Biologicals (1992) 20, 91-100

REVIEW Safety Aspects in the Manufacturing of Plasma-Derived Coagulation Factor Concentrates Thierry Burnouf Service de Fractionnement du Plasma, Centre Regional de Transfusion Sanguine, 19 rue CamilleGuerin, 59012 Lille Cedex, France Abstract. Plasma-derived coagulation factor concentrates, prepared using traditional manufacturing processes, have transmitted viral diseases, especially AIDS, hepatitis B and hepatitis C to patients. To date, more extensive selection of blood donors, improved screening procedures of each plasma donation for direct and indirect viral markers, and newly developed virucidal procedures, especially pasteurization and solvent-detergent, together with extraction technologies of plasma proteins based on high-resolution chromatographic separations, have diminished considerably the risks of transmitting these pathogenic agents. To ensure safety, each production process must be carefully validated, following a rigorous scientific approach to assess its ability to inactivate or eliminate viruses. In addition, Good Manufacturing Practices must avoid any variation from these validated viral inactivation processes and must eliminate risks of potential downstream contamination of purified plasma fractions following viral inactivation or elimination steps. Other side-effects associated with conventional lowpurity preparations, such as acute haemolytic anemia due to contamination by isohaemagglutinins, or immunosuppression possibly due to an overload in fibrinogen and immunoglobulins, have not been reported following infusion of highly purified coagulation factor concentrates. Present state-of-the-art virus inactivation and protein-purification technologies have significantly improved the safety of plasma coagulation factor concentrates. Introduction

Blood-borne viruses

H u m a n plasma is extensively used for extraction of biologically active glycoproteins with applications in the t r e a t m e n t of h u m a n haemorrhagic or thrombotic disorders. ~ They include the concentrates of factor VIII (FVIII), von Willebrand factor (vWF), prothrombin complex (PCC), FIX, FVII, fibrinogen, protein C, antithrombin III (ATIII), FXI, FX and FXIII. Some of these biologicals, especially FVIII, vWF and PCC, prior to the adoption of virus-screening techniques in blood/plasma donations and virus-inactivation treatments, were found to have transmitted viral diseases, most notably AIDS and hepatitis. Sideeffects due to contamination with allogenic proteins, isohaemagglutinins, proenzymes and/or activated coagulation factors, or to discrete denaturation during the purification process, have also been described. The safety of plasma-derived coagulation factor concentrates must be re-assessed in view of the significant progress recently made in plasma viral screening, viral inactivation/elimination procedures, purification technologies and quality control/quality assurance of coagulation factor concentrates.

As covered in detail in several reviews, 2,~ blood may be the host of viruses which may cause problems of a significant magnitude in recipients of blood and/or plasma derivatives. Those which are more relevant to blood transfusion in terms of pathogenicity are briefly described here. More detailed information can be found elsewhere. 4 Hepatitis B virus (HBV) is a double-stranded DNA, lipid-enveloped virus of the hepadnaviridae family; it shows tropism for hepatocytes while persistently circulating in the blood of chronically infected carriers. Plasma from individuals with acute HB may carry up to 10 s infective doses/ml. Hepatitis delta virus (HDV) is a single-stranded, enveloped RNA defective virus, depending on HBV for replication and infectious only in people infected with HBV. Hepatitis A virus (HAV) is an RNA virus belonging to the Picornaviridae family. HAV has been involved in very rare cases of post-transfusion hepatitis. Hepatitis C virus (HCV) is a m e m b e r of the ungrouped togaviruses; its titre in plasma from infected individuals is lower than t h a t found for HBV. Hepatitis

1045-1056/92/020091 + 10 $03.00/0

O 1992 The International Associationof BiologicalStandardization

92

T. Burnouf

C represents the majority of parenteral hepatitis previously attributed to Non A-Non B hepatitis viruses (NANBHV), but additional forms of hepatitis (NANBNC hepatitis) cannot be excluded. H u m a n immunodeficiency virus (HIV- 1 and HIV-2), the agent causing AIDS, is a single-stranded RNA lentivirus t h a t is a member of the retroviridae. It may be present at high concentrations in plasma from individuals before their seroconversion (up to 10.000 infective doses/ml). H u m a n T-cell leukaemia virus 1 (HTLV-1), the prototype h u m a n retrovirus, is an enveloped virus belonging to the oncoviridae group and can be oncogenic. Herpes simplex 1 (HSV- 1), HSV2, varicella-zoster (VZV, or HSV-3), Epstein-Barr (EBV, or HSV-4), cytomegalovirus (CMV, or HSV-5), and h u m a n herpes viruses 6 (HHV 6) and 7 (HHV 7) are h u m a n herpes viruses. All are enveloped, doublestranded linear DNA viruses. EBV may cause infectious m o n o n u c l e o s i s a n d l y m p h o p r o l i f e r a t i v e disorders in immunodeficient patients. Lymphocytes are the reservoir of infection for CMV, thus CMV is rarely present in a free form in plasma. CMV may cause overt, rare clinical diseases in immunocompetent patients. Parvovirus B19 (PVB19) is a small, non-enveloped, DNA virus responsible for acute Table

aplastic crises in patients with chronic haemolytic anaemias. Risks associated with other pathogenic agents, such as neurotropic 'slow viruses' (e.g. Creutzfeld-Jakob agent), have so far been difficult to assess. ~ Several tropical viruses, such as Dengue virus or Lassa virus, can also be present in blood 3 during the viremic phase of the disease. Table 1 lists the more relevant blood-borne viruses, and their respective risks of transmission by whole blood and plasma derivatives. Risks from plasma derivatives are limited to only some of these viruses. CMV, EBV, HHV 6, HTLV-1 and HTLV-2 can be found in whole blood, cell concentrates or possibly in plasma, but there is no reported transmission by plasma derivatives. In contrast, HBV, HCV, HDV, HIV and PVB19 have been transmitted to patients by some coagulation factor concentrates, especially FVIII and PCC. At this stage, transfusion-associated hepatitis (B and C) and AIDS have been of paramount epidemiological importance, especially from the infusion of industrial coagulation factor concentrates, since one single infected plasma donation could contaminate an entire starting plasma pool derived from about 2000 to 20 000 blood or plasma donors.

