1Gla domain A-chain B-chain

Serine protease domain

Prothrombin

ss

F2 A BPrethrombin-1

A’ B

ss

Thrombin

20 Downstream – PPB ’05 abstracts

rhThrombin human thrombin

image overlay

A series of tests was performed to compare the structure of rhThrombin with that of native human a-thrombin (Table 1). Primary structural attributes were examined using peptide map liquid chromatography-mass spectrometry (LC-MS) for amino acid sequence. Secondary structural attributes were studied using peptide map LC-MS, far-ultraviolet (UV) circular dichroism (CD), and X-ray crystallography; these tests also showed disulfide bridging and secondary conformation. Finally, tertiary structural attributes were evaluated using near-UV CD, fluorescence, X-ray crystallography, enzyme kinetic analysis, and inhibition kinetics tests. The results indicated that rhThrombin was virtually identical to plasma-derived human thrombin, except for minor glycosylation differences. Figure 2 presents the X-ray crystallographic comparison; the ribbon-plot images of rhThrombin and native human thrombin demonstrate structural equivalence.

In vivo experiments demonstrated the effectiveness of rhThrombin in achieving hemostasis in a dose-dependent manner in two models of surgically induced bleeding: a rabbit liver excisional wound model and a rat hemi-nephrectomy model (Fig 3). The studies demonstrated that rhThrombin reduced the time to hemostasis and blood loss compared to saline controls. Recombinant human thrombin had similar or greater effects than bovine thrombin in these models.

Table 1. Comparison of rhThrombin to human a-thrombin.

Biophysical characterization

Biological activity

Amino acid analysis Inhibitor profiles

N terminal sequencing Clotting activity

SDS-PAGE

Whole mass analysis

Peptide map

Disulfide bonding pattern

Glycosylation

2° and 3° structural (CD, fluorescence spectroscopy)

X-ray crystallography

Fig 2. X-Ray crystallographic structure of Thrombin.

Fig 3. rhThrombin rnhances hemostasis in a tat heminephrectomy model.

ZymoGenetics reached the milestone of providing Investigational New Drug supplies of rhThrombin in 2003 and, that same year, signed an agreement with a commercial manufacturing organization. In 2004, the commercial scale process was fixed to support Phase 3 clinical studies. The clinical development plan included running Phase 1/2 studies in 2003/2004, which proceeded on schedule, and then initiating a pivotal study in 2005, and, finally, submitting a BLA in 2006.

Recombinant human thrombin was evaluated in four concurrent Phase 2 studies in subjects undergoing spinal surgery, liver resection, peripheral arterial bypass, or arteriovenous graft formation for hemodialysis access. All studies were randomized, double-blinded, and were performed in multiple centers. Subjects were treated with either rhThrombin (1000 U/ml) or placebo in combination with an absorbable gelatin sponge applied to a surgical site requiring a topical hemostatic agent. Application of additional open-label rhThrombin was permitted when medically indicated to achieve hemostasis. The primary objective of the study was to evaluate the safety of rhThrombin as determined by incidence and severity of adverse events. Anti-product antibodies were measured by enzyme-linked immunosorbent

0

50

100

120806040200Time (sec)

Con

tinue

d Bl

eedi

ng (%

) S

140

Bovine Thr 1000 U/mlrhThrombin 1000 U/ml

100

Saline

Downstream – PPB ’05 abstracts 21

assay at baseline and 1 month post-treatment. Time to hemostasis was measured at the surgical site(s) for a maximum of 600 seconds. A total of 130 subjects were enrolled in the four studies. The rate of adverse events observed in subjects treated with rhThrombin was generally comparable to that observed in subjects receiving placebo, and was consistent with the expected rate of adverse events following these types of surgery. There were no serious adverse events that were considered related to administration of rhThrombin. One of 83 subjects (1.2%) exposed to rhThrombin developed low-titer non-neutralizing antibodies to rhThrombin. Both hemostasis at 10 minutes and mean time to hemostasis showed a positive trend in favor of subjects treated with rhThrombin. Recombinant human thrombin appeared to be safe and was well tolerated and minimally immunogenic. These findings will be evaluated in a planned confirmatory Phase 3 study.

In summary, there are inherent potential safety advantages of recombinant products as compared to xenogeneic or plasma-derived products. An integrated clinical/manufacturing development strategy can drive key decisions, such as cell line scale-up technique, conservation of downstream steps, and improvement in final formulation. ZymoGenetics used robust product characterization to confirm product similarity and bridge to commercial scale rhThrombin; not only were nonclinical studies conducted, but tests were run to evaluate molecular integrity and detect product or process impurities. In completed Phase 1/2 studies, rhThrombin appeared to be safe, well tolerated, and minimally antigenic. Phase 3 studies with drug product manufactured with the commercial process are planned. Recombinant human thrombin may be a useful adjunct for achieving hemostasis in a broad population of patients undergoing surgery.

22 Downstream – PPB ’05 abstracts

A comparative study of Cohn and chromatographic fractionation using a novel affinity “Cascade Process”John Curling1, Dev Baines1, Christopher Bryant1, Ruben Carbonell2, Tom Chen3, Patrick Gurgel2, and Timothy Hayes3

1 ProMetic BioSciences Ltd., Cambridge, UK.2 North Carolina State University College of Engineering, Raleigh NC, USA.3 American Red Cross, Plasma Derivatives, Gaithersburg, MD, USA.

e-mail: [email protected]

IntroductionChromatographic fractionation, which was first described in 1977 (1), has been implemented at various production scales up to ca. 500,000 liters annually and incorporated in numerous downstream processes of the Cohn ethanol fractionation backbone. Immunoaffinity chromatography is used in the isolation of coagulation factors from plasma and recombinant sources and antithrombin III is classically isolated by heparin affinity chromatography. A new affinity chromatography Cascade Process (see Bryant et al.) using entirely synthetic, Mimetic® ligands on highly cross-linked, beaded agarose, PuraBead®HF to isolate von Willebrand Factor/Factor VIII complex, IgG, albumin and alpha1-proteinase inhibitor (A1PI) was investigated.

Input dataThe study, using the K-TOPS® Knowledge-based Total Optimised Plant Simulation software package, was undertaken together with Alfa Laval Biokinetics to provide an understanding of the economic viability of the Cascade Process and to compare the process with a standardized Cohn fractionation plant. The

evaluation compared plants with an annual production capacity of 500,000 liters, greenfield facilities and a cost level for the north-east United States.

Input information for the Cascade was developed at the 4 liter scale using an early, non-optimized process. Data from later processes and scale up to the 30 liter scale confirms the assumptions from the small-scale model as shown in Table 1 which also includes the yield input data for the Cohn process. Both processes include downstream processes through to final bulk product/active pharmaceutical ingredient.

Model assumptionsFurther assumptions in the model included a 7 day/week, 24 hour operating schedule with the 3rd shift being used for set-up and cleaning. Plant operation was calculated for 44 weeks, equivalent to >500,000 l/year. For the Cohn process a batch size of 3,500 liters was assumed and a 1,750 liter batch size for the Cascade. Albumin batches are processed every day whereas all other proteins are pooled from 2 batches and processed every 2 days. A batch failure rate of 1.3% equal to 2 batches/year was assumed. Key items that were evaluated included batch size, buffer delivery assuming in-line

Table 1. Comparison of yields used in the model with actual intermediate yields (as of May 2005) combined with calculated yields in downstream processes for the Cascade. Cohn yields are from industry and literature sources and have been summarised elsewhere (2).

Used in model Actual/calculated Cascade yields

Plasma conc. Cohn yields Cascade yields Plasma conc.* Yield, % ~Yield, g/l plasma***

vWF/Factor VIII 10 µg/l 18% 52% 870 IU/l** 40% 350 IU/l

SA >0.2 IU/mg

IgG 8.5 g/l 51% 70% 6.8–7.5 g/l 71% 5 g/l

Albumin 35 g/l 86% 73% 32–33 g/l 84% 26 g/l

A1PI 1.5 g/l 15% 68% 1–1.25 g/l 68% 0.7 g/l

* ProMetic/ARC plasma pool by nephelometry, ** Factor VIII by chromogenic assay, *** Based on actual plasma concentrations

Downstream – PPB ’05 abstracts 23

dilution of buffer concentrates, buffer storage, use of bag technology, and facility configuration. Seven production areas were assumed with the backbone of each process separated from the downstream processes. Pre- and post-viral inactivation areas were assigned for each of the proteins except albumin, which is only processed in a pre-inactivation environment.

Facility designBoth facilities were designed as two storey buildings with utilities and services placed above the fractionation areas. Each facility requires about 100,000 sq. ft. but the Cascade footprint (48,000 sq. ft.) is 15% smaller than the Cohn facility due to the absence of ethanol handling. The Cascade requires about 37,000 sq. ft. of Class 100K operating space and 43,000 sq. ft of utility area. Twice as much area (10,000 sq. ft.) is required for the Cascade Process but the fractionation area is much smaller and the Cascade requires minimal cold room space.

Project costsTotal project costs for both facilities operating at half a million liters plasma capacity are approximately US$ 125 million. Included in this sum are: total direct costs, construction costs, engineering costs, contract administration costs, validation, start-up and commissioning costs, and a 13% contingency.

Operating costsThe K-TOPs simulation not only provides facility and engineering design optimization but also, on the basis of current United States pricing for equipment and consumables, an estimate of the operating costs for the facility. All costs were calculated in US$ using 2004 prices. A source plasma cost of $120 was used in the model. In the following discussion, facility utilities include HVAC, clean steam, plant steam, cooling, water and power. Process utilities are the same but assigned to certain unit operations and contain USP water and Water for Injection (WFI). Consumables includes resin, membranes, filter and bags and a 200 cycle usage was assumed for resins. Raw materials includes buffer salts and solvents: sodium hydroxide at 0.5 M is assumed for sanitization and represents a significant cost in this analysis. Labour and QC costs include direct, supervisory and peripheral labor costs at $25 – $40/hour. Fixed costs contain straight-line depreciation (10 years), overhead at 5% and local taxes. Other costs contains failures, royalties, transportation, and maintenance.

