OR WAIT 15 SECS
Mark Wagner is purification technology and business development specialist at EMD Millipore, 290 Concord Road, Billerica, MA 01821.
Jeffrey Shumway is associate director, bioavailability enhancement at EMD Millipore, 290 Concord Road, Billerica, MA 01821.
Juan Castano is process development scientist at EMD Millipore, 290 Concord Road, Billerica, MA 01821.
Claire Scanlan is process development scientist at EMD Millipore, 290 Concord Road, Billerica, MA 01821.
Ruta Waghmare is group leader, process development sciences at Millipore Corporation, Billerica, MA 01821.
Antibody fragments pose unique challenges in terms of recovery, purification, and formulation.
Monoclonal antibodies (mAbs) are an important class of therapeutics. In recent years, however, increased attention has been devoted to the development of next-generation biologic therapeutics and manufacturing platforms. This shift is being driven by several factors, including the need to develop more efficient and cost-effective processes, reduce side effects, and increase specificity to enhance efficacy.
Fragment antigen-binding (Fab) antibody fragments are one example of these next-generation biologics that are emerging as credible alternatives to the widely accepted mAbs (see Figure 1). Abciximab (ReoPro, Eli Lilly), ranibizumab (Lucentis, Genentech), and certolizumab pegol (Cimzia, UCB) are Fab products currently on the market. There are several more in development by other companies.
Fab fragments can be produced using more economical expression systems (e.g., microbial) while retaining the target specificity of a mAb. Other advantages include elimination of non-specific binding between the fragment crystallizable (Fc) portions of antibodies and the Fc receptor on cells, and the ability to engineer the Fab region for improved specificity and therapeutic efficiency.
Antibody fragments are commonly produced as intracellular products in microbial expression systems such as E. coli. Following fermentation, cell harvest, and disruption, the downstream purification process typically consists of clarification, capture chromatography, polishing chromatography, and ultrafiltration/diafiltration. Final formulation and sterile filtration take place prior to the fill-finish (see Figure 2). In this article, the authors outline some of the unique requirements and challenges posed by antibody fragments in terms of recovery, purification, and formulation.
Recovery and purification
Fab molecules are typically expressed in E. coli or yeast microbial systems. To achieve higher final titers, the amount of solids in these expression systems is increasing, making clarification more challenging. For soluble Fab proteins, the use of normal flow filtration clarification methods post-lysis is not common. Instead, tangential flow filtration (TFF) is often used in place of normal flow filtration.
If normal flow filtration is desired, it is typically incorporated post-centrifugation, using a disk-stack or tubular centrifuge. Because particle size distribution in the batch is small (the particle size is quite homogenous and has small particle-size impurities), a tight depth filter is often employed. Microfiltration can also be used without centrifugation.
With secreted proteins, whole cells are separated from the fermentation broth and the particle size is larger. As a result, microfiltration (using TFF), centrifugation, and normal flow filtration are viable options. Endonuclease agents can also be added prior to clarification to digest DNA and RNA and to aid in the efficiency of the clarification process.
Protein A, followed by cation exchange and anion exchange, can be successfully used for the purification of Fabs containing the Fc region. For a Fab that does not contain an Fc region, capture is typically achieved via cation exchangers or mixed mode resins in bind/elute mode. The advantage of mixed mode is a wider operating window for binding in various conductivity and pH settings. With mixed mode, however, a longer period of process development may be required to develop an elution strategy. To overcome this disadvantage, single mode resins are being developed that operate in higher conductivities with high binding capacities and easy elution processes. A subsequent polishing step, typically using a smaller bead size, generally follows the capture step with the aim of improving the resolution. This polishing step could be ion exchange (IEX) or hydrophobic interaction chromatography (HIC) depending on the previous step and the protein of interest. If further purification is required, a third step could be implemented with HIC or IEX depending on previous steps. A typical purification for non-Fc containing fragments would therefore be cation exchange > HIC > anion exchange.
Fab molecules are typically small in comparison to mAbs (approximately 10 kD-80 kD), and as a result, tighter tangential flow devices are commonly used for the ultrafiltration/diafiltration (UF/DF) step. The membrane molecular weight cut-off recommended for adequate retention is one fifth the size of the product of interest. 1 kD-10 kD molecular weight cut-offs are typically used for Fab molecules. These smaller cut-offs can lead to lower permeate fluxes.