1. Blood-borne viruses Evidence for pathogenic risk from

Virus Major: Hepatitis B virus Hepatitis delta virus Hepatitis C virus HIV 1 and 2 HTLV 1 and 2 Minor: B 19 Parvovirus Hepatitis A virus

Herpes simplex 1 Herpes simplex 2

Family or Group Hepadnaviriadae

Envelope Genome Size (nm) Whole blood

Togaviridae Retroviridae Retroviridae

+ + + + +

DNA RNA RNA RNA RNA

42 36 45 100 100

Parvoviridae

-

DNA

25

Picornaviridae Herpesviridae

Varicella-zoster virus Epstein Barr virus Cytomegalovirus H u m a n herpes virus 6 H u m a n herpes virus 7 Creutzfeld- Jacob agent Unconventional virus or scrapie like agent

-

+ +

RNA DNA

~_

"

.~_

"

.~_

',

+ ..~

"

Plasma Derivatives (clotting factors)

÷ + + +

Only in patients with severe anemia 20-30 Unusal 120-200 + +

÷ + + + Only in patients with severe anemia

Plasma-derived coagulation factors

Donor selection and plasma virus screening Adequate selection of blood or plasma donors and screening procedures of each plasma unit contributes to reducing the virus load in the pooled source material, as recently reviewed2 The selection of donors excludes people not in good health as well as members of high-risk groups such as drug addicts, homosexuals, individuals with multiple sexual partners, or people having travelled to endemic areas of transfusion-associated infectious diseases. Fractionation of plasma from voluntary, unpaid donors, who are less likely to belong to lower socio-economic populations, has reduced the risk of infection, e.g. hepatitis 7 and AIDS. Similarly, a higher incidence of AIDS cases associated with clotting factor deficiency was found from products manufactured in the U.S.A. or from plasma obtained from the U.S.A. 6 Plasma screening eliminates donations which are positive for direct or indirect markers of pathogenic agents. The HBV carrier state is detected by direct screening for HBsAg; despite the sensitivity of '3rd generation' tests using monoclonal antibodies (sensitivity: 0.2 ng/ml HBsAg or less), slightly infectious donations may not be detected. Recently, screening of HCV-positive donations using an immunoserological marker (anti-HCV antibody) has become possible. Previously used surrogate tests, elevated transaminases (ALT or AST), or presence ofanti-HB core (HBc) antibody, may still serve as additional screening measures for carriers of hepatitis viruses. Anti-HIV-1 and -2 antibodies are detected by enzyme immunoassay to eliminate donations from infected donors. Due to similarities in the epidemiology of hepatitis and AIDS, anti-HBc screening is also helpful as a surrogate marker for AIDS during the anti-HIV window period, prior to the appearance of HIV antibodies. HTLV-1 and -2 positive donations can be detected by immunoserology. The regulatory requirements for the screening tests to be performed on plasma donations are determined by national agencies and take into consideration local ethological and epidemiological criteria and cost-benefits ratios. Screening differs between countries but, in most cases, includes at least HBsAg and anti HIV-1. In France, where the safety of the blood supply, exclusively from selected voluntary, unpaid donors, has reached very high levels, blood and plasma donations are presently screened for HBsAg (using highly sensitive tests), anti-HIV-1 and -2, antiHCV, anti-HBc, anti-HTLV-1 and -2 and transaminases. Nevertheless, pooled plasma may still carry a limited infectious risk due to residual viral contami-

93

nations occurring from hypothetical technical or h u m a n errors during screening procedures of individual donations, or from the absence of specific and/or sensitive direct viral marker tests. Indeed, although the transmission of hepatitis B has been essentially eliminated since the introduction of HBsAg screening, it has been established that a large proportion of unsterilized low-purity coagulation factor concentrates could transmit hepatitis C 9-'' and AIDS. 12 Similarly, contamination of these concentrates by HDV and PVB 19 has been reported, '3,~4fortunately with generally no pathogenic incidence in most patients. A part of the viral safety of coagulation factor concentrates thus relies upon the viral inactivation and the extraction methods.