In the model it is estimated that the cost of fractionation by Cohn is about $120/l, whereas the Cascade estimate is $134/l. An overall cost comparison is given in Figure 1, and Figure 2 shows the Cascade costs per liter plasma fractionated and how the costs are distributed between the trunk (backbone process) and the downstream processes in the four protein model.

Fig 1. Cost breakdown comparing the Cascade Process with a standardized Cohn process operating at a 3,500 liter batch size or equivalent. Each process contains the backbone fractionation as well as the downstream processes to final bulk.

Cascade Process

Plasma40%

Plasma50%

Fixed costs16%

Other6%Other

5%

Fixed costs13%

Labour/QC19%

Consumables 4%

Consumables 9%

Raw materials 3%

Raw materials15%

Facility utilities 2%Facility utilities 1% Process utilities 0%Waste streams 0%

Process utilities 1%Waste streams 0%

Labour/QC16%

Cohn process

24 Downstream – PPB ’05 abstracts

5550454035302520151050

Facility utilities

Process utilities

Waste streams

Labour/QC

Consumables

Raw materials

Failures

Total

Trunk

A1PI

Albumin

IgG

vWF/FVIII

Utilities

US$ US$160140120100806040200

Cascade costs/liter plasma fractionated Cascade costs per protein fractionated

11%

7%

13%

23%

11%

34%

66%

The total operating costs are in the range $125–$150 million. Major differences in costs are associated with the use of complex buffers for binding and elution as well as sanitization but resin costs do not dominate the cost picture. Both processes consume large volumes of water.

RevenuesWith significant differences in yield between the two processes the Cascade Process is estimated to provide a revenue stream of about $650 compared to the Cohn equivalent of $325, assuming all proteins are manufactured at maximum capacity and are sold. In the analysis it was assumed that vWF/Factor VIII commanded a price of $10,000/g, IVIG was calculated at $38/g, albumin at $2.25/g and A1PI at $330/g. The yield premium for Cascade fractionated immunoglobulin alone provides an additional estimated revenue of $62 per liter of plasma fractionated.

Fig 2. Cascade costs per liter fractionated exclude fixed costs and maintenance and thus reflect processing costs only. Resins costs range from $500 per liter assigned to ion exchangers and up to $1,500 per liter for bulk affinity resins. Costs per protein show dedicated processing costs for downstream processes. The analysis shows that one third of the costs are associated with the trunk and two thirds with individual proteins with albumin costing the most to produce due to the large quantity of protein processed.

References1. Curling, J.M. et al. A chromatographic procedure for the

purification of human plasma albumin. Vox Sang. 33, 97–110 (1997).

2. Curling, J.M. and Bryant, C. The plasma fractionation industry. New opportunities to move forward? BioProcess International March 2005, 18–25.

Downstream – PPB ’05 abstracts 25

Production of a virally safe despeciated equine botulinum antitoxin productHugh Price*, Bill Bees, Lori Soluk, and Andrew Griffiths

Cangene Corporation, 104 Chancellor Matheson Rd, Winnipeg, Manitoba, Canada, R3T 5Y3.

e-mail: [email protected]

Botulism is a disease caused by the toxin produced by C. botulinum. It is normally found in contaminated food products, but recently the Center for Disease Control classified it as a Category A Agent. It is defined as a high priority agent that poses a risk to national security, can be easily disseminated or transmitted from person to person, could result in high mortality rates, might cause public panic and social disruption, and requires special action for public health preparedness. Cangene Corporation decided that production of the Botulinum antitoxin could most easily be produced from equine plasma as there is a good supply of horses and an ample supply of toxin and toxoids to use to immunize horses. Horses can provide large amounts of plasma and eliminates the need for large numbers of human donors. Using equine plasma also reduces the risk of transmitting human borne viruses. The only drawback to using equine plasma is the need to de-speciate the antitoxin using pepsin to remove the antigenic Fc section of the equine IgG molecule. This reduces the chance of serum sickness and increases tolerability. The de-speciation, however, reduces the circulating half-life of the product as compared to intact IgG.

Cangene Corporation has developed a manufacturing process for a de-speciated equine immunoglobulin product. The product is a blend of seven anti-botulinum serotypes (A–G). The process consists of the chromatographic purification of IgG from equine plasma of a single Botulinum serotype, collected from hyperimmunized horses. The horses are immunized with toxoid to a single serotype, followed by the toxin when the horses’ antibodies are at a protective level. This whole immunization process takes more than nine months. The purified IgG is then enzymatically digested with pepsin, which cleaves the antigenic Fc portion from the equine IgG monomer. The pepsin also digests unwanted plasma protein contaminants, which are then removed, with the pepsin, by diafiltration and anion exchange. After virus filtration

the F(ab')2\Fab is diafiltered, formulated, and bulk filtered. Later the bulks of the seven Botulinum serotypes are blended and filled. The end product is a highly purified liquid F(ab')2\Fab blend with ≤2% monomeric IgG fraction, which may be safely administered to humans.

As the plasma is animal derived, there is the potential for transmission of animal viruses by crossing the species barrier and becoming pathogenic to the user. The manufacture of this product incorporates the following for control of potential viral contamination:

1) Utilization of proper herd management – Herd management, like human donor selection management, is critical to the safety of the finished product. The collection of the plasma at the two sites is under GMP with all animals under veterinarian care. There is a complete health record for each horse that includes regular physical examination, daily observation and a written procedure for follow-up of unexpected death. After a 21-day quarantine period, each horse undergoes 9 CFR Adventitious testing before joining the herd (tested for equine infectious anemia, piroplasmosis, glanders, dourine, and brucelosis). All horses are vaccinated against rabies, tetanus, equine influenza, equine encephalitis, West Nile virus, equine herpes, and Streptococcus equi. The horses undergo a deferral from plasma collection after vaccination as well as treatment with equine drugs. The plasma collection facility is set up so that the horses enter, are washed down with a disinfectant and are kept in a pre-collection area prior to plasmapheresis. The actual collection area is segregated and is operated similarly to a human plasmapheresis center. Autopheresis units are used to collect 20 l of plasma per donation. The plasma is aliquoted into bottles and transferred into controlled freezers for storage.

26 Downstream – PPB ’05 abstracts

SD treatent

Diafilter and depth filter plasma

Cation exchange chromatography

Pepsin digestion at pH 3.5

Filtration/Concentration diafiltration

Anion exchange chromatography

Nanofiltration

Plasma pool

Blend serotypes, sterile filter and fill

200 l

Diafilter, formulate, and filter

200 l

200 l

185 l

200 l

50 l

50 l

110 l

10 l

2) Testing of the plasma to confirm the absence of detectable viruses – Testing is performed to ensure that the plasma pool meets the criteria for viral safety, as outlined in the Code of Federal Regulations (9 CFR part 113.53). The plasma manufacturing pool obtained from various equine plasma donations is tested to detect adventitious agents included: rabies, reovirus (REO-3), bovine viral diarrhea virus (BVDV), equine herpes virus (EHV-1), equine arteritis virus (EAV-1). In addition, West Nile Virus (WNV) and Eastern Equine Encephalitis virus (EEEV) are also tested.

3) The incorporation of specific virus clearance steps in the manufacturing process – The first step, which is used on the pooled plasma, is the addition of solvent and detergent (SD). The SD chemicals are added to plasma that has been filtered to 1.0 µ to ensure the removal of particles that could harbor viruses. The plasma with the SD chemicals is incubated >4 hours at room temperature. The SD step is a well-established virus clearance step, and highly effective for the inactivation of lipid-enveloped viruses. The second step is virus filtration of the purified product, employing Millipore’s NFP (normal flow

Fig 1. Process Flow – 200 l Scale.

Table 1. Viral validation.

Model virus Envelope (DNA/RNA)

Family Size (nm)

Resistance to inactivation

Model for:

Porcine Parvovirus No (DNA) Parvo 18–24 Very High Equine Parvovirus small non-enveloped viruses

Xenotropic Murine Leukemia Virus Yes (RNA) Retro 80–110 Low Equine infectious anemia

West Nile Virus (WNV), Bovine Viral Diarrhea (BVDV)

Yes (RNA) Flavi 40–70 Medium Equine encephalitis viruses, Equine viral arteritis, WNV

Parainfluenza III Virus Yes (RNA) Paramyxo 100–200 Low Orthomyxoviruses such as Equine influenza virus

Pseudorabies Virus Yes (DNA) Herpes 150–200 Medium Equine herpes viruses

Adenovirus No (DNA) Adeno 70–90 High Equine adenovirus

Encephalomyocarditis Virus No (RNA) Picorna ~30 High Equine rhinovirus

parvovirus) filter. The virus filtration is performed from one manufacturing area through the wall into a “virus safe” to prevent re-contamination. The virus filtration step does not distinguish between lipid or non-lipid enveloped viruses, but removes viruses based on the virus size. Both virus clearance steps have been validated by scale-down studies using a panel of model viruses. The panel was selected to represent viruses that are potential contaminants of equine plasma, and to represent a wide range of physical and chemical properties in order to ensure clearance of new and emerging viruses.

The bulk product has been tested to meet all our in-process and finished product specifications such as potency, purity, size distribution, residual solvent detergent chemicals, formulation chemicals, residual pepsin, bacteria, and endotoxin. The product has also been characterized by SDS PAGE and the removal of the Fc by Western Blotting.

To summarize, Cangene Corporation has produced a blended equine Botulinum antitoxin against all seven Botulinum serotypes that is virally safe as well as meeting GMP requirements.

