Next Generation Protein Engineering and Drug Design

Next Generation Protein Engineering and Drug Design

Publisher:	Business Insights Ltd (Datamonitor)
Year of publication:	2007
Type of publication:	Management report
Publisher's reference (if any):	RBDD0013
Author(s):	Sreten Bogdanovic and Beata Langlands
Approximate page count:	210
Price when published:	$2,875
Remarks: Page numbers, where given, refer to the draft manuscript (which may differ from the published version). The copyright in this report is owned by the publisher, to whom any requests for copies should be addressed. The price shown is for a single copy of the print version. Multiple copies and electronic copies usually have different prices.
TABLE OF CONTENTS Executive Summary Chapter 1 Engineering next generation protein drugs 1.0 Summary 1.1 Introduction 1.2 Background on protein structure and function 1.3 Patenting of proteins 1.4 Regulatory requirements 1.5 Commercial imperatives in protein engineering 1.5.1 Introduction 1.5.2 Application markets 1.5.3 Manufacturer markets 1.5.4 Product markets 1.5.5 Geographical markets 1.5.6 Patenting considerations 1.5.6.1 Usefulness of patent metrics 1.5.6.2 The protein engineering patent data set 1.5.6.3 Analysis by assignee patent count 1.5.6.4 Analysis by forward citation count 1.6 Commercial outlook for engineered proteins Chapter 2 Strategies and technologies for protein engineering 2.0 Summary 2.1 Introduction 2.2 Recombinant protein production 2.2.1 Site-directed mutagenesis 2.3 Post-translational modifications (PTMs) 2.3.1 Glycosylation of natural proteins 2.3.2 Manufacture of glycoproteins 2.3.3 Glyco-remodeling 2.3.4 Other PTMs 2.4 Protein characterization 2.5 Directed evolution 2.5.1 Display technologies 2.5.2 Use of protein scaffolds 2.6 Peptide and protein synthesis 2.6.1 Chemoselective ligation 2.7 In silico protein design 2.8 Technology-related patents 2.8.1 Monoclonal antibodies 2.8.2 Other proteins and peptides Chapter 3 Engineering improved monoclonal antibodies 3.0 Summary 3.1 Background on natural antibodies 3.1.1 Human IgG 3.1.2 Generation of antibody diversity 3.2 Introduction to monoclonal antibodies (mAb) 3.2.1 Evolution of mAbs 3.2.2 Drivers for innovation 3.3 The mAb business landscape 3.3.1 Products on the US market 3.3.2 Products in development 3.4 Improving mAb production systems 3.5 Manipulating mAb glycosylation profiles 3.6 Enhancing mAb serum stability 3.7 Engineering fully human mAbs 3.7.1 Human mAbs from recombinant antibody libraries 3.7.1.1 Immune and nonimmune antibody libraries 3.7.1.2 Optimization 3.7.1.3 Phage display libraries 3.7.1.4 Ribosome and mRNA display antibodies 3.7.1.5 Yeast-display antibodies 3.7.2 Human mAbs from transgenic mice and chickens 3.7.3 Human mAbs on the market and in development 3.8 Engineering novel antibody fragments 3.8.1 Monovalent fragments 3.8.2 Multivalent fragments 3.8.3 Fragments on the market and in development 3.9 Engineering for specific therapeutic applications 3.9.1 Cancer 3.9.1.1 Unconjugated mAbs 3.9.1.2 Conjugated mAbs and fusion proteins 3.9.2 Immune and inflammatory disorders 3.9.3 Other areas Chapter 4 Engineering alternatives to antibodies 4.0 Summary 4.1 Introduction 4.2 Comparison with monoclonal antibodies 4.3 Combinatorial scaffold libraries 4.3.1 Scaffolds used in library construction 4.3.1.1 Beta-sheet frameworks 4.3.1.2 Mixed/irregular secondary structures 4.3.1.3 Alpha-helical frameworks 4.3.1.4 Repeat proteins 4.3.2 Scaffold optimization and diversification 4.3.3 Selection technologies 4.4 Recognition proteins as therapeutics 4.4.1 Products in commercial development Chapter 5 Engineering other protein and peptide drugs 5.0 Summary 5.1 Introduction to protein/peptide drugs 5.1.1 Drivers for innovation 5.2 The business landscape 5.2.1 Products on the market and in development 5.3 Improving production systems 5.3.1 Recombinant methods 5.3.2 Transgenic methods 5.3.3 Chemical methods 5.3.4 Tackling immunogenicity 5.4 Manipulating PTMs 5.4.1 Glycoengineering 5.4.1.1 Glycoprotein profiling 5.4.1.2 Glyco-engineering 5.4.2 Other PTMs 5.5 Altering plasma half-lives 5.5.1 Case study: Erythropoiesis-stimulating agents 5.6 Expediting peptide drug discovery 5.6.1 Phage and other display technologies 5.6.2 Rational design 5.6.3 Case study: Antimicrobial peptide discovery 5.7 Exploring the role of pharmacogenomics Chapter 6 Engineering protein therapeutics for delivery 6.0 Summary 6.1 Introduction 6.2 Injectable protein delivery 6.2.1 Half-life extension technologies 6.2.1.1 PEGylation 6.2.1.2 Other approaches 6.2.2 Depot systems 6.3 Pulmonary delivery 6.4 Nasal delivery 6.5 Oral and other forms of