How Innovative Technology Is Moving Biologics Into the 21st Century

Clinical Pharmacology & Therapeutics (2010) 87 3, 261–263. doi:10.1038/clpt.2009.294 Embracing biologics has meant embracing biology to a greater extent than ever before in the pharmaceutical industry. The fundamental difference between synthetic chemistry and biotechnology is that the latter depends on the agency of a living organism other than the chemist. This has a profound impact not only on how biological medicines are made but also on the nature of the final product. Unlike chemical reactions, which yield predictable and reproducible end results if they are done with care, living organisms can be very persnickety. Also, different organisms process proteins, which are the basis of most biologics, in different ways. Thus, the production (expression) system that may be most attractive and most efficient may not turn out the right kind of protein. Even the best-suited systems for expressing particular therapeutic proteins can have significant drawbacks, such as expensive and lengthy cultivation processes, low yield, and the risk of viral contamination. Many therapeutic proteins are complex molecules that contain multiple structural folds, may be composed of two or more individual proteins or subunits connected by disulfide chains, and may have other molecules, such as sugar chains, attached to their amino acid “backbone.” Each of these factors—which can impact the potential safety, structure, activity, cost, and availability of a therapeutic protein—is influenced by the method, in particular, the living organism, used to produce the protein itself. In recent years, there has been a great push to develop new technologies that address the shortcomings of standard methods of manufacturing protein-based therapies.1 Biotechnology and pharmaceutical companies alike have invested in these new technologies to improve existing ventures or facilitate their full entry into the development of biologic medicines, respectively. Here we discuss some of the innovative platforms that could help advance the science of biologics into the 21st century and realize the full potential of innovative protein-based therapies and biosimilars. Protein-based therapies differ from small molecules in many ways. They are larger and much more complex. In addition, an estimated 70% of the proteins are glycoproteins, which means they are studded with various sugar molecules.1 The type and number of sugar chains that are attached to the protein's backbone play a critical role in how glycoproteins function in the body. Relatively minor changes in the composition and structure of the sugar chains (glycosylation) can dramatically affect the in vitro stability and in vivo circulatory half-life of a glycoprotein and its ability to locate the right clinical target, bind to receptors, and evade the immune system (reviewed in ref. 1). All these factors contribute to the overall efficacy and safety profile of a therapeutic protein. The sequence encoding a glycoprotein does not dictate the composition of the final protein. Context is critical because posttranslational modifications are organism specific. Workhorse bacteria such as Escherichia coli, for example, do not add sugars to the vast majority of proteins they produce. Bacterial systems are therefore typically reserved for simple and nonglycosylated therapeutic proteins.1 Mammalian cell lines—most notably, Chinese hamster ovary (CHO) cells and murine myeloma cells (NS0)—are the current industry standard. However, mammalian cells are generally difficult to grow in culture and generally require complex culture media, including animal growth factors. Because the cells themselves can be a growth source for animal viruses, contamination with viruses and other agents is a concern. In addition, mammalian cell culture systems require a relatively long time because mammalian cells do not yield high quantities (by volume) of recombinant proteins. Finally, although therapeutic glycoproteins produced in mammalian cell culture have sugars attached at the same sites, these glycans may differ from those found in humans. For example, therapeutic proteins produced in murine cell lines such as NS0 will often have a particular sugar chain that contains two galactose sugar molecules in sequence (“gal-α-1,3-gal”), whereas the same protein found in humans has sugar chains that have only a single galactose sugar molecule in sequence. Therefore, the gal-α-1,3-gal sugar chain found on glycoproteins expressed in NS0 cells can be recognized by circulating immunoglobulin E in humans. This may be a factor in infusion reactions from hypersensitivity as in the case of an antiepidermal growth factor receptor monoclonal antibody.2 Additionally, therapeutic proteins produced in murine and CHO cells contain the terminal sialic acid N-glycolylneuraminic acid (Neu5Gc), which is not produced in humans because of a genetic mutation. People who consume red meat and dairy products typically have a diverse spectrum of circulating anti-Neu5Gc antibodies, which present a risk of immunogenic response to therapeutic proteins produced in CHO or murine cell lines.3,4 Alternative expression systems, including insect cells, plant cells, and yeast, have been evaluated. Yeast, in particular, is