Table 2. Virus results.

Virus Nanofiltration log reduction

SD log reduction

Porcine Parvovirus (PPV) 4.5–6.0 NT

Xenotropic Murine Leukemia Virus (XmuLV)

≥ 2.3 ≥ 4.7

West Nile Virus (WNV) ≥ 2.1 ≥ 5.3

Bovine Viral Diarrhea (BVDV) ≥ 4.5 ≥ NT

Parainfluenza III Virus (PI3) NT ≥ 5.6

Pseudorabies Virus (PRV) NT ≥ 5.4

Adenovirus (Ad2) ≥ 4.7 NT

Encephalomyocarditis Virus (EMC) ≥ 4.5 NT

Downstream – PPB ’05 abstracts 27

A high-yield IVIG process with efficient viral clearanceJaakko Parkkinen*, Anne Rahola, Leni von Bonsdorff, Hannele Tölö, and Esa Törmä

Red Cross Finland Blood Service, Kivihaantie 7, FI-00310 Helsinki, Finland.

e-mail: [email protected]

IntroductionIgG has traditionally been separated in large scale from human plasma by the cold ethanol fractionation method developed in the 1940s (1,2) and its subsequent modifications. The early IgG preparations could only be administered intramuscularly or subcutaneously because of adverse effects associated with their intravenous infusion (3). These adverse effects were mainly caused by immunoglobulin aggregates inducing complement activation. Therefore, other manufacturing steps were added for further purification of IgG and removal of aggregates (4). Until the 1980s, IVIG preparations were thought not to transmit viral infections. However, reports of the transmission of HCV by a variety of IVIG preparations lead to serious concern about the safety of IVIG with respect to virus infections (5–6). This necessitated the addition of specific virus inactivation steps to the manufacturing.

The addition of multiple steps to manufacturing of IVIG lowers the yield of IgG and raises the manufacturing costs. At the same time, the increasing demand of IVIG has made the yield even more important. Therefore, the emphasis has lately been to develop completely knew IVIG processes. The development of a chromatographic purification process for IVIG has efficiently improved the yield of IgG (7). Recently, Lebing et al. described a novel process for IVIG that starts from Cohn fraction II+III paste and utilizes caprylic acid treatment and chromatography for purification of IgG (8). Caprylic acid precipitation serves both as an effective virus inactivation and purification step.

The introduction of sensitive screening assays of donated blood and plasma for viral markers and implementation of effective virus inactivation methods has greatly improved the safety of the current plasma products. However, a risk of viral transmission may still exist with physico-chemically resistant agents, which are not effectively inactivated by current chemical virus inactivation methods. Parvovirus B19 is an example of a physico-chemically resistant virus transmitted by plasma products (9). Parvovirus B19 antibodies present in IVIG are useful in the treatment of severe complication of parvovirus infection (10). On the other hand, the virus itself was detectable in IVIG preparations using PCR and could theoretically pose an infectious threat to recipients (9). A case of parvovirus B19 infection transmitted by heat-treated IVIG preparation that led to pure RBC aplasia was recently reported (11) as well as a possible superinfection with a new strain of parvovirus B19 in an already B19-infected IVIG recipient (12). Although the risk of parvovirus transmission with current IVIG products is already low based on the wide presence of protecting antibodies in plasma pools and limitation of virus load by PCR testing (13), other non-enveloped viruses with less commonly occurring neutralizing antibodies may still pose a threat.

Another consideration when planning new IVIG processes is that the original Cohn process has proven effective in removing prions. In particular, the precipitation of fraction II+III effectively removes prions (14,15) and when this step is omitted from IgG manufacturing when aiming at a higher yield, other process steps with corresponding efficacy should be considered for the new process to maintain the same level of safety. The caprylic acid precipitation has been effective in this respect (16).

28 Downstream – PPB ’05 abstracts

Considering additional reduction steps for physico-chemically resistant viruses, virus filtration (nanofiltration) offers an efficient means for removing non-enveloped viruses from solutions of biologically active proteins (17). However, efficacious virus filtration of IVIG preparations with filters that could remove even small viruses, such as parvovirus, has been difficult because of clogging tendency of the filter. This reduces the filtration capacity, decreases yield of IgG and increases the filtrations costs.

We describe a modified caprylic acid process for purification of IgG with high-yield from human plasma. Owing to optimization of filtration conditions and lack of polymeric proteins, the product can be efficiently filtered through a small pore size virus filter. The described process has exceptionally high capacity to remove non-enveloped viruses.

ResultsCaprylic acid treatment followed by ion exchange chromatography yielded pure IgG, but some polymeric IgG was present, which prevented efficacious filtration. By combining caprylic acid treatment with PEG precipitation and a single anion exchange chromatography on ANX Sepharose™ it was possible to purify polymer-free IgG with high yield. The purified IgG solution could be filtered with a composite virus filter (Viresolve™ NFP) using a load of up to 11 kg IgG/m2 with only moderate (less than 50%) decrease in flux under optimized conditions. The yield of IgG in the virus filtration was close to 100% and the yield of purified IgG from plasma about 4.5–5.0 g/kg. As another indication of the high purity of the IgG, the solution could be concentrated above 20% without polymerization. When concentrated to a 10% IgG solution, the product proved stable in accelerated stability studies in conventional formulations. A new formulation with trehalose had even better stability.

When assessed with parvovirus B19 spiking and PCR, the process proved to have high efficacy in removal of non-enveloped viruses. Two process steps (CA+PEG precipitation, virus filtration) both removed about 4 log of parvovirus, total reduction being more than 12 log. The enveloped model virus, bovine viral diarrhea virus (BVDV), was completely inactivated in the caprylic acid treatment.

ConclusionsThe results indicate that it is possible to manufacture with high yield from Cohn fraction II+III stable, polymer-free IgG that can be filtered with high capacity through a small pore size virus removal filter. The polymer removal step also serves as an effective virus reduction step and as a whole the process has exceptionally high capacity to remove even physico-chemically stable viruses.

References1. Cohn, E.J. et al. Preparation and properties of serum and plasma

proteins III. A system for the separation into fractions of the protein and lipoprotein components of biological tissues and fluids. J. Amer. Chem. Soc. 68, 459–475 (1946).

2. Oncley, J.L. et al. The separation of the antibodies, isoagglutinins, prothrombin, plasminogen and beta-lipoprotein into subfractions of human plasma. J. Amer. Chem. Soc. 71, 541–550 (1949).

3. Barandun, S. et al. Intravenous administration of human gamma globulin. Vox Sang. 7, 157–174 (1962).

4. Romer, J. et al.: Characterization of various immunoglobulin preparations for intravenous application. II. Complement activation and binding to staphylococcus protein A. Vox Sang. 42, 74–80 (1982).

5. Lane, R.S. Non-A, non-B hepatitis from intravenous immunoglobulin. Lancet 1983, 974–975 (1983).

6. Bjoro, K. et al. Hepatitis C infection in patients with primary hypogammaglobulinemia after treatment with contaminated immune globulin. New Engl. J. Med. 331, 1607–1611 (1994).

7. Lontos, J. Chromatographic purification of immunoglobulins at CSL Bioplasma: a manufacturing perspective. Plasma Product Biotechnology Meeting 2005.

8. Lebing, W. et al Properties of a new intravenous immunoglobulin (IGIV-C, 10%) produced by virus inactivation with caprylate and column chromatography. Vox Sang. 84, 193–201 (2003).

9. Prowse, C. et al. Human parvovirus B19 and blood products. Vox Sang. 72, 1–10 (1997).

10. Young, N.S. Parvovirus infection and its treatment. Clin. Exp. Immunol. 104, 26–30 (1996).

11. Hayakawa, F. et al. Life-threatening human parvovirus B19 infection transmitted by intravenous immune globulin. Brit. J. Haematol. 118, 1187–1189 (2002).

12. Erdman, D.D. et al. Possible transmission of parvovirus B19 from intravenous immune globulin. J. Med. Virol. 53, 233–236 (1997).

13. Schmidt, I. et al. Parvovirus B19 DNA in plasma pools and plasma derivatives. Vox Sang. 81, 228–235 (2001).

14. Reichl, H.E. et al. Studies on the removal of a bovine spongiform encephalopathy-derived agent by processes used in the manufacture of human immuno-globulin. Vox Sang. 83, 137–145 (2002).

15. Trejo, R. et al. Evaluation of virus and prion reduction in a new intravenous immunoglobulin manufacturing process. Vox Sang. 84, 176–87 (2003).

16. Trejo, S.R. et al. Evaluation of virus and prion reduction in a new intravenous immunoglobulin manufacturing process. Vox Sang. 84, 176–87 (2003).

17. Burnouf, T. and Radosevic, M. Nanofiltration of plasma-derived biopharmaceutical products. Haemophilia 9, 24–37 (2003).

Downstream – PPB ’05 abstracts 29

Emerging technology in hyperimmune therapeutic manufacturingWolfgang Ruesseler

Life Therapeutics, 736 Park North Blvd., Suite 100, Clarkston, Georgia 30021, USA.

e-mail: [email protected]

The plasma fractionation market centers on the core therapeutic products Immunoglobulin, factor VIII, and albumin, which together account for over 70% of the total plasma protein market. The drivers behind these products center on availability of source plasma and the technical ability to produce a high yielding commercial product that meets all safety and regulatory standards. There are in excess of 1000 proteins found in plasma. Only 10 to 15 of these are currently used as therapeutics, although the FDA licensed more than 100 assays for identification of therapeutic candidates. While only three products account for 70% of revenue in the industry (1), there is considerable scope for growth through the development and commercialization of niche plasma therapeutic products.