delivery 6.6 Other approaches Chapter 7 Trends and opportunities 7.0 Summary 7.1 Creating non-immunogenic monoclonal antibodies 7.2 The next wave of monoclonal antibody-based agents 7.3 Beyond monoclonal antibodies 7.4 The challenge of follow-on biologics 7.5 The promise of synthetic peptides and proteins 7.6 New frontier: de novo protein design Index List of Tables Table 1.1 World Pharma Market by Indication in 2006 and 2011 Table 1.2 Protein Engineering Markets by Application in 2006 and 2011 Table 1.3 Protein Engineering Markets by Company in 2006 and 2011 Table 1.4 Protein Engineering Markets by Product in 2006 and 2011 Table 1.5 World Pharma Market by Region in 2006 and 2011 Table 1.6 Protein Engineering Market by Region in 2006 and 2011 Table 1.7 US Protein Engineering Patents and Published Applications by Filing and Publication Years. Table 1.8 Protein Engineering US Patents and Published Applications by Leading Assignees. Table 1.9 Top 50 Protein Engineering Patent Assignees by Forward Citation Count Table 2.1 Protein engineering patents; Technologies and Applications Table 2.2 Top 50 Cited Protein Engineering Patents (US Filings, 1992-2006) Table 2.3 Fully Human Engineered Immunoglobulin Patents (US Filings, 1992-2006) Table 3.1 mAb Products by Name Table 3.2 mAb Products by Therapy Area Table 3.3 mAb Products by Stage Table 3.4 mAb Products by Indication Table 3.5 mAb Products by Company Table 3.6 mAb Products by Antigen Table 3.7 mAb Products by Antibody Origin Table 3.8 mAb Products by Format Table 4.1 Non-immunoglobulin binding proteins in development Table 5.1 Examples of launched engineered human recombinant therapeutic proteins Table 5.2 Therapeutic human proteins produced in animal bioreactors; products in development Table 5.3 Engineered small peptide and peptidomimetic drugs: examples from antimicrobial R&D Table 6.1 Injectable protein delivery: technologies facilitating reduced administration frequency EXECUTIVE SUMMARY Chapter 1 Engineering next generation protein drugs This Report will analyse the current market for engineered proteins and peptides and the opportunities for the development of next-generation products. Proteins and genes can be patented as composition-of-matter, processes, or combinations of both. Separate claims can be made on new uses of a therapeutic protein or any of its variants even if the parent gene is already patented. The regulations issued by the FDA in 1996 placed greater emphasis on characterization techniques for biologicals. The BLA currently requires detailed analytical description of the protein product, which includes biopotency and immunogenicity testing. We review the major clinical applications for engineered proteins and we assess the companies which are currently making most progress in targeting these markets, including an examination of patent metrics. We also briefly cover the leading geographical markets. The protein engineering market in 2006 was worth about $67 billion, 10% of total pharma sales, and will rise to a forecast $118 billion, or 12% of pharma sales, in 2011. Monoclonal antibodies currently comprise 27%, but will rise to 36% in 2011 because their sales are increasing more rapidly (14.9% vs 7.5% annually) Oncology is the dominant application for both mAbs and other types of engineered protein, accounting for one-third of sales overall (and over 50% of all mAbs). In value terms, engineered proteins now account for about two-thirds of the worldwide oncology market ($23 billion of $34 billion). CNS represents the most rapidly growing mAb segment, although sales are small in absolute terms. Amgen and Roche have the biggest manufacturer market shares overall, accounting for 40% of total sales. The leading products are Amgen's Aranesp engineered erythropoietin (EPO) and Roche's Rituxan anticancer mAb, with 14% of the entire market between them. North America and Europe are the largest geographical markets, with shares of 44% and 28% respectively, while Latin America (and, in particular, Brazil) is the most rapidly growing. Genentech, owned by Roche, has by far the most protein engineering- related US patents assigned to it (192, 7.4% of the total), followed by Human Genome Sciences (75 patents, 2.9%). Genentech is also the most frequently cited assignee, although over half its patents have never been referenced by subsequent US patent applicants. In terms of references per cited patent, a number of other companies such as Protein Design Labs (now PDL BioPharma) and Bristol-Myers Squibb do as well or better. Chapter 2 Strategies and technologies for protein engineering Both genetic and chemical techniques can be utilized