Virus inactivation treatment of coagulation factor concentrates Several methods to abolish or reduce infectivity of coagulation factor concentrates have been developed. One problem in establishing these techniques is to ensure a high level of virus inactivation (typically at least 4-5 logs of HIV, HBV and HCV) while preserving the biological activity of the coagulation factors. Most current viral inactivation procedures are based on physical or chemical treatments. '5 Treatment with fl-propiolactone (fl-PL) and ultraviolet (UV) combines chemical and photochemical sterilization respectively, resulting in the destruction of the viruses' nucleic acids. '~ FVIII is too sensitive to withstand the original fl-PL/I_W method used in the manufacturing process of a PCC. Chimpanzee studies have shown that fl-PL/UV may inactivate 7 log 10 o f H B V ~7and more than 4 log 10 o f N A N B H V 's but a recent batch of fl-PL/l.W-treated PCC has been implicated in the HIV-seroversion of several haemophilia B patients. ~9 Heat-treatment of lyophilized FVIII concentrate and PCC at 60-68°C for 30-96 h has been investigated since the early 1980s in an attempt to inactivate hepatitis viruses. The adoption of this so-called dry heat treatment started in 1984 to stop the transmission of HIV, as recommended by the National Haemophilia Foundation 2° and the Centers for Disease Control. 21 Validation using an FVIII and a PCC concentrate demonstrated that a 68°C treatment for at least 96 h was necessary to inactivate more than 3.5 log of HIV, whereas a 72-h treatment was not sufficient. 22 In fact, most dry heat-treatment c o n d i t i o n s failed to fully i n a c t i v a t e HBV, 23 NANBHV, 24.~5 and for some FVIII concentrates

94

T. Burnouf

treated 24-30 h at 60°C, HIV. '6 Efficacy is largely influenced by protein composition, formulation, and residual moisture of the product. 27 More severe conditions (80°C for 72 h) appear to yield a FVIII concentrate free of t r ans m i t t i ng HCV, 28while 10-30 min at 100°C might inactivate parvovirus B19. 29 Heating products in the presence of organic solvents such as chloroform or heptane or steam could enhance safety against NANBHV; nevertheless, the transmission of hepatitis has been reported from such FVIII concentrates 20.3t Heating in the liquid state, usually at 60°C for 10 h (pasteurization) in the presence of specific or nonspecific stabilizers to protect coagulation factors, is considered effective for inactivating various viruses. Conditions should be carefully selected, however, so t h at stabilizers do not protect the viruses from inactivation. The efficacy of this process reaches more than 5 logs for several model viruses; 32 at least 4 logs of HIV can be inactivated in less than 5 min. :~:~This procedure, as applied to FVIII and FIX concentrates, has good safety records, 3~,:~4,'35but pasteurized FVIII has been reported to t r a n s m i t hepatitis B 36albeit very infrequently. This procedure has been applied to other coagulation factor concentrates, including FIX, fibrinogen and FXIII, 32 as well as ATIIIY S o l v e n t - d e t e r g e n t (SD) t r e a t m e n t consists of incubating a protein solution with a combination of an organic solvent (usually tri (n-butyl) phosphate) and a detergent (e.g. sodium cholate, Tween 80 or Triton X-100) to disrupt lipid-enveloped viruses. '~8 This method is thus potentially effective at least against HIV, HBV, HCV and HDV. Inactivation rates of more than 10, 6 and 5 logs ofHIV-1, HBV and HCV, respectively, have been reported29 SD-treated FVIII

concentrates have been shown not to t r a n s m i t AIDS or hepatitis in clinidal trials. 4°-42As this t r e a t m e n t is directed towards lipids and has a low d e n a t u r i n g effect on proteins, thus preserving their biological activity, it has been applied to various therapeutic concentrates including FVIII, 4:3 vWf, 44 FIX, PCC, FVII, fibrinogen and fibrin glue, .5 FXI, 46 protein C and thrombin, showing excellent efficacy against model lipid-enveloped viruses (Table 2). G a m m a radiation doses of 2.5-10 Mrads (depending upon the t e m p e r a t u r e ) can inactivate 5-6 logs of HIV in plasma. However, efficient virucidal doses have denat uri ng effects on FVIII and PCC biological activity27 To date, clinical data indicate t h a t pasteurization and SD t r e a t m e n t s (possibly along with severe dry heat) can ensure a high level of safety for coagulation factor concentrates regarding hepatitis and AIDS. Table 3 summarizes the efficacy of the major virus sterilizing methods against HIV, HBV and HCV, grouped according to w h e t h e r they are performed 'in process' or on the final product, and presents their respective advantages and disadvantages. Purification technologies: Role in viral safety Purification obviously plays a significant role in the safety of plasma biologicals. The generation of FVIII and PCC concentrates involved in AIDS and hepatitis transmissions, in addition to not being submitted to any viral inactivation steps, had a purification factor (25-100-fold) from plasma similar to their concentration factor (with a parallel risk of concentrating viruses); production methods were based on nonspecific precipitation or adsorption steps at close to

T a b l e 2. R es ul t s o b t a i n e d in t he i n a c t i v a t i o n of model l i p i d - e n v e l o p e d v i r u s e s by SD t r e a t m e n t of commercial clotting factor concentrates Product* Virus

Time

F VIII v WF Fibrinogen*

PC C t

FIX

F VII

Protein C

Fibrin Glue

FIX

Thrombin

VSV

30 min 1h 2h

4.6 - 5.6 -> 5.6

4-6 -> 5.6 -> 5-6

4.6 -- 5.6 -> 5.6

-> 5-6 -> 5.6 -> 5-6

4.6 --- 5-6 -> 5.6

-> 5.0 -> 5.0 -> 5.0

5-5 -> 5.6 -> 5.6

4.6 -> 5.6 -> 5.6

Sindbis virus 30 min I h

-> 4.8 -> 4.8

-> 4-8 -> 4.8

-> 4.8 -> 4-8

-> 4-8 -> 4.8

-> 4.8 -> 4.8

-> 5-2 -> 4.8

-> 4.8 --- 5-2

-> 4.8 -> 4.8

* Treated at 25 _+0.5°C with 0.3r/( TnBP + 1~ "Ihveen80. PCC = Prothrombin ComplexConcentrate.