Manufacturing capacity and capability is widely regarded as a bottleneck in bringing new therapeutic products to market in a cost-effective and timely manner. The intrinsic cost of operating large fractionation facilities demands that the number of products sold per liter of plasma be high in order to provide economic stability for the industry (2). A few plasma products are produced at large scale whereas other proteins are low volume products. Hyperimmunes can comprise the lowest volume plasma products, however, they often have to be manufactured in few, large batches due to facility/equipment size. Consequently, one failure in production can put a major percentage of the supplies required to meet annual demand at risk.

While one might favor hyperimmune production in separate small-scale facilities, this approach can only be profitable if manufacturing cost per dose can be established at or below the cost level of large-scale facilities. As the investment required for current

state-of-the-art fractionation facilities is prohibitive for most market players, the approach of choice is to increase product yield per liter of plasma by deploying new technologies. One technology that fits this bill is Gradiflow™, a patented separation technology offered by Life Therapeutics. Gradiflow provides significantly higher yields and offers new capabilities in product safety when compared with the technologies and processes that have become the backbone of the fractionation industry.

Gradiflow is a preparative electrophoresis system that uses the inherent molecular characteristics of size and charge to separate individual proteins from complex biological mixtures. Gradiflow technology combines traditional membrane tangential flow filtration with native electrophoresis. In operation, the biological material is pumped tangentially over a separation membrane where voltage applied across the sample, facilitates movement of charged species. Thus by selection of buffer pH and membrane pore size, individual proteins can be isolated by means of both size and charge (Fig 1).

Fig 1. Gradiflow technology combines traditional membrane tangential flow filtration with native electrophoresis.

30 Downstream – PPB ’05 abstracts

Layers 2 to 5

Layer 1

Stream 1 grid

Retaining clip

Separation membrane

Stream 2 grid

Restriction membrane

Housing

When applied to plasma fractionation, Gradiflow can be used in a classic ‘cascade’ process where many products can be extracted from a single plasma volume. When applied to hyperimmune production, a single step purification process is employed producing exceptional product recoveries with excellent purity. For IgG purification, yield is consistently above 90%, while purity exceeds 98%. When Gradiflow is combined with an ion-exchange chromatography step, yield remains high with a purity that meets requirements for a therapeutic grade product. The results shown in Table 1 demonstrate Gradiflow capabilities against a number of specifications relevant to hyperimmune products.

Gradiflow’s ability in pathogen clearance is also significant for both enveloped and non-enveloped viruses. Table 2 demonstrates PPV removal calculated using PCR analysis and TCID50 results. In addition, Gradiflow has demonstrated its capacity to remove infectious prions during the purification process. This capability was validated at Q-One Biotech in Glasgow, Scotland. Gradiflow’s virus removal of greater than 4 logs, together with its capabilities in prion removal, offers the unique advantage of a ‘built-in’ safety step. With the regulatory drive to continually improve the safety of blood based products, this feature alone could drive new standards across the fractionation industry such as the opportunity for Gradiflow to be one of several safety steps incorporated into existing purification schemes.

Proof of Principle projects undertaken with major blood fractionators have demonstrated the significant advantages of Gradiflow technology in the area of hyperimmune manufacturing. The technology continually proves its ability to deliver greater yields of plasma protein products with high purity; simultaneous fractionation and pathogen removal; and few processing steps resulting in faster processing time. The technology is also fully scalable. Current pilot facilities are capable of processing 10 l batches of crude plasma. The plate and frame mechanism employed for membrane ‘stacking’ allows classical modular scale up with membrane separation units (not entire instruments) to run either in series or parallel to achieve the processing volume required. (Fig 2 illustrates the plate and frame approach to scale up).

Table 2. PPV removal using PCR and TCID50 analysis.

Samle Nested PCR (log10)

Infectivity (log10 TCID50)

S1 0 hr 7 7.5

S1 residual 6 6.4

IgG product 1 < 1.9

Viral reduction > 6 > 5.6

Table 1. Results obtained for hyperimmune products.

Specification/Assay Averaged ResultsYield (immunonephelometry) Purity – 1D SDS PAGE – 2D electrophoresis – IEF – HPLC

> 90% > 98%

Total Protein (BCA) 49 mg/ml

IgG Aggregation (HPLC) 0.28%

IgG Subclasses (Nephelometry) Corresponds

IgA (Nephelometry) 0.12 mg/ml 5% solution

IgM (RID) < 0.07 mg/ml

Plasminogen/Plasmin 7.3 CTAu/ml 5% solution

pH 4.92 (formulated)

Osmolality 630 (formulated)

Residual Acrylamide BLD

Fig 2. The plate and frame approach to scale up.

Downstream – PPB ’05 abstracts 31

Life Therapeutics is already a well-known supplier of specialty plasma. It plans to combine its plasma collection capabilities with Gradiflow’s competitive advantages in order to become a manufacturer of niche therapeutic products for the hyperimmune market. This vertical integration strategy will allow Life Therapeutics to lever characteristics of the hyperimmune market (i.e. high value products; small production batches) to its own advantage. Further, it will be able to take control of the regulatory approvals required for Gradiflow, which until recently, relied on the business development plans and investment capabilities of potential customers. Joint ventures and strategic partnerships will consolidate the company’s path to becoming a therapeutic manufacturer. In this regard, Life Therapeutics has signed a joint venture agreement with Kedrion, an Italian based fractionator, to develop a Hepatitis C hyperimmune. Additional development opportunities are being pursued such as Gradiflow applications in the biodefence area.

ConclusionThe recovery of IgG from human plasma using conventional techniques is approximately 40%. The increasing demand for hyperimmunes and the opportunities to develop products for new indications necessitates the development of more efficient processes for improved recovery and safety. Gradiflow lifts the bar in terms of product yield and in-process viral and prion reduction. It is a rapid fractionation process with capture rates that make the most of rare and expensive raw plasma.

References1. Lebing, W. PPTA Conference, June 2003.

2. Turner, P. IPPC, 2004

32 Downstream – PPB ’05 abstracts

Development of an in vitro TSE infectivity assay: application to validation of manufacturing processesB. You1, G. Le Hir1, S. Arrabal1, H. Laude2, J. T. Aubin1, and B. Flan1

1 Laboratoire du Fractionnement et des Biotechnologies. Unité de Sécurisation Biologique. Les Ulis – France.2 Institut National de la Recherche Agronomique. Unité de Virologie Immunologie Moléculaires. Jouy-en-Josas, France.

e-mail: [email protected]

IntroductionThe emergence of the new variant Creutzfeldt-Jakob Disease (vCJD), mostly in the United Kingdom, but also in France and other European countries, raised concerns about its possible transmission through exposure to blood and medicinal plasma derived products (MDP). Therefore, regulatory authorities require that manufacturers of MDP perform an assessment of the capacity of their manufacturing processes to remove transmissible spongiform encephalopthies (TSE) infectivity, based on data published in the literature. In the absence of scientific data, they are required to perform specific studies regarding the process efficacy to remove TSE agents. Such studies imply titration experiments of the infectious agent, involving either time-consuming infectivity protocols that consist of intra-cerebral inoculation of laboratory rodents, or less sensitive immunochemical PrP-res detection, mostly by Western Blot (WB).

The Laboratoire Français du Fractionnement et des Biotechnologies (LFB) presents the development of a sensitive in vitro infectivity titration method, named “tissue culture infectivity assay” (TCIA), based on the infection of highly TSE sensitive cells. TCIA was evaluated as a possible alternative to conventional titration methods.

MethodTCIA uses MovS6 cells, a murine Schwann cell line over-expressing the bovine PrP (VRQ/VRQ genotype; (1), and comprises several steps: (i) infection of TSE permissive cells, seeded in multi-well plaques, by serial dilutions of infectious material, (ii) culturing of cells for 6–8 passages, (iii) at each passage, estimating the number of positive wells by WB, and (iv) determination of the infectious titre by the Spearman – Karber method. The titre is expressed in tissue culture infectious dose (TCID50).

ResultsTo assess the feasibility of this approach and to determine optimal conditions for the assay, a brain homogenate (PG 127 sheep scrapie strain transmitted to transgenic mice) was titrated using the TCIA method. Results, summarized in Table 1, showed that at first passage, no positive well was detected, clearly indicating that no background signal due to residual inoculum was evidenced. After each passage, the number of positive wells increased until reaching a plateau after 6–8 passages. At passage 7, the final titre was estimated at 106.4 TCID50/ml. Importantly, all wells corresponding to the negative control, consisting of cells incubated with a normal brain homogenate, showed no positive result for the duration of the assay.

Downstream – PPB ’05 abstracts 33

To determine the TCIA analytical sensitivity, a calibrated brain homogenate previously titrated by bioassay (108.3 ID50/ml, H. Laude), was analyzed by TCIA and WB. Results are shown in Table 2. The titre of the sample was estimated at 107.7 TCID50/ml by TCIA, and at 105.8 wbU/ml by WB. By comparing the results observed for each method, we came to the conclusion that TCIA was ~ 80 times more sensitive than WB, and comparable to bioassay (1 TCID50 = 4 ID50).

TCIA was also shown to be suitable for evaluating the TSE infectivity removal in manufacturing processes used for the manufacture of biological products, using a 15 nm nanofiltration step as a model. A reduction factor of more than 4.2 log10 was found for this step, which is similar to previously reported data from experiments performed by bioassay or WB.

Table 2. Titration of a calibrated brain homogenate using TCIA. The number of positive / total wells is expressed as a function of the dilution (top line) and the passage number (Pn Column). The corresponding titre, calculated by the Sperman – Karber method, is mentioned in the “TCID50/ml” column.