to engineer change in the structure and function of proteins and peptides. Mammalian cell culture, which is technically complex and expensive, is used for the production of recombinant glycoproteins. Site-directed mutagenesis can be used to optimise proteins and introduce unnatural amino acids. Post-translational modifications (PTMs) of a protein affect its structure and function. Glycosylation represents one of the most common modifications. Other PTMs, including carboxylation and hydroxylation, are characteristic of some therapeutic products. Glyco-engineering of proteins employs bioengineered bacteria, yeast, plant, insect, and mammalian cells with altered glycosylation pathways or can be carried out outside of bioengineered cells. Recombinant protein characterization has been transformed by the application of analytical technologies such as mass spectrometry, high performance liquid chromatography, capillary electrophoretic and biological assays for potency. Directed evolution strategies are being used to increase potency, stability, and other protein characteristics. Directed evolution entails the isolation and optimization of proteins through cycles of iterative random mutagenesis and functional screening. Functional screening is typically accomplished through the use of display technologies. These include phage display, yeast display, ribosome and mRNA display, and CIS display. In the commonly-used phage display a protein is displayed on the surface of a phage. Combinatorial chemistry can be used to generate novel binding molecules with optimized properties using protein frameworks with good biophysical properties as scaffolds. Advances in chemoselective ligation, which uses unprotected peptide fragments and reactive functional groups, has enabled synthesis of medium sized proteins and their modification with polymers in a site-specific manner. In silico protein design approaches use analysis of the 3-D structure of a therapeutic protein to guide the selection of appropriate amino acid sequences to create desired properties. Patents claiming recombinant proteins form the most numerous category of US protein engineering patents filed between 1992-2006. With respect to clinical applications claimed, patents citing cancer form the most numerous category. The top most cited US patent claims humanized immunoglobulins. Chapter 3 Engineering improved monoclonal antibodies The FDA defines a monoclonal antibody (mAb) as a clonal product which may be intact antibody, antibody fragment, conjugate, fusion protein, or bispecific antibody. mAbs form the majority of therapeutic proteins in development and optimal engineering of new mAbs is coming into play as a competitive factor. The therapeutic effect of mAbs is obtained either by blocking a target, or by exerting effector functions residing in the Fc region to activate cytotoxic cells or the complement system. Chimeric, humanized and human mAbs are under development. In chimeric mAbs, the murine Fc region has been replaced with one of human sequence. Humanized mAbs have been created through CDR grafting or variable domain resurfacing. The development of phage display technology, followed by transgenic mice, enabled the development of fully human mAbs. All approved recombinant mAbs have been produced in mammalian cell cultures. Several companies are developing non-mammalian production systems utilizing yeast, insect and other cell types, as well as plants. The aim is to produce glycoforms with predictable or improved effector functions. It is possible to extend the plasma half-life of mAbs by engineering Fc regions to alter the binding affinity for the FcRn receptor. Antibody fragments with a wider range of half-lives can also be generated through PEGylation. Human MAbs are being derived from recombinant antibody libraries, which can be based on immune or nonimmune fragments. Many mutagenesis and selection strategies have been used to provide subnanomolar antibody affinities. Selection strategies can also be applied to improve other properties, such as enhanced stability, resistance to proteases, or aggregation behavior. The binding affinity of human antibodies generated from transgenic mice is often high, reflecting in vivo affinity maturation. In Abgenix's XenoMouse and Medarex's HuMab-Mouse, murine germ-line antibody genes have been replaced with human ones. For many applications, the Fc-mediated effector functions are not required. Directed evolution selection strategies have been used to generate a wide variety of antibody fragments. Two popular monovalent fragment formats are Fab and scFv. Monovalent fragments are being engineered