Academy eBriefings

Follow-on Biologics Workshop

Scientific Issues in Assessing the Similarity of Follow-on Protein Products

Organizers: Kurt A. Brorson (FDA Center for Biologics Evaluation and Research), David Bunk (National Institute of Standards and Technology), Barry W. Cherney (FDA Center for Drugs Evaluation and Research), Curtis Meuse (National Institute of Standards and Technology), and Emily Shacter (FDA Center for Biologics Evaluation and Research)Presented by the New York Academy of Sciences and the U.S. Food and Drug Administration, in collaboration with the National Institute of Standards and Technology (NIST)

Reported by Angelo DePalma | Posted May 3, 2006

Overview

The New York Academy of Sciences, in conjunction with the FDA and the National Institute for Standards and Technology (NIST), held a workshop on December 12–14, 2005 that focused on the scientific issues involved in assessing the similarity of follow-on protein products. At the conference, scientists from academe, industry, and government surveyed the scientific issues surrounding protein structure and function, focusing on specific analytical methodologies for characterizing protein products.

The first half of the conference focused on the challenges posed by the complexities of protein structure. Day one covered primary, secondary, and tertiary structure, and analytical methods for characterizing those structures. On the first half of day two, speakers discussed quaternary structure, protein-protein interactions, and protein aggregates. Speakers from FDA and NIST introduced these sessions, laying out the scientific and regulatory challenges presented by each protein structure element. During the second half of day two, several speakers discussed the impact of processing on protein properties.

Office of Pharmaceutical Science Perspectives

Speaker:
Keith Webber, U.S. Food and Drug Administration, Center for Drug Evaluation and Research

Highlights

Guidelines for follow-on biologics are evolving based on regulatory experience and the best current analytical and manufacturing science.
Developers of follow-on products will employ a range of analytical methods to analyze proteins' primary and higher-order structures.
Comparisons between follow-on products and originator molecules is complicated by the diversity of protein structure, including higher-order folding and associations as well as post-translational modifications.

"We have a mission..."

Keith Webber (FDA, CDER, Office of Pharmaceutical Science) opened the program with the current regulatory perspective, including FDA's expectations for analytical methods used to bring FOBs to market. "We have a mission," Webber stated, "not just the agency, but the industry and academic scientists working on analytical methods related to protein pharmaceuticals, to assure the safety, efficacy, and availability of biopharmaceuticals."

Generally, FDA's expectations for FOBs would be no different than for any other drug: assurance on the part of developers that their products are safe, effective, and of high quality. This high quality applies not just to source materials and final product, but to a rigorous standard for Good Manufacturing Practices, related analytical methods, and process controls.

Under current FDA guidelines, manufacturers who alter a manufacturing process post-approval must determine and document how production changes affect the properties of their product, especially with respect to safety and efficacy. It is widely accepted that potential developers of FOBs will be subject to conditions at least as stringent since the process (and perhaps the expression system as well) will almost certainly differ, and quite significantly, from the originator's.

Which characteristics should be used to compare protein products, and which analytical tools provide the best means for doing so?

Among the pivotal scientific issues facing potential FOBs is the choice of characteristics through which to compare them to the original protein, and of which analytics are the best means for doing so. These ideas, although not mentioned specifically by every speaker, underscored the entire conference. Since FOBs would be drugs which are normally injected, the stakes—and therefore the standards—for approving them would be quite high. Clearly, determining similarity between potential FOBs and originator proteins involves much more rigorous characterization than for proteins used in basic research.

Webber outlined two classes of similarity. An interchangeable product is intended to be chemically, structurally, and functionally similar to a currently marketed product; for example, a generic drug or a product made by the licensed manufacturer using a new manufacturing process. Scientists sometimes refer to these as "comparable products." The important point is that these proteins need to be highly similar chemically and structurally, but not necessarily identical, to be considered interchangeable. However, when analytical methods discern differences, Webber explained, developers should be prepared to relate, through structure-function relationships or other studies, that the differences will not adversely affect safety or efficacy of the product.

Non-interchangeable products, by contrast, would be proteins with the same basic structural features, but which may have one or more intentional modifications for the purpose of improving safety, efficacy, or stability. An example of a non-interchangeable protein pair might be a native protein and the identical material modified by chemical addition of polyethylene glycol (PEG) residues. FDA has already approved a number of PEGylated protein therapeutics—e.g., α-interferon and erythropoietin—which have enjoyed market success.

If the concept of FOBs is successful, manufacturers of FOBs would ultimately face the challenge of producing marketable products. Beginning with a product from the market, they will need to reverse-engineer a process, to the extent that this is possible, to meet targets for chemical and biological similarity as well as for safety, purity, and efficacy. The difficulty of this task depends on the complexity of the original comparator protein and its formulation, the availability of appropriate analytical methods and reference standards, and the ability to establish clear, reasonable values for both absolute properties and those deemed comparable to original molecules. Later presentations showed, however, that several factors confound straightforward comparisons.

Uncertainty plays a role in all drug development, and FOBs are no exception.

Uncertainty plays a role in all drug development, and FOBs would be no exception. Manufacturers often possess an incomplete understanding of the impact of process on product, an area of risk best minimized by maintaining strict control over process-related events. "We always have uncertainty because there is never time to get a complete data set," Webber observed. In traditional pharmaceutical development, even clinical efficacy is expressed in terms of statistics, with the efficacy "goal post" placed arbitrarily at an uncertainty level of 0.05. Moreover, because drug companies don't have an unlimited amount of time or resources, safety data is almost never accumulated for a perfectly representative patient population. "But we don't stop approval for these reasons," Webber said, "we deal with them from a risk-benefit perspective."

Webber is referring to FDA's recent risk-based approach initiative, which allows drug developers (in this case, potential developers of FOBs) to deal with uncertainty by identifying its sources, projecting possible scenarios and hazards resulting from incomplete understanding, and seeking to reduce the risk-benefit ratio by providing additional or supportive information.

Webber stressed the importance of following a path to the development, manufacture, and approval of FOBs that is rooted in science. Quoting Adam Smith, he noted that "Science is the great antidote to the poison of enthusiasm and superstition." He concluded by urging all stakeholders in FOBs to maintain a scientific, objective view of reality. "We need to work together ... to maintain an even keel as we move forward on this path."

Analytical Techniques to Examine Molecular Heterogeneity: Primary Structure

Highlights

Protein structure, in all its manifestations, provides numerous opportunities for comparing follow-on biologicals with the original product.
Post-translational modifications to protein backbones operate combinatorially to create hundreds or thousands of variants of the same basic protein.
Techniques for characterizing protein glycosylation are evolving to the point that, with extensive work, protein sequence, post-translational modifications, and glycoform distribution and structure can largely be defined, although the latter still represents a difficult analytical challenge.

Multiple structural domains

Following Webber's introduction, the first group of speakers discussed analytical techniques for quantifying heterogeneity of protein drug products. In biochemistry, heterogeneity is a measure of dissimilarity and variants, referring to the numbers and types of different protein forms present in a given preparation. All protein preparations are heterogeneous to some degree.

Differences in any structural domain increase biochemical heterogeneity, and can affect a protein's safety and efficacy.

David Bunk from the National Institute of Standards and Technology (NIST), one of the conference organizers, provided a brief overview of the elements of protein structure, including primary structure (amino acid sequence plus post-translational modifications), secondary structure (three-dimensional attributes such as α-helices, β sheets, loops, and turns), tertiary structure (subunit 3D structure) and quaternary structure (spatial orientation and arrangement of protein subunits, and protein-protein interactions). Differences in any structural domain increase biochemical heterogeneity. From a regulatory and scientific perspective it should be presumed—in the absence of contrary evidence—that these dissimilarities can affect a protein's safety and efficacy as well.

Bunk spent some time outlining the significance of post-translational modifications—chemical add-ons to a protein's primary structure that are unique for each expression system and which affect proteins' chemical, biological, and therapeutic properties. The most common post-translational modifications are glycosylation (addition of sugar-containing molecules), acetylation (acetyl groups), methylation (methyl groups), and phosphorylation (phosphate groups). In their natural environment, organisms use these modifications to alter the properties of proteins. In a biomanufacturing environment, cultured cells carry out post-translational modifications according to their own requirements. Potential developers of follow-on biologics must be aware of these modifications, and dozens of others that might affect protein properties.

Following Bunk's introduction, four speakers demonstrated the capabilities of mass spectrometry (MS) for elucidating protein structure, especially with respect to post-translational modifications. MS is an analytical technique that determines a molecule's molecular weight and chemical composition. Many types of MS analysis use fragmentation to produce smaller species, which provide information on localized chemical structure, such as that of post-translational modifications. Because it is highly sensitive to even small mass changes, MS can help scientists determine post-translational modification patterns.

One protein, many forms

The sheer number and combinatorial possibilities for post-translational changes challenges any analytical effort. Donald F. Hunt (University of Virginia) showed, using MS, that the N-terminus of histone H3 alone possessed a "horrendous number of post-translational modifications," including acetylations, phosphorylations, and mono-, di-, or trimethylation on arginine and lysine. "If you add up all these modifications, this one protein would exist in more than 50,000 forms," Hunt said. Nonetheless, through a systematic series of analyses using different fragmentation techniques and state-of-the-art tandem MS, it is possible to define the entire primary structure of all of the isoforms within this difficult class of proteins.

Hunt explained the histone code hypothesis, which states that specific N-terminus post-translational modifications on histones, proteins found in the nucleus, can lead to activation or deactivation of protein production in the cell. Specifically, acetylations activate transcription of DNA to RNA, while methylation deactivates this process. He went on to demonstrate the ability of tandem MS to pick out these rather subtle changes in protein structure.

Despite the huge combinatorial possibilities in post-translational modification, it is more likely that, for a given protein, a much smaller number of chemical modifications would predominate. The challenge for analytical biochemists is to purify proteins exhibiting these changes prior to MS analysis. William Hancock (Northeastern University) described how chromatography, either alone or in tandem with subsequent MS analysis, might resolve these protein variants by first purifying protein components and then analyzing them for mass. Hancock discussed several types of chromatography—the workhorse purification method in biochemistry—and the specific chemical modifications they work best with. Chromatography is a purification method through which molecules containing various chemistries bind to specialized resins as they pass through them.

Jonathan Amster (University of Georgia) continued on the theme of mass spectrometry with his presentation on fourier-transform MS (FTMS), which boasts high resolving power (>1,000,000 at m/z 1000), part-per-million mass accuracy, and compatibility with many popular ionization and fragmentation methods. Fourier-transform (FT) techniques apply advanced mathematical methods to obtain more information from data than is normally available through straightforward analysis.

Until relatively recently, FTMS was thought to be too complex to be used in an industrial setting. Its computational and instrumentation requirements were not nimble enough for a manufacturing facility. However, advances in hardware and software have pushed FT methods into mainstream analytical applications. Notable advances include superconducting magnet technology (up to 14 tesla), external ion source technology for coupling high pressure sources (in particular electrospray ionization coupled to FTMS), hybrid quadrupole MS/FTMS, and ion fragmentation techniques (e.g. electron capture dissociation, infrared multiphoton dissociation, and collisional dissociation). Moreover, ion dissociation methods can be combined synergistically to provide important structural details of follow-on protein products.

Vernon Reinhold (University of New Hampshire) completed the main group of MS presentations with a talk about MS techniques for resolving protein heterogeneity due to glycosylation, which has thwarted countless protein analysis projects over the years. "Since more than half of all proteins are glycosylated, a strategy for structural determination of carbohydrates is seen as a next step in understanding genomic function," he said.

A comprehensive strategy for sequencing glycan adducts to proteins has been elusive.

Reinhold called for "an improved accounting of structure" for glycan adducts based on appreciation of the specific functions of carbohydrate epitopes, but admitted that a comprehensive strategy for carbohydrate sequencing has been elusive. More than a century after Hermann Emil Fischer received the Nobel Prize for his work on sugars, protein researchers still lack a comprehensive sequencing strategy for characterizing these organic polymers.

According to Reinhold little attention has been given to carbohydrates due to the recent excitement over proteins and genes. Even before the Human Genome Project few chemists worked with carbohydrates due to difficulties in purification, intractable chemistry, and nearly endless combinations of forms (branching patterns, connectivity to each other and to proteins and genes). A single gene mutation in a protein glycosylation pathway, however, can profoundly affect the eventual pattern by which oligosaccharides add to proteins. The mutation is amplified to all glycosylated proteins an organism creates. "Selective strategies to assign components of carbohydrate structure show little focus toward congruency," said Reinhold, who then presented his vision for achieving this.