Plasma-derived coagulation factors

95

T a b l e 3. Evidence for efficacy of virus inactivation treatments of clotting factor concentrates (After ref 15 and 31). Advantages and inconveniences. (Starting plasma screened for HBsAg.) Hepatitis Procedure

HIV

A - - I n process 1--Heat t r e a t m e n t Dry product - - H o t vapour (60°C, 10 h, 1080 mbar) + - - H e a t + chloroform (60°C, 72 h) + - - H e a t + Heptane (60°C, 20 h) + Liquid product --pasteurization (60°C, 10 h)

2--Solvent-detergent

3--fl-propiolactone-UV B--Terminal Dry product --60°C, 30 h --60°C, 72 h --60°C, 96 h --80°C, 72 h

C

B

--Potential risks of downstream contamination

+(?)

-

+

+

+

+

+

+

-b

+(?)

+

+

m

_

-

-

+

Advantages and inconveniences

+

+

--Requires sophisticated equipment - - M a y induce some loss of protein activity - - N o t always efficient against hepatitis viruses --Potentially effective against lipid-enveloped and non-enveloped viruses --Requires protein stabilizers t h a t may also protect viruses --Stabilizers must be subsequently eliminated - - M a y induce some loss of protein activity --Highly effective only against lipid-enveloped viruses --The virus sterilizing agents must be eliminated --Protein function is generally unaffected. --Potential carcinogenicity of fl-propiolactone (when not hydrolysed) --Instability of fl-propiolactone --No risk of downstream contamination --Most conditions do not ensure inactivation of hepatitis viruses --Efficacy highly dependent on freeze-drying conditions (especially final moisture of product) and formulation

+ Efficient. - Not efficient.

neutral pH. 4s For example, in those FVIII preparations and in PCC, FVIII and FIX represented less than 1% of the total protein, respectively. Thus, no major segregation from any infectious agent could realistically occur so as to ensure a sufficient level of safety. In contrast, an anti-inhibitor coagulant complex ('activated' PCC) whose preparation involves a 20% cold ethanol precipitation step shown to inactivate or eliminate more t h a n 3-5 log in HIV infectivity,49 did not t r a n s m i t AIDS to patients. 5° In the last few years, however, manufacturing processes of trace plasma proteins have evolved quite impressively, ensuring a degree of purification of c. 10 000-fold or more, ~1 as compared with a concentration factor of only 30-100, and providing inherently increased viral safety. The production of new coagulation factor concentrates often relies upon a combination of centrifuga-

tion, precipitation, chromatography, ultrafiltration and freeze-drying steps. Each of those can, at least in principle, eliminate or inactivate viruses. However, chances t h a t filtration on current media, individual low-speed centrifugation and non-denaturing precipitation could induce a significant removal of viruses are remote. 5~ New generations of filters (some with very small pores of 40 nm or less) may be helpful at eliminating viruses, 5~ but filterability and the recovery of coagulation factors in terms of antigen and activity must be assessed. Freeze-drying may inactivate about 1 log of HIV in some products23 Ultrafiltration is generally used for concentrating and formulating a product prior to sterile filtration and dispensing. In most cases, membranes with cutoff values from 10 000-100 000 daltons are used to remove low-molecular-weight components and retain the protein fraction of interest; in such conditions, no

96

T. Burnouf

extensive elimination of viruses, which are large entities, can be expected unless uncontrolled binding of viruses happens during the process. However, it has recently been reported that ultrafiltration applied to a FIX preparation eliminates more than 5 logs of HIV. 54 Chromatographic adsorption is increasingly used in the manufacture of new, highly purified therapeutic preparations. When used under conditions providing high specificity and selectivity, chromatographic purification can significantly separate infectious agents from therapeutic fractions. Clearance factors of more than 4 and 5 logs of HIV have been found during the immunopurification of FVIII s5 and FIX 54 respectively. More than 4-5 logs of various model viruses are eliminated during immunoaffinity chromatography of FVIIIY Similar clearance factors (unpublished) have been found during ion-exchange chromatography of vWF on DEAE-Fractogel. 44 More than 6 logs of HIV P24Ag were eliminated by purification of FIX on two ion-exchangersY A 500fold reduction in encephalomyocarditis virus, a nonenveloped virus, was found after QAE-sepharose chromatography of an immunopurified factor VIII. 56 Hydrophobic interaction chromatography of PCC on octanohydrazide-Sepharose 4B may adsorb some HBV 5s and NANBHV, 59 but this step alone is not sufficient to ensure viral safety. 31 The use of immunopurification techniques based on murine antibodies requires careful validation of the absence of pathogenic viruses in the cell lines and culture media of the antibodies, and demonstration that the purification process eliminates and inactivates cellular contaminants, proteins and nucleic acids. Because the chemistry of chromatographic media may change after repeated use, potentially modifying the extent of virus removal, manufacturing processes should necessarily include a specific step to inactivate viruses. For example, an immunopurified FVIII preparation that was dry-heated at 60°C for 30 h was found to transmit hepatitis C, showing that both efficient viral inactivation and efficient elimination procedures may need to be combined to ensure safety, s° Validation of virus inactivation and elimination procedures