into multivalent molecules and bispecific fragments. The three major areas of therapeutic application are: cancer; diseases of the immune system; and infectious diseases. In cancer, engineering strategies vary depending on whether activation of effector functions or inhibition of signaling is the desired outcome. In contrast to unconjugated mAbs, a shorter blood residence time is frequently desirable for mAbs conjugated with toxic compounds, and can be achieved either by appropriate engineering strategies. Chapter 4 Engineering alternatives to antibodies In addition to antibodies, many other natural proteins mediate specific high-affinity interactions. Engineered antibodies are facing increasing competition from alternative binding proteins derived from combinatorial scaffold libraries. They are mainly small single-chain proteins. Several non-immunoglobulin binding proteins are under commercial investigation; most are still in an early phase of development. Developers of alternative binding proteins avoid the royalty stacking that burdens monoclonal antibodies today. However, not many alternative scaffolds have been shown to bind peptides yet. Many alternative binding proteins are stable without disulfide bonds, enabling cheap and efficient production in the reducing cytoplasm of bacteria. Unlike whole IgG-based MAbs, alternative binding proteins do not have effector functions, although engineering efforts are underway to confer such functions on them. A wide range of different non-immunoglobulin scaffolds with diverse origins and characteristics are currently used for combinatorial library display. Beta-sheets are the dominating secondary structural elements in scaffolds used for library construction. Alternative scaffolds need to be engineered for library generation, and scaffold optimization to improve chemical robustness, thermodynamic stability or recombinant expression yields is typically introduced early on in the process. The choice of a selection technology best suited for a given binding protein library depends on different parameters, such as the library diversity, the properties of the scaffold and the intended applications. In general, very high affinity is desirable for in vivo applications of alternative binding proteins. This can be achieved by a high intrinsic monovalent affinity of each binding site and/or by oligomerization. Chapter 5 Engineering other protein and peptide drugs Over 140 therapeutic proteins and peptides have been approved by the FDA and over 1,000 candidate products are under investigation. While the first generation of protein and peptide therapeutics entered development with little or no optimisation, recent approvals include products which have been re-engineered to have either an altered amino acid sequence or post-translational modification (in particular altered carbohydrate content). A variety of approaches are under investigation to improve cell production systems in order to allow the expression of effective amounts of soluble, functional proteins. Transgenic milk production offers a cost-effective system for the manufacturing of complex therapeutic proteins. Several companies have generated transgenic animal herds that yield large amounts of candidate therapeutic proteins. There is growing interest in modifying protein and peptide drugs to extend their plasma half-lives. Recent developments concerning erythropoiesis-stimulating agents are discussed; they provide a striking example of how the market need for longer-lasting therapeutic proteins is driving the development of novel engineered products. In addition to proteins, short agonist peptides have been utilized extensively as drugs in medicine. Novel peptides can be naturally derived or chemically synthesized, with the latter method currently more prevalent. Progress has also been made in the synthesis of peptide inhibitors. By altering their amino acid sequence, peptides are readily permutated into large libraries that lend themselves to rapid combinatorial analysis. Phage display derived peptides are often completely novel. Peptides can be designed de novo, but most peptides of biological interest are derived from sequences of native proteins. Peptide analogs with lower susceptibility to proteolysis have considerable potential as therapeutics. The current focus is on optimizing and expanding analog traits with chemistry. Selection of an optimal synthesis strategy depends in large part on the characteristics of the peptide being produced. Re-engineering may involve using computational methods to identify and replace antigenic epitopes. In recent years, levels of microbial resistance to