Pn 10-4 10-4.7 10-5.4 10-6.1 10-6.8 10-7.5 Neg TCID50/ml

1 0/5 0/5 0/5 0/5 0/5 0/5 0/5 < 4.1

5 NT 5/5 5/5 2/5 1/5 0/5 0/5 7.0

6 NT 5/5 5/5 5/5 1/5 0/5 0/5 7.4

7 NT 5/5 5/5 5/5 2/5 0/5 0/5 7.6

8 NT 5/5 5/5 5/5 3/5 0/5 0/5 7.7

9 NT NT NT 5/5 3/5 0/5 0/5 7.7

10 NT NT NT 5/5 3/5 0/5 0/5 7.7

Table 1. Titration of a brain homogenate using TCIA. The number of positive / total wells is expressed as a function of the dilution (top line) and the passage number (Pn Column). The corresponding titre, calculated by the Sperman – Karber method, is mentioned in the “TCID50/ml” column.

Pn 10-4 10-4.7 10-5.4 10-6.1 10-6.8 Neg TCID50/ml

1 0/5 0/5 0/5 0/5 0/5 0/5 < 3.7

2 1/5 0/5 0/5 0/5 0/5 0/5 3.4

3 5/5 3/5 0/5 0/5 0/5 0/5 5.6

4 5/5 5/5 3/5 0/5 0/5 0/5 6.3

5 NT 5/5 3/5 0/5 0/5 0/5 6.3

6 NT NT 4/5 0/5 0/5 0/5 6.4

7 NT NT 4/5 0/5 0/5 0/5 6.4

ConclusionsThe TCIA method has shown several characteristics that make it an interesting alternative to conventional titration methods in the study of TSE: it is highly sensitive/specific, it is fast (less than 8 weeks) and low cost, it measures the presence of infectivity, and is suitable for the validation of manufacturing processes. Further developments are ongoing, including comparison with bioassay in terms of sensitivity, standardization, sensitivity enhancement, and reducing time of duration of the assay. Such improvements should lead to the development of a standardized assay that will represent a reliable alternative to current bioassay techniques.

Reference1. Archer, F. et al. Cultured peripheral neuroglial cells are highly

permissive to sheep prion infection. J Virol. 78, 482–90 (2004).

34 Downstream – PPB ’05 abstracts

Biochemical and pre-clinical characterization of a new 10% liquid triple virally reduced human intravenous immune globulin (IGIV, 10% TVR)Wolfgang Teschner*1, Harald Arno Butterweck1, Wilfried Auer2, Eva Maria Muchitsch2, Alfred Weber3, and Hans-Peter Schwarz4

1 Baxter AG, Industriestr. 131, A-1220 Vienna.2 Baxter AG Industriestr. 20, A-1220 Vienna.3 Baxter AG, Benatzkygasse 2 –6, A-1220 Vienna.4 Baxter AG, Wagramerstr. 17–19.

e-mail: [email protected]

The purity, efficacy, safety, pharmacokinetics, and toxicity of a new 10% liquid intravenous immune globulin (IGIV,10%TVR) from human plasma was investigated and compared with Gammagard S/D, a licensed lyophilized IGIV.

The new process for the IGIV,10%TVR combines the well-established ethanol fractionation with advanced chromatographic purification of the IgG by ion exchange chromatography in the downstream process, resulting in a high quality product. Three dedicated virus clearance steps, S/D-treatment, 35 nm nanofiltration and incubation of the final container formulated at low pH for an extended time period, are incorporated into the downstream process.

Purity is demonstrated by a g-globulin content of ~ 100% as measured by cellulose acetate electrophoresis, by absence of amidolytic (plasmin-like) activity as measured by the chromogenic substrate PL-1 (Pentapharm) and by low content of other Ig-classes (Table 1). The use of valuable raw material, i.e. human plasma, is optimized since the new process offers the possibility to purify coagulation factors and inhibitors from the same starting material without any impact on the properties of the final container (Table 1). IgG subclass distribution is found comparable to the physiologic range in normal plasma. Molecular size distribution analysis demonstrates an IgG monomer and dimer content of ~100%.

Table 1. IGIV,10%TVR final container purity (Mean values of manufacturing lots).

Parameter Method Units Pre-adsorption of blood coagulation factors and antithrombinNone (3 Lots) FEIBA, Antithrombin

(2 Lots)F-IX, F-VII, Antithrombin

(2 Lots)F-IX, F-VII (3 Lots)

g-globulin CAE % 100 100 100 100

IgG Immuno-nephelometry

% of total Protein

104 103 103 100

IgA ELISA mg/ml 0.03 0.03 0.04 0.02

IgM RID mg/ml <0.016 <0.016 <0.016 <0.016

Amidolytic Activity

PL-1 nmol/ml min <10 <10 <10 <10

Downstream – PPB ’05 abstracts 35

Efficacy is demonstrated in vivo by mouse protection test using Klebsiella pneumoniae and Streptococcus pneumoniae and in vitro by opsonization, by the EP test method for Fc functional integrity, and by a FACS method.

The protective activity of IGIV,10%TVR against systemic bacterial infections in mice is at least as good as of Gammagard S/D. The opsonophagocytosis effect of IGIV, 10 % is determined as the capacity of the product to promote phagocytosis and intracellular killing of either Gram negative (E. coli O55:K59; ATCC #12014) or positive (S. agalactiae, Group B; ATCC #13813) bacteria by human neutrophils (PMNs) in the presence of a complement source. The Lot Values show that IGIV,10% TVR lots give values similar to the reference Gammagard S/D (Table 2). The European Pharmacopoeia test method for Fc functional integrity of IGIV preparations uses the hemolysis of rubella antigen coated red blood cells by IgG and complement. As the EP method requires specific anti-rubella antibodies in the IGIV preparation to mediate the Fc-dependent biological effect, the Fc function was also evaluated by a flow cytometric binding assay. This test is based on a method originally described by Ramasamy et al. (Vox Sang 78, 1785–193 [2000]) and measures the binding of IGIV to Fc receptors of the human monocytic cell line

THP-1. In both tests measuring the Fc functionality the average % value of ten manufacturing batches is close to 100% (98.6%; Table 2).

The blood pressure lowering effect in spontaneously hypertensive rats and the bronchospastic effect in guinea pigs are used as indicators of the anaphylactoid potential. The bronchospastic activity is measured after rapid intra-arterial injection of a test article into an anesthetized guinea pig. A guinea pig model is chosen because of the known sensitivity to substances which could cause anaphylactoid reactions resulting in an increase in pulmonary inflation pressure. Positive reactions are not observed after administration of IGIV,10% lots, Gammagard S/D lot 99H25B11 or formulation buffer. Only one animal showed a positive reaction after injection of 1000mg/kg Gammagard S/D lot 00G07AX11.

Using the hypotensive rat model a potential drop in blood pressure after administration of the test or control items is measured. Statistically, no significant difference can be detected for the decrease in blood pressure between the lots of IGIV,10% and the lots of Gammagard S/D or the formulation buffer. The results demonstrate that the risk of anaphylactoid reactions after the administration of the IGIV,10% is low and similar to Gammagard S/D.

Table 2. Functional integrity of IgG in IGIV, 10% TVR. (Mean values of final container manufacturing lots).

Parameter Average Range

Opsono-phagocytosis Ratio to reference Gammagard S/D

E. coli 1.2 1.0–1.6

S. agalactiae 1.1 0.9–1.2

Fc Function EP-method % 98.6 84–110

FACS-method % 98.6 82–121

36 Downstream – PPB ’05 abstracts

Table 3. Influence of IGIV, 10% TVR and Gammagard S/D on vital functions in dogs.

Article Dose Lot No. Cardiovascular function

Respiratory function

DIC

(Positive reactions) (Positive reactions) (Positive reactions)

IGIV, 10%TVR 500mg/kg 01C21AN11 0/4 0/4 0/4

01C21AN21 0/4 0/4 0/4

01D05AN11 0/4 1/4 0/4

Gammagard S/D (active control)

500mg/kg 99H25AB11 1/4 1/4 0/4

Formulation buffer (negative control)

5ml/kg 01D26AT11 0/4 0/4 0/4

The influence of the products on vital functions like cardiovascular, respiratory, and blood coagulation variables is tested in dogs. This study suggests that the risk of a positive reaction related to cardiovascular or respiratory function or the development of a disseminated intravascular coagulation (DIC) after administration of IGIV,10% is at least as rare as that observed after infusion of Gammagard S/D (Table 3).

The thrombogenic potential is measured in rabbits by determining thrombus formation. The formation of thrombi is evaluated using a scoring system proportional to thrombus formation (0 = liquid blood without thrombi; 0.5–1 = a few small thrombi; 2 = several middle-sized thrombi or many small thrombi; 3 = a greater number of middle-sized thrombi; 3.5 = few larger thrombi; 4 = one large thrombus). In none of the animals treated with IGIV,10% was a score higher than 2 observed. The overall mean scores of 0.60 for IGIV,10% TVR suggests that the thrombogenic potential is low for IGIV,10%.

Pharmacokinetics in rats is evaluated by the variables “in vivo recovery” and “half-life” after a single intravenous dose of 1000 mg/kg. Blood samples were drawn from each individual animal one week before application and 1, 4, 8, 24, 48, 72, 96 and 168 hours after injection of the immune globulin. In summary, the results of the pharmacokinetic study show that the values of the in vivo recovery and the half-life of the a- and b-phase are very similar for the lots of IGIV,10% TVR and for Gammagard S/D.

In acute toxicity studies carried out in mice and rats, IGIV,10%TVR compares favorably to Gammagard S/D. The “No observed adverse effect level” (NOAEL) for IGIV, 10%TVR for this study in mice was 5000 mg/kg, and 2500 mg/kg for the reference Gammagard S/D. In rats the NOAEL was 2000 mg/kg for IGIV, 10% and below 2000 mg/kg for Gammagard S/D.

In conclusion, liquid IGIV,10%TVR combines high purity with excellent efficacy, safety, and tolerability in pre-clinical studies.