His approach is based on multi-stage disassembly, in an ion trap mass spectrometer, of methylated polysaccharides. In the highly energetic MS environment, the complex carbohydrate chains can be fragmented systematically and predictably. These fragments are then analyzed for mass and number through multidimensional MS. Identification of glycoforms removed from the protein is assisted by input from an ion fragment library for sugars, associated high throughput analysis tools, and a mathematical formula that correlates where the sugar-containing molecules were located from their MS fragmentation patterns. Not surprisingly, Reinhold showed that this approach, combining multidimensional MS with sophisticated data analysis tools, could unravel the structures of the multiple glycoforms from a single glycoprotein species.

Secondary and Tertiary Structure

Highlights

Every condition and environment experienced by protein therapeutics may affect higher-order (secondary, tertiary, quaternary) structure.
The impact of higher-order structural changes on therapeutic efficacy and safety is unpredictable and difficult to measure.
It is likely that analytical strategies for determining secondary and tertiary structure will employ multiple, orthogonal methods.

Additional layers of structure differentiation

Primary structure, including post-translational modifications, represents only one level of complexity when comparing two proteins thought to be similar. Where the recombinant gene, expression system and fermentation conditions are the principal determinants of primary structure (including post-translational modifications), every operation and condition experienced during manufacture of therapeutic proteins can potentially affect higher order protein structure.

Russ Middaugh, who opened up the session on secondary and tertiary protein structure, posited a theme that recurred repeatedly throughout this conference: The use of a single, low resolution technique such as circular dichroism, fluorescence, fourier-transform infrared, and Raman spectroscopies, light scattering, and calorimetry is not sufficient to demonstrate "congruence of structure" between two proteins.

Empirical phase diagram of recombinant protein antigen. Representing the relationship between individual pH vs. temperature experiments through color allows a comparison between two protein preparations.

Recently, multidimensional data analysis has emerged to provide a partial solution to this problem. Middaugh uses the term empirical phase diagram to describe his adaptation of multidimensional methods. An empirical phase diagram differs from classic thermodynamic phase diagrams in that reversibility across the phase boundaries is essential for the latter but absent in the former. Instead of depicting thermodynamic phase behavior, then, empirical phase diagrams provide visualizations of protein properties using a combination of analysis techniques (e.g. ultraviolet, infrared, and fluorescence spectroscopy) under multiple conditions for each technique (e.g., ultraviolet spectra acquired at varying pH or temperature).

To construct an empirical phase diagram Middaugh represents a protein as a vector in a very highly dimensioned space. The components of the vector are the individual spectral values based on thousands or tens of thousands of measurements. Next, each vector is represented as a visual representation so that two proteins can be compared, side by side, without referring to the raw data.

An empirical phase diagram produces a more detailed picture of a protein than one gleaned from individual analytical methods.

"Even a simple technique like absorption spectroscopy can have its power dramatically enhanced by these approaches," Middaugh said. For example, the six or seven derivative peaks of a protein's near-ultraviolet absorption spectrum can be used to generate descriptive vectors, and to monitor simultaneously the environments of the widely dispersed tryptophan, tyrosine, and phenylalanine residues in a protein. When these vectors are created as a function of appropriate solution variables such as temperature, pH, ionic strength, redox potential, agitation, freeze/thaw stress, etc., a detailed picture of the protein emerges which is greater than the information gleaned from the analytical methods individually, or by their simple sum.

Middaugh noted that the key to advancing the empirical phase diagram approach to follow-on biologics is to determine how much information (in the formal sense of information theory) is required to define the structure of a biological macromolecule. Only then, he says, will developers of potential follow-on products be certain of a protein's identity within a defined degree of uncertainty.

NMR: no longer only for small molecules

Organic chemists have used nuclear magnetic resonance (NMR) spectrometry for three decades to measure simple connectivity in organic molecules. For many years, the technique lacked the resolving power to tackle large complex molecules, and distinguishing protein structures by NMR was considered impossible. Today, steady improvement in instrumentation coupled with exponential improvements in computing power have made protein NMR possible, although it remains in the expert realm.

Daron Freedberg (FDA Center for Biologics Evaluation and Research) provided a tutorial on the capabilities and applicability of NMR for characterizing follow-on biologics.

NMR belongs to the toolbox of quantitative spectroscopic methods of protein characterization that include MS, circular dichroism, Raman, and infrared spectroscopy. Like these techniques NMR is nondestructive and applicable to a wide range of macromolecules and ingredients, but provides the additional benefit of atom-level resolution. NMR's principal drawback is the size limitation of generally about 30 kDa for proteins, a number that is constantly being revised upwards as methods improve.

"Multidimensional" NMR of proteins provides information on connectivity and chemical environment that is otherwise inaccessible.

In a typical NMR analysis, atoms within a chemical compound (normally hydrogen, carbon, nitrogen, and phosphorus) are elevated to an excited energetic state. As the atoms relax to the ground state the energy released is plotted as a spectrum in parts per million units. These energy values correspond to the chemical environments and connectivity of each atom.

Because proteins are so large and contain so many atoms, their NMR spectra are too complex to parse in a straightforward manner. Researchers therefore apply multidimensional techniques through which the energy released by one atom during relaxation causes energetic changes in neighboring atoms as a function of the distance between them. So-called "multidimensional" NMR of proteins takes days, sometimes weeks, for a single protein but provides information on connectivity and chemical environment that is otherwise inaccessible.

Optical methods: tried and true

Keith Oberg (MannKind Corp.), the only industrial speaker on the program, made the case for optical spectroscopy, which can help demonstrate equivalence between two proteins. Optical methods are extremely sensitive to minor structural changes. When differences in optical spectra are observed during manufacture, they can point to process changes that will produce the right material. Similarly, optical spectroscopy can be used in a quality setting, after manufacturing, for characterization purposes.

The most commonly used optical methods are ultraviolet and fluorescence spectroscopy, circular dichroism, and infrared spectroscopy. Optical spectroscopy's advantages include high sensitivity, simplicity, low cost, amenability to generic methods, and applicability throughout the biopharmaceutical development and manufacturing process. On the minus side, optical techniques provide relatively low resolution (compared, say, to NMR), present a learning curve, and cannot be relied upon as stand-alone characterization methods.

Despite its limitations, optical spectroscopy provides insights into higher-order structural status for proteins, and may provide insight into problems early in development and even during manufacturing. Through difference optical spectroscopy, two products or formulations may be compared for subtle conformation and stability differences.

Curtis Meuse (NIST) continued coverage of optical spectroscopy with a discussion of tertiary structure analysis, emphasizing solution structure. Tertiary structure is significant for several reasons. Since proteins utilize their three-dimensional shape for molecular recognition, a protein's primary amino acid sequence and atomic composition alone are insufficient for demonstrating similarity to other proteins, not to mention safety and efficacy. Moreover, since the information required to achieve higher-order structures is contained in the amino acid sequence, from an analytical perspective the tertiary structure implies the correct primary and secondary structures.

Optical methods for protein folding will need to be considered for comparing follow-on biologicals because competing methods are also limited.

Most analytical methods, including physical methods like chromatography and calorimetry, provide some insight into tertiary structure. But making structural sense of spectra and chromatographs, alone or in combination, challenges even computer-aided analysts. Sampling limitations, the multiplicity of tertiary properties, the need to detect relatively small changes due to amino acid substitutions, post-translational modifications, interference from excipients, lack of methods harmonization between laboratories, and difficulty quantifying similarities and differences all contribute to the confusion. Despite these limitations, optical methods for studying protein folding have long been used to shed light on the structural effects of site directed mutations. Optical methods for protein folding will need to be considered for comparisons of follow-on biologicals because competing methods are also limited.

To some degree orthogonal analysis methods can mitigate the limitations of optical spectroscopy, as Frederick Schwarz (NIST) demonstrated in his discussion of calorimetry. This technique, which measures minuscule quantities of heat absorbed or released as a molecule undergoes stress-induced changes, readily detects subtle differences in proteins. Schwarz demonstrated, for example, how calorimetry might determine differences in protein binding, denaturation, and aggregation.

Quantitative structure-property relationships

The a priori prediction of protein affinity and behavior during preparative chromatography has been a longstanding goal for protein manufacturers. Steve Cramer (Rensselaer Polytechnic Institute) discussed quantitative structure-property relationship (QSPR) models for predicting protein affinity in hydrophobic interaction chromatography (HIC) systems. QSPRs are mathematical relationships that relate chemical or molecular properties to an anticipated property or behavior. Cramer also covered QSPRs for elucidating the underlying mechanisms of chromatographic behavior, and presented a method for predicting this important characteristic directly from protein sequence and crystal structure data.

Cramer constructs QSPRs by first obtaining experimental data that will be used as the dependent variable (e.g., retention data, isotherm parameters, etc.), calculating a large number of molecular property descriptors for each protein in the experimental data set, and selecting features that correlate best with the experimental response. After evaluating the model on a training set of molecules, he tests it on new molecule sets.

Protein-Protein Interactions: Quaternary Structure

Highlights

Aggregation is highly undesirable in therapeutic proteins in part because aggregates tend to elicit powerful, sometimes dangerous immune responses.
A variety of precise methods exist for measuring aggregation, among them light scattering, ultracentrifugation, field-flow fractionation, and atomic force microscopy.
Clinically, the immune response to aggregates is triggered by molecular components and can be enhanced with non-protein substances (adjuvants), interacting with the individual's unique immune response.

Aggegation = aggravation

Tuesday's session on protein-protein interactions focused for the most part on protein aggregation and quaternary structure. Amy Rosenberg (DTP, CDER, FDA) presented a useful overview of these interactions, offering valuable background on aggregation and its impact on therapeutic protein safety.

Protein aggregates, which can be soluble or insoluble, have far-ranging effects on protein products, from manufacturability, to product potency, to safety. Approaching this topic from the perspective of an immunologist, Rosenberg concentrated her comments on safety, specifically the immunogenicity of undesirable protein aggregates.

Higher-order protein aggregates provoke immune responses that profoundly affect therapeutic protein quality.

The enhanced ability of protein higher-order aggregates to provoke an immune response profoundly affects therapeutic protein quality. As experience with human growth hormone, interleukin-2, interferon-α, and various plasma proteins has shown, the impact can be predictable in inducing robust immune responses. Principal mechanisms of immunogenicity include enhanced antigen presenting activity (which may be specific or nonspecific) and B-cell activation, possibly independent of T cell help.

According to Rosenberg, the human immune system is primed for response to invasive threats; protein aggregates can resemble some of these through their macromolecular structure. This is because "the threat signature of microbes and bacteria is defined by the patterns and structures of their proteins, polysaccharides, and lipids," Rosenberg said. The extent to which protein aggregates resemble bacteria and viruses—through high molecular weight and display of repetitive and conformational epitopes—determines how robustly aggregates will activate the antibody generating mechanism.

The threat signature of microbes is defined by patterns in their surface proteins, polysaccharides, and lipids.

It would be a mistake to view aggregates in isolation, however. Many product- and host-related factors are essential for triggering an immune response. Among the most prominent are other product impurities, whether the aggregate is derived from a "foreign" protein or formed from an endogenous protein homolog, and the product's inherent immunomodulatory characteristics. Dose, frequency of dose, immune competence, route of administration and, for therapeutic counterparts of endogenous proteins, the extent of underlying immune tolerance to the endogenous protein, are among the patient-related factors determining immunogenicity.

Most protein therapeutic companies use a single assay, based on size exclusion chromatography, to analyze for aggregate type and concentration. Size exclusion chromatography analyzes molecules according to size. Dimers, trimers, etc. separate from the original proteins because their masses are multiples of the monomer molecular weight. For innovator products, the aggregate profile is related to patient outcomes through clinical trials so developers know the implications of a particular aggregate profile. Follow-on comparability analyses will demand some sort of relative quantitation of aggregates as well as qualitative analyses in the follow-on product and the original.

Almost any departure from the original manufacturing operations can affect aggregation profile.

Almost any departure from the original manufacturing operations can affect aggregation profile, including changes in purification steps, virus removal methods, source materials, or cell lines. Even switching from multi-dose vials to single-dose syringes can change the quantity and types of aggregates. Rosenberg cited examples of syringe products that leached metals that caused aggregation of several protein therapeutic products or leaching of vulcanizing agents associated with unanticipated immune activity.

"The impact of differences in aggregates and other product qualities on clinical behavior has to be understood clinically," Rosenberg said, "so we often require clinical studies for high-risk manufacturing changes."

Rosenberg concluded by presenting a wish list for the ideal analytic for aggregation. Such a method would improve currently available technology (especially for formulated product), capture the full range of potential aggregates including species with very high molecular weight, discriminate reversible from irreversible aggregates, and do so in automated, quantitative, high-throughput fashion.