antibiotics have risen dramatically injecting a strong impetus into development of new antimicrobial peptides which are surveyed here. They reflect different approaches to peptide drug engineering. Pharmacogenomic studies of protein drugs generally focus on pharmacodynamics. In new protein drug development, an understanding of the natural variation in both the drug and its target may guide engineering efforts aimed at the optimization of the lead molecule. Chapter 6 Engineering protein therapeutics for delivery Until recently, all approved biopharmaceuticals required to enter the bloodstream were administered parenterally, which is associated with a number of disadvantages, including reduced patient compliance and potential complications. The majority of sales currently originate from protein therapeutics that are immediate release formulations, with subcutaneous injection the most popular route of delivery, followed by intravenous infusion. Drug delivery can offer an injectable protein product a competitive edge in a number of ways. Technologies that result in a more stable plasma drug concentration can improve the efficacy and safety profile of the drug. The need to administer products less frequently can improve patient compliance and reduce costs. Persistence in circulation is amenable to engineering by a number of approaches, including half-life extension technologies and depot technologies. A prominent half-life extension technology involves the covalent attachment of polyethylene glycol (PEG) to the compound of interest. Enzon licensed this technology to Nektar Therapeutics and several refinements and proprietary approaches have recently been developed. For glycoproteins, post-translational carbohydrate remodeling has emerged as a strategy for increasing in vivo duration of action. Other half-life extension technologies include conjugation of polysialic acid, approaches exploiting human serum albumin's long circulation half- life, and (for peptides) introduction of non-natural D-amino acids or peptide cyclization. Several companies have focused on reducing the frequency of injection through the use of depot systems. Drug carriers include biodegradable microparticles, natural and synthetic polymers, microcapsules and liposomes. In the last three years advances in nonparenteral systemic drug delivery culminated in first approvals of products for pulmonary and nasal delivery. Recent advances, particularly particle engineering and formulation methods to control particle size, consistency, and stability, are expanding the applications for pulmonary delivery to proteins and peptides. Nasal delivery is attractive because of its convenience and the large surface area for absorption generated by the nasal microvilli. However, nasal delivery generally results in low bioavailability. The next frontier is oral delivery of therapeutic peptides and proteins, but this has always presented a significant challenge. Several proprietary technologies are under investigation. Chapter 7 Trends and opportunities Although it is accepted in the field that the risk of immunogenicity may be reduced by using fully human antibodies, mabs generated by combinatorial approaches may still be immunogenic and strategies need to be developed to eliminate antigenic epitopes. Given the increasing competition for therapeutic antibodies to the same target, new monoclonal antibody products with enhanced functionality, specificity and/or effector functions would gain a competitive edge. Antibody formats face increasing competition from alternative binding proteins, although few have been shown to bind peptides as yet. Developments are awaited in screening nonimmunoglobulin libraries and increasing the binding of alternative binding proteins to peptides. The advent of follow-on biologics (FOBs) highlights the significance of engineering improvements to proteins which would allow products to be differentiated on the basis of superior characteristics. Developers of FOBs will need to employ sophisticated analytical techniques to characterize their products. The future looks promising for synthetic peptides, and eventually synthetic proteins, since synthesis of medium-sized proteins is already possible. For patented proteins, substatial re-engineering of the protein during chemical synthesis may allow the product to be commercialized without risking patent infringement. The recent advances in the understanding of the process by which proteins fold and design of artificial proteins may permit de novo synthesis of therapeutic proteins. Novel protein molecules may have therapeutic potentials that are absent in natural proteins.