Since the viral safety of coagulation factor concentrates depends on the manufacturing process, major manufacturing steps that may play a role in the inactivation and elimination of pathogenic infectious agents must be assessed and validated. Validation is

intended to establish the efficacy of a given manufacturing step to inactivate or eliminate different types of viruses and to d e f n e the physico-chemical conditions important in ensuring reproducibility of performance. Parameters influencing the rate and extent of viral inactivation during virus inactivation treatments include (but are not limited to) the moisture content of the product and potential protective effects of salts during dry-heat treatment, the type of protein stabilizer and protein concentration during SD treatment. Some important factors to control during chromatographic steps are the quality of gel packing, pH, osmolarity and the flow rate of buffer and plasma fractions, and the reproducibility of separation after recycling the gel. Validation studies must use conditions that mimic as closely as possible those used on the production scale. In order to standardize and validate the safety of plasma-derived coagulation factor concentrates manufactured or distributed in Europe, recommendations on virus validation studies, including the choice of viruses to use, have been defined by the Biotechnology/Pharmacy Working Party (BPWp) of the EEC. 6',62 Other adverse effects associated with coagulation factor concentrates

Apart from viral risks, coagulation factors may be associated with specific adverse effects which have been covered in a recent review. 63 Factor VIII

High dose infusions of low-purity FVIII preparations may induce acute haemolytic anaemia due to the presence of isohaemagglutinins against blood group type A, B and AB cells. Isogroup products (with a titre of anti-A and anti-B antibodies at 2 or less) prepared from selected plasma obtained from blood group types A and AB, or B and AB, avoid such complications. 1 Also, contamination with allogenic proteins such as fibrinogen, immunoglobulins (Ig) A, G and M and fibronectin, has been said to be responsible for anaphylactic reactions 64 or hypersensitivity. 65 Some foreign proteins, e.g. IgG aggregates 66 or fibrinogen, might also induce immunological aberrations in m u l t i p l y - t r a n s f u s e d h a e m o p h i l i a c s , although exposure to viruses might have played a significant role in these pathologies. Fortunately, current high-purity concentrates,43.55.s6 avoid such potential adverse effects. However, one major complication remains the possible development of anti-FVIII alloantibodies in 10-15% of patients with major haemophilia A, 67mainly in response to an infused fac-

Plasma-derived coagulation factors

tor recognized as 'foreign' by the host. Anti-FVIII antibodies have been reported to occur at higher frequency in patients treated with an immunopurified FVIII concentrate. 6s This stresses the need for purification and viral inactivation procedures to be designed in such a way as to avoid the alteration of the three-dimensional structure of proteins and the exposure of new antigenic determinants in the therapeutic protein (or in a contaminating protein).

PCC Thrombotic complications such as disseminated intravascular coagulation have been specifically associated with high doses of some PCC preparations in patients suffering from haemophilia B or liver disease. 69 Causes may be hypercoagulable states due to high contents in FII and X, or the presence of activated coagulation factors and/or coagulant-active phospholipids in the productJ ° New, highly purified FIX preparations have been shown to be much less thrombogenic than PCC in animal models 71 and in m a n J 2 Inhibitors to FIX have also been found in haemophilia B patients following infusion of PCC. '

Other products Other haemostatic products (FVII, vWF, ATIII, protein C, FX, fibrinogen etc) have been developed recently to allow selective haemotherapy in deficient patients. Because of their purity and mode of viral inactivation, they usually do not carry most of the side-effects inherent to crude preparations, including the viral risks of whole plasma or pooled cryoprecipitate. Quality control/quality assurance

Each batch of coagulation factor concentrates is subjected, following Pharmacopoeia requirements, to a set of tests including those to determine pyrogen (and endotoxin) levels and isohaemagglutinin titres, and to confirm the absence of anti-HIV antibodies and HBsAg; the absence of these viral markers, especially in highly purified preparations, may only indicate their elimination in side-fractions, and not the lack of contamination of the starting plasma. When relevant, the absence of HCV or HIV contamination in batches of final products, using the polymerase chain reaction (PCR), may be checkedJ 3'74 It is extremely important to exercise full control of the production process to avoid uncontrolled deviation from an established and validated production procedure. The absence of the risk of downstream viral contamination for those products virally inacti-

97

vated during the purification process is paramount. Products subjected to final dry-heat treatment must have constant formulation and residual moisture to ensure reproducible viral inactivation efficacy. Conclusion

To date, most coagulation factor concentrates are manufactured by procedures which are well established and proven to provide products with a high level of safety. Indeed, manufacturing processes have evolved dramatically in the last few years. Most current products are subjected to efficient viral inactivation procedures t h a t can inactivate pathogenic plasma-borne viral agents, most notably HIV and hepatitis viruses, in levels t h a t far exceed the potential infectious risk from screened source plasma. In addition, modern products are obtained by purification processes which have reached a high degree of sophistication. Present purification levels have decreased the risk of transmission of any pathogenic agents that could contaminate the source plasma material and have ensured the elimination of unwanted protein contaminants. In spite of the advent of recombinant DNA products, h u m a n plasma should remain an invaluable source material for the preparation of therapeutic biologicals including coagulation factors.

