CAF-DCF Product Development Division

Contact person: Ruth Laub PhD

The CAF-DCF Product Development Division (former R&D) is located in Brussels. For its staff, research and development means ensuring both the efficacy of plasma-derived medicinal products and their biological safety as regards pathogens, pollutants, and accompanying proteins. Focusing on therapeutic proteins (IVIG, albumin, AGP, FVIII) and their excipients in plasma and concentrates, the division develops both immunological methods and biochemical-biophysical techniques and exploits them in industrial applications. The paradigm of plasma derivative safety is approached through projects in different areas: NAT screening, collection and statistical evaluation of critical virus epidemiological data, neutralization by specific antibodies, virus infectivity testing in a cell model, virus inactivation/virus elimination validation studies, and pathogen reduction techniques (including UVC irradiation developed in our R&D Division).

Our activities:


Intravenous immunoglobulins (IVIGs) and clinical immunoprotection.

Sero-epidemiology in EU and US plasmas and levels of protective anti-pneumococcal polysaccharide antibodies (PnPsAbs) in plasma pools for fractionation and therapeutic IVIGs.

Streptococcus pneumoniae is one of the most common pathogens in children and the elderly, causing severe diseases such as invasive pneumonia. A few S. pneumoniae serotypes are responsible for all pathologies. Although the protective antibody contents of intravenous immunoglobulins are ill defined, estimating them is crucial to treating immunodeficient patients and to ensuring clinical protection.

IVIGs are produced from large pools of plasma (>10,000 donations) and contain a wide range of PnPsAbs. Standardized stenotype-specific ELISAs selectively measure PnPsAbs after pre-incubation with the common cell-wall polysaccharide (CWPS), with or without 22F. Under these conditions, WHO13 recommends thresholds of 0.35 and 0.2 µg /ml respectively in vaccinated pediatric patients. Using an ICH-Q2(R1) validated high-throughput antibody quantification procedure (HTQ), we quantified 16 major PnPsAbs in different IVIG preparations, including Multigam® and Nanogam® batches, and evaluated their epidemiology in donations collected in Belgium, The Netherlands, France, and the US.

All the PnPsAb contents appeared remarkably stable in Multigam® batches produced over the 2002-2008 period. For a dose of 400 mg IVIG, the major PnPSAb titres were between 219 µg and 109 µg (5-fold enrichment as compared to plasma pools). Although a comparable distribution was found in all IVIG concentrates and plasma pools from different origins, qualitative differences were observed according to the brand product and plasma origin. In pools from healthy human donors, PnPsAbs against some serotypes failed to reach the putatively protective titre of 1.3 µg/ml recommended by Sorensen et al (1998). This suggests a need to discuss the relevance of this threshold. IVIG’s meet most of the requirements of WHO13 for PnPsAb contents and constitute a powerful tool for anti-infective immunoprophylaxis.

Multi-centre clinical study in 22 pediatric patients with primary deficiency treated with Multigam®

Immunoglobulins are plasma-derived products used in children for a variety of well-established indications. Dosing is by body weight and often tailored to individual needs. Continual IgG replacement therapy reduces the frequency and severity of infections in these patients. The correlation between IVIG dose and immunity is a subject of continued interest for both theoretical and clinical reasons. The serum IgG concentration immediately preceding the next scheduled IVIG infusion is designated the through level. Through levels have been used to evaluate the adequacy of a particular dosage.

A non-interventional prospective multicenter study of PID and BMT pediatric patients has been launched according ICH-E11 guidelines. It is conducted within the context of everyday medical practice. The study has been approved by the central and local Ethics Committees. A total of 22 PID patients and 18 BMT patients have been enrolled at 8 clinical centers. One PID and 4 BMT patients have been excluded. All patients must have been under regular treatment with Multigam® (CAF-DCF, Brussels) for at least 5 consecutive infusions. Patients were monitored over a period spanning 6 Multigam® infusions/patient.

Serotype-specific APabs are measured by 16 specific ELISAs (serotypes: 1, 3, 4, 6, 6B, 7F, 8, 9N, 9V, 10A, 12F, 14, 18C, 19A, 19F, and 23F) according to WHO guidelines. The ELISAs are performed by an automated robotic platform and validated according to ICH-Q2(R1) guidelines. Traceability of samples, tests, and related results is carefully ensured with computer tools. More than 50,000 assays have been performed to determine trough levels of total IgG, the four IgG subclasses, and specific antibodies protecting against S. pneumoniae infection (see above). At the present time, these data are being statistically analyzed.

 

The redox state of albumin (Alb) and its clinical benefit

The therapeutic benefit of pharmaceutical-grade albumin containing acetyl-tryptophan may be limited by reduction of its antioxidant capacity. Human albumin (Alb) is widely used for volume replacement or correction of hypoalbuminemia. It is also used for extracorporeal liver support procedures based on albumin-recycling or single-pass albumin dialysis.