John Carpenter (University of Colorado) described protein aggregation as "ubiquitous" in the production, shipping, storage, and delivery of therapeutic proteins. Potential manufacturers would need to be especially aware of aggregation resulting from heterogeneous nucleation of protein aggregates by foreign particles. During fill/finish operations, the container, closure, and filling pump are potential sources of aggregation as well. Carpenter noted that aggregate formation can not only be induced by stress vectors, but also occurs in the absence of obvious stress, when the protein's native state is favored. Conversely, maximizing thermodynamic and colloidal stabilities can reduce aggregation.

According to Carpenter, aggregates in therapeutic proteins can cause adverse effects in patients, ranging from immune response to anaphylactic shock, even if only small fraction of product is aggregated. Furthermore, the impact of aggregation on patient health is difficult to predict.

Quantifying aggregation

Karin Caldwell (Uppsala University, Sweden) discussed field-flow fractionation (FFF), a family of separation techniques that sort molecules and particles primarily according to size (flow FFF) or mass (sedimentation FFF). According to Caldwell FFF may be used as an alternative to analytical chromatography.

Field-flow fractionation can replace chromatography for some types of protein analysis.

The separation takes place in an open, unobstructed channel where the sample components are transported from injection site to exit by the laminar mobile phase flow. FFF separations depend not on the choice of mobile phase, but rather on the magnitude of a perpendicularly applied field—for example, of flow or sedimentation type—selected to accomplish a desired resolution. "The operator is therefore free to perform the separation in a buffer of choice," Caldwell noted. The transport takes place under low shear, and sample contact with channel walls is minimal. Both these effects promote high sample recovery with minimal denaturation.

FFF's size resolution dynamic range spans four orders of magnitude. Complex mixtures of protein aggregates are readily separable and identified as monomer, dimer, trimer, tetramer, etc. Moreover, combining FFF with light scattering methods provides real-time evidence for mass/size selectivity, which is especially suited for aggregate detection.

In her presentation as well as in the subsequent panel discussion, Ewa Folta-Stogniew (Yale University) made a strong case for light scattering to assess protein aggregation. Folta-Stogniew described two types of scattering experiments: static and dynamic, each with its own strengths and weaknesses. Static light scattering measures aggregates' weight-average molar mass but the technique is limited to certain solvents, requiring adaptation with aggregates that absorb laser light (vs. simply scattering it). Dynamic light scattering determines hydrodynamic radius, a measure of size which is affected by the particle's shape, but the technique cannot discriminate between shape effects and changes in oligomeric states. Thus, a non-aggregated protein may give the same reading as a different molecule which is aggregated.

In light scattering experiments, the scatter angle depends on the size and shape of the analyte.

Among the most novel suggestions for protein analysis methods was Roger Marchant's (Case Western Reserve University) atomic force microscopy technique. Normally used for materials analysis, atomic force microscopy images proteins in three dimensions, measures interatomic adhesion forces, and quantifies surface-protein and protein-cell dynamics.

Concluding the session on quaternary structure, Thomas Laue (University of New Hampshire) and Igor Kaltashov (University of Massachusetts) discussed ultracentrifugation and mass spectrometry, respectively, for determining a protein product's aggregation status.

Analytical ultracentrifugation is a technique for characterizing sizes and shapes of molecules in solution. As a "first-principle" method, ultracentrifugation requires no reference standards since all that matters is a particle's sedimentation velocity—how rapidly it travels downward. Ultracentrifugation is particularly useful for characterizing the particle size distribution of proteins, vaccines, viruses, and aggregates.

Sedimentation velocity determines two quantities: the rate of the boundary movement, and the rate of boundary spreading. And measuring sedimentation velocity provides the highest resolution for determining particle size distribution, molecular weights, asymmetry, purity, and solution conformations of soluble proteins and aggregates. Laue claims that with sedimentation the resolution of molecular components exceeds any size-dependent chromatography method, even gel permeation. One can also use the concentration dependence of the sedimentation coefficient to assess the asymmetry of a molecule.

Data from an analytical run using ultracentrifugation can be used to develop industrial-scale preparative methods.

Ultracentrifugation is gentle, insensitive to solute/solvent type, requires no modification of the protein under investigation, operates under several complimentary instrument platforms, and has low operating costs. On the minus side, the technique only determines molecular weights to within 3% accuracy—"certainly not mass spec resolution," Laue admitted. Instrumentation is expensive, and as a "technique of accuracy" ultracentrifugation requires good and consistent operator analytical skills and a reasonable understanding of solution thermodynamics. And while there is no simple way to recover the fractionated sample after ultracentrifugation, the data from an analytical run can be used to develop industrial-scale preparative centrifugation methods.

Kaltashov mainly discussed electrospray ionization (ESI) and matrix assisted laser desorption ionization (MALDI) MS techniques for characterizing tertiary and quaternary protein structure. These two MS methods are particularly attractive because they can handle very large masses of aggregates and measures how these structures fall apart to their component proteins. ESI MS allows analysis of proteins directly, in solution. However, with non-polar solvents that are incompatible with ESI, the protein must be purified and processed before analyzing with either ESI or MALDI.

Although MS provides many advantages for direct analysis of higher order structures, each of its benefits carries a potential drawback as well. ESI's gentle ionization preserves weak, non-covalent complexes but allows formation of non-specific complexes as well; and while the technique's high sensitivity is particularly suited for analyzing dilute protein solutions, high concentrations of protein or excipient can interfere with analysis.

Effect of the Manufacturing Process on the Product

Highlights

Manufacturing profoundly affects protein structure and quality.
Small and/or micro-scale optimization experiments can assist in identifying process conditions most likely to lead to desirable product characteristics.
Choosing the most appropriate chromatography system can greatly improve isolation of the desired protein while minimizing impurities.
Chemical and protein additives, as well as pressure, may reduce the tendency of proteins to aggregate.

Roadmap to follow-on products

Charles Cooney (Massachusetts Institute of Technology) opened the Tuesday afternoon session with the first in-depth discussion of the conference on the impact of manufacturing on properties of protein biologics.

Cooney began by challenging a long-held tenet of the regulatory dogma surrounding biotechnology: "the process defines the product." Rather, he said, the process primarily affects product quality. A therapeutic protein's potency and safety can therefore be maximized through a process known as quality by design. The operative word here is design, which Cooney defined as "a thoughtful exercise, thinking forward, about how things will respond to the environment you put them in." Key to this process is a knowledge of the goals of the particular protein project—for example a research-grade protein, reference material, therapeutic protein, or FOB—and all the quality and analytic documentation associated with that designation. Quality by design entails predetermined quality standards, the analytic capability to measure progress, and an understanding of how the process affects important product properties—"not all the properties, but the ones that are important," Cooney said. "It's important that we not lose sight of that objective."

Cooney's presentation defined the needs and goals for follow-on biologicals in concrete terms. He called for stakeholders to arrive at a clear scientific foundation and roadmap for arriving at methods to characterize and demonstrate similarity of biotherapeutic proteins. Inherent in this exercise is the identification and assessment of uncertainties associated with producing biological products produced by different processes, different companies, in different locations.

A working model for introducing follow-on biological products already exists in other countries. Multiple manufacturers at different facilities around the world already produce versions of human growth hormone, interferon, chemically modified biotherapeutics, and other "biogenerics" today. Cooney recognizes that processes for follow-on products might not exactly follow this (temporary) model due to issues of intellectual property, the use of differing analytical methods, introduction of new process technology, changes in scale of operation, and incorporation of prior experience. He illustrated this point with a grid indicating the continuum of change brought about by improved process, new location, new process, and a combination of these three factors. And this set of possibilities raises a crucial question: Can dissimilar processes produce similar products?

Can dissimilar processes produce similar products?

The answer lies in the ability to measure molecular complexity as it relates to protein structure and function. Barriers to understanding this complexity result from the adequacy of analytic methods, recognizing the reactive (product-sensitive) steps in a process, and acknowledging the relationship between structure and function. Underlying these imperatives is a thoughtful evaluation and management of risks based on the inherent uncertainties of biomanufacturing. "It's ok to take risks, but you have to manage them one step at a time," Cooney said.

In the second segment of his talk, Cooney presented a strategy for process design and optimization based on microscale reactor technology. Normally biomanufacturers model processes in small bioreactors of up to 100 liters in volume, a scale which consumes a good deal of time, buffer, cells, and lab resources. The micro-bioreactor he described, manufactured by BioProcessors (Woburn, MA), features reactors of about 300 microliters in volume connected to nutrient, buffer, and reagent reservoirs through microfluidic channels. The microreactors mimic large-scale fermenters in every discernable way, Cooney said, and have very low inter-reactor variability.

Cooney concluded with a renewed call for stakeholders—industry, academia, and regulators—to agree on the critical analytic and process requirements for promoting follow-on products. He called for:

a common vocabulary between industry, regulators, and academics
analytical methods to measure protein properties both absolutely and in comparison
real-time analytics
alignment of process and manufacturing science with clinical needs
points of reference for purity, structure, and efficacy
understanding of acceptable uncertainty with respect to structure, efficacy, and analytic measurements of impurities, variants and contaminants
a strategy for managing a continuum of risk
building on converging technologies
avoiding restrictions that preclude product and process improvement.

Impact of manufacturing

Continuing in the spirit of Cooney's presentation, Erik Fernandez (University of Virginia) discussed how chromatography affects a protein product's quality through the variable separation and elution of contaminants and various forms of the product (aggregates, post-translational modifications, folding variants, etc.). Fernandez noted that while biomanufacturers document chromatography-induced variations, many of these trends are unpredictable. He called for the development of new analytical tools for analyzing the impact of chromatography on product quality and, ultimately, tools for anticipating variations.

Sarah Harcum (Clemson University) also focused on manufacturing processes in her discussion of the impact of bioreactor environment on product quality. The sheer number of expression systems available for expressing recombinant proteins today, she noted, is both a blessing and a curse. Bacterial, yeast, insect cell culture, transgenic animals and plants, and mammalian cell culture systems exist in dizzying variety. Each system generates proteins with distinctive degradation profiles, post-translational modifications, and structural features, any of which can affect protein similarity and, ultimately, safety and efficacy. Moreover, cell culture, the workhorse of protein manufacturing, offers a variety of culture types (batch, fed batch/perfusion, continuous), each expressing product carrying a unique profile.

Product quality, particularly glycosylation, is sensitive to environmental conditions such as temperature, nutrient limitations and feeding strategies, pH and ammonium concentration, and carbon dioxide concentration and osmolality. Because of this, Harcum added, "We really need to understand what's going on inside the cells a lot better before we can understand glycosylation." Since product quality changes during a production run, achieving a constant, controlled process is the secret to reproducible quality.

François Baneyx (University of Washington) concluded the manufacturing session with a presentation on protein renaturation and folding. "It is remarkable that proteins can fold within the tremendously crowded environment within the cell," he said. Baneyx described folding as a kind of isomerization or first-order reaction, which under highly stressful cellular conditions should be disfavored energetically.

High pressure can sometimes be used to discourage aggregation and coax proteins into folding.

Nevertheless, with assistance from folding modulators known as holding chaperones and folding chaperones, folding is actually favored and occurs readily in vivo. The in vitro story is quite different. Although folding produces a thermodynamically favorable state and the information required for folding is inherent in the amino acid sequence, natural chaperones are absent during manufacturing, formulation, and packaging. Moreover, high protein concentration can favor aggregation at the expense of folding.

Manufacturers must therefore inhibit aggregation and encourage folding by addition of aggregation suppressors (amino acids, cyclodextrins, detergents), folding enhancers (sugars, polyols, salts, and the amino acids glycine and alanine), protein additives that mimic the activity of natural chaperones, and chemical or protein oxidation-reduction agents to encourage disulfide bond formation. Another approach which can partially restore folding is matrix-assisted refolding, which relies on ion exchange, size exclusion, or hydrophobic interaction chromatography media. For proteins at gram-per-liter concentration or higher, high pressure can induce refolding over a period of hours or days.

Impurities and Contaminants

Highlights

Characterizing impurities, while difficult, is essential for approval of any biological product. Impurity profiling is especially important for follow-on products since impurities represent a significant element of comparability with originator proteins.
Techniques of research proteomics may play a role in streamlining analysis of follow-on biologicals.

Facing the inevitable

Many factors will determine the eventual success of the FOB concept and individual follow-on biologicals, among them the product's impurity profile. In her introduction to the Tuesday afternoon session on impurities in biological products, Kathleen Clouse (FDA, Center for Drug Evaluation and Research) outlined the causes and significance of major classes of contaminants.

Process-related impurities may be thought of as those carried through from various manufacturing operations: for example, media components, chemical additives, leachables, cellular components, and adventitious infectious agents. Product-related impurities are analogous to what a chemist might call "side products"—substances, proteins in this instance that are related to the protein product but that may or may not equal the desired product in terms of safety and efficacy.

The impact of impurities is amplified in non-recombinant follow-on biologics, which are often administered at higher doses and frequency than recombinant proteins.