Acknowledgements Thanks are expressed to Professor M. Goudemand and to Dr M. Maniez-Montreuil for helpful discussions. References

1. Goudemand M. Les fractions coagulantes. Encyl M~d Chir (Paris, France) Sang 1987; 3: 1-12. 2. Schilt U. Overview of viruses relevant to blood transfusion. In: Morgenthaler JJ, ed. Virus Inactivation in Plasma Products. Curr Stud Hematol Blood Tranfus. Basel: Karger, 1989; 56: 1-8. 3. Seidl S, Kfihnl P. Viral disease transmissible by blood transfusion. In: Greenwalt TJ, ed. Methods in Hematology. Vol 17. Blood Transfusion. Edinburgh Churchill Livingstone: 1988; 234-256. 4. Fields BN, Knipe DM. Fields Virology,Vol 2. New York: Raven Press; 1990. 5. Prusiner SB" M°lecular bi°l°gy °f pri°ns causing infec" tious and genetic encephalopathies of humans as well as scrapie sheep and BSE of cattle. Develop Biol Stand 1991; 75: 55-74. 6. KOhnlP, Seidl S, BShm BO. Reduction of virus load in blood donations by screening methods. In: Morgenthaler JJ, ed. Virus Inactivation in Plasma Products. Curr Stud Hematol Blood Transfus. Basel: Karger, 1989; 56: 9-22.

98

To Burnouf

7. Alter HA. Post-transfusion hepatitis: clinical features. risk and donor testing. In: Petricciani JC, Gust ID, Hoppe PA, Krijnen HW, eds. The Safety of Blood and Blood Products. Chichester: WHO, John Wiley & Sons Ltd 1987; 47-61. 8. Allen JR. Epidemiologic considerations in the transmission of LAV/HTLV-III by blood and blood products. In: Petricciani JC, Gust ID, Hoppe PA, Kl-ijnen HW, eds. The Safety of Blood and Blood Product. Chichester: WHO, John Wiley & Sons Ltd, 1987; 33-47. 9. Sugg U, Schnaidt M, Schneider W, Lissner R. Clotting factors and non-A, non-B hepatitis. N Engl J Med 1980; 303: 493. 10. Norkrans G, Widell A, Teger-Nilsson A, Kjellman H, Frosner G. Acute hepatitis non-A, non-B following administration of factor VIII concentrates. Vox Sang 1981; 41: 129-133. 11. Fletcher M, Trowell J, Craske J, Pavier K, Rizza C. Non-A, non-B hepatitis after transfusion of factor VIII in infrequently treated patients. Br Med J 1983; 287: 1754-1757. 12. Goedert J, Sarngadharan M, Eyster M, Weiss S, Bodner A, Gallo R, Blattner W. Antibodies reactive with human T-cell leukemia viruses in the serum of hemophiliacs receiving factor VIII concentrate. Blood 1985; 66: 742-744. 13. Mortimer PP, Luban N, Kelleher J, Cohen B. Transmission of serum parvovirus-like by clotting factor concentrates. Lancet 1983; ii: 482-484. 14. Bartolomei Corsi O, Azzi A, Morfini M, Fanci R, Rossi FelTini P. Human parvovirus infection in haemophiliacs first infused with treated clotting factor concentrates. J Med Virol, 1988; 25: 151-153. 15. Prince AM, Horowitz B, Horowitz MS, Zang E. The development of virus-free labile blood derivatives--A review. Eur J Epidemiol 1987; 3: 103-118. 16. LoGrippo G, Hal~man F. Chemical and combined methods for plasma sterilization. Bibl Haematol 1958; 7: 225-230. 17. Prince AM, Stephan W, Brotman B. flpropiolactone/ultraviolet irradition: a review of its effectiveness for inactivation of viruses in blood derivatives. Rev Infect Dis 1983; 5: 92-107. 18. Stephan W, Dichtelmuller H, Kotitschke R, Prince AM, Friis RR, Bauer H. Inactivation of hepatitis viruses (B, Non-A/Non-B) and retroviruses in pooled human plasma by means of fl-propiolactone-and UV-treatment. In: Smit Sibinga CTh, Das PC, Seidl S, eds. Plasma Fractionation and Blood Transfusion. Boston: Martinus NijhoffPublishers 1985; 147-153. 19. Kleim JP, Bailly E, Schneweis KE, Brackmann HH, Hammerstein U, Hanfland P, van Loo B, Oldenburg J. Acute HIV-1 infection in patients with hemophilia B treated with fl-propiolactone-UV-inactivated clotting factor. Thromb Haemostas 1990; 64: 336-337. 20. National Haemophilia Foundation Medical and Scientific Advisory Council. Recommendations concerning AIDS and therapy of hemophilia (revised October 13, 1984) New York: National Hemophilia Foundation 1984. 21. Centers for Disease Control. Update: acquired immunodeficiency syndrome (AIDS) in persons with hemophilia. N.M.W.R. 1985; 33: 589-591. 22. Burnouf T, Martinache L, Goudemand M. L'inactiva-

23.

24.

25. 26.

27.

28.

29.

30.

31. 32.

33.

34.

35.