Downstream – PPB ’05 abstracts 37

Efficacy and safety of subcutaneous immunologlobulin replacement therapy at home in patients with primary immunodeficiency diseases: combined analysis of two clinical studies, one in North America and one in EuropePeter Kiessling1, Hans D. Ochs2, Michael Borte3, Michaela Praus4, and Hubert Heinrichs1

1 Clinical Research & Development, ZLB Behring GmbH, Marburg, Germany.2 Department of Pediatrics, University of Washington, Seattle, WA, United States.3 Department of Pediatrics, Municipal Hospital “St. Georg”, Leipzig, Germany.4 Biostatistics, Accovion GmbH, Marburg, Germany.

e-mail: [email protected]

Rationale of the studyIntravenous infusions of immunoglobulin (IVIG) every 2–6 weeks have been the standard therapy for patients with primary immunodeficiencies (PID). Complications of IVIG therapy such as poor venous access, inability to perform self-infusion, and systemic reactions such as headache, fever and chills limit the use of IVIG at home. Weekly self-administered subcutaneous immunoglobulin infusions (SCIG) at home are becoming an alternative treatment regime. In the present study, we evaluated the safety and efficacy of a 16% pasteurized, preservative-free liquid human IgG preparation intended for subcutaneous use.

Table 1. Participating study sites and Main investigators.

North American Trial European and Brazilian Trial

USAArthur H. Althaus Louisville, KYPedro C. Avila, San Francisco, CAMelvin Berger, Cleveland, OHSudhir Gupta, Irvine, CARobert W. Hostoffer, South Euclid, OHLisa J. Kobrynski, Atlanta, GARobyn Levy, Atlanta, GALaurie Myers, Durham, NCHans Ochs, Seattle, WA

G. Wendell Richmond, Oak Park, ILRobert L. Roberts, Los Angeles, CASuzanne Skoda-Smith, Gainesville, FLRalph Shapiro, Plymouth, MNMark Stein, North Palm Beach, FLCanadaChaim Roifman, TorontoDonald F. Stark, Vancouver, BCPeter Vadas, Toronto

AustriaAndreas Böck, ViennaBrazilBearitz Costa Carvalho, Sao PauloGermanyMichael Borte, LeipzigHartmut Peter, FreiburgIlka Schulze, BerlinTim Niehues, Düsseldorf

PolandEwa Bernatowska, WarsawJan Kus, WarsawSpainNuria Matamoros, Palma de MallorcaDolores Hernandez, ValenciaOscar Asensio, SabadellSwedenCarl Granert, Stockholm

Methods and design1. DesignIn two prospective studies, one in the US and Canada (NA study), the other in Europe and Brazil (EU study), 125 PID patients between 3 and 74 years of age self infused SCIG (Vivaglobin®, ZLB Behring) on a weekly basis at home. The patients began SCIG therapy one week after their last IVIG infusions, and entered a 3-month wash-in/wash-out period followed by the 12-month efficacy period in the NA study and a 6-month efficacy period in the EU study. Clinical endpoints included the rate of serious bacterial infections (SBI), rates of all types of infections, as well as serum (S) IgG levels observed during the study. Safety variables comprised local and systemic reactions, laboratory investigations, and vital signs.

38 Downstream – PPB ’05 abstracts

2. Product characteristics• Vivaglobin is a 16% liquid, pasteurized, polyvalent

human normal immunoglobulin G preparation manufactured by ZLB Behring, Marburg, Germany. The manufacturing of Vivaglobin contains two virus inactivation steps: precipitation by ethanol in the presence of fatty alcohols and pasteurization.

• Vivaglobin is prepared from plasma donations that have been tested negative for antibodies to HIV-1, HIV-2, HCV, and for HBs antigen. The plasma pools were tested non-reactive for HBV DNA, HCV RNA, and HIV-1 RNA utilizing sensitive and specific NAT/PCR NA

Table 4. Annual rate of infections.

NA Study (n=51)

EU Study (n=47)

Annual rate of serious bacterial infections (SBIs/subject year)

0.04 0.04

Upper 1-sided 99% CI for SBIs 0.14 0.21

Annual rate of any infections (infections/subject year)

4.4 4.3

Table 2. Patient’s demographic and disease characteristics.

Population characteristic

NA study(N=65)

EU study(N=60)

Pooled(N=125)

Age (years): Mean 33.9 27.2 30.7

Gender: Male/Female 37/28 43/17 80/45

Duration of disease (years): Mean

12.8 10.4 11.6

IgG level at baseline (mg/dl): Mean

223 236 230

Table 3. Patient’s primary diseases (number of subject’s).

NA study(N=65)

EU study(N=60)

Pooled(N=125)

CVID 49 (75.4%) 36 (60.0%) 85 (68.0%)

Congenital hypo - or agammaglobulinemia

16 (24.6%) 18 (30.0%) 34 (27.2%)

Others* 0 6 (10%) 6 (4.8%)

* Include IgG2 Subclass Deficiency (1), M. Bruton (1), Nijmegen - Breakage Syndrome (1), Subclass deficiency IgG (1), Severe combined immunodeficiency (1), Wiskott - Aldrich – Syndrome (1)

ResultsA total of 5,953 infusions were administered to 125 patients in the course of the two studies. The patients received a weekly mean dose of 158 mg/kg in the NA study and 89 mg/kg in the EU study during the efficacy phase of the studies. Only three serious bacterial infections (pneumonias) were reported during the efficacy phase in the two studies, two in the NA study and one in the EU study, resulting in an identical annualized rate of 0.04 SBI per patient year (Table 4). The annualized rate for any kind of infection was similar in both studies with 4.4 episodes/patient year in the NA study and 4.3 episodes/patient year in the EU study (Table 4). Sinusitis and upper respiratory infections were the most frequently reported types of infection. Mean S-IgG levels increased from 837 mg/dl to 922 mg/dl at 101% of the previous IVIG dose in the EU study and from 786 mg/dl at start of SCIG to 1040 mg/dl during the efficacy phase at 136% of the previous IVIG dose in the NA study (Table 5). No related serious adverse events were reported in any of the studies. Local injection site reactions (Table 7), of mostly mild or moderate intensity, dropped rapidly in both studies from initially 85% to about 40% during the course of the NA study and from 65% to about 20% in the EU study.

Downstream – PPB ’05 abstracts 39

Table 5. Vivaglobin dose and IgG levels.

NA Study (n=51)

EU Study (n=47)

Vivaglobin dose (% of IVIG) 136 101

IgG increase during Vivaglobin Therapy (mg/dL)

255 86

Vivaglobin IgG level (mg/dL) 1040 922

Fig 1. The percentage of patients reporting local tissue reactions after SCIG infusion. NA trial: Patients who have been pre-treated with IGIV [ • ] EU trial: Patients who have been pre-treated with IGIV [ º ] Patients who have been treated with SCIG before the trial [ * ].

Table 7. Most frequent adverse events by patient, irrespective of relationship to the study drug.

Adverse event ( ≥ 10% subjects)*

NA Study (n=65)

EU Study (n=60)

Injection site reaction 60 (92%) 44 (73%)

Headache 31 (48%) 21 (35%)

Gastrointestinal disorder 24 (37%) 26 (43%)

Fever 16 (25%) 25 (42%)

Nausea 12 (18%) 0 (0%)

Sore throat 11 (17%) 5 (8%)

Rash 11 (17%) 2 (3%)

Allergic reaction 7 (11%) 5 (8%)

Skin disorder 5 (8%) 8 (13%)* Excluding infections

ConclusionsTwo major clinical trials have demonstrated that weekly self-administration of SCIG with a 16% IgG preparation is safe and effective in patients with PID, resulting in normalized stable IgG levels and providing satisfactory protection against severe (e.g. bacterial) infections.

Table 6. Vivaglobin Dosage and SBIs*.

NA Study (n=51)

EU Study (n=47)

Mean Vivaglobin dose (mg/kg/week)

158 89

Range Vivaglobin dose (mg/kg/week)

34–352 51–147

Annual rate of serious bacterial infections (SBIs)

0.04 0.04

NA

EU, excl. previous SCIG

Swedish, previous SCIG

Subj

ects

with

inje

ctio

n si

te re

actio

ns (%

)0

10

20

30

40

50

60

70

80

90

100

Infusion

706050403020100

40 Downstream – PPB ’05 abstracts

Alpha-1 acid glycoprotein and apotransferrin effectively protect against ischemia-reperfusion injuryJaakko Parkkinen1, Bart de Vries2, and Wim A Buurman2

1 Finnish Red Cross Blood Service, Kivihaantie 7, FI-00310 Helsinki, Finland.2 Department of Surgery, Nutrition and Toxicology Research Institute Maastricht, Academic Hospital Maastricht, Maastricht, The Netherlands.

e-mail: [email protected]

IntroductionIschemia reperfusion injury takes place when blood flow to one or several organs is abruptly decreased or completely stopped (= ischemia) and the supply of oxygen-rich blood is subsequently restored (= reperfusion). Clinical conditions in which ischemia-reperfusion (I/R) injury plays a pathophysiological role include organ transplantations, myocardial infarction and revascularisation procedures, cardiac and angioplastic surgery, stroke and restoration of adequate blood flow after shock.

Currently, there are no effective means of preventing I/R injury in clinical practice other than restriction of the ischemic period to a minimum. Several drug candidates have failed in clinical trials, particularly in neuroprotection, and no single therapeutic concept has proved effective enough so far. It appears that targeting several pathogenic mechanisms at the same time would be needed for therapeutic efficacy. Plasma proteins offer an interesting option in this respect as there are several cytoprotective plasma proteins already known and many plasma proteins have several interaction sites with cells and other molecules. Two major plasma proteins have turned out to be promising candidates in prevention of I/R injury: alpha-1 acid glycoprotein (AGP) and apotransferrin (apo-Tf) (1,3).