Alb is prepared from pooled human plasma by ethanol precipitation. For blood-borne pathogen inactivation, albumin is pasteurized for 10 h at 60°C according to the European Pharmacopoeia and then incubated at 25°C. In commercial albumin, caprilate alone (Capr) or caprilate + N-acetylryptophan (AcTrp) are used to stabilize the monomeric form of HSA by increasing its denaturation temperature and calorimetric enthalpy. N-Ac-Trp has only a minor stabilizing effect. Furthermore, Capr but not N-AcTrp, when used alone, exerts a pronounced effect by preventing Alb polymerization during heat treatment. Capr and N-Acetyl Tryp bind to Alb, but very little is known about their respective effects on its capacity to bind oxidant drug molecules, and no crystallographic model has yet been published.

We have further evaluated the impact of these stabilizers on Alb binding and antioxidant capacity after pasteurization by measuring the binding of well-known molecules (haemin, bilirubin, warfarin, copper) and probes (dansylsarcosine, cis-parinaric acid).

Commercial Alb preparations containing Capr or Capr + N-AcTrp were studied and compared with lab-scale-produced albumins. The results show that addition of N-AcTrp to Alb in the presence of Capr, strongly decreases the total antioxidant capacity of Alb and binding of the powerful oxidant haemin (Ka and number of binding sites), cis-parinaric acid, and copper. The free Cyst-34 (1/3 of the molecules) is blocked in the presence of N-AcTrp, as demonstrated with Ellman’s reagent and by binding of florescent dyes reacting with free sulfhydryl radicals, followed by SDS-PAGE and ESI-Q-TOF. The bilirubin affinity constant (Ka) of lab-scale preparations was restored after the post-pasteurization incubation at 25°C required by the Pharmacopoeia. After addition of stabilizer(s) to the HSA preparation, the binding affinity for dansylsarcosine (used as a probe) was slightly reduced. An in silico model was established, in which AcTrp binds to the IIIA domain and to the IA/IIA domain, shared with Capr. In contrast to Capr, AcTrp appears to drastically modify the shape of HSA by allostery, as shown by the altered binding of haemin, cis-parinaric acid, and copper. This may result in a drastically reduced HSA antioxidant capacity.

Properties of pharmaceutical albumin stabilized with Capr (CAF-DCF) as compared to albumin stabilized with Capr and N-AcTrp (Baxter) in an in vitro MARS-detoxifier-mimicking device. The Molecular Adsorbent Recirculation System (MARS) is an extracorporeal liver dialysis system consisting of two separate dialysis circuits. The first circuit contains human serum albumin, is in contact with the patient's blood through a semi-permeable membrane, and has two special filters, charcoal and IEX, to clean the albumin after it has absorbed toxins from the patient's blood. The second circuit is a hemodialysis machine, and is used to clean the albumin of the first circuit before it is recirculated to the semi-permeable membrane in contact with the patient's blood.

In collaboration with De Bruyn and coworkers at the KUL, we aimed to determine the impact of exposure to activated charcoal and IEX, both present in the MARS device used for hemodialysis, by measuring the binding of warfarin, diazepam and salicylate. This study was performed by the KUL lab and confirmed that addition of N-AcTrp to Alb modifies the kinetic binding parameters of Alb. Only after a long incubation (7 h) in the presence of charcoal/IEX did Baxter Alb recover its binding activity, to a level depending on the substrate and the adsorbing material in the device. The data support the view that changes in ligand-binding affinity and albumin capacity depend on the Alb supplier, the contact time, the ligand, and the use of IEX or charcoal. 

 

Other studies

Proteases in plasma intermediate fractions and consequence on safety and efficacy - New FVIII: vWF concentrate process

To ensure product safety and efficacy, protein therapeutics must meet defined quality criteria both immediately after manufacture and the end of their shelf lives. Proteins are typically sensitive to slight changes in solution chemistry or to subtle changes in product processing. Even lyophilized products undergo enzymatic or chemical degradation, resulting in oxidation and aggregation. Aggregates are hard to identify because most classical techniques such as chromatography or electrophoresis require a pre-filtration step. First assays have shown that dynamic light scattering allows quantification of large and very large aggregates present in cryoprecipitate and FV preparations obtained in different ways.

On the other hand, co-purifying proteins such as proteases may drastically affect product safety. A recent worldwide thromboembolic adverse event outbreak linked to two specific IVIG brands has raised serious safety concerns. Contaminating activated coagulation factor FXI, a serine protease (or an FXIa-like protease) could have a relevant role. Very low but significant content of FXI was found in intermediate products but are reduced (traces) in IVIGs. A battery of tests was set up to quantify and identify potential proteases in intermediate fractions. About 4 different protease activities were distinguished. Studies identifying these proteins are in progress.

The process yielding FVIII concentrate rich in vWF was also studied, in collaboration with GJ Derksen, J Penninkx, and A Koenderman, starting with cryoprecipitates of different origins. A third virus inactivation step was also investigated. The results highlight the impact of cryoprecipitate quality on the total yield.