36. 37.

tion virale des ddrivds plasmatiques. Nouv Rev Fr hematol 1987; 29: 93-96. Hollinger B, Dolana G, Thomas W, Gyorkey F. Reduction in risk of hepatitis transmission by heattreatment of a human factor VIII concentrate. J Infec Dis 1985; 150: 250-262. Colombo M, Carnelli V, Gazengel C, Mannucci P, Savidge G, Schimpf K. Transmission of non-A, non B hepatitis by heat-treated factor VIII concentrate. Lancet 1985; ii: 1-4. Preston F, Hay C, Dewar M, Greaves M, 2higer D. NonA, Non-B hepatitis and heat-treated factor VIII concentrates. Lancet 1985; ii: 213. Dietrich SL, Mosley JW, Lusher JM, Hilgartner MW, Operskalsky EA, Habel L, Aledorf LM, Gjerset GF, Koerper MA, Lewis BH, Pegelow CH. The Transfusion Safety Study Group Transmission of human immunodeficiency virus type 1 by dry-heated clotting factor concentrates. Vox Sang 1990; 59: 129-135. Horowitz B, Wiebe M, Lippin A, Vandersande J, Stryker M. Inactivation of viruses in labile blood derivatives. II. Physical methods. Transfusion 1985; 25: 523-527. Study group of the UK Haemophilia centre. Effect of dry-heating of coagulation factor concentrates at 80°C for 72 hours on transmission of Non-A, Non B hepatitis. Lancet 1988; ii: 814-816. Rubinstein AI, Rubinstein DB. Sequential dry heat treatments at sterilizing temperatures (100°C) to produce sterile factor VIII concentrates. Vox Sang 1989; 57: 272. Mannucci PM, Zanetti AR, Colombo M, Chistolini A, De Biasi R, Musso R, Tamponi G. Antibody to hepatitis C virus after a vapour-heated factor VIII concentrate. Thromb Haemostas 1990; 64: 232-234. Mannucci PM, Colombo M. Virucidal treatment of clotting factor concentrates. Lancet 1988; ii: 782-785. Heimburger N, Karges HE. Strategies to produce virussafe blood derivatives. In: Morgenthaler JJ, ed. Virus Inactivation in Plasma Products. Curr Stud Hematol Blood Transfus. Basel: Karger 1989; 56: 23-33. Gust ID, Deinhardt F, Hollan SR, Rey F. The safety of blood products. In: Petricciani JC, Gust ID, Hoppe PA, Krijnen HW, eds. The Safety of Blood and Blood Products. Chichester: WHO, John Wiley & Sons Ltd 1987; 69-90. Schimpf K, Mannucci P, Kreutz W, Brackman HH, Auerswald G, Ciavarella N, M6sseler J, DeRosa V, Kraus B, Bruekmann Ch, Mancuso G, Mittler U, Haschke F, Morfini M. Absence of hepatitis after treatment with a pasteurized factor VIII concentrate in patients with hemophilia and no previous transfusions. N Engl J Med 1987; 316: 918-922. Mannucci PM, Schimpf K, Brettler DB, Ciavarella N, Colombo M, Haschke F, Lechner K, Lusher J, Weissbach G. The International Study group. Low risk for hepatitis C in hemophiliacs given a high-purity, pasteurized factor VIII concentrate. Ann Intern Med 1990; 113: 27-32. Brackman HH, Egli H. Acute hepatitis B infection after treatment with heat-inactivated factor VIII concentrate. Lancet 1988; ii: 967. Roggendorf M, Delfos U, Novakova-Banet A. Virus safety of pasteurized antithrombin III concentrate. Thromb Haemostas 1991; 65: 212.

Plasma-derived coagulation factors

38. Horowitz B, Wiebe ME, Lippin A, Stryker MH. Inactivation of viruses in labile blood derivatives. I. Disruption of lipid-enveloped viruses by tri-(n-butyl) phosphate detergent combinations. Transfusion 1985; 25: 516-522. 39. Horowitz MS, Bolmer SD, Horowitz B. Elimination of disease-transmitting enveloped viruses from human blood plasma and mammalian cell culture products. Bioseparation 1991; 1: 409-417. 40. Horowitz MS, Rooks C, Horowitz B, Hilgartner MW. Virus safety of solvent/detergent-treated antihaemophilic factor concentrate. Lancet 1988; ii: 186-189. 41. Gazengel C. Viral safety of solvent-detergent treated FVIII cocentrate. Results of a French Multicenter Study. XIIth Congress ISTH, Toyo, August 1989. Thromb Haemost 1989; 62: 455. 42. Noel L, Gu~rois C, Maisonneuve P, Verroust F, Laurian Y. Antibodies to hepatitis C virus in haemophilia. Lancet 1989; ii: 560. 43. Burnouf T, Burnouf-Radosevich M, Huart JJ, Goudemand M. A highly purified factor ViII:c concentrate prepared from cryoprecipitate by ion-exchange chromography. Vox Sang 1991; 60: 8-15. 44. Burnouf-Radosevich M, Burnouf T. Chromatographic preparation of a therapeutic highly von Willebrand factor concentrate from human cryoprecipitate. Vox Sang 1992; 62: 1-11. 45. Burnouf T. Examples of virus inactivation treatment in plasma derivatives manufacturing. In: Stoltz JF, Rivat C, eds. Biotechnology of plasma proteins. Colloque INSERM, 1989; 175: 373-382. 46. Burnouf-Radosevich M, Burnouf T. A therapeutic highly purified factor XI concentrate from human plasma. (submitted). 47. Hiemstra H, Tersmette M, Vos A.HV, Over J, Van Berke] MP, De Bree H. Inactivation of human immunodeficiency virus by gamma radiation and its effect on plasma and coagulation factors. Transfusion 1991; 31: 32-39. 48. Newmann J, Johnson AJ, Karpatkin MH, Puszkin S. Methods for the production of clinically effective intermediate- and high-purity factor VII concentrates. Br J Haematol 1971; 21: 1-20. 49. Piszkiewicz D, Kingdon H, Apfelzeig R, McDougal JS, Cort SP, Andrews J, Hope J, Cabradilla CD. Inactivation of HTLV-IIULAV during plasma fl'actionation. Lancet 1985; ii: 1188-1189. 50. Gazengel C, Larrieu MJ. Lack of seroconversion for LAV/HTLV-III in patients exclusively given unheated activated prothrombin complex prepared with ethanol step. Lancet 1985; ii: 1189. 51. BurnoufT. Integration of chromatography with traditional plasma protein fractionation methods. Bio-separation 1991; 1: 383-396. 52. Einarsson M, Morgenthaler JJ. Removal of viruses from plasma fractions by chromatography. In: Morgenthaler JJ, ed. Virus Inactivation in Plasma Products. Curr Stud Hematol Blood Transfus. Basel: Karger 1989, 56: 138-145. 53. Hamamoto Y, Harada S, Kobayashi S, Yamaguchi K, Lijima H, Manabe SI, Tsurumi T, Aizawa H, Yamamoto N. A. novel method for revmoal of human immunodeficiency Virus: filtration with porous polymeric membranes. Vox Sang 1989; 56: 230-236.