"Characterizing the impurities can be one of the most difficult things to do," Clouse said, "particularly when you consider the potential role impurities would play in determining the sameness or similarity of the products." Contamination is even more troublesome, she observed, for plasma-derived products, where even trace impurities can cause adverse events. The impact of impurities tends to be amplified in these products, which are often administered at higher doses and frequency than recombinant proteins.

Potential FOB manufacturers should therefore set impurity specifications based on the known safety, toxicity, and bioactivity of the impurities in question, taking into account factors such as dose, route of administration, and the impact of the impurity on product quality and stability. Finally, potential FOB manufacturers need to institute manufacturing capability that provides adequate consistency to minimize the adverse effects of impurities and product alike.

Clouse then outlined a plan for minimizing impurities that includes determining potential impurities as early in the process as possible, determining risks for various levels of contaminants, and building validatable impurity removal into the manufacturing process.

Clouse concluded by posing questions fundamental to dealing with protein impurities. Namely, how well analytics discriminate among impurities and product variants, the impact of undetectable impurities, and how impurities contribute to product complexity and negatively affect characterization of the desired protein. While progress is ongoing towards resolving these issues, the questions remain unresolved.

Purification: less than perfect

Next, Timothy Veenstra (NCI and SAIC-Frederick) illustrated how impurities can mask the identity of therapeutically relevant proteins. In analyzing a protein implicated in interstitial cystitis, Veenstra discovered protein activity in a fraction with molecular weights below 10 kDa. However, the MS fragmentation pattern suggested that the peptide contained 200 glutamine residues, which indicated a much higher molecular weight than 10 kDa. The culprit, previously undetected folate, was the confounding factor.

An unfortunate aspect of biopharmaceutical manufacturing is that purification is less than perfect. Because of their immunogenicity, host cell proteins (HCPs) passing through filters and chromatography columns alongside the desired product require careful evaluation, according to Nadine Ritter (The Biologics Consulting Group). She described how strategies employing commercially available HCP immunoassays now augment classical, customized, single-product assays.

Like many aspects of impurity testing in biologics, assaying HCPs is an inexact science. Generally, HCPs should be minimized to the extent possible (ppm levels). However, no exact limit for HCPs can be established because of the differences in dosing and production processes. Manufacturers employing immunoassays to measure HCPs, she explained, should consequently evaluate the assays' capability to quantify HCPs of various sizes that could potentially co-purify with the product, and characterize the antibody reagent using gels and immunoblots.

Bioassays and Potency

Highlights

Bioassays are the principal means of determining preclinical activity and potency for protein therapeutics.
Comparability refers to the similarity between proteins produced by the same company. For follow-on biologicals, the operative term is similarity.
Binding assays and functional assays each play a role in protein characterization.

How to measure activity?

Steven Kozlowski (FDA, Office of Biotechnology Products) opened the final day of the conference with an overview of assays for evaluating the biological activity of therapeutic proteins.

Higher-order structure can be inferred from a molecule's biological activity, with potency as the quantitative measure of that effect.

Biomanufacturers acquire mountains of physicochemical information about products, but —despite technologies described earlier in this eBriefing—total knowledge of higher-order structure remains elusive. Nevertheless, it is possible to infer functional aspects of the higher-order structure from the molecule's biological activity, which is defined as a specified, desired biological effect, with potency as the quantitative measure of that effect. Potential FOB developers will be expected to demonstrate activity/potency through a combination of biochemical, tissue-based, animal-based, or cell-based assays, as are developers of original biotherapeutics.

In some cases it may be appropriate to conduct traditional cell binding assays as a first measure of potency. Increasingly, protein chemists utilize physical methods like surface plasmon resonance (SPR), ultracentrifugation, and calorimetry to measure binding events. These techniques provide more information than the single data point from a binding assay at a particular concentration. SPR, for example, measures multiple binding events from one molecule in a single experiment.

For the concept of FOBs to be successful, potency assays will be part of every stage of FOB development. Early on, meaningful laboratory assays indicate whether a product has the expected activity and might cause toxicity problems. In late-stage development, a potency assay needs to be validated, use a well-defined reference standard, and have specification variances that are defined and justified. "A non-validated potency assay can prevent a product approval," Kozlowski said. Finally, assays guide late-stage development and product release through in-process testing, stability programs, and by monitoring manufacturing process changes.

Kozlowski finished with a brief discussion of comparability vs. similarity for potential FOBs. Although these terms are often used interchangeably, there are subtle differences. Comparability normally refers to proteins produced by the same manufacturer by different processes, for example after substituting a 15-day fermentation for a 12-day fermentation. Biopharmaceuticals are said to be comparable when there are no significant differences in the quality, safety, and efficacy of the two products. Similarity has essentially the identical meaning, except it refers to products from two different manufacturers. FDA has issued guidances on comparability but none on similarity. Potential developers of FOBs will need to demonstrate similarity by applying sound scientific principles based on standards at least as stringent as those for comparability.

Binding vs. functional assays

Following Kozlowski's FDA perspective, the next two speakers expanded on the theme of bioassays. Laureen Little (Bioquality) discussed the special case of enzyme therapeutics, while C. Jane Robinson (NIBSC, UK) compared binding assays versus funcational bioassays.

The ICH (International Conference on Harmonization) Q6B guideline for testing biopharmaceuticals defines potency as "the measure of the biological activity using a suitably quantitative biological assay" and biological activity as "the specific ability or capacity of the product to achieve a defined biological effect." But binding assays assess product binding to receptors, antibodies, or other molecules without necessarily inducing any functional response. "Whether a binding assay can provide information on potency depends on the individual product and assays," Robinson said.

Within the context of trying to ensure that a follow-on biologic, or biosimilar, might be expected to have the same clinical effect and safety profile as the innovator product, both functional bioassays and binding assays can contribute to the comparability exercise. Binding assays can demonstrate dissimilarities in epitopes of the molecule, which even if they do not affect the potency as determined by a particular bioassay, can affect other clinically relevant properties such as immunogenic potential.

Both binding assays and functional bioassays can contribute to characterizing biomolecules for purposes of comparability or similarity. Bioassays measure functional response, biological activity and potency, and can permit prediction of potency in other systems. Functional bioassays are also sensitive to differences that affect signal transduction and response such as epitopes involved in receptor binding and subunit association.

By contrast, binding assays measure association (not necessarily a biological response), generally to one or more relevant biological molecules, and are sensitive to differences in epitopes involved in the particular interaction. Correlation between a given functional bioassay and a given binding assay depends on the individual product and assays. Clearly, the most complete analysis would include both types of assays.

Future protein concentration studies might use a proteomics approach.

David Bunk (NIST) followed with a discussion on options for assaying protein concentration, a key element of determining potency. After describing standard methods, Bunk gave several examples of concentration studies and listed reference material sources for principal protein product types. He wrapped up his talk with his view on the future of protein concentration studies using a "proteomics" approach. Bunk is referring to methods used by protein researchers to analyze protein content from very complex mixtures such as human tissues. One such method, which employs isotopically labelled protein standards, eliminates problems associated with incomplete enzymatic digestion of protein in analytic samples.

Reference Standards

Highlights

Follow-on protein developers will have difficulty obtaining reference standards to which to compare their products.
Since many analytical techniques used with proteins generate measurements more appropriate for comparison (vs. absolute values), the availability of reference standards is essential.
Developers should look outside conventional protein analytics to strengthen, accelerate, and make analysis of protein therapeutics more reliable.

Standardizing standards

The final two technical presentations covered issues and challenges in developing reference materials for biological products. Emily Shacter (FDA/CDER) spoke first on the importance of reference standards.

Because most analytical techniques do not provide absolute values, the only way to compare proteins is on a relative basis, through side-by-side comparison. "So many of these techniques are so highly influenced by the operator carrying out the assays, it is unlikely that most of these analytical tests can be carried out in the absence of a comparator," Shacter said.

Because potential FOB developers will lack access to the originator drug except in its formulated state, analytic comparisons will be inherently difficult.

But therein lies a thorny scientific problem. Because FOB developers will lack access to the originator drug except in its formulated state, analytic comparisons will be inherently difficult. The final drug product will contain excipients, detergents, protein and chemical stabilizers, sugars, and buffers, all of which can interfere with characterization. But removing these components might subject the protein to physicochemical changes that could make comparisons meaningless, so developers will probably need to reverse-engineer the formulation and duplicate it with their product. Even then, they will still not enjoy access to the identical bioassays involved in product release, and the reference standards for measuring impurities. "Coming up with a different assay may mean that there will be different endpoints," Shacter warned.

The solution is reference standards and materials, which was the topic of Adrian Bristow's (NIBSC, UK) talk. Providing and validating reference materials is one of the most critical issues in FOB development, he explained.

The World Health Organization maintains more than two million ampoules containing 400 reference materials for a wide range of biologicals. These support a range of analytical (biological and physicochemical) methods such as those described in this eBriefing. These methods, however, are comparative and their results are meaningless without reference materials. The second significant set of international reference materials are pharmacopoeial materials, which are limited to molecules for which pharmacopoeial monographs exist.

Bristow used the example of erythropoietin to illustrate differences between the WHO reference standard to the European Pharmacopoeia standard. The WHO standard, which defines the international unit, is available in very low concentration (under one microgram per ampoule) and contains a protein excipient. Therefore, it is more appropriate for bioassays than for physicochemical testing. The pharmacopoeial standard contains no protein excipient and is distributed at 250 micrograms per ampoule, therefore supports more material-intensive physicochemical analysis. Neither standard, nor the two together, are adequate for supporting the surge in FOB activity expected over the next few years.

FOB stakeholders should define essential molecular and biological properties and settle on analytics to demonstrate those attributes.

Bristow called for development of method-independent reference standards and greater harmonization between those standards and formulated products. More importantly, he urged FOB stakeholders to define essential molecular and biological properties and settle on analytics to demonstrate those attributes.

"We've seen some really fantastic science [here] but I am not so sure if we haven't asked these people to solve the wrong problems," Bristow wondered. "They're coming up with ever-increasingly minor modifications to the molecule, in vanishingly small quantities, where the clinical relevance is either going to be very difficult to prove or, let's face it, trivial or non-existent. These proteomics guys are talking about developing analytical methods for picking out a nanogram per milliliter of a protein in serum, and identifying it. Why the hell can't they do this in a pharmaceutical preparation? It's surely got to be less of a problem."

Conclusion

Emily Shacter and Steven Kozlowski provided a conference summary and outlined future goals in their concluding remarks.

Shacter called for a battery of orthogonal analytical methods for determining the structure, function, and purity of protein products. FDA is interested not only in primary test results, but in the algorithm used to determine similarity through side-by-side comparisons. "The more testing we do, the more confidence we'll have in the conclusions we're drawing about the degree of similarity," Shacter said.

In his wrap-up, Kozlowski presented the positive historical trends in biotechnology, from early fears of genetically engineered "superbugs" to today, when biotechnology is generally accepted. He presented his "iceberg" model of biological products consisting of release tests, characterization, and process, with the caption asking "How much of the iceberg (desired product) can we see?"

"We've shown in this meeting that characterization is advancing, and that part of the iceberg is getting bigger while process is getting smaller," Kozlowski observed. "But at the same time FDA wants to push the importance of process in a different way. By defining critical attributes of products, and really homing in on what makes our product our product, and what variants are really irrelevant, we can take that map and hone our process. And that gives us a better process and more flexibility. That's a very hard task, but it's certainly something to aspire to."

Open Questions

Primary Structure

Which methods are most appropriate for analyzing post-translational modifications?

How will MS analysis overcome problems with PEGylated molecules?

In what way will potential FDA policy for FOB product development differ from current guidances for today's "comparable" products (e.g. growth hormone, insulin)?

How important is amino acid sequence for biotherapeutics protein safety and efficacy?

Secondary and Tertiary Structure

What is the clinical relevance of tertiary structure?

How must analytical methods differ to qualify as "orthogonal?"

Is it possible to quantify two proteins' similarity or dissimilarity based on differences in some number of analytical methods?

Quaternary Structure

Does it make more sense during potential FOB development and manufacture to compare aggregation in absolute terms, or in comparison to the originator product?

Are proteins with high aggregation rates but very low immunogenicity approvable?

Is it possible, or even desirable, to compare aggregates in a potential FOB drug product with higher-order components of a formulated product?

Effects of Manufacturing

How would regulators treat potential FOB manufacturing processes that differ significantly and fundamentally (e.g. different expression system) from the originator's process?

Will requirements for clinical testing be based solely on physicochemical or process differences?

Who would determine specific problems in analytics and processing that need to be overcome for potential approval of FOBs, and when will this happen?

Impurities and Contaminants

Would regulatory treatment of impurity and contaminant profiles differ for potential FOBs than for originator products?

Which analytical methods are most appropriate for measuring non-protein impurities?