MethodsThe plasma proteins were tested in a mouse model of renal ischemia-reperfusion injury comprising 45 minutes of unilateral ischemia of the kidney followed by contralateral nephrectomy (1). At

reperfusion, mice were administered intraperitonally with the substances studied, or sterile PBS. Animals were killed at indicated time points after reperfusion, blood samples collected, and the ischemia-damaged kidney harvested for biochemical and immunohistochemical analysis addressing reactive oxygen species, apoptosis, integrity of renal epithelia, kidney function, neutrophil influx and complement activation.

Apo-Tf was purified from Cohn fraction IV paste by the process described before (4) and AGP from fraction V supernatant by ion exchange chromatography and ultrafiltration.

ResultsStudies with alpha-1 acid glycoproteinBoth bovine (1) and human AGP (2) effectively protected against loss of renal function when given at the point of reperfusion. Whereas 0.1 mg AGP already provided a significant protection as reflected by the reduced blood urea levels; maximal protection was provided using 5mg of AGP (Fig. 1). To unravel the mechanisms, we examined whether high levels of endogenous AGP could protect against renal I/R injury. Transgenic mice over-expressing rat AGP were no better protected than the wild-type controls, indicating that administration of purified AGP was needed for the effect. AGP dose-dependently reduced the influx of neutrophils induced by renal I/R injury. Interestingly, fucose-depleted AGP appeared to have the same inhibitory effect on neutrophil infiltration as naturally fucosylated AGP, indicating that the effect was not mediated by inhibition of selectin-mediated neutrophil adhesion. Renal I/R led to significant renal

Downstream – PPB ’05 abstracts 41

deposition of complement C3, which was dose-dependently inhibited by AGP. The apoptotic cell death in the course of renal I/R injury is localized in the tubular cells, and these cells displayed severe actin cytoskeletal derangements in the control mice. In contrast, AGP-treated animals showed minimal injury to the tubular brush border and a well-preserved tubular structure. Treatment with AGP also prevented I/R-induced destruction of tight junctions between the epithelial cells.

Studies with apotransferrinRenal I/R injury induced a significant release of redox-active iron into the circulation after 2–24 hours of reperfusion. This indicated that endogenous iron-binding capacity, which in physiologic situations prevents the availability of redox-active iron in the circulation, is overwhelmed in the course of renal I/R injury. Treatment with apo-Tf (0.5 mg) resulted in a significant decrease of circulating redox-active iron. Apo-Tf but not iron-saturated Tf (holo-Tf) also reduced formation of reactive oxygen radicals in the ischemic kidney after reperfusion (4).

Apo-Tf reduced the influx of neutrophils in a dose-dependent manner, whereas holotransferrin had no evident effect on neutrophil infiltration. On the other hand, both Apo-Tf and holo-Tf significantly

reduced C3 deposition after renal I/R. These data indicated that transferrin exerts different types of protective effects, some of which are dependent on iron binding and others not. In contrast to AGP, apo-Tf did not prevent apoptosis in the renal I/R model and in the holo-Tf case even worse apoptosis. Thus, the protective effects of apo-Tf and AGP depended on discernable mechanisms. Finally, administration of apo-Tf dose-dependently reduced the loss of renal function in contrast to holo-Tf, which did not affect renal function as compared with PBS treatment (Figure 1).

Figure 1. Administration of apotransferrin and alpha-1 acid glycoprotein effectively reduce loss of renal function after ischemia-reperfusion (I/R). Renal function was measured by blood urea nitrogen (BUN) 24 hr after I/R (n=6 per group). BUN values are expressed as mean±SEM. Statistical significance as compared with control-treated animals denoted as (*) at P <0.05 and (**) at P <0.01.

Table 1. Properties of alpha-1 acid glycoprotein (AGP)

• Most abundant immunocalin in plasma, acute phase reactant

• Like other immunocalins, has a hydrophobic binding cleft and potential other interaction sites, but its physiological ligands are not known

• AGP effectively inhibits pathogenic events during renal I/R injury:– prevents apoptosis– prevents neutrophil infiltration– prevents complement activation– protects integrity of tubular epithelia– protects kidney function– exogenous purified AGP required for the

protective effect• In other animal models AGP effectively protects

against hemorrhagic and septic shock• Molecular mechanisms under investigation:

neutralisation of inflammatory mediators – evidence for several mechanisms

Table 2. Properties of apotransferrin (apo-Tf)

• Iron transport protein in plasma, binds iron in redox-inactive form

• Displays additional iron-independent cytoprotective mechanisms

• Apo-Tf effectively inhibits pathogenic events during renal I/R injury:– binds redox-active iron liberated from cells

during ischemia– prevents generation of reactive oxygen

radicals in renal tissue– prevents neutrophil infiltration– prevents complement activation– protects kidney function

• Proved safe and well tolerated in phase I/IIa clinical trials in iron chelation indication in hematologic stem cell transplantation patients

42 Downstream – PPB ’05 abstracts

DiscussionWe have shown that two major plasma proteins, AGP and apo-Tf, exert potent protective effects against reperfusion injury by targeting different mechanisms. The finding that the protective effect of AGP depended on administration of purified protein whereas high endogenous level of AGP was not protective suggests that AGP may accumulate and neutralise endogenous inflammatory mediators. This would fit to its structural homology to other immunocalins which typically bind small hydrophobic proteins. In fact, we have identified high affinity binding of pathophysiogically relevant bioactive lipids to AGP (Ojala et al., unpublished data). Earlier data suggest that anti-inflammatory effects of AGP are partially dependent on its carbohydrate chains and cells also exhibit high affinity receptors for AGP (5). Thus, AGP probably exerts different type of protective effects mediated by its different functional sites.

Redox-active iron has been implicated in the pathophysiology of I/R injury in several organ systems. Apo-Tf, the endogenous iron-binding protein, significantly reduced the amount of circulating redox-active iron after renal I/R, reduced oxygen free radical formation and protected against acute renal failure. Interestingly, the complement–inhibitory effect of apo-Tf was not mediated by iron binding. Lesnikov et al. (6) have recently described potent cytoprotective effects for apotransferrin, which were also partially iron independent. Apo-Tf has proved well tolerated in phase I/IIa clinical trials in leukaemia patients receiving allogenic bone marrow transplantation, and apo-Tf effectively bound the liberated iron in the circulation of the patients (7).

Taken together, AGP and apo-Tf are promising candidates as effective and safe plasma protein therapeutics for disease states mediate by I/R injury. Both proteins can be recovered by simple processes from side fractions of plasma fractionation with a yield of 0.2–1.0 g/kg plasma. Both proteins are also relatively simple and stable proteins, suitable for most virus inactivation and removal methods. The estimated effective doses of the proteins are in the order of several grams, which makes their purification from plasma a competitive manufacturing approach as compared to transgenic or recombinant expressions technologies.

References1. Daemen, M., et al: Functional protection by acute phase proteins

alpha(1)-acid glycoprotein and alpha(1)-antitrypsin against ischemia/reperfusion injury by preventing apoptosis and inflammation. Circulation 102 1420–6 (2000).

2. de Vries B, et al.: Exogenous alpha-1-acid glycoprotein protects against renal ischemia-reperfusion injury by inhibition of apoptosis and inflammation. Transplantation 78 1116–24 (2004).

3. de Vries, B. et al. Reduction of circulating redox-active iron by apotransferrin protects against renal ischemia-reperfusion injury. Transplantation 77 669–675 (2004).

4. von Bonsdorff, L. et al. Development of a pharmaceutical apotransferrin product for iron binding therapy. Biologicals 29 27–37 (2001).

5. Hochepied, T., Berger, F.G., Baumann, H., Libert C: Alpha(1)-acid glycoprotein: an acute phase protein with inflammatory and immunomodulating properties. Cytokine Growth Factor Rev 14 25–34 (2003).

6. Lesnikov, V.A. et al Prevention of Fas-mediated hepatic failure by transferrin. Lab Invest. 84 342–52 (2004).

7. Sahlstedt, L, et al Effective binding of free iron by a single intravenous dose of human apotransferrin in haematological stem cell transplant patients. Br. J. Haematol, 119 547–53, (2002).

Downstream – PPB ’05 abstracts 43

44 Downstream – PPB ’05 abstracts

List of posters presented at PPB 2005, Crete, Greece* denotes author for correspondence

1. Development of a Method to Detect Contaminant Proteins in Immunoglobulin G Preparations IV. A Proteomics ApproachPeter Gomme*1, Judith Lysaght2, Bernie McInerney2, and Joe Bertolini11CSL Bioplasma 189–209 Camp Road, Broadmeadows, Victoria 3047, Australia. 2Australian Proteome Analysis Facility, Level 4 Building F7B, Macquarie University, North Ryde, NSW 2109, Australia.

2. An Abbreviated Affinity Chromatography Cascade Process for FVIII/von Willebrand Factor Complex and Immunoglobulin GTom Chen*1, Sonday Allen1, Dev Baines2, Jason Betley2, Davida Blackman1, Timothy Hayes1, Guy Harris2, Bastian Lobezoo2, and Keith Watson2

1American Red Cross, Plasma Derivatives, Gaithersburg, MD, USA. 2ProMetic BioSciences Ltd., Cambridge, UK.

3. A Combined Proteomics and Well-Characterized Approach to Comparability for the Development of Cascade-Fractionated- and other Plasma-Derived ProductsTimothy Hayes*1, Tom Chen1, Dev Baines2, Tom Busby1, Kevin Carrick1, Dale Schmidt1, and Nadine Ritter3

1American Red Cross, Plasma Derivatives, Gaithersburg, MD, USA. 2ProMetic BioSciences Ltd., Cambridge, UK. 3Biological Consulting Group, Alexandria VA, USA.