99

54. Kim HC, McMillan CW, White GC, Bergman GE, Saidi P. Clinical experience of a new monoclonal antibody purified factor IX: half-life, recovery, and safety in patients with hemophila B. Semin Hematol 1990; 27: 3O-35. 55. Hrinda ME, Tart C, Curry W, Newman J, Schreiber AB, D'Alisa R. Monoclonal antibody affinity purification of plasma proteins removes viral contaminants. In: Stoltz JF, Rivat C, eds. Colloque INSERM 1989; 175: 413-418. 56. Liu S, Addiego J, Gomperts E, Kessler C, Garanchon L, Neslund G, Foster V, Berkebile R, Courter S, Lee M, Kingdon H, Griffith M. A plasma factor VIII:c concentrate purified by immunoaffinity. In: Stoltz JF, Rivat C, eds. Biotechnology of plasma proteins. Colloque 1989; 175: 263-270. 57. Burnouf T, Michalski C, Goudemand M, Huart JJ. Properties of a highly purified human plasma factor IX:c therapeutic concentrate prepared by conventional chromatography. Vox Sang 1989; 57:225-232 58. Iwarson S, Einarsson M, Smallwood L, Snoy P, Gerety RJ. Chromatographic removal of hepatitis B virus from a factor IX concentrate--Experimental studies in chimpanzees. Transfusion 1987; 27: 171-173. 59. Einarsson M, Prince A, Brotman B. Removal of non-A, non-B hepatitis virus from a concentrate of the coagulation factors II, VII, IX, and X by hydrophobic interaction chromatography. Scand J Infect Dis 1985; 17: 141-146. 60. Berntorp E, Nilsson IM, Ljung R, Widell A. Hepatitis C virus transmission by monoclonal antibody purified factor VIII concentrate. Lancet 1990; i: 1531-1532. 61. Horaud F. European project concerning the guidelines on medicinal products derived from human blood and plasma. Dev Biol Stand 1991; 75: 237-340. 62. CPMP Guidelines. Ad hoc working party ofbiotechnology/pharmacy-Note for guidance 'Validation of virus removal and inactivation procedures (III/8115/89). February 11-13, 1991. 63. Bloom AL. Progress in the clinical m a n a g e ment of haemophilia. Thromb Haemostas 1991; 66: 166-177. 64. Helmer RE, Alperin JB, Yunginger JW, Grant JA. Anaphylactic reactions following infusions of factor VIII in a patient with classic hemophilia. Am J Med 1980; 69: 953-957. 65. Eyster ME, Bowman HS, Averstick JN. Adverse reactions to factor VIII infusions. Ann Intern Med 1977; 87: 248. 66. Wadsworth C, Hanson LA, Kjellman H, S6derstr6m T, Blomb~ck M. Some characteristics ofaggregates of IgG and plasma proteins in heat-treated factor VIII concentrates. Blut 1989; 58: 133-141. 67. Kasper CK. Treatment of factor VIII inhibitors. In: Coller BS, ed. Progress in hemostasis and thrombosis. London: WB Saunders Co. 1989; 9: 57-86. 68. Bell BA, Kurczynski EM, Bergman G. Inhibitors to monoclonal antibody purified factor VIII. Lancet 1990; ii: 638. 69. Kasper CK. Thromboembolic complications. Thromb Diath Haemorrh 1975; 33: 640-644. 70. Aronson DL, M~nach~ D. Thrombogenicity of factor IX complex: in-vivo investigation. Dev Biol Standard, 1987; 67: 149-155.

100

T. Burnouf

71. McGregor IR, Ferguson JM, McLaughlin LF, Burnouf T, Prowse CV. Comparison of high purity factory IX concentrates and a prothrombin complex concentrate in a canine model of thrombogenicity. Thromb Haemostas 1991; 66: 609-613. 72. Mannucci PM, Bauer KA, Gringeri A, Barzegar S, Santagostino E, Tradati FC, Rosenberg RD. No activation of the coagulation cascade after large doses of a high-purity factor IX concentrate. Br J Haematol 1991; 79: 606-611.

73. Semple MG, Loverday C, Preston E, Tedder RS. Detection of HIV-1 RNA in factor VIII concentrate. AIDS 1991; 5: 597-609. 74. Zhang LQ, Simmonds P, Ludlam CA, Brown AJL. Detection, quantification and sequencing of HIV-1 from the plasma of seropositive individuals and from factor VIII concentrates. AIDS 1991; 5: 675-681.

Received for publication 10 December 1991; accepted 23 April 1992.