Additionally, some Fab molecules are PEGylated, which leads to higher viscosities as concentration increases. Because of this, the influence of the type of screen (screens help create turbulence and promote mass transfer) in flat-sheet devices is paramount and should be taken into account in small-scale optimization studies. A final consideration for E. coli-expressed Fab molecules is downstream endotoxin removal, which is often achieved through anionic membrane adsorbers or charged membrane filters.
The primary consideration in formulation design for Fab protein therapeutics is stability. Protection of the protein from chemical degradation during storage as well as degradation from proteolytic influences in vivo is crucial in the formulation for all protein-based therapeutics. A second and equally important function of formulation design related to protein activity is to inhibit aggregation tendency. Aggregation may result in decreased efficacy due to inactivation associated with intermolecular interaction and the effect of steric inhibition.
When considering bioavailability, Fabs present both opportunities and challenges. Being significantly smaller than native monoclonal antibody molecules, Fabs designed as cancer therapeutics have the ability to provide better tumor penetration as compared to larger molecular species. Conversely, Fabs suffer the disadvantage of significantly shorter circulation and half-life in vivo, two characteristics that can be addressed in formulation.
PEGylation, the chemical modification of a protein by conjugation to poly(ethylene glycol) (PEG) polymers offers, in many cases, an appropriate solution to the challenge associated with shortened circulation and half-life of the Fab therapeutic. The most common chemical modification of Fabs with PEGs involves reaction of an activated PEG-maleimide with the cysteine residues on the Fab. PEGylated proteins offer the advantage of appearing 5-10 times larger in volume than proteins of comparative molecular weight because of the hydration effects associated with PEG. PEGylation results in improved stability and solubility while maintaining efficacy associated with the antibody fragment. Additional benefits of PEGylation include reduced toxicity, decreased clearance rates and proteolysis effect, and reduced immunogenicity and antigenicity although recent understanding of PEGylation effects has indicated an immune response associated with PEG that must be considered and monitored (1, 2).
Another effect of PEGylation is the tendency of a PEGylated protein to demonstrate slower sedimentation behavior, even with larger volume as compared to the native Fab. In fact, the protein-associated PEG tends to dominate hydrodynamic properties associated with the bound protein and may also contribute to a decrease in aggregation-related effects.
Despite the aforementioned benefits, however, PEGlyation is not the only methodology that can yield stabilizing effects on therapeutic Fab protein formulations. Non-ionic detergents, such as polysorbates, are commonly employed in Fab formulation strategies as additional means of promoting solubility and reducing aggregation. Careful constructs of excipient ingredients (salts and buffers) may also be considered in terms of viscosity-related effects. Salt-enhanced repulsive protein-interaction effects based on the Debye theory of diffusive charge indicates a correlation to improved repulsive effects associated with monovalent salts (e.g., NaCl) as opposed to divalent (e.g., MgCl2) ionic ratios (3). Monovalent versus divalent cations and/or anions may also be considered as stabilizing species in formulation.
Fab molecules are generating increased interest as the demand for target therapeutics with improved efficacy continues to grow. These molecules present unique challenges for the “traditional” processes used for the development and manufacture of monoclonal antibodies.
Each of the steps in recovery and purification must be optimized based on the process requirements and the molecule characteristics to ensure a robust, stable, and scalable Fab production process.
The same careful consideration must be applied to development of the optimum formulation. As with all complex protein formulation development efforts, the goal of Fab formulation is preservation of product stability and efficacy. All potential technologies can and should be considered to assure a robust formulation. Care should also be exercised to ensure the appropriate level of both purity and consistency in all materials used in formulation to minimize the impact of impurities and the possible effect on stability and activity.
The authors would like to thank Robert Blanck, market research & analytics manager at EMD Millipore, for his contributions to this article.
1. Y. Lu, et al., J. Pharm. Sci., 97 (6) 2062-2079 (2008).
2. R.P. Garay and J.P. Labaune, The Open Conference Proceedings Journal, 2, 104-107 (2011).
3. A.S. Parmar and M. Muschol, Biophys J., 97 (2) 590-598 (2009).
4. Chugai Seiyaku Kabushiki Kaisha, “Stabilizer for protein preparation comprising meglumine and use thereof,” EP 1908482 A1, April 2008.
About the Authors
Claire Scanlan is process development scientist; Jeffrey Shumway is associate director, bioavailability enhancement; Juan Castano is process development scientist; Mark Wagner is purification technology and business development specialist; and Ruta Waghmare, PhD, is associate director, emerging biotechnology, [email protected], all at EMD Millipore, 290 Concord Road, Billerica, MA 01821.