Bioassays and Potency

Can molecules be considered "similar" based solely on biological activity?

Program

Monday, December 12, 2005
7:00 – 8:30 AM	Registration and Continental Breakfast
8:30 – 9:00 AM	Introduction and Goals of the Workshop Welcoming Remarks Rashid Shaikh, New York Academy of Sciences Current Regulatory Directions Keith Webber, FDA Meeting Goals and Agenda Emily Shacter, FDA
9:00 AM – 5:30 PM	Session I – Analytical Techniques to Examine Molecular Heterogeneity of Active Ingredient: Comparisons, Strengths and Weaknesses Primary Structure Session Moderator: David Bunk, NIST Overview of Primary Structure and Related Issues David Bunk, NIST Analysis of Post-Translationally Modified Peptides and Proteins by Mass Spectrometry: New Technology and Applications Donald F. Hunt, University of Virginia Chromatography Techniques William Hancock, Northeastern University Intermission
	Fourier Transform MS Jonathan Amster, University of Georgia Towards a Goal of Automated Glycoproteomic Analysis Vernon Reinhold, University of New Hampshire *Panel Discussion* Panel Moderator: Barry Cherney, FDA
12:30 – 2:00 PM	Luncheon
	Secondary and Tertiary Structure Session Moderator: Blair Fraser, FDA Analysis of the Secondary and Tertiary Structure of Proteins by a Multi-dimensional Phase Diagram Approach Russ Middaugh, University of Kansas Characterizing Biologics by NMR Spectroscopy Daron Freedberg, FDA/CBER Spectroscopic Techniques — FTIR, Fluorescence, Other — for Secondary Structure Analysis Keith Oberg, Medical Research Products - A Spectroscopic Techniques for Tertiary Structure Analysis Curtis Meuse, NIST Intermission Thermodynamic Characterization of Protein Pharmaceutical Products by Calorimetry Frederick P. Schwarz, CARB/NIST Surface Hydrophobicity/HIC Steve Cramer, Rensselaer Polytechnic Institute *Panel Discussion* Panel Moderators: Blair Fraser, FDA and Curtis Meuse, NIST
5:30 – 7:00 PM	Wine and Cheese Reception
Tuesday, December 13, 2005
7:30 – 8:30 AM	Registration and Continental Breakfast
8:30 – 12:00 PM	Session I Continued Session Moderator: Amy Rosenberg, FDA Protein-Protein Interactions — Quaternary Structure Overview and Related Issues Amy Rosenberg, FDA Critical Factors Governing Aggregation of Proteins in Aqueous Solution John F. Carpenter, University of Colorado Health Sciences Center Field-Flow Fractionation (FFF) in Protein Purification and Characterization Karin D. Caldwell, Uppsala University Light Scattering as a Tool for Assessing Protein Aggregates Ewa Folta-Stogniew, Yale University Imaging Proteins Using Atomic Force Microscopy Roger E. Marchant, Case Western Reserve University Intermission Uses of Analytical Ultracentrifugation Thomas M. Laue, University of New Hampshire Mass Spectrometry of Higher Order Protein Structures Igor Kaltashov, University of Massachusetts *Panel Discussion* Panel Moderator: Barry Cherney, FDA
12:00 – 1:30 PM	Luncheon
1:30 PM – 3:00 PM	Session II — Effect of the Manufacturing Process on the Product Session Moderator: Stephen Moore, FDA Product Definition by Process Design Charles l. Cooney, Massachusetts Institute of Technology Impact of Changes in Chromatographic Operation on Biopharmaceutical Product Quality Erik Fernandez, University of Virginia Effects of the Bioreactor Environment on Product Quality Sarah W. Harcum, Clemson University Renaturation and Folding François Baneyx, University of Washington *Panel Discussion* Panel Moderator: Kurt Brorson, FDA Intermission
3:30 – 5:45 PM	Session III — Impurities and Contaminants Session Moderator: Andrew Chang, FDA Overview — What Types of Impurities are of Concern and Why Impurities Matter? Kathleen A. Clouse, FDA Detection of Impurities in Proteomic Research Timothy D. Veenstra, SAIC-Frederick, Inc. Immunoassays for Residual HCP Analysis Nadine M. Ritter, The Biologics Consulting Group, LLC *Panel Discussion* Panel Moderator: Andrew Chang, FDA
Wednesday, December 14, 2005
7:30 – 8:30 AM	Registration and Continental Breakfast
8:30 – 11:30 AM	Session IV — Bioassays and Potency Session Moderator: Marjorie Shapiro, FDA Overview Steven Kozlowski, FDA Case Studies Example 1: Enzyme Assays — Single Function vs. Pleiotropy Laureen Little, Bioquality Example 2: Binding Assays Versus Functional Bioassays C. Jane Robinson, National Institute for Biological Standards and Control Intermission Example 3: Challenges to Assaying Protein Concentration David Bunk, NIST *Panel Discussion* Panel Moderator: Steven Kozlowski, FDA
11:30 AM – 1:00 PM	Luncheon
1:00 – 2:00 PM	Session V — Assessing Similarity of Active Ingredients Session Moderator: Emily Shacter, FDA Overview of Issues Emily Shacter, FDA Challenges in Developing Reference Materials for Biotech Products Adrian Francis Bristow, National Institute for Biological Standards and Control Case Studies on Structure-Activity-Stability Relationships with Therapeutic Proteins Chris Jones, National Institute for Biological Standards and Control Intermission
2:30 – 4:00 PM	Roundtable Discussion — How to Compare Products/Proteins in the Absence of Reference Standards Moderator: Emily Shacter, FDA
4:00 – 4:45 PM	Session VI — Workshop Wrap-Up Wrap-Up Emily Shacter, FDA Closing Remarks Steven Kozlowski, FDA

Analytical Techniques to Examine Molecular Heterogeneity of Active Ingredient: Comparisons, Strengths and Weaknesses

Comparative Analysis of Post-translationally Modified Proteins and Peptides by Mass Spectrometry: New Technology and Applications

Speaker: Donald F. Hunt, PhD
Departments of Chemistry and Pathology, University of Virginia, Charlottesville, Virginia

For the direct analysis of proteins on a chromatographic time scale, we use monolithic nanocolumns, a modified linear ion trap mass spectrometer, and sequential ion/ion reactions to fragment the intact protein and convert all fragments to singly charged species.^1,2 Proteins are converted to gas-phase multiply charged positive ions by electrospray ionization and then allowed to react with fluoranthene radical anions. Electron transfer to the multiply charged protein promotes random fragmentation of amide bonds along the protein backbone. Multiply charged fragment ions are then deprototonated in a second ion/ion reaction with the carboxylated anion of benzoic acid. The m/z values for the resulting singly and doubly charged ions are used to read a sequence of 15-40 amino acids at both the N and C termini of the protein. This information, along the the measured mass of the intact protein is used to identify unknown proteins, to confirm amino acid sequence of a known protein, to detect posttranslational modifications, and to determine the presence of possible splice variants. For the comparative analysis of chemically or post-translationally modified proteins, two samples are digested proteolytically and the resulting peptides from each are then converted to d₀- and d₃-methyl esters, respectively. The two samples are then mixed together and analyzed by nano flow HPLC interfaced to electrospray ionization on a tandem linear ion trap-Fourier transform mass spectrometer (LTQ-FTMS). This instrument operates at a resolution of 100,000, measures masses to three decimal places, and records the molecular masses of peptides in each sample at the high attomole level.³ Changes in the structures of peptides between the two samples can be detected at the attomole level with a dynamic range of 5,000/1. For the analysis of phosphorylated proteins, immobilized metal affinity chromatography (IMAC) is employed to enrich the sample for phosphopeptides prior to analysis by nanoflow HPLC.⁴

¹Syka, J. E. P., J. J. Coon, M. J. Schroeder et al. Peptide and protein sequence analysis by electron transfer dissociation mass spectrometry. Proc. Natl. Acad. Sci. USA 101: 9528-9533. Full Text

²Coon, J. J., B. Ueberheide, J. E. P. Syka, et al. 2005. Protein identification using sequential ion/ion reactions and tandem mass spectrometry. Proc. Natl. Acad. Sci. USA 102: 9463-9468.

³Syka, J. E. P., J. A. Marto, D. L. Bai et al. 2004. Novel linear quadrupole ion trap/FT mass spectrometer: performance characterization and use in the comparative analysis of histone H3 post-translational modifications. J. Proteome. Res. 3: 621-626.

⁴Schroeder, M. J., D. J. Webb, J. Shabanowitz et al. 2005. Methods for the detection of paxillin post-translational modifications and interacting proteins by mass spectrometry. J. Proteome. Res. 4: 1832-1841

Towards a Goal of Automated Glycoproteomic Analysis

Speaker: Vern Reinhold, PhD*
Center for Structural Biology, University of New Hampshire, Durham, New Hampshire

The proliferation of reports attributing biological function to oligosaccharide epitopes continues unabated. Unfortunately, a comprehensive strategy for carbohydrate sequencing is lacking. This brief report outlines three focused efforts to establish congruent strategies for carbohydrate sequencing, (i) analytical considerations that account for all aspects of small oligomer structure by MSⁿ disassembly, (ii) database support using an ion fragment library and associated tools for high throughput analysis, and (iii) a concluding algorithm for defining oligosaccharide topology from MSⁿ disassembly pathways. The data mining effort focuses on correlating the fragments of small oligomers to stereospecific glycan structures, an outcome attributed to a combination of metal ion adduction and analyte conformation. Product masses and ion intensities vary with inter-residue linkage, branching position, and monomer stereochemistry. This bottom-up approach to achieve full oligosaccharide and glycan characterization will be supplemented with an MSⁿ fragment spectral library and associated tools. As a third component of this effort we introduce the operational details of an algorithm used to assign the topology (branching and linkage) of permethylated glycans using MSⁿ data. Because the algorithm is de novo (and therefore not limited to previously-characterized glycans) and high-performance, they represent a step toward high-throughput glycan analysis.

Acknowledgements: NIH grants GM 45054, NCRR-BRIN RP016459.
Keywords: Fragment library, MSⁿ-disassembly, HTP-analysis, Sequencing, Algorithm.

* Coauthors: Dave Ashline, Hailong Zhang, and Tony Lapadula; Center for Structural Biology, University of New Hampshire, Durham, NH

Analysis of the Secondary and Tertiary Structure of Proteins by a Multi-dimensional Phase Diagram Approach

Speaker: Russ Middaugh, PhD
Department of Pharmaceutical Chemistry, University of Kansas, Lawrence, Kansas

A method is described in which proteins can be represented as vectors in a highly dimensioned phase space. A series of measurements using methods such as uv derivative absorption spectroscopy, intrinsic and extrinsic probe fluorescence, circular dichroism, FTIR and Raman spectroscopies is performed as a function of solution variables (temperature, pH, ionic strength,protein concentration, redox potential, etc.). The data is normalized and then analyzed by an eigenvalue type approach. The final data is truncated based on relative contributions of each technique to the final vector and the three most important contributors coded by an RGB color scheme.

This yields a detailed phenomenological phase diagram-like picture of a protein that can be used in a wide variety of applications that will be discussed. Preliminary details can be viewed at J. Pharm. Sci. 92: 1805–20, 94: 1893–1911.

Characterizing Biologics by NMR Spectroscopy

Speaker: Daron I. Freedberg, PhD
Center for Biologics Evaluation and Research, Food and Drug Administration, Rockville, Maryland

Biologics are complex and varied, but many can be well-characterized. Though NMR (nuclear magnetic resonance) is routinely used to identify small molecules and their structures, its application to regulatory uses has been limited. The underutilization in regulatory uses stands in contrast to the increasing number of biological macromolecules currently being studied with NMR. This talk will address the advantages and disadvantages of using NMR to characterize biologics for regulatory purposes.

Spectroscopic Techniques — FTIR, Fluorescence, Other — for Secondary Structure Analysis

Speaker: Keith A. Oberg, PhD
Medical Research Products - A, Sylmar, California

Optical spectroscopic techniques are low-resolution but rapid and reproducible means for global protein structure evaluation. They are useful orthogonal methods for comparing the structure of proteins in different formulations. Infrared (IR) and Circular Dichroism (CD) spectroscopy are most common, with fluorescence and UV absorption of some use in stability determination. CD can differentiate alpha helix and non-periodic structures; it also reveals structure around aromatic residues. IR can readily distinguish beta sheet from alpha helix. It can also probe subtle changes in backbone and side-chain conformation at the single-residue level. Neither technique provides absolute structural information but they are sensitive to changes in structure, and are useful in guiding further studies. In combination, IR and CD spectra provide a structural fingerprint for proteins that can be used to demonstrate conformational equivalence. Both methods can be used for drug-substance evaluation. IR is particularly useful for evaluation of drug product because spectra can be collected from complex non-transparent mixtures, including lyophilized solids. Both methods are subject to interference and require thoughtful application.