an attractive protein-expression system because the cells do not require any animal-based growth factors; they grow rapidly and produce large amounts of proteins that are secreted directly into the growth medium. Although yeasts are superior to bacteria for making glycoproteins, native yeasts do not add or process the sugars in the same way that human cells do.5 Thus, they historically have not been used to produce therapeutic proteins in which glycosylation is important for efficacy. Recently, the ability to engineer not just the amino acid backbone of a protein but also the structure of the sugar units, or glycans, has increased tremendously, heralding a new era of “glycoengineering.”6 Protein researchers and biotechnology companies have long recognized the possibilities of optimizing therapeutic proteins by changing and controlling the structure of these sugar units. Such manipulation has been shown to improve various critical characteristics of therapeutic glycoproteins, including circulatory half-life, potency, tissue distribution, and immunogenicity. For example, modifying recombinant human erythropoietin (epoetin alfa, marketed by Amgen (Thousand Oaks, CA) as Epogen) to include two new sites for the addition of sugar molecules significantly extended the half-life of the resulting product (darbepoetin alfa, marketed by Amgen as Aranesp).6 Another successful example of glycoengineering is the enzyme replacement therapy Cerezyme (marketed by Genzyme (Cambridge, MA)), which was modified for better liver targeting by reconfiguring the attached mannose sugars, increasing binding of the protein to macrophages predominant in liver tissue.7,8,9 Major biopharmaceutical companies have acquired or formed partnerships with various glycosylation technology companies to develop therapeutic proteins with enhanced properties. For example, Roche (Basel, Switzerland) acquired Glycart (Schlieren-Zürich, Switzerland), which has genetically engineered mammalian cells to produce antibodies studded with bisected nonfucosylated sugar chains. These modified therapeutic antibodies purportedly bind with greater affinity to receptors on specific immune cells, leading to improved killing of target cells through antibody-dependent cytotoxic activity.10,11 The ability to genetically engineer various types of cells—not just mammalian cells—to produce modified glycoproteins has created new technology platforms that permit greater control over glycosylation than previously possible. It has also permitted the use of more efficient expression systems to produce therapeutic glycoproteins. In 2003, scientists engineered the yeast Pichia pastoris to glycosylate proteins in a manner similar to human cells.12 As mentioned above, yeast-based expression systems offered certain advantages and are highly tractable to genetic manipulation. Since this initial breakthrough, work continues to refine the genetics and processes of the Pichia system, creating the first yeast-based platform capable of expressing fully humanized therapeutic glycoproteins.1,5 The Pichia system has several potential advantages over mammalian expression systems (1). Unlike mammalian expression systems, engineered Pichia allow for tight control of glycosylation. Yeasts that produce glycoproteins with the right signature of sugar molecules (glycoform) can be selected and grown in large quantities. Thus, the end product consists of a much more homogeneous population of glycoproteins compared with glycoproteins manufactured in mammalian cells.1,5 This has important implications for scaling up production, manufacturing therapeutic proteins at multiple sites (portability), and performing quality assurance/control (analytical comparability). As an increasing number of therapeutic proteins approach the expiration of their patent life, there has been a growing interest in the production of biosimilars and the development of innovative/chemically modified proteins with the potential to improve on the therapeutic properties of the originator molecules. The replication and characterization of small-molecule drugs for generic pharmaceuticals are routinely undertaken. In contrast, the consistent production and characterization of large, high-quality, more complex therapeutic proteins with assured safety and efficacy represent a considerably more challenging task. Process continues to define the product for protein therapeutics, and small process changes may lead to substantive variations in protein product characteristics. Technological innovations have the potential to advance the production and development of biosimilars, perhaps even creating more robust systems that improve product consistency and homogeneity. Prior to incorporation of such technologies, numerous legislative, regulatory, and patient- and industry-related challenges must still be overcome to create a robust biosimilars industry. Manufacturers, governments, regulators, and patient-advocacy organizations will need to collaborate in a way that will encourage further innovation, embrace new technologies to be applied to protein-based therapies with established mechanisms of action, and, ultimately, provide patients and physicians with a greater range of therapeutic choices. The authors are employed by Merck Research Laboratories.