Human Parvovirus B19 (B19V) interactions with cell surface receptors in a hepatoblastoma cell model

Blood-borne human parvovirus has a worldwide distribution and typically causes a mild childhood febrile illness known as the fifth disease. In patients with underlying immunological and hematological disorders, B19 has been associated with more severe manifestations such as arthropathy and hydrops foetalis. Previous studies have demonstrated that hepatoblastoma and hepatocarcinoma cells can be used successfully for in vitro production of infectious B19 particles. Globoside, KU80 autoantigen, and α5β1 integrin have been identified as cell receptors/coreceptors for B19V in the Epo-dependent bone marrow megakaryoblastic leukemia cell line (UT7/Epo), also described to support B19V multiplication in vitro. First studies performed with Prof A Op De Beek’s team have shown that B19V proliferation is strongly modified in the presence of specific antibodies recognizing these receptors/co-receptors. 

Monitoring viral infection markers in the donor population in Belgium in 2009 - critical evaluation and annual assessment update

Besides careful donor selection and precautionary exclusion, surveillance of infectious markers in the donor population is important in recognizing trends in the prevalence and incidence of transfusion-related infections.

The guideline EMEA/CPMP/BWP/125/04 Rev.1 recommends collecting epidemiological data over the last year and the three previous years and identifying any overall trends in the rates of infectious markers in the donor population. To assess the residual risk of agents for which screening is performed, we use the statistical approach known as the incidence rate/window period model (Kleinman et al., 1997; Busch et al., 2006). The residual risk is calculated as equal to the incidence rate of infection in donors, multiplied by the length of the infectious window period during acute infection. These epidemiological data for the period 1 January to 31 December 2009 were obtained from the following regional transfusion organizations and blood centres: Dienst voor het Bloed (DvB), Service du Sang (SdS), ASBL La Transfusion du Sang Charleroi, ETS de Mont-Godinne and AZ Sint-Jan AV Brugge. Each of these is accredited function as a blood establishment and is inspected by the National Health Authority. Blood and plasma are collected only from voluntary, unpaid donors.

Infection incidence rates cannot be measured by observation of seroconversion in an FD population, as they are in an RD population (Soldan, 2009). Direct measurement of the incidence (the frequency of ’NAT pos Ab neg‘ donations) among FDs is problematic in our case, as NAT screening has a very low yield (1 HIV-positive, 2 HCV-positive, no HBV-positive donation in 2009). Therefore, the indirect method described by Busch (2005) was used to calculate overall incidence rates (for FDs and RDs combined), although O’Brien et al., 2007 have shown that this method overestimates the HCV risk. For the RD population, incidence rates were derived by dividing the number of known confirmed incident cases detected by serological and NAT screening by the number of person-years. As the study covered a rather long period (7 years) and as candidates were included in the FD group, the frequency of donation (among FDs and RDs) was rather low. Because of the low positive rates and the small sizes of the centers, the analyses for all centers were performed for 6 successive overlapping 2-year periods: 2003-2004, 2004-2005, 2005-2006, 2006-2007, 2007-2008, 2008-2009 to provide meaningful estimates.

FDs, including candidates who have not yet donated, represented 17.7% of the total donor population. Numbers of FDs and RDs were very stable over the period 2003 to 2009 (7 years), with a mean of 279,614 donors. This represents about 2.6% of the total Belgian population. A total 590,347 donations from RDs were collected, of which 83% from whole blood. As for the incidence rate trends, the risk estimates for all three viral markers remained within in the confidence interval of the previous 2-year period, 2007-2008. Since January 2010, to improve blood product safety as regards HBV, all the Belgian blood transfusion organizations will perform the HBV NAT test on smaller mini-pools (mini-pools of maximum 24 donations). This will shorten the window period and thus reduce the residual risk estimate for HBV. 

Collaborations

Op De Beeck A1, Caillet-Fauquet P1, Draps ML1, Kubler L1, Tuerlinckx D2, Martin B2, Duclos M2, Laureys G3, Haerynck F3, Florkin B4, Ferster A5, De Schutter I6, Malfroot A6, Philippet P7, Chantrain C8, De Bruyn T9, Augustijns P9, Annaert P9, Meijers B10, Evenepoel P10, Willems L11, Koenderman A12, Derksen G-J12 and Penninkx J12.

  1. ULB, Medecine Faculty, Epigénétique du Cancer, Brussels, Belgium
  2. Cliniques Universitaires U.C.L. de Mont-Godinne, Mont-Godinne, Belgium
  3. UZ Gent, Gent, Belgium
  4. CHR La Citadelle, Liège, Belgium
  5. HUDERF, Brussels, Belgium
  6. UZ Brussel, Brussels, Belgium
  7. CHC Espérance, Liège, Belgium
  8. UCL Saint-Luc, Brussels, Belgium
  9. KUL, Laboratory for Pharmacotechnology & Biopharmacy, Department of Pharmaceutical Sciences, Leuven, Belgium
  10.  University Hospital Leuven, Division of Nephrology, Faculty of Medicine, Leuven, Belgium
  11.  Univestity Hospital Gasthuisberg, Hospital Pharmacy, Department of Pharmaceutical Sciences, Leuven, Belgium
  12.  Sanquin Plasma Products, Product Development, Amsterdam, The Netherlands

Key publications

Patents