Spectroscopic Techniques for Tertiary Structure Analysis

Speaker: Curtis W. Meuse, PhD
Biochemical Science Division, National Institute of Standards and Technology, Gaithersburg, Maryland

Proteins must have appropriate tertiary structures to function. This is true in biology and this is true in drug formulation. Optical spectroscopy methods are an important class of analytical methods for describing the tertiary structural properties of proteins. These biophysical techniques rely on changes in protein optical properties between the native and denatured states of proteins. By following changes in the optical properties of proteins as functions of time or degrees of folding, optical spectroscopy can be used to determine kinetic and thermodynamic properties of protein tertiary structure. While generally lower resolution than X-ray crystallography and nuclear magnetic resonance, optical methods are more rapid and can be applied to most protein in many different environments. In addition, they directly address the dynamical nature of protein structure. Recently, there have been efforts to standardize these techniques, for example Protein Science (2005), 14: 602–616, to allow the creation of a detailed database to develop and test models for protein folding. This talk will describe how these methods work and discuss their strengths and weaknesses.

Thermodynamic Characterization of Protein Pharmaceutical Products by Calorimety

Speaker: Frederick P. Schwarz, PhD
Center for Advanced Research in Biotechnology/National Institute of Standards and Technology, Rockville, Maryland

Two calorimetric methods, isothermal titration calorimetry (ITC) and differential scanning calorimetry (DSC), have been employed to thermodynamically characterize the properties of proteins. ITC is widely used to determine the thermodynamics of biological interactions in solution from measurements of the heat detected upon titrating a solution of one reactant into a solution of the other reactant of a binding interaction. Since the range of interactions investigated by ITC is very general, it is well-suited to quantitatively compare the interaction of a ligand with its protein pharmaceutical product. ITC results of an antigen binding to monoclonal antibodies mutated to enhance their binding affinities will be presented to show the potential and limitation of ITC for protein drug research. DSC is used to determine the conformational stability of the active state of proteins from measurements of changes in the heat capacity of the protein in solution as a function of temperature. DSC measurements can not only access the thermal stability of a protein, but also thermodynamic structural properties and the solution conditions, which would enhance its stability. DSC results on the thermodynamic stability of several proteins including mutated proteins will be covered to illustrate the potential and limitation of DSC for protein pharmaceutical product characterization.

Critical Factors Governing Aggregation of Proteins in Aqueous Solution

Speaker: John F. Carpenter, PhD
Center for Pharmaceutical Biotechnology, University of Colorado Health Sciences Center, Denver, Colorado

Even a low level of aggregates (e.g., 1%) can compromise the safety of a therapeutic protein product. The rate of aggregation in aqueous solution can be reduced by increasing the thermodynamic stability of the native state ("conformational stability") and charge-charge repulsion between protein molecules ("colloidal stability"). For some proteins, the rate limiting step for precipitation is the accumulation of soluble aggregates (e.g., dimers) that serve as prenuclei and, at a threshold level (e.g., 2%), trigger gross precipitation. In other cases, a small fraction of protein molecules form visible and subvisible particles, even though the formulation confers conformational and colloidal stability, and greatly inhibits soluble aggregate formation. Such particle formation can be induced via heterogeneous nucleation of protein aggregation by foreign particles that are shed from vials, syringes, stoppers and filling pumps. The foreign materials include glass, silicone oil, rubber, tungsten and stainless steel. Specific actions can be taken to reduce the degree of aggregation caused by a given factor. However, ultimately the aggregate levels in a dose of a therapeutic protein depend on all aspects of manufacturing, formulation, container, closure, shipping, storage and delivery to the patient.

Field-Flow Fractionation (FFF) in Protein Purification and Characterization

Speaker: Karin D. Caldwell
Department of Surface Biotechnology, Uppsala University, Uppsala, Sweden

The rapid evolution of protein-based pharmaceuticals has created an urgent need for process control and careful product evaluation. Due to the size of proteins, and their often significant structural fragility, specific care is required in their downstream processing to produce biologically active, pure monomeric species in high yield.

Field-flow fractionation is a hydrodynamically based family of separation methods especially well suited for the analysis of fragile macromolecular samples, and samples with odd shapes, that span a broad range of sizes/molecular weights. This is because the separation takes place in open unobstructed channels with mobile phase flow under low shear. The channel in an FFF system corresponds to the packed column in HPLC, and all the system's ancillary equipment is the same as that used in conventional chromatography. In the channel, sample separation is accomplished through the combined interaction of a laminar mobile phase flow with a perpendicular, externally applied field.

The sedimentation and flow fields are most commonly utilized in protein characterization, and the results of separations under well chosen conditions of field strength and flow rate are not only pure fractions, but fractions characterized with respect to size or mass, depending on the choice of field.

The presentation will introduce instrumentation applicable to the two techniques and compare their selectivities. Their performances will be illustrated by protein separations in which monomers, dimers, and higher order aggregates are separated and their masses evaluated. A recently initiated effort to analyze protein glycoforms will also be discussed.

Reference:
Caldwell, K. D. & K.-G Wahlund. 2005. Field-flow fractionation in protein analysis. In Methods for Structural Analysis of Protein Pharmaceuticals, W. Jiscoot & D. Crommelin, Eds.

Light Scattering as a Tool for Assessing Protein Aggregates

Speaker: Ewa Folta-Stogniew, PhD
W. M. Keck Biotechnology Resource Laboratory, Department of Molecular Biophysics and Biochemistry, Yale University School of Medicine, New Haven, Connecticut

Static and dynamic laser light scattering are discussed as tools for studying protein oligomerization and aggregation. Dynamic light scattering (DLS) is very well suited for detection of small quantities of aggregates in protein samples because DLS can easily analyze samples with broad distribution of sizes. No fractionation is used and sample is analyzed in batch mode without any chromatographic separation. DLS detector can also be used as an "on-line" detector coupled with a fractionation step.

Static light scattering (SLS) is best utilized for measurement of molar masses as an "on-line" detector coupled with size exclusion chromatography (SEC), refractive index (RI) and ultraviolet (UV) detection. Since static light scattering provides only the weight-average molar mass, M_w, of the species in solution, the SEC separation plays an integral role in the overall analysis, albeit, the elution from SEC does not need to correlate with the molar masses of the species being studied. SEC/LS allows determination of molar mass of unmodified proteins with a precision of +5% in a single experiment that uses ~100 µg of protein; for DLS "on-line" measurement, ~400 µg is needed. Monitoring the elution from SEC by three detectors, UV, LS and RI, provides an excellent tool for detection of sample heterogeneity. Potential loss of protein on the SEC column, sample dilution, and restriction on elution solvent are major limitations of SEC/LS analysis. SEC/LS is suitable for analysis of glycoproteins, proteins modified by polyethylyne glycol as well as membrane proteins solubilized in non-ionic detergents. SEC/LS analysis represents a fast and robust approach to determining M_w and quantifying aggregates in protein preparations.

Imaging Proteins Using Atomic Force Microscopy

Speaker: Roger E. Marchant, PhD
Department of Biomedical Engineering, Case Western Reserve University, Cleveland, Ohio

Atomic force microscopy (AFM) has proven a very useful approach for imaging and studying proteins in air or under near physiologic conditions, and for visualizing protein interactions on nanoscale lengths over time. AFM has been used to obtain molecular and sub-molecular visualization of many biological molecules and assemblies, including plasma and structural proteins and cell membrane proteins. Here, the use of nanoscale probes techniques, such as AFM, for achieving molecular level visualization of proteins, and multimer assemblies will be described. High-resolution imaging (nanometer scale), combined with the measurement of biomechanical properties and surface-dependent intermolecular interactions, makes AFM very powerful technique. This presentation will outline the basic principles and utility of AFM for imaging measurements, and offer objective perspectives on both the advantages and disadvantages of the technique. This will include addressing key issues of image resolution, typical artifacts, and morphological methods for improving image accuracy. This will be followed by examples of the progress in the use of AFM for high-resolution visualization of proteins, and multimer assemblies. Surface-dependent molecular processes also can be followed including adsorption phenomena and subsequent tertiary level conformational changes, assembly formation, and interactions with cellular components. We shall also briefly explore the use of AFM for obtaining sensitive picoNewton measurements of intermolecular forces and for mapping of surface properties.

Uses of Analytical Ultracentrifugation

Speaker: Tom Laue, PhD
Department of Biochemistry and Molecular Biology University of New Hampshire, Durham, New Hampshire

For over 75 years, analytical ultracentrifugation (AUC) has proven to be a powerful method for characterizing solutions of macromolecules, and an indispensable tool for the quantitative analysis of macromolecular interactions. Because it relies on the principle property of mass and the fundamental laws of gravitation, AUC has broad applicability and can be used to analyze the solution behavior of any sort of molecule in a wide range of solvents and over a wide range of solute concentrations. Sedimentation velocity, in which the force due to gravitation is opposed by the frictional force, is particularly useful for characterizing the aggregation state of proteins. No modification of the protein is required, and the analysis software is developed from hydrodynamic first principles, making the analysis of data relatively straightforward and objective. A brief description will be presented of the theory of sedimentation velocity, along with some applications of sedimentation velocity to the analysis of protein aggregation.

Mass Spectrometry of Higher Order Protein Structures

Speaker: Igor A. Kaltashov, PhD
University of Massachusetts at Amherst, Amherst, Massachusetts

Characterization of higher order structure of proteins and their assemblies using mass spectrometry-based methods enjoyed an explosive growth of popularity in the past decade. Continuous technological advances in both methods of ionizing biopolymers (particularly, electrospray ionization, ESI) and their mass analysis resulted in a dramatic expansion of the scope of systems that can be characterized, as well as the quality and reliability of information deduced from such studies. Conformational heterogeneity of proteins and their assemblies can be evaluated by monitoring ionic charge state distributions in ESI mass spectra. More detailed information on protein higher order structure and conformational stability can be obtained by measuring kinetics of hydrogen/deuterium exchange in solution using MS. Since the soft nature of ESI process allows in many cases intact non-covalent assemblies to be transferred from solution to the gas phase, ESI MS measurements often provide an efficient way to determine protein quaternary structure. Analytically useful data can be obtained on MDa non-covalent assemblies and beyond. ESI MS is also very successful in addressing a serious problem inherent to most other biophysical tools, namely the great difficulty associated with the analysis of protein structure, conformational stability and interactions in multi-component systems. Most of the work in the field focused so far on proteins and their interactions with each other, as well as small ligands (such as metal ions, small organic molecules, etc.). However, it becomes clear that MS has also a great potential for characterizing protein interactions with other biopolymers, particularly oligonucleotides. Furthermore, noticeable progress was done recently in characterizing conformation and stability of highly heterogeneous biopolymers (such as extensively glycosylated proteins, as well as proteins conjugated to synthetic polymers).

Effect of the Manufacturing Process on the Product

Impact of Changes in Chromatographic Operation on Biopharmaceutical Product Quality

Speaker: Erik J. Fernandez, PhD
Department of Chemical Engineering, University of Virginia, Charlottesville, Virginia

Chromatography is a keystone of biopharmaceutical purification. Variations in chromatographic design and operation can lead to changes in amounts of contaminants not related to the target molecule as well as the distribution of product-related variants. No simple, generalized, robust tools for predicting these variations currently exist. Examples of these variations in chromatographic behavior will be presented, and the emerging opportunities for analysis and prediction of these trends in chromatography behavior will be discussed.

Effects of the Bioreactor Environment on Product Quality

Speaker: Sarah W. Harcum, PhD
Department of Bioengineering, Clemson University, Clemson, South Carolina

Therapeutic proteins are very labile molecules, where it is well recognized that bioreactor conditions greatly influence the quality of the recombinant protein. For simple proteins expressed in bacteria, protein modifications are often observed under stressful conditions. For proteins made in yeast, fungi, or insect cultures, less is know about the impact of the environmental factors on protein quality; however, these species are selected over bacterial systems to due the ability of the host to mediate post-translation modifications. For recombinant glycoprotein therapeutics, which are most often made using mammalian hosts in order to better mimic human glycoproteins, the protein glycosylation reactions within a cell are strongly influenced by environmental factors, such as nutrients, dissolved oxygen, pH, waste by-products, and temperature. Since recombinant glycoproteins are highly expressed relative to native proteins, the influence of environmental factors on the expression or function of glycosylation-related enzymes is potentially exacerbated. This talk will present information regarding bioreactor environmental changes that normally occur and methods used to control the bioreactor environment. Bacterial, yeast, fungi, and insect culture conditions with respect to protein quality will be briefly presented. Case studies will be presented that highlight the known impact of mammalian culture conditions on protein glycosylation.