EXECUTIVE SUMMARY
Chapter 1 Engineering next generation protein drugs

This Report will analyse the current market for engineered proteins and peptides and the opportunities for the development of next-generation products.

Proteins and genes can be patented as composition-of-matter, processes, or combinations of both. Separate claims can be made on new uses of a therapeutic protein or any of its variants even if the parent gene is already patented.

The regulations issued by the FDA in 1996 placed greater emphasis on characterization techniques for biologicals. The BLA currently requires detailed analytical description of the protein product, which includes biopotency and immunogenicity testing.

We review the major clinical applications for engineered proteins and we assess the companies which are currently making most progress in targeting these markets, including an examination of patent metrics. We also briefly cover the leading geographical markets.

The protein engineering market in 2006 was worth about $67 billion, 10% of total pharma sales, and will rise to a forecast $118 billion, or 12% of pharma sales, in 2011. Monoclonal antibodies currently comprise 27%, but will rise to 36% in 2011 because their sales are increasing more rapidly (14.9% vs 7.5% annually)

Oncology is the dominant application for both mAbs and other types of engineered protein, accounting for one-third of sales overall (and over 50% of all mAbs). In value terms, engineered proteins now account for about two-thirds of the worldwide oncology market ($23 billion of $34 billion).

CNS represents the most rapidly growing mAb segment, although sales are small in absolute terms.

Amgen and Roche have the biggest manufacturer market shares overall, accounting for 40% of total sales.

The leading products are Amgen's Aranesp engineered erythropoietin (EPO) and Roche's Rituxan anticancer mAb, with 14% of the entire market between them.

North America and Europe are the largest geographical markets, with shares of 44% and 28% respectively, while Latin America (and, in particular, Brazil) is the most rapidly growing.

Genentech, owned by Roche, has by far the most protein engineering- related US patents assigned to it (192, 7.4% of the total), followed by Human Genome Sciences (75 patents, 2.9%).

Genentech is also the most frequently cited assignee, although over half its patents have never been referenced by subsequent US patent applicants. In terms of references per cited patent, a number of other companies such as Protein Design Labs (now PDL BioPharma) and Bristol-Myers Squibb do as well or better.
Chapter 2 Strategies and technologies for protein engineering

Both genetic and chemical techniques can be utilized to engineer change in the structure and function of proteins and peptides.

Mammalian cell culture, which is technically complex and expensive, is used for the production of recombinant glycoproteins. Site-directed mutagenesis can be used to optimise proteins and introduce unnatural amino acids.

Post-translational modifications (PTMs) of a protein affect its structure and function. Glycosylation represents one of the most common modifications. Other PTMs, including carboxylation and hydroxylation, are characteristic of some therapeutic products.

Glyco-engineering of proteins employs bioengineered bacteria, yeast, plant, insect, and mammalian cells with altered glycosylation pathways or can be carried out outside of bioengineered cells.

Recombinant protein characterization has been transformed by the application of analytical technologies such as mass spectrometry, high performance liquid chromatography, capillary electrophoretic and biological assays for potency.

Directed evolution strategies are being used to increase potency, stability, and other protein characteristics. Directed evolution entails the isolation and optimization of proteins through cycles of iterative random mutagenesis and functional screening.

Functional screening is typically accomplished through the use of display technologies. These include phage display, yeast display, ribosome and mRNA display, and CIS display. In the commonly-used phage display a protein is displayed on the surface of a phage.

Combinatorial chemistry can be used to generate novel binding molecules with optimized properties using protein frameworks with good biophysical properties as scaffolds.

Advances in chemoselective ligation, which uses unprotected peptide fragments and reactive functional groups, has enabled synthesis of medium sized proteins and their modification with polymers in a site-specific manner.

In silico protein design approaches use analysis of the 3-D structure of a therapeutic protein to guide the selection of appropriate amino acid sequences to create desired properties.

Patents claiming recombinant proteins form the most numerous category of US protein engineering patents filed between 1992-2006. With respect to clinical applications claimed, patents citing cancer form the most numerous category. The top most cited US patent claims humanized immunoglobulins.
Chapter 3 Engineering improved monoclonal antibodies

The FDA defines a monoclonal antibody (mAb) as a clonal product which may be intact antibody, antibody fragment, conjugate, fusion protein, or bispecific antibody. mAbs form the majority of therapeutic proteins in development and optimal engineering of new mAbs is coming into play as a competitive factor.

The therapeutic effect of mAbs is obtained either by blocking a target, or by exerting effector functions residing in the Fc region to activate cytotoxic cells or the complement system.

Chimeric, humanized and human mAbs are under development. In chimeric mAbs, the murine Fc region has been replaced with one of human sequence. Humanized mAbs have been created through CDR grafting or variable domain resurfacing. The development of phage display technology, followed by transgenic mice, enabled the development of fully human mAbs.