4. Manufacturing Processes of Plasma-Derived Products Reduce Emerging Pathogens Albrecht Gröner. ZLB Behring GmbH, Marburg, Germany.

5. Screening of Industrial Plasma by HBV NAT: Increase in Viral Safety?G. Zerlauth* and M. Gessner. Baxter AG, Plasma Control Europe, Vienna, Austria.

6. Study on the rHSA Quality SpecificationJia Qian*, Li Mei Yan, and Xingjun ZhouNew drug R&D center, North China Pharmaceutical Group Corp., shijiazhuang, Hebei, P. R. of China.

7. Study on Serological Reaction to SARS-CoV of IVIG and Original Plasma from Qualified DonorsGuo Zhongping, Wang Yu*, and Gou HongtaoChengdu Rongsheng Pharmaceutical Co., Ltd, Waidong Baojiang Bridge 610023, Chengdu, Sichuan, P. R. of China.

8. Characterization of IgG Fragments in Liquid Intravenous Immunoglobulin ProductsRobert V. Diemel1,2, Hendricus G. J. ter Hart2, Gerardus J. A. Derksen2, Anky H. L. Koenderman*2, and Rob C. Aalberse1

1Sanquin Research. 2Sanquin Plasma Products, Sanquin Blood Supply Foundation, Amsterdam, The Netherlands.

9. Characterization of Plasminogen Removal GelIngela Blomqvist1, Inger Lagerlund*1, Anna Mattsson1, Roberto Meidler2, and Israel Nur2

1GE Healthcare, Uppsala, Sweden. 2Omrix Biopharmaceuticals Inc., Rehovot, Israel.

10. Removal of Viruses and Pathogen Agents in the Downstream Processing of Plasma and Biopharmaceutical Products by Nanofiltration with Planova® FilterS. Vynck*1, E. Kederer1, H. Hisada2, and T. Sato2

1Asahi Kasei Planova Europe, Brussels, Belgium. 2Asahi Kasei Pharma Corporation, Japan.

Downstream – PPB ’05 abstracts 45

11. Advantages of the EBA Method for the First Stage Chromatography Plasma Proteins Re-purificationG. L. Volkov*1, S. I. Andrianov, A. Yu. Slominskiy, O. S. Havrylyuk, and T. V. GoroshnikovaPalladin Institute of Biochemistry of the NAS of Ukraine. 1Application and Training Laboratory of Amersham Biosciences Representative Office in Ukraine. Leontovicha str., 9, 01601, Kyiv, Ukraine.

12. Plasma Economics – A Brighter Outlook in the Short Term but Changing Fundamentals Suggest Future ChallengesRosemary Cummins. Equity Research, Citigroup, Melbourne, Australia.

13. Electrophoretic Analysis of Clotting Factor VIII ConcentratesN. V. Odinokova, M. A. Azhigirova*, and R. N. KhametovaNational Research Centre of Haematology, Moscow, Russia.

14. The Possibility of One-Stage Clotting Method for Determination of Factor VIII Activity in the Purified Antihaemophilic ConcentratesA. Berkovskiy*, E. Sergeeva, A. Suvorov, M. Ajiguirova, and A. VorobiovHematology Centre, Moscow, Russia.

15. Is Citrate Deficiency a Common Problem in Transfusion Medicine?Edward Shanbrom1 and William J. Owens*2

1Santa Ana, CA, USA. 2University of California, Irvine, Orange CA, USA.

16. Blood Bank Preparation of Safe Immunoglobulin ConcentrateEdward Shanbrom1 and William J. Owens*2

1Santa Ana, CA, USA. 2University of California Irvine, Orange, CA, USA.

17. Innovative Method for Producing Therapeutic Proteins in PlantsKimmo Koivu1 and Timo Virkajärvi*2

1UniCrop Ltd, Helsinki, Finland. 2Rintekno Oy, Espoo, Finland.

18. UVivatec – A Scalable Technology for UV Virus Inactivation in Laboratory and ProductionDr Sebastian Schmidt*, DI Markus Zamponi, and DI Jörg KaulingBayer Technology Services GmbH, D-51368 Leverkusen, Germany.

46 Downstream – PPB ’05 abstracts

Author index

Arrabal, S. 32 Aubin, J. T. 32 Auer, W. 34

Baines, D. 8, 10, 22 Beacom, B. 10 Bees, B. 25 Betley, J. 10 Borte, M. 37 Bregenholt, S. 14 Bryant, C. 8, 22 Burton, S. 8 Butterweck, H. A. 34 Buurman, W. A. 40

Carbonell, R. 8, 22 Chen, T. 8, 10, 22 Curling, J. 8, 22

Daehler, A. 6 de Vries, B. 40

Echelard, Y. 16

Flan, B. 32 Forstrom, J. 19

Griffiths, A. 25 Gurgel, P. 22

Hammond, D. 8 Hayes, T. 8, 10, 22 Heinrichs, H. 37 Hermans, P. 12

Jensen, A. 14

Kiessling, P. 37

Laude, H. 32 Le Hir, G. 32

Martinelli, T. 6 Meade, H. M. 16 Minchinton, R. 6 Muchitsch, E. M. 34

Ochs, H. D. 37 Ohrstrom, J. 19

Parkkinen, J. 27, 40 Pearson, J. 10 Pike, R. 6 Praus, M. 37 Price, H. 25

Rahola, A. 27 Reynolds, T. 19 Ruesseler, W. 29

Schwarz, H-P. 34 Sierkstra, L. N. 12 Soluk, L. 25

ten Haaft, M. 12 Teschner, W. 34 Tölö, H. 27 Törmä, E. 27

Vazquez, P. 10 Weber, A. 34 von Bonsdorff, L. 27

You, B. 32

Zuckerman, L. 19

imagination at work28-4094-52

Sepharose is a trademark of General Electric companies. GE, imagination at work, and GE monogram are trademarks of General Electric Company.

ATryn is a trademark of GTC Biotherapeutics.

CaptureSelect is trademark of BAC.

E-PAGE is a trademark of Invitrogen

Gradiflow is trademark of Life Therapeutics

K-TOPs is a trademark of Biokinetics

MAbsorbent, Mimetic, and Purabead are trademark of ProMetic Biosciences.

Symplex and Sympress are trademarks of Symphogen A/S.

Viresolve of Millipore Corp.

Vivaglobin is a trademark of ZLB Behring, part of the CSL Group.

The Polymer Chain Reaction (PCR) is covered by patents owned by Roche Molecular Systems and F Hoffmann-La Roche Ltd. A license to use the PCR process for ceratin research and development activities accompanies the purchase of certain reagents from licensed suppliers such as Amersham Biosciences and affiliates when used in conjunction with an authorized thermal cycler.

GE Healthcare Bio-Sciences AB, a General Electric Company

GE Healthcare UK Ltd Amersham Place, Little Chalfont, Buckinghamshire, HP7 9NA, UK

GE Healthcare Bio-Sciences Corp 800 Centennial Avenue, Piscataway, New Jersey 08855, USA

GE Healthcare Europe GmbH Munzinger Strasse 5, D-79111 Freiburg, Germany

GE Healthcare Bio-Sciences K.K. Sanken Building 3-25-1, Shinjuku-ku, Tokyo 169-0073, Japan

All goods and services are sold subject to the terms and conditions of sale of the company within GE Healthcare which supplies them. GE Healthcare reserves the right, subject to any regulatory and contractual approval, if required, to make changes in specifications and features shown herein, or discontinue the product described at any time with notice or obligation. Contact your local GE Healthcare representative for the most current information © General Electric 2006 – All rights reserved.

The views expressed by the contributors and correspondents are their own and do not necessarily reflect the views of GE Healthcare Bio-Sciences.

www.gehealthcare.com/protein-purification http://www.bo-conf.com/ppb07

GE Healthcare Bio-Sciences AB

Björkgatan 30

751 84 Uppsala

Sweden

Asia Pacific Tel: +85 65 6 275 1830 Fax: +852 2811 5251 • Australasia Tel: +61 2 9899 0999 Fax: +61 2 9899 7511 • Austria Tel: 01/57606-1619 Fax: 01/57606-1627 • Belgium Tel: 0800 73 888 Fax: 03 272 1637 • Canada Tel: 800 463 5800 Fax: 800 567 1008 • Central, East, & South East Europe Tel: +43 1 982 3826 Fax: +43 1 985 8327 • Denmark Tel: 45 16 2400 Fax: 45 16 2424 • Finland & Baltics Tel: +358 (0)9 512 39 40 Fax: +358 (0)9 512 39 439 • France Tel: 01 69 35 67 00 Fax: 01 69 41 96 77 • Germany Tel: 0761/4903-490 Fax: 0761/4903-405 • Italy Tel: 02 27322 1 Fax: 02 27302 212 • Japan Tel: +81 3 5331 9336 Fax: +81 3 5331 9370 • Latin America Tel: +55 11 3933 7300 Fax: +55 11 3933 7304 • Middle East & Africa Tel: +30 210 9600 687 Fax: +30 210 9600 693 • Netherlands Tel: 0165 580 410 Fax: 0165 580 401 • Norway Tel: 815 65 555 Fax: 815 65 666 • Portugal Tel: 21 417 7035 Fax: 21 417 3184 • Russia & other C.I.S. & N.I.S Tel: +7 (095) 232 0250 Fax: +7 (095) 230 6377 • South East Asia Tel: 60 3 8024 2080 Fax: 60 3 8024 2090 • Spain Tel: 93 594 49 50 Fax: 93 594 49 55 • Sweden Tel: 018 612 1900 Fax: 018 612 1910 • Switzerland Tel: 0848 8028 12 Fax: 0848 8028 13 • UK Tel: 0800 616928 Fax: 0800 616927 • USA Tel: 800 526 3593 Fax: 877 295 8102


Recommended