Renaturation and Folding

Speaker: François Baneyx, PhD
Department of Chemical Engineering, University of Washington, Seattle, Washington

This presentation will review the major differences between the in vivo folding of recombinant protein in bacterial systems and their in vitro renaturation from the aggregated (inclusion body) state. How refolding variables (e.g., folding modulators and chemical additives) and processes can be manipulated to enhance the recovery yields of bioactive protein products.

Impurities and Contaminants

Immunoassays for Residual HCP Analysis

Speaker: Nadine M. Ritter, PhD
Biologics Consulting Group, LLC, Alexandria, Virginia

One of the unique aspects of biopharmaceutical products that are derived from cellular expression systems is that purification process used to obtain the target active protein can yield co-purifying proteins from the host cell itself. These "host cell-derived proteins" (HCPs) represent one category of process-related impurities that require careful analytical evaluation to assure patient safety and product consistency. To date, the most widely-used technology for measuring and monitoring residual HCPs has been immunological methods. Because of its potential specificity and sensitivity, an immunoassay can be the ideal analytical method for the determination of HCPs. However, there are several elements of an HCP immunoassay that can impact the accuracy and reliability of the data it generates. The nature of HCP immunoassays, and some of the key factors involved in utilizing them with biopharmaceutical products, will be discussed.

Detection of Impurities in Proteomic Research

Speaker: Timothy D. Veenstra, PhD
Laboratory of Proteomics and Analytical Technologies, SAIC-Frederick, Inc., Frederick, Maryland

The continuing development of mass spectrometry (MS) instrumentation has provided very sensitive ways to detect proteins complexed with other proteins as well as opening up the potential to identify biomarkers within human biofluids. Analytical parameters such as analysis speed and sensitivity have enabled MS to play a large role in both of these endeavors. While sensitivity has enabled the direct detection of proteins and peptides at levels not previously possible by MS, it also can confound the analysis through the detection of non-specifically bound proteins or other "impurities." These detected impurities can often confound the analysis and prevent valuable information from being easily recognized.

Assessing Similarity of Active Ingredients

Challenges in Developing Reference Materials for Biotech Products

Speaker: Adrian Francis Bristow, PhD
National Institute for Biological Standards and Control, UK

Quality control testing of medicines is required to demonstrate compliance with criteria of identity, potency and purity. This remains the case for biologics, including biotech products, and inevitably, follow-on biologics. Most analytical techniques used in control of biotech products are comparative, and require reference materials. Thus, quantitative biological or physico-chemical assays of potency depend on suitably calibrated reference materials, traceable to higher order International standards or to the SI, and qualitative tests of identity similarly depend on comparisons with reference materials. Although significant progress has been made in the development of absolute, or reference material-independent methodology applicable to biotech products, the backbone of pharmacopoeial or similar release specifications remains bioassay, chromatographic and electrophoretic methods which cannot operate without reference materials.

Available reference materials for biotech products include WHO International Biological standards, Pharmacopeial reference substances, other national or regional standards and manufacturers working standards. The specific requirements of follow on biologics; demonstration of identity with originator products , usually by comparison of formulated preparations, present significant challenges both to analytical methodology and to the provision of suitable reference materials. These include: extending the scope of reference materials from biological methods to physico-chemical methods; balancing the need to use excipients ensuring long term stability with the requirements of analytical methods; meeting the requirements of physico-chemical methods in terms of significantly increased quantities of active substance per issue of the reference material, and ensuring continued supply of donated candidate materials from product originators. Whilst it may be reasonable to expect solutions to the scientific problems, it is possible that the non-scientific problems may severely restrict the utility of international reference materials in the development of follow-on biologics.

Web Sites

Biotechnology Industry Organization (BIO)
BIO has collected their comments on follow-on biotechnology product regulation as full-text pdf files here.

EMEA/DIA Joint Workshop on EMEA New Guidelines for Development and Approval of Biosimilars
Program for a December 2005 conference on European regulation of follow-on biologics (PDF, 368 KB).

European Medical Association Guidelines
Emerging Biopharmaceutical Enterprises, an organization of European pharmaceutical companies, has posted documents from the European Medical Association here related to the regulation of follow-on biologics, called "biosimilars" in Europe. See in particular:

FDA Center for Biologics Evaluation and Research
Web site of the FDA division responsible for regulating blood-derived biological products.

FDA Center for Drug Evaluation and Research
Web site of the FDA division responsible for regulating most protein products.

The Law of Off-Patent Biopharmaceuticals
A statement by Senator Orrin Hatch on steps Congress must take to provide the regulatory framework for follow-on biologics.

PhRMA Comments to the FDA on follow-on biologics (PDF, 805.1 KB)
PhRMA (The Pharmaceutical Research and Manufacturers of America) advocates for public policies on behalf of pharmaceutical/biotechnology research companies.

Scientific Considerations Related to Developing Follow-On Protein Products (February 2005)
Collected slide presentations for a February 14-16, 2005, FDA Meeting

Scientific Considerations Related to Developing Follow-On Protein Products (September 2004)
Slide presentations for the earlier, September 14-15, 2004 FDA Meeting. This online docket also contains a transcript of the meeting.

Journal & Magazine Articles

Ashline, D., S. Singh, A. Hanneman & V. Reinhold. 2005. Congruent strategies for carbohydrate sequencing. 1. Mining structural details by MS(n). Anal. Chem. 77: 6250-6262.

Baneyx, F. & M. Mujacic. 2004. Recombinant protein folding and misfolding in Escherichia coli. Nat. Biotechnol. 22: 1399-1408.

Baneyx, F. 2004. Keeping up with protein folding. Microb. Cell Fact. 3: 6. Full Text

Bristow, A., P. Berger, J. M. Bidart, et al. 2005. Establishment, value assignment, and characterization of new WHO reference reagents for six molecular forms of human chorionic gonadotropin. Clin. Chem. 51: 177-182. Full Text

Bunk, D. M. & M. J. Welch. 2006. Characterization of a new certified reference material for human cardiac troponin I. Clin. Chem. 52: 212-219.

Chamberlain, P. 2004. Biogenerics: Europe takes another step forward while the FDA dives for cover. Drug Discov. Today 9: 817-820.

Charles, S. A. 2005. SuperGenerics: a better alternative for biogenerics. Drug Discov Today. 10: 533-535.

Chi, E. Y., S. Krishnan, T. W. Randolph & J. F. Carpenter. 2003. Physical stability of proteins in aqueous solution: mechanism and driving forces in nonnative protein aggregation. Pharm. Res. 20: 1325-1336.

Clark, K. J., F. W. Chaplin & S. W. Harcum. 2004. Temperature effects on product-quality-related enzymes in batch CHO cell cultures producing recombinant tPA. Biotechnol. Prog. 20: 1888-1892.

Combe, C., R. L. Tredree & H. Schellekens. 2005. Biosimilar epoetins: an analysis based on recently implemented European medicines evaluation agency guidelines on comparability of biopharmaceutical proteins. Pharmacotherapy. 25: 954-962.

Coon, J. J., B. Ueberheide, J. E. Syka, et al. 2005. Protein identification using sequential ion/ion reactions and tandem mass spectrometry. Proc. Natl. Acad. Sci. USA 102: 9463-9468. Full Text

Defelippis, M. R. & F. S. Larimore. 2005. The role of formulation in insulin comparability assessments. Biologicals (not yet in print).

Donohue, M. J., M. B. Satterfield, J. J. Dalluge, et al. 2005. Capillary electrophoresis for the investigation of prostate-specific antigen heterogeneity. Anal. Biochem. 330: 318-327.

Eyles, S. J. & I. A. Kaltashov. 2004. Methods to study protein dynamics and folding by mass spectrometry. Methods 34: 88-99.

Feldbaum, C. B. BIO and biogenerics. Nat. Biotechnol. 23: 17.

Freedberg, D. I., R. M. Venable, A. Rossi, et al. 2004. Discriminating the helical forms of peptides by NMR and molecular dynamics simulation. J. Am. Chem. Soc. 126: 10478-10484.

Fromell, K. M. Andersson, K. Elihn & K. D. Caldwell. 2005. Nanoparticle decorated surfaces with potential use in glycosylation analysis. Colloids Surf. B. Biointerfaces 46: 84-91.

Gale, B. K., K. D. Caldwell & A. B. Frazier. 1998. A micromachined electrical field-flow fractionation (mu-EFFF) system. IEEE Trans. Biomed. Eng. 45: 1459-1469.

Garcia, B. A., C. M. Barber, S. B. Hake, et al. 2005. Modifications of human histone H3 variants during mitosis. Biochemistry 44: 13202-13213.

Griffiths, S. 2004. Betting on biogenerics. Nat. Rev. Drug Discov. 3: 197-198.

Hake, S. B., B. A. Garcia, E. M. Duncan, et al. 2006. Expression patterns and post-translational modifications associated with mammalian histone H3 variants. J. Biol. Chem. 281: 559-568.

Herrera, S. 2004. Biogenerics standoff. Nat. Biotechnol. 22: 1343-1346.

Hitchens, K. 2003. Push for generic biologics gaining momentum. Drug Topics (August 1). Full Text

Hoerner, J. K., H. Xiao & I. A. Kaltashov. 2005. Structural and dynamic characteristics of a partially folded state of ubiquitin revealed by hydrogen exchange mass spectrometry. Biochemistry 44: 11286-11294.

Issaq, H. J., K. C. Chan, G. M. Janini, et al. 2005. Multidimensional separation of peptides for effective proteomic analysis. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 817: 35-47.

Johnson, K. A., B. A. Shira, J. L. Anderson & I. J. Amster. 2001. Chemical and on-line electrochemical reduction of metalloproteins with high-resolution electrospray ionization mass spectrometry detection. Anal. Chem. 73: 803-808.

Johnson, L. S. Mollah, B. A. Garcia, et al. 2004. Mass spectrometry analysis of Arabidopsis histone H3 reveals distinct combinations of post-translational modifications. Nucleic Acids Res. 32: 6511-6518.

Kawanishi, T. 2005. Regulatory perspectives from Japan: comparability of biopharmaceuticals. Biologicals (not yet in print).

Lapadula, A. J., P. J. Hatcher, A. J. Hanneman, et al. 2005. Congruent strategies for carbohydrate sequencing. 3. OSCAR: an algorithm for assigning oligosaccharide topology from MS(n) data. Anal. Chem. 77: 6271-6279.

Louet, S. 2003. Lessons from Eprex for biogeneric firms. Nat. Biotechnol. 21: 956-957.

Nieminen, O. & K. Nordstrom. 2004. Regulation of biogenerics: a survey of viewpoints. BioDrugs. 18: 399-406.

Rafferty, B., P. Maile, P. Rigsby, et al. 2001. International Standards for hepatocyte growth factor/scatter factor: initial assessment of candidate materials and their evaluation by multicentre collaborative study. J. Immunol. Methods 258: 1-11.

Schachter, H., S. Chen, W. Zhang, et al. 2002. Functional post-translational proteomics approach to study the role of N-glycans in the development of Caenorhabditis elegans. Biochem. Sox. Symp. 69: 1-21.

Schellekens, H. 2005. Follow-on biologics: challenges of the "next generation." Nephrol. Dial. Transplant. 20 Suppl 4: iv 31-36.

Thienpont, L. M., K. Van Uytfanghe, J. Marriott, et al. 2005. Feasibility study of the use of frozen human sera in split-sample comparison of immunoassays with candidate reference measurement procedures for total thyroxine and total triiodothyronine measurements. Clin Chem. 51: 2303-2311.

Tsumoto, K., D. Ejima, Y. Kita & T. Arakawa. 2005. Why is arginine effective in suppressing aggregation? Mini Rev. Med. Chem. 5: 613-619.

Vanderah, D. J., T. Parr, V. Silin, et al. 2004. Isostructural self-assembled monolayers. 2. Methyl 1-(3-mercaptopropyl)-oligo(ethylene oxide)s. Langmuir 20: 1311-1316.

Zhang, H., S. Singh & V. N. Reinhold. 2005. Congruent strategies for carbohydrate sequencing. 2. FragLib: an MS(n) spectral library. Anal. Chem. 77: 6263-6270.

Organizers

Kurt A. Brorson, PhD

FDA Center for Biologics Evaluation and Research
email | web site | publications

Kurt Brorson is a staff scientist in CDER's division of monoclonal antibodies, Office of Biotech Products. Kurt Brorson received a BA in biology from the University of Chicago in 1984 and PhD in molecular biology from the California Institute of Technology in 1990. After a two-year postdoctoral fellowship at the NIH, he joined the FDA in 1992 as a staff fellow and was converted to a staff scientist in 1999. In addition to review and policy activities, he conducts research on viral safety issues associated with biotechnology products.