All approved recombinant mAbs have been produced in mammalian cell cultures. Several companies are developing non-mammalian production systems utilizing yeast, insect and other cell types, as well as plants. The aim is to produce glycoforms with predictable or improved effector functions.

It is possible to extend the plasma half-life of mAbs by engineering Fc regions to alter the binding affinity for the FcRn receptor. Antibody fragments with a wider range of half-lives can also be generated through PEGylation.

Human MAbs are being derived from recombinant antibody libraries, which can be based on immune or nonimmune fragments. Many mutagenesis and selection strategies have been used to provide subnanomolar antibody affinities. Selection strategies can also be applied to improve other properties, such as enhanced stability, resistance to proteases, or aggregation behavior.

The binding affinity of human antibodies generated from transgenic mice is often high, reflecting in vivo affinity maturation. In Abgenix's XenoMouse and Medarex's HuMab-Mouse, murine germ-line antibody genes have been replaced with human ones.

For many applications, the Fc-mediated effector functions are not required. Directed evolution selection strategies have been used to generate a wide variety of antibody fragments. Two popular monovalent fragment formats are Fab and scFv. Monovalent fragments are being engineered into multivalent molecules and bispecific fragments.

The three major areas of therapeutic application are: cancer; diseases of the immune system; and infectious diseases. In cancer, engineering strategies vary depending on whether activation of effector functions or inhibition of signaling is the desired outcome.

In contrast to unconjugated mAbs, a shorter blood residence time is frequently desirable for mAbs conjugated with toxic compounds, and can be achieved either by appropriate engineering strategies.
Chapter 4 Engineering alternatives to antibodies

In addition to antibodies, many other natural proteins mediate specific high-affinity interactions. Engineered antibodies are facing increasing competition from alternative binding proteins derived from combinatorial scaffold libraries. They are mainly small single-chain proteins.

Several non-immunoglobulin binding proteins are under commercial investigation; most are still in an early phase of development. Developers of alternative binding proteins avoid the royalty stacking that burdens monoclonal antibodies today. However, not many alternative scaffolds have been shown to bind peptides yet.

Many alternative binding proteins are stable without disulfide bonds, enabling cheap and efficient production in the reducing cytoplasm of bacteria. Unlike whole IgG-based MAbs, alternative binding proteins do not have effector functions, although engineering efforts are underway to confer such functions on them.

A wide range of different non-immunoglobulin scaffolds with diverse origins and characteristics are currently used for combinatorial library display. Beta-sheets are the dominating secondary structural elements in scaffolds used for library construction.

Alternative scaffolds need to be engineered for library generation, and scaffold optimization to improve chemical robustness, thermodynamic stability or recombinant expression yields is typically introduced early on in the process.

The choice of a selection technology best suited for a given binding protein library depends on different parameters, such as the library diversity, the properties of the scaffold and the intended applications.

In general, very high affinity is desirable for in vivo applications of alternative binding proteins. This can be achieved by a high intrinsic monovalent affinity of each binding site and/or by oligomerization.
Chapter 5 Engineering other protein and peptide drugs

Over 140 therapeutic proteins and peptides have been approved by the FDA and over 1,000 candidate products are under investigation.

While the first generation of protein and peptide therapeutics entered development with little or no optimisation, recent approvals include products which have been re-engineered to have either an altered amino acid sequence or post-translational modification (in particular altered carbohydrate content).

A variety of approaches are under investigation to improve cell production systems in order to allow the expression of effective amounts of soluble, functional proteins.

Transgenic milk production offers a cost-effective system for the manufacturing of complex therapeutic proteins. Several companies have generated transgenic animal herds that yield large amounts of candidate therapeutic proteins.

There is growing interest in modifying protein and peptide drugs to extend their plasma half-lives. Recent developments concerning erythropoiesis-stimulating agents are discussed; they provide a striking example of how the market need for longer-lasting therapeutic proteins is driving the development of novel engineered products.

In addition to proteins, short agonist peptides have been utilized extensively as drugs in medicine. Novel peptides can be naturally derived or chemically synthesized, with the latter method currently more prevalent. Progress has also been made in the synthesis of peptide inhibitors.

By altering their amino acid sequence, peptides are readily permutated into large libraries that lend themselves to rapid combinatorial analysis. Phage display derived peptides are often completely novel.

Peptides can be designed de novo, but most peptides of biological interest are derived from sequences of native proteins. Peptide analogs with lower susceptibility to proteolysis have considerable potential as therapeutics. The current focus is on optimizing and expanding analog traits with chemistry.

Selection of an optimal synthesis strategy depends in large part on the characteristics of the peptide being produced. Re-engineering may involve using computational methods to identify and replace antigenic epitopes.

In recent years, levels of microbial resistance to antibiotics have risen dramatically injecting a strong impetus into development of new antimicrobial peptides which are surveyed here. They reflect different approaches to peptide drug engineering.

Pharmacogenomic studies of protein drugs generally focus on pharmacodynamics. In new protein drug development, an understanding of the natural variation in both the drug and its target may guide engineering efforts aimed at the optimization of the lead molecule.
Chapter 6 Engineering protein therapeutics for delivery

Until recently, all approved biopharmaceuticals required to enter the bloodstream were administered parenterally, which is associated with a number of disadvantages, including reduced patient compliance and potential complications.

The majority of sales currently originate from protein therapeutics that are immediate release formulations, with subcutaneous injection the most popular route of delivery, followed by intravenous infusion.

Drug delivery can offer an injectable protein product a competitive edge in a number of ways. Technologies that result in a more stable plasma drug concentration can improve the efficacy and safety profile of the drug. The need to administer products less frequently can improve patient compliance and reduce costs.

Persistence in circulation is amenable to engineering by a number of approaches, including half-life extension technologies and depot technologies.

A prominent half-life extension technology involves the covalent attachment of polyethylene glycol (PEG) to the compound of interest. Enzon licensed this technology to Nektar Therapeutics and several refinements and proprietary approaches have recently been developed.

For glycoproteins, post-translational carbohydrate remodeling has emerged as a strategy for increasing in vivo duration of action.

Other half-life extension technologies include conjugation of polysialic acid, approaches exploiting human serum albumin's long circulation half- life, and (for peptides) introduction of non-natural D-amino acids or peptide cyclization.

Several companies have focused on reducing the frequency of injection through the use of depot systems. Drug carriers include biodegradable microparticles, natural and synthetic polymers, microcapsules and liposomes.

In the last three years advances in nonparenteral systemic drug delivery culminated in first approvals of products for pulmonary and nasal delivery.

Recent advances, particularly particle engineering and formulation methods to control particle size, consistency, and stability, are expanding the applications for pulmonary delivery to proteins and peptides.

Nasal delivery is attractive because of its convenience and the large surface area for absorption generated by the nasal microvilli. However, nasal delivery generally results in low bioavailability.

The next frontier is oral delivery of therapeutic peptides and proteins, but this has always presented a significant challenge. Several proprietary technologies are under investigation.
Chapter 7 Trends and opportunities

Although it is accepted in the field that the risk of immunogenicity may be reduced by using fully human antibodies, mabs generated by combinatorial approaches may still be immunogenic and strategies need to be developed to eliminate antigenic epitopes.

Given the increasing competition for therapeutic antibodies to the same target, new monoclonal antibody products with enhanced functionality, specificity and/or effector functions would gain a competitive edge.

Antibody formats face increasing competition from alternative binding proteins, although few have been shown to bind peptides as yet. Developments are awaited in screening nonimmunoglobulin libraries and increasing the binding of alternative binding proteins to peptides.

The advent of follow-on biologics (FOBs) highlights the significance of engineering improvements to proteins which would allow products to be differentiated on the basis of superior characteristics. Developers of FOBs will need to employ sophisticated analytical techniques to characterize their products.

The future looks promising for synthetic peptides, and eventually synthetic proteins, since synthesis of medium-sized proteins is already possible. For patented proteins, substatial re-engineering of the protein during chemical synthesis may allow the product to be commercialized without risking patent infringement.

The recent advances in the understanding of the process by which proteins fold and design of artificial proteins may permit de novo synthesis of therapeutic proteins. Novel protein molecules may have therapeutic potentials that are absent in natural proteins.