David Bunk, PhD

National Institute of Standards and Technology
email | publications

David Bunk received his doctorate in chemistry from Texas A&M University in 1992. He then was awarded a National Research Council Postdoctoral Fellowship at the National Institute of Standards and Technology. After finishing his postdoctoral fellowship, he continued his research at NIST in the analytical chemistry division.

His research interests at NIST focus on the application of chromatographic and mass spectrometric techniques to the analysis of proteins. Recently he has been investigating the use of proteomics method for the quantification of clinically-relevant proteins in serum, specifically cardiac troponin I, c-reactive protein, and B-type natriuretic peptide.

Bunk is a member of the American Society for Mass Spectrometry and the American Association for Clinical Chemistry and actively serves on committees for the American Association for Clinical Chemistry, the International Federation of Clinical Chemistry, and the Clinical Laboratory Standards Institute.

Barry W. Cherney, PhD

FDA Center for Drugs Evaluation and Research
email | web site | publications

Barry Cherney is deputy director at the division of therapeutic proteins in the office of biotechnology products at FDA's Center for Drugs Evaluation and Research (CDER). His regulatory duties have included oversight of reviews for INDs and BLAs for recombinant and naturally derived proteins, product reviewer training, and development of regulatory guidance to address scientific issues related to characterization, manufacture, and control of specified and naturally derived protein products.

Cherney obtained his PhD in biology from Case Western Reserve University. His dissertation work involved the purification and characterization of the nuclear enzyme, Poly (ADP-ribose) synthetase. In 1986, he began postdoctoral training in the department of biochemistry and molecular biology at Georgetown University where he worked on molecular approaches toward defining the structure/function of this enzyme. He joined CBER in 1991 as a senior staff fellow in the division of hematological products and in addition to his regulatory duties, conducted research on the molecular mechanisms leading to tumorigenesis. Since 1998, Cherney has held positions as a full time regulatory reviewer, an expert biologist and became deputy director of the division of therapeutic proteins in March 2001. He was FDA's topic lead expert on ICH Q5E for comparability of biotech products and has been involved with issues concerning the comparison of protein products throughout his FDA career.

Curtis Meuse, PhD

National Institute of Standards and Technology
email | publications

Curtis Meuse is a scientist in the biochemical science division of Chemical Science and Technology Laboratory at the National Institute of Standards and Technology (NIST). He received his PhD in chemistry at the University of Massachusetts, Amherst studying ultra-thin polymer layers using infrared spectroscopy. As a postdoctoral fellow in the Laboratory of Chemical Physics at the National Institutes of Health, he synthesized specifically fluorinated phospholipids and studied their nano-domain structure using Raman spectroscopy. In 1995, he was hired as part of a NIST-wide initiative to develop expertise in the development and characterization of new biomimetic materials. In the Biomolecular Materials Group, he developed methods and models for the characterization of the molecular structure of cell membrane components organized on surfaces using neutron reflectivity and optical techniques. In addition, a state-of-the-art infrared spectroscopic ellipsometer was built and methods for the analysis of spectroscopic data to determine the thickness, optical constants and molecular structure of various cell monolayer and bilayer constructs were developed.

Currently in the biospectroscopy group, Meuse’s work is focused on protein conformation measurements. The goals of his work are to develop and standardize methods to characterize the biologically active state, to allow the measurement of structural changes, and to characterize physical processes that contribute to biological inactivation of proteins such as aggregation and macromolecular association/dissociation. For example, infrared methods have been developed to quantify protein structural stability and protein binding interactions by quantifying the extent of the exchange of deuterium for hydrogen in the amide bonds of proteins in solutions, as solids, in membranes or immobilized on surfaces. In addition, collaborations with the United Kingdom’s National Physics Laboratory are underway on a pilot study for the standardization of biomolecular circular dichroism measurements under the auspices of the International Committee for Weights and Measures.

Emily Shacter, PhD

FDA Center for Biologics Evaluation and Research
email | publications

Emily Shacter is chief of the laboratory of biochemistry in the division of therapeutic proteins of the Office of Biotechnology Products, Office of Pharmaceutical Science, CDER. She received her PhD in biochemistry from Johns Hopkins University in 1982, carried out basic cell regulation and cancer research at the National Institutes of Health for 12 years, and then joined the FDA in 1994.

Shacter oversees the review of INDs and BLAs for novel therapeutic proteins covering a wide range of clinical indications, including cancer, hematopoiesis, tissue repair and remodeling, thrombosis and thrombolysis, inflammatory disease, and bioterrorism. She is engaged in establishing CDER policy regarding product manufacture and characterization and performs inspections of biotechnology manufacturing facilities.

Shacter also runs an active laboratory research program to support the scientific review of therapeutic proteins that have wide-ranging mechanisms of action and unique biological and physicochemical characteristics. Her laboratory research focuses on studying the cytotoxic mechanisms of cancer chemotherapy drugs and understanding how elements of the immune system, such as oxidants and activated phagocytes, influence killing and clearance of dying cells. Her research expertise in the area of protein oxidation is applied regularly to the review of protein therapeutics.

Speakers

Jon Amster, PhD

University of Georgia
Department of Chemistry
email | web site | publications

François Baneyx, PhD

University of Washington
Department of Chemical Engineering
email | web site | publications

Adrian Bristow, PhD

National Institute for Biological Standards and Control
email | publications

Karin D. Caldwell, PhD

Uppsala University
Center of Surface Biotechnology
email | web site | publications

John F. Carpenter, PhD

University of Colorado Health Sciences Center
Center for Pharmaceutical Biotechnology
email | web site | publications

Kathleen A. Clouse, PhD

Food and Drug Administration
Office of Biotechnology Products
email | publications

Charles Cooney, PhD

Massachusetts Institute of Technology
Department of Chemical Engineering
email | web site | publications

Steve Cramer, PhD

Rensselaer Polytechnic Institute
Department of Chemical and Biological Engineering
email | web site | publications

Erik J. Fernandez, PhD

University of Virginia
Department of Chemical Engineering
email | web site | publications

Ewa Folta-Stogniew, PhD

Yale University School of Medicine
W.M. Keck Foundation Biotechnology Resource Laboratory
email | publications

Daron I. Freedberg, PhD

U.S. Food and Drug Administration
Center for Biologics Evaluation and Research
email | publications

William Hancock, PhD

Northeastern University
Department of Chemistry and Chemical Biology
email | web site | publications

Sarah W. Harcum, PhD

Clemson University
Department of Bioengineering
email | web site | publications

Donald Hunt, PhD

University of Virginia
Department of Chemistry
email | web site | publications

Igor Kaltashov, PhD

University of Massachusetts
Department of Chemistry
email | web site | publications

Steven Kozlowski, MD

Food and Drug Administration
Office of Biotechnology Products
email

Tom Laue, PhD

University of New Hampshire
Department of Biochemistry and Molecular Biology
email | web site | publications

Laureen Little, PhD

Bioquality
email | web site

Roger E. Marchant, PhD

Case Western Reserve University
Department of Biomedical Engineering
email | web site | publications

Russ Middaugh, PhD

University of Kansas
Department of Pharmaceutical Chemistry
email | web site | publications

Keith Oberg, PhD

Alfred Mann Foundation
publications

Vernon Reinhold, PhD

University of New Hampshire
Department of Chemistry
email | web site | publications

Nadine M. Ritter, PhD

Biologics Consulting Group, LLC
email | web site | publications

C. Jane Robinson, PhD

National Institute for Biological Standards and Control
Division of Immunology and Endocrinology
email

Amy Rosenberg, MD

Food and Drug Administration
Division of Therapeutic Proteins
email

Frederick Schwarz, PhD

National Institute of Standards and Technology
Center for Advanced Research in Biotechnology
email | web site | publications

Tim D. Veenstra, PhD

SAIC-Frederick, Inc.
email | web site | publications

Keith Webber, PhD

U.S. Food and Drug Administration
Center for Biologics Evaluation and Research
email | web site | publications

Angelo DePalma

Angelo DePalma is a freelance writer based in Newton, New Jersey. In 1984, he received a PhD in chemistry from the State University of New York, Stony Brook. His work appears in a dozen pharmaceutical industry trade magazines, and he is the author of a bestselling book on vitamins and supplements.

Presented by the New York Academy of Sciences and the U.S. Food and Drug Administration, in collaboration with the National Institute of Standards and Technology (NIST)

EmailPrint

Topics
Close
Browse NYAS.org for information in your field of interest.

Life Sciences & Biomedical Research

Behavioral Sciences

Biochemistry, Biophysics & Structural Biology

Biotechnology

Cell & Molecular Biology

Clinical Medicine

Computational Biology

Developmental Biology

Drug Development

Evolutionary Biology

Genetics & Genomics

Neuroscience

Microbiology, Immunology & Infectious Diseases

Physiology

Public Health & Epidemiology

Translational Research

Physical Sciences & Engineering

Chemistry & Chemical Engineering

Civil Engineering

Computer Science

Materials Science

Mathematics

Nanotechnology

Physics

Environmental Sciences & Sustainability

Climate Change

Energy

Environmental Policy

Green Buildings

Transportation & Infrastructure

Urban Sustainability

Water

Social Sciences

Anthropology

Economics

Psychology

Science, Society & Culture

Art & Science

Bioethics

History & Philosophy of Science

Science Policy

Innovation & Economic Development

Career Development

Science Education
About Academy eBriefings
eBriefings let our worldwide audiences stay informed about the new research discussed at our meetings. They feature multimedia presentations of speakers’ slides and audio, meeting summaries, and links to related resources.
About | All eBriefings | Sign-up for eBriefing Alerts
Get Free Access to Publications

Members of the Academy have unlimited online access to our entire library of eBriefings and the complete catalogue of the Annals of the New York Academy of Sciences. As a member you'll also receive one free print Annals volume of your choosing each year and you can purchase additional volumes at discounted rates. There are many other benefits to membership, including free attendance at over 100+ events and webinars.
join today

Your Name
Your Email
Recipient's Email
Your Comments

Comments*

Follow-on Biologics Workshop

Scientific Issues in Assessing the Similarity of Follow-on Protein Products

Follow-on Biologics Workshop

Scientific Issues in Assessing the Similarity of Follow-on Protein Products

Overview

Office of Pharmaceutical Science Perspectives

Highlights

"We have a mission..."

Analytical Techniques to Examine Molecular Heterogeneity: Primary Structure

Highlights

Multiple structural domains

One protein, many forms

Secondary and Tertiary Structure

Highlights

Additional layers of structure differentiation

NMR: no longer only for small molecules

Optical methods: tried and true

Quantitative structure-property relationships

Protein-Protein Interactions: Quaternary Structure

Highlights

Aggegation = aggravation

Quantifying aggregation

Effect of the Manufacturing Process on the Product

Highlights

Roadmap to follow-on products

Impact of manufacturing

Impurities and Contaminants

Highlights

Facing the inevitable

Purification: less than perfect

Bioassays and Potency

Highlights

How to measure activity?

Binding vs. functional assays

Reference Standards

Highlights

Standardizing standards

Conclusion

Open Questions

Primary Structure

Secondary and Tertiary Structure

Quaternary Structure

Effects of Manufacturing

Impurities and Contaminants

Bioassays and Potency

Reference Standards

Media

Program

Monday, December 12, 2005

Tuesday, December 13, 2005

Wednesday, December 14, 2005

Analytical Techniques to Examine Molecular Heterogeneity of Active Ingredient: Comparisons, Strengths and Weaknesses

Comparative Analysis of Post-translationally Modified Proteins and Peptides by Mass Spectrometry: New Technology and Applications

Towards a Goal of Automated Glycoproteomic Analysis

Analysis of the Secondary and Tertiary Structure of Proteins by a Multi-dimensional Phase Diagram Approach

Characterizing Biologics by NMR Spectroscopy

Spectroscopic Techniques — FTIR, Fluorescence, Other — for Secondary Structure Analysis

Spectroscopic Techniques for Tertiary Structure Analysis

Thermodynamic Characterization of Protein Pharmaceutical Products by Calorimety

Critical Factors Governing Aggregation of Proteins in Aqueous Solution

Field-Flow Fractionation (FFF) in Protein Purification and Characterization

Light Scattering as a Tool for Assessing Protein Aggregates

Imaging Proteins Using Atomic Force Microscopy

Uses of Analytical Ultracentrifugation

Mass Spectrometry of Higher Order Protein Structures

Effect of the Manufacturing Process on the Product

Impact of Changes in Chromatographic Operation on Biopharmaceutical Product Quality

Effects of the Bioreactor Environment on Product Quality

Renaturation and Folding

Impurities and Contaminants

Immunoassays for Residual HCP Analysis

Detection of Impurities in Proteomic Research

Bioassays and Potency

Binding Assays Versus Functional Bioassays

Assessing Similarity of Active Ingredients

Challenges in Developing Reference Materials for Biotech Products

Web Sites

Journal & Magazine Articles

Organizers